Topical anaesthetics for repair of dermal laceration

  • Review
  • Intervention

Authors


Abstract

Background

Topical local anaesthetics are recognized as providing effective analgesia for numerous superficial procedures, including repair of dermal lacerations. The need for cocaine in topical anaesthetic formulations has been questioned due to concern about adverse effects, and so novel preparations of cocaine-free anaesthetics have been developed.

Objectives

To compare the efficacy and safety of infiltrated local anaesthetics with those of topical local anaesthetics for repair of dermal lacerations and to evaluate the efficacy and safety of various single or multi-component topical anaesthetics to identify cocaine-free topically applied local anaesthetics that may provide equivalent analgesia to those containing cocaine.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2010, Issue 10); MEDLINE (1966 to November 2010); EMBASE (1980 to November 2010); CINAHL (1982 to November 2010); and reference lists of articles. We also handsearched selected journals, reviewed abstracts presented at international society meetings, reviewed metaregisters of ongoing trials and contacted manufacturers and researchers in the field.

Selection criteria

We included randomized controlled trials (RCTs) that evaluated the efficacy and safety of topical anaesthetics for repair of torn skin in adult and paediatric patients.

Data collection and analysis

Two authors independently assessed trial quality and extracted data. We contacted study authors for additional information. We collected adverse event information from the trials.

Main results

We included 23 RCTs involving 3128 patients. The small number of trials in each comparison group and the heterogeneity of outcome measures precluded quantitative analysis of data in all but one outcome, pain scores using a visual analogue scale. The majority of trials that compared infiltrated and topical anaesthetics are at high risk of bias, which is likely to affect the interpretation of the results. Several cocaine-free topical anaesthetics were found to provide effective analgesic efficacy. However, the data regarding the efficacy of each topical agent is mostly based upon single comparisons, in trials that have unclear or high risk of bias. Mild, self-limited erythematous skin induration occurred in one case after application of topical tetracaine-adrenaline-cocaine (TAC) where a total of 1042 patients were exposed. No serious complications were reported in any of the patients treated with either cocaine-based or cocaine-free topical anaesthetics.

Authors' conclusions

Based on mostly descriptive analysis, topical anaesthetics are possibly an efficacious, non-invasive means of providing analgesia prior to suturing of dermal lacerations. However, additional well designed RCTs with low risk of bias are necessary before definitive conclusions can be made.

Résumé scientifique

Anesthésiques topiques pour la réparation d'une lacération cutanée

Contexte

Les anesthésiques topiques locaux sont réputés apporter une analgésie efficace pour de nombreuses interventions superficielles, y compris la réparation des lacérations cutanées. La nécessité de la présence de cocaïne dans les formulations d'anesthésiques topiques a été remise en question en raison d'inquiétudes relatives aux effets indésirables, et de nouvelles préparations d'anesthésiques dépourvus de cocaïne ont été développées.

Objectifs

Comparer l'efficacité et l'innocuité des anesthésiques locaux en infiltration à celles des anesthésiques topiques locaux pour la réparation des lacérations cutanées et évaluer l'efficacité et l'innocuité de divers anesthésiques topiques mono ou multi-composants pour identifier des anesthésiques locaux topiques dépourvus de cocaïne pouvant fournir une analgésie équivalente à ceux contenant de la cocaïne.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre Cochrane des essais contrôlés (CENTRAL) (The Cochrane Library 2010, Numéro 10), MEDLINE (de 1966 à novembre 2010), EMBASE (1980 à novembre 2010) et CINAHL (de 1982 à novembre 2010), et dans les bibliographies des articles. Nous avons aussi effectué des recherches manuelles dans certaines revues, examiné des résumés présentés lors des réunions des Sociétés internationales, examiné les métaregistres des essais en cours et contacté les fabricants et chercheurs dans ce domaine.

Critères de sélection

Nous avons inclus les essais contrôlés randomisés (ECR) ayant évalué l'efficacité et l'innocuité des anesthésiques topiques pour la réparation d'une lacération cutanée chez des patients adultes et pédiatriques.

Recueil et analyse des données

Deux auteurs ont, de manière indépendante, évalué la qualité des essais et extrait les données. Nous avons contacté les auteurs des études afin d'obtenir des informations complémentaires. Dans les essais, nous avons recueilli des informations sur les événements indésirables.

Résultats principaux

Nous avons inclus 23 essais contrôlés randomisés (ECR) portant sur 3128 patients. Le petit nombre d'essais dans chaque groupe de comparaison et l'hétérogénéité des mesures des critères de jugement ont empêché la réalisation d'analyses quantitatives des données pour tous les résultats sauf un, à savoir les scores de douleur évalués à l'aide d'une échelle visuelle analogique. La majorité des essais ayant comparé les anesthésiques topiques et les anesthésiques en infiltration présentent un risque élevé de biais, qui est susceptible d'avoir une incidence sur l'interprétation des résultats. Il a été constaté que plusieurs anesthésiques topiques dépourvus de cocaïne contribuent effectivement à l'efficacité analgésique. Toutefois, les données concernant l'efficacité de chaque agent topique reposent principalement sur des comparaisons simples, dans des essais qui présentent un risque incertain ou élevé de biais. Un cas d'induration cutanée érythémateuse locale, légère s'est produit après l'application topique de tétracaïne-adrénaline-cocaïne (TAC), sur un total de 1042 patients exposés. Aucune complication grave n'a été signalée chez aucun des patients traités avec des anesthésiques topiques à base de cocaïne ou dépourvus de cocaïne.

Conclusions des auteurs

D'après une analyse essentiellement descriptive, les anesthésiques topiques sont vraisemblablement un moyen non invasif efficace de procurer une analgésie avant une suture de lacérations cutanées. Toutefois, d'autres essais cliniques randomisés (ECR) bien conçus, avec un faible risque de biais, sont nécessaires pour pouvoir tirer des conclusions définitives.

Plain language summary

Local anaesthesia (numbing medicine) that is directly applied to the skin may be used to provide pain control for repair of lacerations

Pain control for suturing of torn skin is conventionally achieved by injecting medication into the skin, which may itself cause pain. Topical anaesthetics are directly applied to the skin and are painless to administer. Cocaine was one of the first anaesthetics to be successfully applied topically. Concerns over adverse effects with cocaine and the administrative burdens of dispensing a controlled substance led to the development of cocaine-free anaesthetics. There are numerous cocaine-free topical anaesthetics and these were found to be effective for enabling repair of dermal lacerations. We included 23 randomized controlled trials involving 3128 patients in this review. The small number of trials in each comparison group and the heterogeneity of outcome measures precluded quantitative analysis of data in all but one outcome, pain scores using a visual analogue scale. Additional studies are necessary to directly compare the effectiveness of different formulations of topical anaesthetics. No serious side effects were reported in the studies included in the review following use of cocaine-containing or cocaine-free topical anaesthetics.

Résumé simplifié

Une anesthésie locale (traitement insensibilisant), directement appliquée sur la peau, peut être utilisée afin de soulager la douleur pour la réparation des lacérations.

Afin de soulager la douleur pour suturer une peau déchirée, il est classique d'injecter des médicaments dans la peau, ce qui en soi peut provoquer une douleur. Les anesthésiques topiques sont appliqués directement sur la peau et sont indolores à administrer. La cocaïne a été l'un des premiers anesthésiques à être appliqués avec succès par voie topique. Les inquiétudes quant aux effets indésirables de la cocaïne et les charges administratives liées à la délivrance d'une substance contrôlée ont conduit au développement d'anesthésiques dépourvus de cocaïne. Il existe de nombreux anesthésiques topiques dépourvus de cocaïne et ils se sont avérés efficaces pour permettre la réparation des lacérations cutanées. Nous avons inclus 23 essais contrôlés randomisés portant sur 3128 patients dans cette revue. Le petit nombre d'essais dans chaque groupe de comparaison et l'hétérogénéité des mesures des critères de jugement ont empêché la réalisation d'analyses quantitatives des données pour tous les résultats sauf un, à savoir les scores de douleur évalués à l'aide d'une échelle visuelle analogique. Des études supplémentaires sont nécessaires pour comparer directement l'efficacité des différentes formulations d'anesthésiques topiques. Dans les études incluses dans la revue, aucun effet secondaire grave n'a été rapporté suite à l'utilisation d'anesthésiques topiques contenant de la cocaïne ou dépourvus de cocaïne.

Notes de traduction

Traduit par: French Cochrane Centre 24th January, 2013
Traduction financée par: Instituts de Recherche en Sant� du Canada, Minist�re de la Sant� et des Services Sociaux du Qu�bec, Fonds de recherche du Qu�bec Sant� et Institut National d'Excellence en Sant� et en Services Sociaux

Laički sažetak

Lokalni anestetici koji se primjenjuju na koži mogu se koristiti za ublažavanje boli prilikom liječenja razderotina kože

Ublažavanje boli prilikom šivanja razderotina kože obično se postiže davanjem injekcija lijekova protiv boli u kožu, što samo po sebi može uzrokovati bol. Topikalni anestetici primjenjuju se izravno na kožu i bezbolni su za primjenu. Kokain je jedan od prvih anestetika koji se uspješno primijenio topikalno. Bojazan o nuspojavama kokaina i administrativnim problemima vezanim za korištenje kontroliranih lijekova kao što je kokain doveli su do razvoja anestetika koji ne sadrže kokain. Postoje brojni topikalni anestetici koji ne sadrže kokain, a za njih je utvrđeno da su djelotvorni za anesteziju kože prilikom liječenja razderotina. U ovaj Cochrane sustavni pregled uključena su 23 kontrolirana klinička istraživanja u kojima je sudjelovalo 3128 ispitanika. Mali broj sudionika nađen je u svakoj usporedbi i studije su se značajno razlikovale, zbog čega nije bilo moguće zajedno analizirati rezultate tih istraživanja osim za pokazatelj intenziteta boli na vizualno-analognoj ljestvici. Potrebne su dodatne studije u kojima će se izravno usporediti djelotvornost različitih formulacija topikalnih anestetika. Nisu zabilježene ozbiljne nuspojave u studijama uključenima u ovaj sustavni pregled, bez obzira na to je li lokalni anestetik sadržavao kokain ili ne.

Bilješke prijevoda

Hrvatski Cochrane
Prevela: Livia Puljak
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Summary of findings(Explanation)

Summary of findings for the main comparison. topical local anaesthetics compared to infiltrated local anaesthetics or other topical agents for repair of dermal lacerations
  1. 1 Each of the trials had high risk of bias in multiple domains or unclear risk of bias in three domains

    2 Two of the four trials had at least one domain that was high risk of bias

    3 Two of the trials had unclear risk of bias in multiple domains and the other two studies had high risk of bias in two domains

    4 Six of the studies had high risk of bias for at least one domain and the other five studies had unclear risk of bias for one or more domains

    5 Each of the five trials had unclear risk of bias in one or more domains. However, none of the trials contained any domains that were clearly high risk

topical local anaesthetics compared to infiltrated local anaesthetics or other topical agents for repair of dermal lacerations

Patient or population: patients with repair of dermal lacerations

Settings:

Intervention: topical local anaesthetics

Comparison: infiltrated local anaesthetics or other topical agents

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
  infiltrated local anaesthetics or other topical agents topical local anaesthetics    
Cocaine-containing topical anaesthetics versus infiltrated local anaestheticsSee commentSee commentNot estimable1006
(6 studies)
⊕⊕⊝⊝
low 1
Unable to mathematically combine results because of heterogeneity of outcome measures
Comparisons between different cocaine-containing topical anaestheticsSee commentSee commentNot estimable530
(4 studies)
⊕⊕⊝⊝
low 2
Unable to mathematically combine results because each topical anaesthetic comparison limited to a single study
Cocaine-free topical anaesthetics compared to infiltrated local anaestheticsSee commentSee commentNot estimable393
(4 studies)
⊕⊕⊝⊝
low 3
Unable to mathematically combine results because of heterogeneity of outcome measures
Cocaine-free topical anaesthetics compared to cocaine-containing topical anaestheticsSee commentSee commentNot estimable1231
(11 studies)
⊕⊕⊝⊝
low 4
Two of the eleven trials studied a common topical anaesthetic and could be mathematically combined
Comparisons between different cocaine-free topical anaestheticsSee commentSee commentNot estimable656
(5 studies)
⊕⊕⊕⊝
moderate 5
The trials could not be mathematically combined because each study compared a different cocaine-free topical anaesthetic
Anaesthetic related adverse effects Study population RR 0
(0 to 0)
1686
(11 studies)
  
1 per 1000 0 per 1000
(0 to 0)
Medium risk population
  
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

CI: Confidence interval; RR: Risk ratio;
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Pain caused by repair of torn skin may be an unpleasant experience for patients. Analgesia or pain control is conventionally achieved through local anaesthetic infiltration. Local anaesthetics are a class of drugs that interrupt the transmission of electrical impulses along sensory nerves by inactivating sodium channels (Stoelting 1999). However, the infiltration of local anaesthetics, which involves injecting the medication into the skin, may itself cause significant pain (Kundu 2002). Many patients, especially children, fear or dislike needles. Topical anaesthetics are not injected. Rather, the agent is directly applied to a local area or the skin. Therefore, topical anaesthesia may be preferable to infiltration anaesthesia for pain control during skin laceration repair. There are several different forms of topical anaesthetics, including solutions, gels, creams, ointments and skin patches. Adverse reactions to topical local anaesthetics include local responses (rash, stinging) or systemic allergic reactions (diffuse swelling, difficulty breathing, anaphylaxis) (Drug Facts 2003). An overdose with topical local anaesthetics may adversely affect the cardiovascular or central nervous systems (Drug Facts 2003). Untoward effects from high systemic levels of local anaesthetics include hypotension, cardiac arrhythmias (bradycardia, ventricular fibrillation, asystole), light-headedness, double vision, a metallic taste, drowsiness and seizures (Stoelting 1999).

In 1980, Pryor et al published the initial report of successful topical anaesthesia for repair of torn skin (Pryor 1980). The initial formulation, tetracaine-adrenaline-cocaine (TAC), gained widespread acceptance in North America and has largely supplanted infiltration anaesthesia for this purpose (the term 'epinephrine' rather than 'adrenaline' is used in the USA) (Grant 1992). However, the necessity to employ cocaine in topical anaesthetic formulations has been questioned due to concern over possible adverse effects (Bush 2002; Grant 1992). Although the application of TAC to skin lacerations results in undetectable or low systemic cocaine levels (Terndrup 1992, Vinci 1999), inadvertent mucosal application or overdose may cause significant cocaine absorption, resulting in serious consequences such as seizures (Dailey 1988; Daya 1988;Tipton 1988; Wehner 1984). Moreover, there are administrative and financial burdens of dispensing a controlled substance that is widely abused in the community. Accordingly, in the past decade novel preparations of cocaine-free topical anaesthetics have been developed. Analysis of the efficacy and safety of established and recently developed topical anaesthetics is needed.

Anaesthetic efficacy (capacity for producing desired anaesthetic effect) during procedures such as wound repair is reflected by the patient's self report of pain intensity during the intervention. Acceptable tools to quantify pain intensity include the visual analogue scale (VAS), numeric rating scale, verbal rating scale, faces scale or other validated descriptors of pain intensity. Non-concordance has been demonstrated between patients' and practitioners' assessments of procedure-related pain (Choiniere 1990; Singer 1999; Stephenson 1994).

Objectives

  1. To compare the efficacy of infiltrated local anaesthetic agents with that of topically applied local anaesthetic agents for repair of dermal (skin) lacerations. We determined anaesthetic efficacy by the patient's self report of pain intensity during repair of the wound.

  2. To compare the efficacy of various single or multi-component topical anaesthetic agents for repair of dermal lacerations.

  3. To identify cocaine-free, topically applied local anaesthetics that are potentially as efficacious as cocaine-containing topical anaesthetics.

  4. To compare the safety of the various topically applied local anaesthetic agents and preparations.

Methods

Criteria for considering studies for this review

Types of studies

We included only randomized, controlled trials (RCTs) or quasi-randomized trials. We included relevant trials that were published in abstract format or presented at national or international society meetings. We attempted to locate unpublished studies by contacting relevant manufacturers and investigators. We did not consider data from review articles, case reports or letters to the editor.

Types of participants

We included adult and paediatric patients of either sex. We did not set a minimum age threshold, so that we could identify as many relevant studies as possible.

Types of interventions

We only included trials that evaluated the efficacy of topical local anaesthetics for repair of dermal (skin) lacerations. We included comparisons between:

1. infiltrated local anaesthetic agents with topically applied local anaesthetic agents;

2. different topical local anaesthetic formulations.

We defined topical anaesthetics as any agents that are directly applied to the skin in order to produce numbness. We included both amide and ester local anaesthetics. We accepted topical preparations that contain more than one local anaesthetic. We also included multi-component topical anaesthetics that contain vasoconstrictors (that is, cocaine, adrenaline). Acceptable forms of topical local anaesthetics included solution, gel, cream, ointment, lotion, jelly, balm, or aerosol spray. We excluded studies that administered local anaesthetics using iontophoresis (a mild electrical current).
We excluded papers that applied topical anaesthetics to mucous membranes (moist linings of the mouth, nose, and eyes). In order that the procedures that were evaluated involved an approximately equivalent intensity and quality of pain, we limited the technique of skin closure to instrumentation including suture placement or stapling. Less invasive approaches to repair lacerations, such as application of tape or tissue adhesives, were excluded. We included only superficial injuries that involved the epidermis or dermal layers. We did not consider deeper wounds involving the fascia or non-skin structures. We set no limitations on the dimensions of the laceration, but we excluded procedures on infected wounds. We also excluded studies that administered systemic analgesics or sedatives that may alter the accuracy of the patients' perceived pain intensity or report.

Types of outcome measures

Primary outcome

Patient report of pain intensity during wound repair

Secondary outcomes

1. Incidence of topical anaesthetic failure, requirement for supplemental anaesthesia, patient's acceptance of anaesthesia, patient behavioural responses, and surrogate pain scores provided by the physician, parent or other observers

2. Topical anaesthesia-related acute toxicity and adverse effects (allergic reaction, neurological and cardiovascular toxicity)

Search methods for identification of studies

Electronic searches

We searched the Cochrane Central Registrar of Controlled Trials (CENTRAL) (The Cochrane Library 2010, Issue 10); MEDLINE (1966 through November 2010); EMBASE (1980 through November 2010) and CINAHL (1982 through November 2010).

We sought unpublished studies by directly contacting the primary investigator for the included trials. We searched for additional papers by reviewing the references of each retrieved study.

We searched MEDLINE, CENTRAL and CINAHL using the search strategy described in Appendix 1, Appendix 2 and Appendix 3. We combined the MEDLINE search with the first two levels of the optimal trial search strategy (Higgins 2011). We searched EMBASE using the search strategy found in Appendix 4.

We searched metaregisters of ongoing trials (http://www.controlled-trials.com/ and http://clinicaltrials-dev.ifpma.org/).

All trials were limited to human studies. We did not implement any language restrictions in the literature search.

Searching other resources

We manually searched the following journals (1980 through 2009):

  1. Academic Emergency Medicine,

  2. Annals of Emergency Medicine,

  3. Emergency Medicine Clinics of North America,

  4. Journal of Emergency Medicine,

  5. Emergency Medicine Australasia (formerly known as Emergency Medicine).

We reviewed abstracts presented at the following national or international society meetings:

  1. American Academy of Pain Medicine (AAPM),

  2. American Pain Society (APS),

  3. American College of Emergency Physicians (ACEP),

  4. American Society of Anesthesiologists (ASA),

  5. American Society of Regional Anesthesia and Pain Medicine (ASRA),

  6. The European Society of Regional Anaesthesia and Pain Therapy (ESRA),

  7. Society for Academic Emergency Medicine (SAEM).

We contacted the following manufacturers of topical anaesthetics for ongoing or unpublished trials:

  1. Astra Zeneca,

  2. Endo Pharmaceuticals,

  3. Ferndale Laboratories,

  4. New England Compounding Center,

  5. Smith and Nephew,

  6. Topicaine.NET.

Data collection and analysis

Assessment of studies for inclusion in the review

Two authors (AE and either CB or IE) independently reviewed the study titles and abstracts identified by the search strategy. We obtained the full publications if at least one author decided that the study potentially met the inclusion criteria. Two authors (AE and CB or IE) independently examined the complete articles and selected trials that met the inclusion criteria. In the event of disagreement, another author (DC) was consulted.

Data extraction

We compared the retrieved information to verify accuracy and resolved disagreements by consensus. In studies that presented the results in bar graph format (Anderson 1990; Ernst 1990; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a), two authors (AE, IE) independently extracted numerical data by measuring graphs with a ruler. The average of the two measurements was then calculated. In one RCT we calculated the standard deviation (SD) for each experimental group's mean pain score by multiplying the standard error of the mean (SEM) by the square root of the sample size (Smith 1997b). In three studies we calculated the mean pain scores and SDs from individual patient data (Anderson 1990; Ernst 1990; Gaufberg 2007). White and associates reported the results in separate groups, according to the characteristics of the laceration (length and location) (White 1986). We pooled pain scores for each anaesthetic group and reported the results collectively. Furthermore, in order to facilitate statistical comparison, we converted VAS pain scores reported on a 10 cm scale to a 100 mm scale by multiplying the scores by 10 (Adler 1998; Kuhn 1996; Zempsky 1997).

Details regarding the data extracted from each study can be found in Appendix 5 (Data extraction form).

Data analysis

The small number of trials in each comparison group and the heterogeneity of outcome measures precluded meta-analysis in the majority of comparisons. Therefore, we performed mostly a descriptive analysis. In the comparison of topical prilocaine-phenylephrine (PP) and topical tetracaine-epinephrine-cocaine (TAC), the reported outcomes (pain intensity measures) were statistically combinable and thus we pooled the data. Statistical calculations were computed with Review Manager (RevMan 5.1). We pooled patient self reported VAS scores (which are continuous outcomes) using the mean scores and SDs to derive a weighted mean difference (WMD) as well as the 95% confidence intervals (CIs). We computed a Chi2 test to test for heterogeneity. In the one statistically combinable comparison there was heterogeneity and thus we used a random-effects model for meta-analysis. However, if the diversity between trials would have been significant, then we planned to perform a descriptive analysis. It was planned to analyse dichotomous data using Review Manager (RevMan 5.1). Specifically, the relative risk would have been computed. However, due to the lack of relevant data in the included studies we did not analyse dichotomous data.

A subgroup analysis was intended to determine whether there were different results between adult and paediatric patients. We considered patients younger than 18 years old to be paediatrics and patients aged 18 years or older to be adults. However, subgroup analysis by age was not possible because of the small number of studies in each comparison group. Also, many trials included only paediatric or only adult patients. Moreover, the studies that included both adult and paediatric patients did not separately report outcomes according to the different age groups. We were unable to obtain the unreported data despite efforts to contact the study authors. No sensitivity analysis was intended or performed for this review.

Missing information

If necessary, an electronic mail or postal mail letter was sent to the contact author in order to acquire missing information. Additional data were sought from eight trials, but we were able to successfully obtain additional information from only one study (Smith 1997a). Furthermore, we contacted by electronic mail and received responses from two primary authors, Drs Amy Ernst and Gary Smith, regarding whether any of the patients' data may have been included in more than one of their studies (Ernst 1990; Ernst 1995a; Ernst 1995b; Ernst 1997; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a).

Results

Description of studies

Independent review of the abstracts and titles identified by the electronic database searches yielded 39 potentially relevant studies. We obtained each of the 39 trials in full and examined them for possible inclusion in the review. Sixteen of the 39 retrieved trials did not meet the inclusion criteria. Furthermore, eight additional potentially relevant papers were identified, either through review of obtained study references (Bass 1990; Bonadio 1988a; Bonadio 1988b; Chipont 2001; Liebelt 1997; Peirluisi 1989; Yamamoto 1997) or manual searches of journals (Bonadio 1992). However, none of the eight papers met the inclusion criteria of the review. A detailed description of each of the 24 studies is provided in the Characteristics of excluded studies table. We were unable to locate any unpublished studies that qualified for the present review, despite direct communication with pertinent manufacturers and investigators. None of the 23 included trials were industry sponsored. A total of 23 RCTs met the inclusion criteria of this review. Detailed descriptions of each trial are provided in Characteristics of included studies. The detailed search results are found in Figure 1.

Figure 1.

Search flow diagram

Participants

The trials included a total of 3128 adult and paediatric patients. Three trials included only adult participants (Ernst 1995b; Gaufberg 2007; White 1986). One trial enrolled only paediatric patients who were 10 years or younger (Schaffer 1985). Another trial was limited to children but the upper age limit was not specified (Bonadio 1990). The remaining 19 studies enrolled both adult and paediatric patients according to the definition provided above. The inclusion criteria applied in 10 of the retrieved trials potentially allowed children who were less than three years old to be enrolled (Anderson 1990; Blackburn 1995; Hegenbarth 1990; Pryor 1980; Schaffer 1985; Schilling 1995; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a). No duplicate patient data were included in the trials by Ernst or Smith (Ernst 1990; Ernst 1995a; Ernst 1995b; Ernst 1997; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a).

Wound closure

Wound closure was performed strictly with sutures in 21 studies. One study repaired lacerations with both sutures and staples (Krief 2002). Another trial repaired lacerations with skin staples in a minority (7%) of patients (Hegenbarth 1990). No alternative techniques of wound repair were employed. The lacerations were located on four anatomical regions: the face, scalp, extremities, and less commonly the trunk. All of the lacerations were superficial and the length of the dermal injuries ranged from less than 1.0 cm to 10.0 cm.

Topical anaesthetics

The 23 included RCTs studied a total of 18 different topical anaesthetics, which are listed in Appendix 6. Four studies had multiple arms that compared more than two different anaesthetic agents (Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a). Smith 1996 had six different groups, including five different topical anaesthetics and an infiltrated local anaesthetic arm. Smith 1997a evaluated two topical anaesthetics and infiltrated local anaesthetic. Smith 1997b compared four different topical anaesthetics and Smith 1998a studied three different topical agents.

Seventeen of the 23 studies compared different forms of topical anaesthetics and only a minority of trials contained arms with infiltrated local anaesthetic groups. Therefore, the main comparison was between the different topical preparations.

Outcome measures

The primary outcome measure was analgesic efficacy, reflected in the patient's self report of pain intensity during repair of the wound. Thirteen of the included trials determined anaesthetic efficacy through the participant's self report of pain intensity (Blackburn 1995; Ernst 1995a; Ernst 1995b; Ernst 1997; Gaufberg 2007; Kendall 1996; Krief 2002; Kuhn 1996; Smith 1996; Smith 1997b; Smith 1998a; White 1986; Zempsky 1997). Unless otherwise specified, discomfort was assessed during suturing or stapling. There were numerous tools that were used for patient self report of pain intensity. Ten studies used VAS pain scale scores (Ernst 1995b; Ernst 1997; Gaufberg 2007; Kendall 1996; Krief 2002; Kuhn 1996; Smith 1996; Smith 1997b; Smith 1998a; Zempsky 1997). Three RCTs used a faces pain scale (Blackburn 1995; Kendall 1996; Kuhn 1996). Furthermore, two trials used numerical pain ratings (0 to 10) (Ernst 1995a; White 1986).

Various secondary outcome measures were extracted from the RCTs. Nine trials provided observer-reported VAS pain scores (Ernst 1995b; Ernst 1997; Kendall 1996; Krief 2002; Kuhn 1996; Smith 1996; Smith 1997a; Smith 1998a; Zempsky 1997). Three studies used observer-rated Likert scores (Smith 1996; Smith 1997a; Smith 1997b). Two RCTs used observer-reported faces pain scales (Blackburn 1995; Kuhn 1996) and one study used an observer-rated multi-dimensional pain intensity scale (Ernst 1995a). Four trials calculated the percentage or absolute number of sutures eliciting pain (Bonadio 1990; Ernst 1995a; Ernst 1995b; Ernst 1997). Ten studies reported the requirement for supplemental lidocaine infiltration (Anderson 1990; Blackburn 1995; Ernst 1995a; Ernst 1997; Hegenbarth 1990; Krief 2002; Schaffer 1985; Vinci 1996; White 1986; Zempsky 1997). Seven RCTs assessed the effectiveness of anaesthesia by probing the laceration with a needle (Anderson 1990; Ernst 1990; Ernst 1997; Hegenbarth 1990; Kuhn 1996; Resch 1998; Schilling 1995). Seven trials included a verbal categorical scale to describe anaesthetic effectiveness (Pryor 1980; Resch 1998; Schaffer 1985; Schilling 1995; Smith 1996; Smith 1997b; Vinci 1996). Two studies employed an observer-reported compliance rating (Anderson 1990; Smith 1996) and two RCTs used an observer-rated acceptability of wound repair (Kendall 1996; Pryor 1980). Two studies reported the total number of topical anaesthetic doses (Gaufberg 2007; Vinci 1996). The following secondary outcome measures were each used by a single trial: the Childrens Hospital of Eastern Ontario Pain Scale (CHEOPS) (Kuhn 1996), observer numerical-rating of anaesthetic effectiveness (Ernst 1990), the Restrained Infants, Children Distress Rating Scale (RICDRS) (Smith 1996), and amount of anaesthetic used (Gaufberg 2007).

Adverse effects

Eleven trials explicitly assessed and reported the nature and incidence of topical local anaesthetic-related acute adverse effects (Blackburn 1995; Bonadio 1990; Ernst 1990; Ernst 1995a; Hegenbarth 1990; Kendall 1996; Kuhn 1996; Resch 1998; Schaffer 1985; Schilling 1995; Vinci 1996).

Risk of bias in included studies

We analysed the risk of bias in the 23 included trials by assessing randomization (sequence generation), blinding, allocation concealment and incomplete data outcome. Further information regarding risk of bias can be found in the 'Risk of bias' graph (Figure 2), summary (Figure 3) and tables (Characteristics of included studies).

Figure 2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Figure 3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

We concluded that there was adequate, random sequence generation in six of the 23 trials (26%) (Ernst 1995a; Ernst 1995b; Ernst 1997; Resch 1998; Vinci 1996; Zempsky 1997) and insufficient information to make a judgement in nine of the studies (39%) (Ernst 1990; Gaufberg 2007; Krief 2002; Kuhn 1996; Schilling 1995; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a).

We concluded that participants and personnel were adequately blinded to the identity of the anaesthetic in 13 of 23 studies (57%) (Blackburn 1995; Bonadio 1990; Ernst 1990; Ernst 1995a; Ernst 1995b; Kuhn 1996; Resch 1998; Schaffer 1985; Schilling 1995; Smith 1996; Smith 1997a; White 1986; Zempsky 1997). There was insufficient information to confirm adequate blinding in four of the papers (17%) (Krief 2002; Smith 1997b; Smith 1998a; Vinci 1996). However, 13 of the 17 studies (76%) that compared different forms of topical anaesthetics were appropriately blinded. Seven of the eight trials that compared topical anaesthetics to infiltrated anaesthetic were not blinded (Anderson 1990; Ernst 1997; Gaufberg 2007; Hegenbarth 1990; Kendall 1996; Pryor 1980; Smith 1996). One trial (Smith 1997a) was adequately blinded because after the topical or local anaesthetic was administered, suturing procedures were videotaped. These videotapes were later reviewed by an observer who was completely blinded to which form of anaesthetic the patient had received.

We concluded that allocation was adequately concealed in five of the 23 studies (22%) (Blackburn 1995; Ernst 1995b; Kuhn 1996; Resch 1998; Schilling 1995) and unclear in six other studies (26%) (Ernst 1990; Krief 2002; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a).

Ten of the trials (43%) appropriately addressed incomplete data outcome (Anderson 1990; Ernst 1990; Ernst 1995a; Gaufberg 2007; Kendall 1996; Kuhn 1996; Pryor 1980; Schilling 1995; Vinci 1996; Zempsky 1997) and it was unclear in 12 studies (52%) (Blackburn 1995; Bonadio 1990; Ernst 1997; Hegenbarth 1990; Krief 2002; Resch 1998; Schaffer 1985; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a; White 1986).

Effects of interventions

See: Summary of findings for the main comparison topical local anaesthetics compared to infiltrated local anaesthetics or other topical agents for repair of dermal lacerations

We first present the evidence regarding cocaine-containing topical anaesthetics. We included comparisons between cocaine-based topical anaesthetics and each of the following: a) infiltrated local anaesthetics; and b) different formulations of cocaine-based topical agents. Next, we summarize the evidence evaluating cocaine-free topical anaesthetics. We compared cocaine-free topical agents with each of the following: a) infiltrated local anaesthetics; b) formulations of cocaine-containing topical agents; and c) different formulations of cocaine-free topical anaesthetics. We also reported the data on acute anaesthetic-related adverse effects. A detailed and inclusive description of each of the 23 trials is provided in the table Characteristics of included studies.

(1) Evaluation of cocaine-containing topical anaesthetics

1a. Cocaine-containing topical anaesthetics versus local anaesthetic infiltration (six studies)

Six studies compared a topical cocaine-based agents with infiltrated local anaesthetic (see Appendix 7 for detailed study information). Five studies compared topical tetracaine-adrenaline-cocaine (TAC) with infiltrated local anaesthetic. Outcomes could not be mathematically combined because of the numerous different measures used to determine anaesthetic efficacy (Anderson 1990; Hegenbarth 1990; Pryor 1980; Smith 1996; Smith 1997a). The results regarding anaesthetic efficacy of topical TAC were inconsistent. Moreover, each of the trials either had a high risk of bias in multiple domains (Anderson 1990; Hegenbarth 1990; Pryor 1980) or unclear bias in three domains (Smith 1996; Smith 1997a). One study found topical adrenaline-cocaine (AC) to provide equivalent analgesia to local anaesthetic infiltration (Kendall 1996). However, the study was not blinded and there was a high risk of bias for both sequence generation and allocation concealment.

1b. Comparisons between different cocaine-containing topical anaesthetics (four studies)
Four studies compared topical tetracaine-adrenaline-cocaine (TAC) with another cocaine-based topical anaesthetic (Appendix 8). No significant difference in anaesthetic efficacy was found between TAC and either topical bupivacaine-adrenaline-cocaine (MAC) (Kuhn 1996) or adrenaline-cocaine (AC) (Bonadio 1990). Kuhn 1996 had unclear risk of bias for sequence generation but a low risk of bias for the other three key domains. Bonadio 1990 did not use a formal pain scoring scale to assess the efficacy of AC and there was a high risk of bias for both sequence generation and allocation concealment. Neither cocaine (C) (Ernst 1990) or tetracaine-cocaine (TC) (Vinci 1996) were found to be effective topical anaesthetics.

(2) Evaluation of cocaine-free topical anaesthetics

2a. Cocaine-free topical anaesthetics versus infiltrated local anaesthetic (four studies)
Four RCTs compared five different cocaine-free topical anaesthetics with infiltrated local anaesthetic (Appendix 9). Smith 1996 found no significant difference in VAS pain scores between infiltrated local anaesthetic and four different noradrenaline-containing topical anaesthetics, including bupivacaine-noradrenaline (BN), etidocaine-noradrenaline (EN), mepivacaine-noradrenaline (MN) and prilocaine-noradrenaline (PN). Smith 1997a also compared topical mepivacaine-noradrenaline (MN) with infiltrated local anaesthetic, but concluded that the later provided better analgesia. The two studies of topical MN could not be mathematically combined because of the heterogeneity in outcome measures, and Smith 1996 did not report standard deviations of the pain scores. Both of the trials by Smith and associates had an unclear risk of bias in at least three key domains. Also, in Smith 1996 the comparisons of infiltrated lidocaine and topical anaesthetics were not blinded. Moreover, Smith 1997a did not employ patient self-reported pain scoring scales, but instead provided observer estimates of pain.

No significant difference was found between infiltrated local anaesthetic and either topical lidocaine-adrenaline-tetracaine (LAT) (Ernst 1997) or topical lidocaine-epinephrine (TLE) (Gaufberg 2007). However, both of the trials were not blinded and did not appropriately conceal allocation.

2b. Cocaine-free topical anaesthetics versus cocaine-containing topical anaesthetics (11 studies)

Eleven trials compared 13 different cocaine-free topical anaesthetics with topical TAC (Appendix 10). Each of the studies employed TAC as the cocaine-containing topical preparation. Smith and associates published four papers relevant to this section (Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a). In comparisons confined to a single trial, Smith and associates found similar analgesic efficacy between topical TAC and each of the following topical agents: bupivacaine-noradrenaline (BN), prilocaine-noradrenaline (PN), tetracaine-lidocaine-phenylephrine (TLP) and tetracaine-phenylephrine (TP) (Smith 1996; Smith 1997b). Two papers compared topical prilocaine-phenylephrine (PP) with topical TAC (Smith 1997b; Smith 1998a). In Analysis 1.1, the patient-reported VAS (100 mm) pain scores were pooled and no difference was found between topical PP and topical TAC (Figure 1) (WMD 5.56, 95% CI -2.20 to 13.32). There were conflicting conclusions from two studies regarding the efficacy of topical mepivacaine-noradrenaline (MN) (Smith 1996; Smith 1997a). The trials could not be statistically combined because different pain scoring scales were used to determine anaesthetic efficacy and Smith 1996 did not report the standard deviations of the outcomes. Each of the four trials by Smith and associates had unclear risk of bias for three or more key domains.

Three studies found similar efficacy between topical lidocaine-adrenaline-tetracaine (LAT) and TAC (Ernst 1995a; Ernst 1995b; Schilling 1995). We could not mathematically combine the results because of the heterogeneity in outcome measures. It was unclear whether Schilling 1995 used appropriate sequence generation but there was low risk of bias for the other domains. The two trials by Earnst and associates each had a high risk of bias for one key domain.

One RCT that did appropriately conceal allocation found no difference in pain scores in children anaesthetized with either EMLA cream (lidocaine 2.5% and prilocaine 2.5%) or topical TAC (Zempsky 1997). Another paper that probably did not have random sequence generation found no difference in the efficacy of topical lidocaine-adrenaline (LE) versus topical TAC (Blackburn 1995).

Topical TAC outperformed etidocaine-noradrenaline (Smith 1996), topical bupivacaine-phenylephrine (Smith 1998a), topical tetracaine-adrenaline (Schaffer 1985) and topical tetracaine (White 1986).

2c. Comparisons between different cocaine-free topical anaesthetics (five studies)
Five RCTs evaluated different cocaine-free topical anaesthetics (Appendix 11). Smith 1996 found no significant difference in anaesthetic efficacy between four different noradrenaline-containing topical anaesthetics, including bupivacaine-noradrenaline (BN), etidocaine-noradrenaline (EN), mepivacaine-noradrenaline (MN) and prilocaine-noradrenaline (PN). Another multi-armed RCT (Smith 1997b) demonstrated no significant difference between three different topical formulations that contained the vasoconstrictor phenylephrine, including prilocaine-phenylephrine (PP), tetracaine-phenylephrine (TP) and tetracaine-lidocaine-phenylephrine (TLP). A third trial by the same primary author concluded that there is similar efficacy between topical prilocaine-phenylephrine (PP) and bupivacaine-phenylephrine (BP) (Smith 1998a). Krief 2002 found no significant difference in pain scores in patients treated with either topical EMLA or LAT. Each of the papers by Smith and associates, as well as Krief 2002, had unclear risk of bias in at least three domains.

A fifth RCT, which had unclear assessment of incomplete data but otherwise was at low risk of bias, concluded that the solution and gel formulations of LAT provided comparable analgesic efficacy (Resch 1998).

(3) Anaesthetic related acute adverse effects

Approximately half of the included trials (12/23, enrolling 1713 patients) reported data regarding the incidence of potential anaesthetic-related acute adverse effects. Further details are displayed in Summary of findings for the main comparison. There was only one reported episode of a local anaesthetic-related complication in any of the studies. In Vinci 1996, a single paediatric patient developed a large indurated, erythematous reaction one day after application of topical TAC. The skin reaction completely resolved with anti-histamine treatment and warm compresses. There were no other reported incidents of local anaesthetic-induced reactions or toxicity. Schaffer 1985 reported that only after being discharged home, 10.7% of children treated with TAC and 7.8% who received topical AC became drowsy or excitable. However, none of the symptoms occurred in the emergency department and there was no evidence that the symptoms were causally related to the topical anaesthetic. Two trials that included an infiltrated local anaesthetic group reported data on acute side effects (Hegenbarth 1990; Kendall 1996). None of the combined 256 participants administered local anaesthesia via infiltration in these two studies reported any adverse effects.

Ten different RCTs that studied cocaine-based topical anaesthetic explicitly reported information about acute adverse effects (Blackburn 1995; Bonadio 1990; Ernst 1990; Ernst 1995a; Hegenbarth 1990; Kendall 1996; Kuhn 1996; Schaffer 1985; Schilling 1995; Vinci 1996). The pooled data from these 10 trials resulted in an acute adverse reaction in 1/1042 patients (incidence 0.096%). This complication was not serious and is described previously. A total of five RCTs that used cocaine-free topical agents reported data on anaesthetic-related toxicity or side effects (Blackburn 1995; Ernst 1995a; Resch 1998; Schaffer 1985; Schilling 1995; Schilling 1995). None of the 358 patients in these five RCTs experienced any acute adverse reactions.

Discussion

The present review consisted of a descriptive analysis. There were two predominant limitations that precluded meta-analysis. First, most of the comparisons between specific anaesthetic agents were confined to a single trial. Only in few instances were similar comparisons duplicated in subsequent studies. Moreover, the trials employed numerous different measures to determine anaesthetic efficacy. In fact, only 13 of the 23 included studies used a formal pain scale. The primary outcome measure was analgesic efficacy, reflected in the patient's self report of pain intensity during repair of the wound. We extracted surrogate pain scores provided by observers, however, non-concordance has been demonstrated between patients' and practitioners' assessments of procedure-related pain (Choiniere 1990; Singer 1999; Stephenson 1994). Therefore, during analysis we only considered surrogate pain scores in instances when patient-reported pain scales were not available.

Our systematic review addressed four principal questions regarding topically applied local anaesthetics for dermal laceration repair. First, we assessed whether the benefits of non-invasive, topical anaesthetic application occur at the expense of decreased analgesic efficacy. The data are from a single study that has unclear risk bias (Smith 1997a) and the remainder of the trials were at high risk of bias (Anderson 1990; Ernst 1997; Gaufberg 2007; Hegenbarth 1990; Kendall 1996; Pryor 1980; Smith 1996). Smith 1997a did not use patient self-reported pain scores to determine anaesthetic efficacy, but instead used observer-estimated pain scores. Therefore, there is a paucity of high quality studies, with low risk of bias, to make definitive conclusions regarding the efficacy topical anaesthetics versus infiltrated local anaesthesia. Future research using well designed RCTs is necessary to answer this question.

Our second objective was to compare the efficacy of various single or multi-component topical anaesthetic agents for repair of dermal lacerations. The data are from studies that have either unclear risk of bias (Ernst 1990; Kuhn 1996; Schilling 1995; Smith 1996; Smith 1997a; Smith 1997b; Smith 1998a) or high risk of bias (Blackburn 1995; Bonadio 1990; Ernst 1995a; Ernst 1995b; Schaffer 1985; Vinci 1996; White 2004; Zempsky 1997). We have summarized the findings of individual trials in Appendix 8; Appendix 10; and Appendix 11. However, the evidence contains bias that may cause some doubt about the findings, or even significantly weaken the results.

The third objective was to determine the clinical necessity for topical application of the ester anaesthetic, cocaine. Thirteen RCTs were included in the review, which assessed the effectiveness of cocaine-free topical anaesthetics. None of the studies were at low risk of bias. Data from two studies were mathematically combined and found topical prilocaine-phenylephrine (PP) to provide effective analgesia (Smith 1997b; Smith 1998a). However, both of the studies had unclear risk of bias for each key domain and therefore there is some uncertainty about the results. Additional formulations of topical cocaine-free anaesthetics were assessed only in a single RCT. Results from the papers with unclear risk of bias found that the following agents may provide effective topical analgesia: lidocaine-adrenaline-tetracaine (LAT) (Schilling 1995), bupivacaine-noradrenaline (BN) (Smith 1996), prilocaine-noradrenaline (PN) (Smith 1996), tetracaine-lidocaine-phenylephrine (TLP) (Smith 1997b), tetracaine-phenylephrine (TP) (Smith 1997b) and lidocaine-prilocaine (EMLA) (Krief 2002). Topical LAT, which exploits the rapid onset of lidocaine and long duration of tetracaine (Altman 1985), has been the most widely studied cocaine-free formulation. However, before definitive conclusions can be made, additional investigation is warranted with trials that are well designed and assess anaesthetic efficacy with validated patient self-reported pain scoring scales.

Finally, we evaluated the safety of both cocaine-containing and cocaine-free topical anaesthetics. Many of the included trials (12 of 23) reported data regarding the incidence of potential anaesthetic-related acute adverse effects. Only one study reported a topical local anaesthetic-related side effect (Vinci 1996). The reaction consisted of a large indurated, erythematous reaction that occurred after topical application of TAC. No serious complications, such as seizures or anaphylactic reactions, were reported in any of the trials. Although the reported data is insufficient to determine the exact incidence of complications, if topical anaesthetics are applied as directed and appropriately dosed then serious adverse effects are probably infrequent. The combined observations from 10 trials that administered cocaine-based agents and explicitly reported data on side effects resulted in one adverse reaction out of 1042 total patients (incidence 0.096%). Six studies that administered cocaine-free anaesthetics agents reported data on toxicity and none of the 418 patients in this group experienced acute adverse reactions.

There are sources of potential bias in the review. The primary outcome was patients' self report of pain intensity during repair of the wound, using validated pain scales. However, a significant number of included trials used observer-reported pain scores or other elementary surrogate outcomes to determine anaesthetic efficacy. Non-concordance has been demonstrated between patients' pain scores and ratings by physician, parents or other proxies (Choiniere 1990; Singer 1999; Stephenson 1994). Moreover, 20 of the included 23 RCTs enrolled paediatric patients and the evaluation of pain in children can be challenging. Several pain scales, including the VAS and faces scale, have been used in a reliable and validated manner in children as young as five years (Berde 1991; Lander 1993; Zeltzer 1991). Also, there is evidence supporting the validity of tools for measuring acute pain in children as young as three years old (Tyler 1993). However, the lower age limit at which children can credibly quantify pain intensity is controversial (Tyler 1993) and there is a need to validate the behavioural pain scales for early verbal and pre-verbal children (Crellin 2007). Therefore, we cannot exclude the possibility that pain assessment in younger paediatric patients may not be accurate.

The present review is limited to repair of dermal lacerations. Therefore, the results may not be generalizable to repair of wounds located on mucosal surfaces. Also, the dermis provides a barrier to penetration of topical anaesthetic, and so our findings may not be applicable to instrumentation of intact skin.

In conclusion, based on mostly descriptive analysis, topical anaesthetics may in fact be an efficacious, non-invasive means of providing analgesia prior to suturing of dermal lacerations. However, the data regarding the efficacy of each topical anaesthetic are mostly based upon single comparisons in trials that have unclear or high risk of bias. In order to make definitive conclusions, additional methodologically well designed studies with low risk of bias are necessary. Future research should focus on the efficacy of cocaine-free anaesthetics in light of the burdens of dispensing cocaine, a controlled substance that is widely abused.

Authors' conclusions

Implications for practice

Injection of anaesthetics per se induces discomfort, and may worsen 'needle anxiety' in paediatric participants and distort the wound site (Kundu 2002). Therefore, topical anaesthetics are preferable if they do in fact provide similar analgesia to injected local anaesthetics. There is a suggestion from individual studies that some topical formulations may have similar efficacy to conventional local anaesthetics. However, because of methodological heterogeneity and a lack of high quality trials, definitive conclusions for clinical practice cannot be made. 

If cocaine-free topical anaesthetics have similar effectiveness as cocaine-containing agents then the latter can no longer be justified in light of their high cost and potential adverse effects. Topical lidocaine-adrenaline-tetracaine (LAT), which exploits the rapid onset of lidocaine and the long duration of tetracaine, has been the most widely studied cocaine-free formulation. However, additional studies with sound methodological design are necessary before drawing definitive conclusions for clinical practice.

No serious complications were reported in any of the patients treated with either cocaine-based or cocaine-free topical anaesthetics. One mild, self-limiting skin reaction did occur in one case after application of topical TAC. Nevertheless clinicians should exhibit caution and apply topical formulations only as directed, avoid mucous membrane contact and follow appropriate dosing regimens.

Implications for research

More investigation is warranted to compare topical lidocaine-adrenaline-tetracaine (LAT) with the other potentially efficacious, cocaine-free topical anaesthetics such as bupivacaine-noradrenaline (BN), prilocaine-phenylephrine (PP) or tetracaine-lidocaine-phenylephrine (TLP). Also, future research could determine additional clinically useful topical local anaesthetics or combinations.

Furthermore, there is a need for additional methodologically sound studies that are less likely to be flawed by bias or confounding variables. Many of the included trials did not determine analgesic efficacy with validated, patient self-reported pain scales, but instead used observer-reported pain scores or other elementary surrogate measures. Future trials should adopt uniform outcomes using the patient's own assessments of procedure-related pain intensity. It is young children that may benefit most from non-invasive, effective topical anaesthesia prior to laceration repair. Therefore, validated behavioral pain and distress scales for pre-verbal or early verbal children will facilitate the study of the efficacy and safety of topical anaesthetics in this patient subgroup.

Acknowledgements

We would like to thank Mathew Zacharias, Marialena Trivella, Ronan O’Sullivan, Stephen Priestley, Sujesh Bansal and Nete Villebro for for their help and editorial advice during the preparation of the systematic review.

We would also like to thank Mathew Zacharias, André Muller, Stephen Priestley, Janet Wale, Kathie Godfrey, Nete Villebro and Jane Cracknell for their help and editorial advice during the preparation of the protocol for the systematic review. We also acknowledge the efforts of Karen Hovhannisyan, Catherine Guarcello, Marybeth Edwards, Richard Kammer and Dr Morton Rosenberg for assistance with the literature search, Dr. Joseph Lau with advise with statistics, and Dr. Mukhtar Zaidi for pharmacological consultation.

Data and analyses

Download statistical data

Comparison 1. Topical prilocaine-phenylephrine (PP) versus topical tetracaine-epinephrine-cocaine (TAC)
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Patient self-reported VAS (0-100 mm) pain scores2240Mean Difference (IV, Random, 95% CI)5.59 [-2.16, 13.35]
Analysis 1.1.

Comparison 1 Topical prilocaine-phenylephrine (PP) versus topical tetracaine-epinephrine-cocaine (TAC), Outcome 1 Patient self-reported VAS (0-100 mm) pain scores.

Appendices

Appendix 1. Search strategy for CENTRAL, the Cochrane Library

#1 MeSH descriptor Lacerations, this term only
#2 MeSH descriptor Wounds and Injuries, this term only
#3 MeSH descriptor Facial Injuries explode all trees
#4 MeSH descriptor Finger Injuries explode all trees
#5 MeSH descriptor Wounds, Penetrating explode all trees
#6 MeSH descriptor Hand Injuries explode all trees
#7 MeSH descriptor Sutures explode all trees
#8 MeSH descriptor Surgical Stapling explode all trees
#9 (laceration* or wound* or suture or stapling or repair*):ti,ab
#10 ((facial or dermal or cutaneous or finger or hand or eyelid) near injur*):ti,ab
#11 (penetrat* near wound*)
#12 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11)
#13 (topical near (an?esthe* or lidocain* or lignocain* or tetracain* or amethocain* or benzocain* or butamben or dibucain* or pramoxin* or prilocain* or etidocian))
#14 emla* or (eutectic mixture of local an?esthe*) or (tetracaine?adrenaline?cocain*) or (tetracaine?epinephrine?cocain*) or (lidocaine?adrenaline?tetracain*) or (lidocaine?epinephrine?tetracain*) or (spray or ointment or gel or cream or lotion or jelly or balm):ti,ab
#15 MeSH descriptor Administration, Topical, this term only
#16 MeSH descriptor Ointments, this term only
#17 MeSH descriptor Gels, this term only
#18 (#13 OR #14 OR #15 OR #16 OR #17)
#19 (#12 AND #18)

Appendix 2. Search strategy for MEDLINE(Ovid SP)

1. laceration.mp. or exp lacerations/ or exp facial Injuries/ or exp finger injuries/ or exp wounds, penetrating/ or exp hand injuries/ or exp sutures/ or exp surgical stapling/ or ((wounds.mp. or exp wounds/) and injuries/) or (injury adj3 (hand or eyelid or finger or facial or dermal)).mp. or cutaneous.mp. or staple.mp. or repair.mp.
2. (topical adj3 (an?esthe* or lidocaine or lignocaine or lidoderm or tetracaine or amethocaine or benzocaine or butamben or pramoxine or prilocaine or topical)).mp. or exp administration, topical/ or topical.ti,ab.or emla.mp.or eutectic mixture of local an?esthe*.mp. or tetracaine-adrenaline-cocaine.mp. or tetracaine-epinephrine-cocaine.mp. or lidocaine-adrenaline-tetracaine.mp. or lidocaine-epinephrine-tetracaine.mp. or spray.ti,ab. or ointment.mp. or exp ointments/ or gel.mp. or exp gels/ or cream.mp. or lotion.mp. or jelly.mp. or balm.mp.
3. 1 and 2
4. ((randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) not (animals not (humans and animals)).sh.
5. 3 and 5

Appendix 3. Search strategy for CINAHL (EBSCO host)

S28 S20 and S27
S27 S21 or S22 or S23 or S24 or S25 or S26
S26 AB random* or controlled trial* or mulicenter or placebo*
S25 (MM "Multicenter Studies")
S24 (MH "Prospective Studies+")
S23 (MM "Double-Blind Studies") or (MM "Single-Blind Studies") or (MM "Triple-Blind Studies")
S22 (MM "Placebos")
S21 (MM "Random Assignment") or (MH "Clinical Trials+")
S20 S12 and S19
S19 S13 or S14 or S15 or S16 or S17 or S18
S18 AB emla* or (eutectic mixture of local an?esthe*) or (tetracaine?adrenaline?cocain*) or (tetracaine?epinephrine?cocain*) or (lidocaine?adrenaline?tetracain*) or (lidocaine?epinephrine?tetracain*)
S17 AB spray or ointment or gel or cream or lotion or jelly or balm
S16 AB ( an?esthe* or lidocain* or lignocain* or tetracain* or amethocain* or benzocain* or butamben or dibucain* or pramoxin* or prilocain* or etidocian* ) and topical
S15 (MH "Gels")
S14 (MH "Ointments")
S13 (MH "Administration, Topical")
S12 S1 or S2 or S3 or S4 or S5 or S6 or S7 or S8 or S9 or S10 or S11
S11 AB laceration* or wound* or injur* or stapl* or repairor or suture*
S10 AB penetrat* and AB wound*
S9 AB ( facial or dermal or cutaneous or finger or hand or eyelid ) and AB injur*
S8 (MH "Surgical Stapling")
S7 (MH "Sutures")
S6 (MH "Hand Injuries")
S5 (MH "Finger Injuries")
S4 (MH "Arm Injuries")
S3 (MH "Facial Injuries")
S2 (MH "Wounds and Injuries+") or (MH "Wounds, Penetrating+")
S1 (MH "Tears and Lacerations")

Appendix 4. Search strategy for EMBASE (Ovid SP)

1. exp laceration/ or injury/ or exp face-injury/ or exp finger-injury/ or exp hand-injury/ or exp suture/ (123072)
2. (laceration* or wound* or injur* or ((facial or dermal or cutaneous or finger or hand or eyelid) adj3 injur*)).ti,ab. or (penetrat* adj3 wound*).mp. or (stapl* or repairor or suture*).ti,ab.
3. 1 or 2
4. (topical adj3 (an?esthe* or lidocain* or lignocain* or tetracain* or amethocain* or benzocain* or butamben or dibucain* or pramoxin* or prilocain* or etidocian)).mp.
5. (emla* or eutectic mixture of local an?esthe* or tetracaine?adrenaline?cocain* or tetracaine?epinephrine?cocain* or lidocaine?adrenaline?tetracain* or lidocaine?epinephrine?tetracain*).mp.
6. (spray or ointment or gel or cream or lotion or jelly or balm).ti,ab.
7. topical-drug-administration/ or ointment/ or gel/
8. 6 or 4 or 7 or 5
9. 8 and 3
10. (placebo.sh. or controlled study.ab. or random*.ti,ab. or trial*.ti,ab.) not (animals not (humans and animals)).sh.
11. 10 and 9

Appendix 5. Data Extraction Form

First authorJournal/Conference Proceedings etcYear

 

 

  

 

Trial characteristics
 Further details
RCT/Quasi 
Study size (number of patients) 
Setting of study (single versus multi-centre, inpatient versus outpatient) 

  

Participant characteristics
 Further details
Age (mean, median, range, etc) 
Sex of participants (numbers, percentages) 
Wound characteristics(length, location of laceration, etc.) 

  

Anaesthetic characteristics
 Further details
Infiltratedanaesthetics (agent, dosage) 
Topical anaesthetics (agent, dosage, duration of application) 
Were systemic analgesics or sedatives given? 

 

Outcomes
 Further details
Primary measure of pain intensity (patient self-report using pain scale such as VAS, numerical rating, etc) 
Secondary measure of pain intensity (incidence of topical anaesthetic failure, requirement of supplemental anaesthesia, patient behavioural responses etc) 
Anaesthetic related toxicity or acute adverse events 

 

 

Methodological quality

 

Sequence Generation
State here method used to generate sequence and reasons for grading Grade (circle)

 

 

Low risk
High risk
Unclear

 

 

Allocation Concealment
State here method used to conceal allocation and reasons for grading Grade (circle)

 

 

Low risk
High risk
Unclear

 

 

Blinding
State here method used to blind study and reasons for grading Grade (circle)

 

 

Low risk
High risk
Unclear

  

Description of Withdrawals and Dropouts
State here method used to address incomplete outcome data Grade (circle)

 

 

Low risk
High risk
Unclear

  

 

Appendix 6. Local anaesthetics and vasoconstrictors including alternative names

COCAINE CONTAINING TOPICAL ANAESTHETICS

AC = Epinephrine-cocaine or Adrenaline-cocaine

C = Cocaine

MAC = Bupivacaine- epinephrine-cocaine or Bupivacaine-adrenaline-cocaine

TAC = Tetracaine-epinephrine-cocaine or Tetracaine adrenaline-cocaine

TC = Tetracaine Cocaine

COCAINE FREE TOPICAL ANAESTHETICS

BN = Bupivacaine-norepinephrine

EMLA = Eutectic mixture of local anaesthetics = lidocaine-prilocaine

EN= Etidocaine-norepinephrine

LAT = LET = Lidocaine-epinephrine-tetracaine or Lidocaine-adrenaline-tetracaine

LE = lidocaine-epinephrine or Lidocaine-adrenaline

MN= Mepivacaine-norepinephrine

PN = Prilocaine-norepinephrine

PP = Prilocaine-phenylephrine

T = Tetracaine

TE = Tetracaine-epinephrine or Tetracaine-adrenaline

TP = Tetracaine-phenylephrine

TLP = Tetracaine-lidocaine-phenylephrine

ALTERNATIVE NAMES TO LOCAL ANAESTHETICS AND VASOCONTRICTORS

Epinephrine is the same as adrenaline

Bupivacaine also called marcaine or sensorcaine

Lidocaine also called xylocaine

Appendix 7. Cocaine-containing topical anaesthetics versus infiltrated local anaesthetics

StudyAnaestheticsPatient self-reported pain scoresSecondary outcome measuresIncidence anaesthetic toxicity
Anderson 1990Topical tetracaine-epinephrine-cocaine (TAC) vs infiltrated lidocaineNone1) Adequate initial anaesthesia (TAC = 89% vs infiltrated local anaesthetic = 79%, P=non-significant)
2) Physician compliance scale (1=complete compliance to 4=continuous resistance) (mean score ± SD: TAC = 1.25 ± 0.57 vs infiltrated local anaesthetic = 1.94 ± 1.12, P<0.002).
3) Requirement of supplemental lidocaine infiltration (topical TAC = 18% vs infiltrated local anaesthetic = 23%, P=non-significant).
Not reported
Hegenbarth 1990Topical TAC vs infiltrated lidocaineNone1) Adequate initial anaesthesia for facial and scalp lacerations (topical TAC = 81% vs infiltrated local anaesthetic = 87%, P=0.005). Adequate initial anaesthesia for the extremity and trunk wounds (topical TAC = 43% vs infiltrated local anaesthetic = 89%, P<0.0001)0/467
Pryor 1980Topical TAC vs infiltrated lidocaineNone1) Verbal rating of anaesthetic efficacy (complete: TAC = 84% vs infiltrated local anaesthetic = 88%, P=not reported).
2) Anaesthetic acceptability: patients 17 years or younger preferred topical TAC (P<0.005), there was no difference between the two anaesthetic groups in patients older then 17 years
Not reported
Smith 1996Topical TAC vs infiltrated lidocaine Patient-reported VAS (100mm) pain scores (mean scores: topical TAC = 12.0 versus infiltrated local anaesthetic = 26.3, P=NS)1) Observer-reported VAS pain scores
2) Observer-reported Likert pain scores
3) Oberver-rated Restrained Infants and Children Disress Rating Scale
4). Suture technician-rated anaesthetic effectiveness
Not reported
Smith 1997aTopical TAC vs infiltrated lidocaineNone

1) Observer-reported VAS pain scores (suture technicians, research assistants, video-tape reviewers)

2) Observer-reported Lickert (1-7) pain scores (parent, suture technicians)

3) Requirement for supplemental lidocaine infiltration

(See Characteristics of included studies for data)

Not reported
Kendall 1996Topical (epinephrine-cocaine) AC vs infiltrated lidocaineThe study pooled patient-reported VAS and Wong-Baker faces pain scores (mean score: topical AC = 4.50 vs infiltrated local anaesthetic = 4.40, P=NS)1) Physician-rated VAS pain scores
2) Parent-rated VAS scores
3) Parent's rating of overall acceptability of procedure
0/107

Appendix 8. Comparisons between different cocaine-containing topical anaesthetics

StudyTopical AnaestheticsPatient self-reported pain scoresSecondary outcome measuresIncidence anaesthetic toxicity
Kuhn 1996Bupivacaine- adrenaline-cocaine (MAC) vs tetracaine- epinephrine-cocaine (TAC)

1) In children < 12 years: Wong-Baker faces (1-9) scale (mean score ± SD: topical MAC = 2.35 ± .50 vs topical TAC = 2.46 ± 2.34, P=0.96)

2) Patients 12 years or older: VAS (100 mm) pain scale (mean score ± SD: topical MAC = 6.9 ± 10.9 vs topical TAC = 12.0 ± 14.5, P=0.16)

1) Adequacy of initial anaesthesia

2) Patients preference for topical anaesthesia in the future

0/180
Bonadio 1990TAC vs adrenaline-cocaine (AC)None1) Physician calculated the total number of "sutures eliciting pain" (topical AC = 7/103 (4%) vs topical TAC = 7/151 (7%), P= not-reported).0/55
Ernst 1990TAC vs cocaine (C)None

1) Incidence of "poor anaesthesia" (topical cocaine = 20% vs topical TAC = 12%, P=not-reported)

2) Physician numerical rating of anaesthetic effectiveness (0=least effective to 10=most effective) (mean scores ± SD: topical cocaine = 6.44 ± 3.48 vs topical TAC = 7.74 ± 3.03, P=0.005).

0/139
Vinci 1996TAC (two different strengths) vs tetracaine-cocaine (TC)None

Topical TAC 1 vs topical TC:
1) Complete anaesthesia (TAC 1 = 73% vs TC = 28%, P<0.001).

2) Requirement second dose of topical anaesthetic (TAC 1 = 30% vs topical TC = 66%, P<0.003).

3) Requirement for supplemental lidocaine infiltration (TAC 1 = 6% vs topical TC = 9%, P=not reported)

Topical TAC 2 vs topical TC:
1) Complete anaesthesia (TAC 2 = 63% vs TC = 28%, P<0.001)

2) Requirement second dose of topical anaesthetic (TAC 2 = 46% vs TC = 66%, P<0.003)

3) Requirement for supplemental lidocaine infiltration (TAC 2 = 2% vs TC = 9%, P=not reported)

1/156 (erythematous rash one day after application of standard topical TAC)

Appendix 9. Cocaine-free topical anaesthetics versus infiltrated local anaesthetics

StudyAnaestheticsPatient self-reported pain scoresSecondary outcome measuresIncidence anaesthetic toxicity
Ernst 1997Topical lidocaine-epinephrine-tetracaine (LAT) vs infiltrated lidocaine VAS (100mm) pain scores (median values: topical LAT = 0 vs infiltrated local anaesthetic = 0, P=0.48)

1) Physician-rated VAS pain scores

2) Requirement supplemental lidocaine infiltration

3) Percentage painful sutures

Not reported
Gaufberg 2007Topical lidocaine-epinephrine (LE) vs infiltrated lidocaine VAS (100mm) pain scores (mean score ± SD: topical TLE = 0.16 ± 0.46 versus infiltrated lidocaine = 0.20 ± 0.49, P=0.59).

1) Amount of lidocaine required (mg)

2) Total number of topical anaesthetic applications

Not reported
Smith 1996Topical bupivacaine-norepinephrine (BN), topical etidocaine-norepinephrine (EN), topical mepivacaine-norepinephrine (MN) and topical prilocaine-norepinephrine (PN) vs infiltrated lidocaine

VAS (100 mm) pain scores (mean scores: BN = 18.3, EN = 46.5, MN = 27.0, PN = 36.0 vs infiltrated anaesthetic = 26.3, standard deviations not reported)

(no significant difference between any of the cocaine free topical agents and infiltrated lidocaine)

1) Observer reported VAS pain scores

2) Observer reported Likert pain scores

3) Oberver-rated Restrained Infants and Children Disress Rating Scale

4) Suture technician-rated anaesthetic effectiveness

Not reported
Smith 1997aTopical mepivacaine-norepinephrine (MN) vs infiltrated lidocaineNone

1) Observer reported VAS pain scales scores

2) Observer reported Lickert pain scores

3) Requirement for supplemental lidocaine infiltration

(See characteristics of included studies for data)

Not reported

Appendix 10. Cocaine-free topical anaesthetics versus cocaine-containing topical anaesthetics

StudyTopical AnaestheticsPatient self-reported pain scoresSecondary outcome measuresIncidence anaesthetic toxicity
Smith 1996Bupivacaine-norepinephrine (BN), etidocaine-norepinephrine (EN), mepivacaine-norepinephrine (MN) and prilocaine-norepinephrine (PN) vs tetracaine-epinephrine-cocaine (TAC)

Patient-reported VAS (100 mm) pain scores (mean scores: BN = 18.3, EN = 46.5, MN, PN = 36.0 vs TAC = 12.0, standard deviations not reported)

(TAC significantly outperformed EN; no significant difference between any other groups)

1) Observer reported VAS and Likert pain scale scores

2)Observer-rated Restrained Infants and Children Disress Rating Scale

3) Suture technician-rated anaesthetic effectiveness

Not reported
Smith 1997aMepivacaine-norepinephrine (MN) vs
TAC
None

1) Observer-reported VAS pain scores (suture technicians, research assistants, video-tape reviewers)

2) Observer-reported Lickert (1-7) pain scores (parent, suture technicians)

3) Requirement for supplemental lidocaine infiltration

(See characteristics of included studies for data)

Not reported
Smith 1997bPrilocaine-phenylephrine (PP),
tetracaine-phenylephrine (TP) and
tetracaine-lidocaine-phenylephrine (TLP) vs TAC
VAS (100 mm) pain scores (mean score ± SD: PP = 29.0 ± 43.4, TP = 24.2 ± 37.2, TLP = 30.6 ± 40.3 versus TAC = 17.6 ± 34.1 (No significant difference between groups, P=0.5)

1) Oberver-reported VAS (100 mm) pain scores

2) Oberver-reported Likert (1-7) pain scores

3) Suture technicians-rated the anaesthetic effectiveness

Not reported
Smith 1998aPrilocaine-phenylephrine (PP) and bupivacaine-phenylephrine (BP) vs TAC VAS (100 mm) pain scores (mean score ± SD: PP = 21.0 ± 28.0 and BP = 41.0 ± 35.0 vs TAC = 18.0 ± 24.0). (No difference reported between groups, P=0.07)Observer-reported VAS pain scores (suture technicians, research assistants and parents)Not reported
Ernst 1995aLAT vs TAC Modified multidimensional pain scale (0-10) (mean ranked sum: LAT = 49.0 vs TAC = 46.9, P=0.71)

1) Physician-rated modified multidimensional pain scale (0-10)

2) Percentage of sutures causing pain

3) Requirement of supplemental lidocaine infiltration

0/95
Ernst 1995bLAT vs TAC VAS (100 mm) pain scores (mean ranked sum: LET = 45.3 vs TAC = 50.8, P=0.27)

1) Physician-reported VAS scores

2) Percentage of sutures causing pain

Not reported
Schilling 1995LAT vs TACNone

1) Adequacy of initial anaesthesia (LAT = 74.4% vs TAC = 79.5%, P=0.46)

2) Anaesthetic effectiveness (complete anaesthesia: LAT = 82.4% vs topical TAC = 75.9%, P=0.18)

0/151
Zempsky 1997Lidocaine-prilocaine (EMLA) vs TAC VAS (100 mm) pain scores (mean score ± SD: EMLA = 46.0 ± 26.0 vs TAC = 40.0 ± 25.0, P=0.50)

1) Observer-rated VAS pain scores

2) Requirement for supplemental lidocaine infiltration

Not reported
Blackburn 1995Lidocaine-epinephrine (LE) vs TAC Faces pain scale (1-9) scores (mean score ± SD: LE = 3.29 ± 1.92 vs TAC = 2.66 ± 1.78, P=0.33).Requirement for supplemental lidocaine infiltration0/35
Schaffer 1985Tetracaine-epinephrine (TA) vs TAC)None

1) Physician-rating of anaesthetic effectiveness (complete anaesthesia: TA = 47.1% vs TAC = 75%, P<0.05)

2) Requirement of rescue lidocaine infiltration (TA = 27.5% vs TAC = 8.9%, P=0.01)

0/107
White 1986Tetracaine (T) vs TAC Numerical pain scale (0-10) score (mean scores: tetracaine = 5.6 vs TAC = 3.53, P<0.05, standard deviations not reported)Requirement of supplemental lidocaine infiltrationNot reported

Appendix 11. Comparisons between different cocaine-free topical anaesthetics

StudyTopical AnaestheticsPatient self-reported pain scoresSecondary outcome measuresIncidence anaesthetic toxicity
Smith 1996Bupivacaine-norepinephrine (BN) vs etidocaine-norepinephrine (EN) vs mepivacaine-norepinephrine (MN) vs prilocaine-norepinephrine (PN) Patient-reported VAS (100 mm) pain scores (mean scores: topical BN = 18.3 vs topical EN = 46.5 vs topical MN = 27.0 vs topical PN = 36.0) (no significant difference between any of cocaine free topical groups).

1) Observer reported VAS and Likert pain scale scores

2)Observer-rated Restrained Infants and Children Disress Rating Scale

3) Suture technician-rated anaesthetic effectiveness

Not reported
Smith 1997bPrilocaine-phenylephrine (PP) vs
tetracaine-phenylephrine (TP) vs
tetracaine-lidocaine-phenylephrine (TLP)
VAS (100 mm) pain scores (mean score ± SD: PP = 29.0 ± 43.4 vs TP = 24.2 ± 37.2 vs TLP = 30.6 ± 40.3 (No significant difference between groups, P=0.5)

1) Oberver-reported VAS (100 mm) pain scores

2) Oberver-reported Likert (1-7) pain scores

3) Suture technicians-rated the anaesthetic effectiveness

Not reported
Smith 1998aPrilocaine-phenylephrine (PP) vs bupivacaine-phenylephrine (BP) VAS (100 mm) pain scores (mean score ± SD: topical PP = 21.0 ± 28.0 vs topical BP = 41.0 ± 35.0, P=0.07)Observer-reported VAS pain scores (suture technicians, research assistants and parents)Not reported
Krief 2002Lidocaine-prilocaine (EMLA) vs lidocaine-epinephrine-tetracaine (LAT) VAS (100 mm) pain scores were not significantly different between the two groups (mean pain scores not provided, P > 0.05)

1) Observer-reported VAS pain scores (legal guardian and physician)

2) Requirement of supplemental lidocaine infiltration

Not reported
Resch 1998Topical LAT gel vs LAT solutionNone

1) Adequacy of initial anaesthesia (adequate anaesthesia: LAT solution = 84% vs LAT gel = 82%, P>0.05)

2) Effectiveness of anaesthesia (complete anaesthesia: LAT solution = 76% vs LAT gel = 85%, P=0.007)

0/194

What's new

DateEventDescription
18 May 2015AmendedDeclaration of interest statement updated in review

History

Protocol first published: Issue 3, 2005
Review first published: Issue 6, 2011

DateEventDescription
11 July 2012AmendedContact details updated.
31 May 2012AmendedContact details updated.
17 April 2012AmendedContact details updated.
18 January 2012AmendedContact details updated.
1 September 2008AmendedConverted to new review format.

Contributions of authors

Conceiving the review: Daniel Carr (DC)
Co-ordinating the review: Anthony Eidelman (AE) and DC
Undertaking manual searches: AE and Ikay Enu (IE)
Screening search results: AE, Cristy Baldwin (CB), IE
Organizing retrieval of papers: AE, IE and Jocelyn Weiss (JW)
Screening retrieved papers against inclusion criteria: AE, CB, IE, and JW
Appraising quality of papers: AE, CB
Abstracting data from papers: AE, CB, IE
Writing to authors of papers for additional information: AE
Providing additional data about papers: AE, IE
Obtaining and screening data on unpublished studies: AE
Data management for the review: AE, IE, Ewan McNicol (EM)
Entering data into Review Manager (RevMan 5.1 ): AE, EM
RevMan statistical data: AE
Other statistical analysis not using RevMan: AE
Double entry of data: (data entered by person one: AE; data entered by person two: EM)
Interpretation of data: AE, CB, EMand DC
Statistical analysis: AE
Writing the review: AE, DC
Securing funding for the review: DC
Performing previous work that was the foundation of the present study: AE, IE, JW, DC
Guarantor for the review (one author): AE
Responsible for reading and checking review before submission: AE and DC

Declarations of interest

DBC was employed by Javelin Pharmaceuticals from 2004 to 2010. None of the product candidates developed or considered for development by Javelin involved local anaesthetics in any formulation or route of delivery. None known for the other authors

Sources of support

Internal sources

  • Evenor Armington Fund, USA.

  • Saltonstall Fund, USA.

External sources

  • No sources of support supplied

Differences between protocol and review

The title was changed from 'Topical anaesthetics for repair of torn skin' to 'Topical anaesthetics for repair of dermal laceration'

Notes

May 2015

Declaration of interest statement updated in review

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Anderson 1990

MethodsSingle-centre RCT
Participants151 patients age less then 18 years old with lacerations located on the scalp (n=31), face (n=79) or extremity (n=41)
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 5-10 minutes (n=56),
2. Intra-dermal infiltration with lidocaine 1% (n=53),
3. Topical placebo solution applied for 5-10 minutes (n=42)
Outcomes

1. Prior to laceration repair, the physician probed the wound with a 25 gauge needle to determine adequacy of initial anaesthesia.
2. The physician graded patient compliance during the suturing process on a four point scale (1-complete compliance, 2-occasional resistance, 3-frequent resistance, 4-continuous resistance).
3. Requirement of supplemental lidocaine infiltration.

Results of topical TAC versus topical placebo:
1. Adequate initial anaesthesia (topical TAC = 89% versus topical placebo = 17%, P<0.0001).
2. Physician compliance scale (1-4) ratings (complete compliance to continuous resistance) (mean score ± SD: topical TAC = 1.25 ± 0.57 versus topical placebo = 1.93 ± 0.96, P<0.002).
3. Requirement of supplemental lidocaine infiltration (topical TAC = 18% versus topical placebo = 83%, P<0.0001).

Results of topical TAC versus infiltrated local anaesthetic:
1. Adequate initial anaesthesia (topical TAC = 89% versus infiltrated local anaesthetic = 79%, P=non-significant)
2. Physician compliance scale (1-4) ratings (complete compliance to continuous resistance) (mean score ± SD: topical TAC = 1.25 ± 0.57 versus infiltrated local anaesthetic = 1.94 ± 1.12, P<0.002).
3. Requirement of supplemental lidocaine infiltration (topical TAC = 18% versus infiltrated local anaesthetic = 23%, P=non-significant).

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "The last digit of the patient's medical record number was used to enter patients into either the intradermal or topical group"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "The last digit of the patient's medical record number was used to enter patients into either the intradermal or topical group"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: "Individual study vials containing 5ml of TAC or placebo were prepared in the pharmacy of University of Massachusetts Medical Center following a standard protocol and assigned numbers"; "The ED staff member evaluating and suturing the patient were blind to the solution contained in the vials"

Comment: Comparisons of topical TAC and topical placebo were probably blinded. However, comparisons between lidocaine infiltration and topical TAC were probably unblinded

Incomplete outcome data (attrition bias)
All outcomes
Low risk153 eligible patients, 2 refused to participate. 151 randomized and no missing outcome data

Blackburn 1995

MethodsSingle-centre RCT
Participants35 adult and paediatric patients (minimum age of 2 years) with facial and scalp lacerations, 6cm or less in length
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 10.4%), applied for 20 minutes (n=18)
2. Topical TLE solution (lidocaine 5% and epinephrine 1:2000), applied for 20 minutes (n=17)
Outcomes

1. The patient reported discomfort using a facial effective pain scale (1-9), which consisted of nine faces with various emotional expressions. However, in a few cases the patient was too young to use the pain scale, so the physician estimated the subject's pain on the same faces scale. The study combined both the self-reported and surrogate faces pain scale scores in the final results.
2. Requirement of rescue lidocaine infiltration.
3. The study reported any acute adverse reactions directly related to the anaesthetic.

Results:
1. Faces pain scale (1-9) scores (mean score ± SD: topical LE = 3.29 ± 1.92 versus topical TAC = 2.66 ± 1.78, P=0.33).
2. Requirement for supplemental lidocaine infiltration (topical LET = 6% versus topical TAC = 6%, P=not reported).
3. No acute anaesthetic related adverse effects

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "The TAC and TLE solutions were arbitrarily assigned to a single-dose (10ml), sequentially-numbered vials by the pharmacist. The vials, with the specific contents unknown to the emergency physician, were forwarded to the ED as requested"

Comment: Probably not done

Allocation concealment (selection bias)Low risk

Quote: "The solutions were made visibly identical by adding methylene blue to the TLE solution so that it matched the intrinsic blue color of TAC"

"The vials, with the specific contents unknown to the emergency physician, were forwarded to the ED as requested"

Comment: Probably done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "The solutions were made visibly identical by adding methylene blue to the TLE solution so that it matched the intrinsic blue color of TAC"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk35 patients in study, but insufficient reporting of attrition or exclusions to permit judgment

Bonadio 1990

MethodsSingle-centre RCT
Participants55 paediatric patients with facial lacerations
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 10-15 minutes (n=24)
2. Topical AC solution (epinephrine 1:2000, cocaine 11.8%), applied for 10-15 minutes (n=31)
Outcomes

1. The physician calculated the total number of "sutures eliciting pain" using the following system. Each suture placed involved two points; an entrance and exit piercing of the wound tissue with the needle. A painful response consisted of either a verbal patient experiencing a painful sensation or a non-verbal patient beginning to cry, or crying with greater intensity. The total number of "sutures placed eliciting pain" was calculated, by dividing the total number of painful responses by two.
2. The study reported any acute adverse effects due to the anaesthetic.

Results:
1. The physician calculated the total number of "sutures eliciting pain" (topical AC = 7/103 (4%) versus topical TAC = 7/151 (7%), P=not-reported).
2. No acute anaesthetic related adverse effects

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "...as in each case an assistant randomly selected one of the two solutions for physician application..."

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "an assistant randomly selected one of the two solutions for physician application..."

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "The managing physician was "blind" to which preparation was being administered,...the physician was informed of the solution composition only after the suturing procedure and pain scoring were completed"

Comment: Probably done, assuming the two solutions were visually identical

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk55 patients in study, but insufficient reporting of attrition or exclusions to permit judgment

Ernst 1990

MethodsRCT
Participants139 adult and paediatric patients, age greater then 5 years, with laceration of the face (n=53), scalp (n=33), extremity (n=52) or trunk (n=1), less then 5 cm in length
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 5-10 minutes (n=69)
2. Topical cocaine solution 11.8%, applied for 5-10 minutes (n=70)
Outcomes

1. The physician assessed the adequacy of initial anaesthesia by pricking the wound with a pin. If pain was elicited with pinprick, then 1% lidocaine was infiltrated and the patient was assigned to the "poor anaesthesia" group.
2. In patients who did not require infiltrated lidocaine, the physician rated the effectiveness of anaesthesia during suturing on a numerical scale (0-10). 3. The trial reported acute adverse reactions directly related to the anaesthetic.

Results:
1. Incidence of "poor anaesthesia" (topical cocaine = 20% versus topical TAC = 12%, P=not-reported)
2. Physician-rating of anaesthetic effectiveness on a numerical (0 to 10) scale (least effective to most effective) (mean scores ± SD: topical cocaine = 6.44 ± 3.48 versus topical TAC = 7.74 ± 3.03, P=0.005).
3. No acute anaesthetic related adverse effects

NotesContacted author for additional information, but no reply.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "TAC and cocaine solutions were randomly distributed with only a number from 1-150 appearing on each vial"

Comment: Unclear. The exact mechanism of randomization not described

Allocation concealment (selection bias)Unclear risk

Quote: "TAC and cocaine solutions were randomly distributed with only a number from 1-150 appearing on each vial"

"The investigator was blinded as to the identity of the agent. The code was kept in the pharmacy and was available to the investigators only in case of emergency"

Comment: Unlear. There may be allocation concealment if there was a pharmacy controlled randomization process. However, this is not explicitly reported, so we decided upon unclear

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "The investigator was blinded as to the identity of the agent. The code was kept in the pharmacy and was available to the investigators only in case of emergency"

Comment: Probably done, assuming the local anaesthetic solutions are identical in colour.

Incomplete outcome data (attrition bias)
All outcomes
Low risk148 patients were enrolled and 9 patients excluded from the study before un-blinding and analysis (4 improper application, 4 patient age less then 5 and one because laceration too large). We concluded low risk of bias because the number of excluded patients was balanced between the two interventions and the reasons for exclusion unlikely to be related to pain scores during suturing.

Ernst 1995a

MethodsSingle-center RCT
Participants95 patients age 5 to 17 years with lacerations on the face (n=64) or scalp (n=31), 7cm or less in length
Interventions1. Topical LAT gel (lidocaine 4%, epinephrine 1:2000, tetracaine 0.5%), applied for 10-30 minutes (n=48)
2. Topical TAC gel (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 10-30 minutes (n=47)
Outcomes

1. Patient-rated modified multi-dimensional pain scale (0-10).
2. Physician-rated modified multi-dimensional pain scale (0-10).
3. Percentage of sutures causing pain.
4. Requirement of supplemental lidocaine infiltration.
5. The trial reported acute adverse reactions directly related to the anaesthetic.

Results:
1. Patient-reported modified multi-dimensional pain scale (0-10) scores (mean ranked sum: topical LAT = 49.0 versus topical TAC = 46.9, P=0.71).
2. Physician-assigned multi-dimensional pain scale (0-10) scores (mean ranked sum: topical LAT=48.7 versus topical TAC = 47.3, P=0.80)
3. Percentage of sutures placed causing pain (mean ranked sum: topical LET = 49.57 versus topical TAC = 46.39, P=0.51).
4. Requirement supplemental lidocaine infiltration (topical LET = 4%, topical TAC = 6%, P=not-reported)
5. No acute anaesthetic related adverse effects.

NotesContacted author for additional information, but no reply.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote "Gels were randomized according to a random numbers table"

Comment: Probably done

Allocation concealment (selection bias)High risk

Quote: "randomized according to a random numbers table"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Patients and physicians performing suturing were blinded to which gels were being used. Only the numbers 1-100 appeared on the capped syringes"

Comment: Probably done, assuming the two gels were visually identical

Incomplete outcome data (attrition bias)
All outcomes
Low risk100 patients entered into trial, but 5 patients were excluded before statistical analysis because topical anaesthesia was inadequate and lidocaine infiltration was required. Two patients in the LAT group and 3 in the TAC group were excluded. We judged low risk of bias because the number of excluded patients were balanced between the two interventions.

Ernst 1995b

MethodsSingle-centre RCT
Participants95 adult patients with lacerations of the face (n=81) or scalp (n=13), 7cm or less in length
Interventions1. Topical LAT solution (lidocaine 4%, epinephrine 1:2000, tetracaine 0.5%), applied for 10-30 minutes (n=48)
2. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 10-30 minutes (n=47)
Outcomes

1. Patient-rated VAS (100 mm) pain score.
2. Physician-rated VAS (100 mm) pain score.
3. Percentage of sutures eliciting pain.

Results:
1. Patient-reported VAS (100 mm) pain scores (mean ranked sum: topical LET = 45.3 versus topical TAC = 50.8, P=0.27).
2. Physician-reported VAS scores (mean ranked sum: topical LAT = 41.6 versus topical TAC = 54.6, P=0.01).
3. Percentage of sutures causing pain (mean ranked sum: topical LET = 42.8% versus topical TAC = 53.3%, P=0.36).

NotesContacted author for additional information, but no reply.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote: "Solutions were randomized according to a random numbers table"

Comment: Probably done

Allocation concealment (selection bias)Low risk

Quote: "The solutions were prepared by a pharmacist and were available in coded sterile, capped 3ml syringes"

"Both TAC and LAT were clear solutions..."

"Patients and physicians performing wound closure were blinded"

Comment: Probably done.

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "The solutions were prepared by a pharmacist and were available in coded sterile, capped 3ml syringes with a cotton ball for application"

"Both TAC and LAT were clear solutions mixed from powders"

"Patients and physicians performing wound closure were blinded"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
High risk100 total patients enrolled but only 95 included in final data analysis. Four patients excluded because they required additional injected lidocaine (1 LAT group, 3 in TAC group) and one patient excluded because of improper data collection. We judged "no" (high risk of bias) because requirement of additional lidocaine is directly related to pain intensity during laceration repair.

Ernst 1997

MethodsSingle-centre RCT
Participants66 paediatric and adult patients, age greater then 5 years with lacerations located on the face (n=30), scalp (n=10) or extremity (n=24), 1.5-10 cm in length
Interventions1. Topical LAT gel (lidocaine 4%, epinephrine 1:2000, tetracaine 0.5%), applied for 10-20 minutes (n=33),
2. Intra-dermal infiltration with lidocaine 1%, epinephrine, buffered with 8.4% NaHCO3 (n=33)
Outcomes

1. Patient-rated VAS (100 mm) pain scale scores.
2. Physician-rated VAS (100 mm) pain scale scores.
3. Requirement of supplemental lidocaine infiltration
4. Percentage of sutures placed eliciting pain.

Results:
1. Patient self-reported V AS (100 mm) pain scores (median values [interquartile range]: topical LAT = 0 [0-1.35] versus infiltrated local anaesthetic = 0 [0-0.6], P=0.48, standard deviations not reported).
2. Physician-reported VAS (100 mm) pain scores (median values [interquartile range]: topical LAT = 0 [0-0.55] versus infiltrated local anaesthetic = 0 [0-0.35], P=0.83, standard deviations not reported).
3. Percentage of sutures causing pain (topical LAT = 13% versus infiltrated local anaesthetic = 6%, P=0.28)
4. Requirement of supplemental infiltrated anaesthesia (LAT = 6% versus infiltrated anaesthetic = 0%, P=not reported).

NotesContacted author for additional information, but no reply.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote: "The doses of anaesthetic were numbered 1-66 according to a computer generated random table of numbers prepared before the study"

Comment: Probably done

Allocation concealment (selection bias)High risk

Quote: "physicians and patients were not blinded to the form of anaesthesia"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: "Because of the obvious differences in form and application, physicians and patients were not blinded to the form of anaesthesia"

Comment: Probably not done

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk66 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Gaufberg 2007

MethodsSingle-centre RCT
Participants100 adult patients greater then 18 years old with lacerations involving scalp (n =15), face (n = 15), lower extremity (n =13), upper extremity (n = 15 ) or hands (n = 42). Laceration length ranged from <1 cm to >5cm.
Interventions

1. TLE solution (lidocaine 5%, epinephrine 0.025%), applied for 10-15 minutes for 1 to 4 sequential layered applications (n = 50)

2. Intra-dermal infiltration with lidocaine (n = 50)

Outcomes

1. Patient-rated VAS (100 mm) pain scale scores

2. The amount of lidocaine to required(mg)

3. Number of applications of the topical anaesthetic

4. Difficulty with wound healing or infections

Results:

1. Patient-reported VAS (100 mm) pain scores (mean score ± SD: topical TLE = 0.16 ± 0.46 versus infiltrated lidocaine = 0.20 ± 0.49, P=0.59).

2. Amount of lidocaine required (mean score: TLE = 135mg versus infiltrated lidocaine = 124mg, P=0.90, SD not reported)

3. Number of anaesthetic applications of TLE (mean score = 2.7; 2 patients (4%) required 1 layer, 17 patients (34%) required 2 layers, 26 patients (52%) required 3 layers, 5 patients (10%) required 4 layers.

4. No patients had poor wound healing or infections

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "We performed a prospective, randomized controlled trial.."

Comment: Unclear, study reported to be randomized but method of sequence generation not described

Allocation concealment (selection bias)High riskComment: Probably not done. The interventions of topical anaesthesia versus infiltrated anaesthesia are visually different. No mechanism to conceal the intervention from patients or study personnel was described
Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: "100 patient were enrolled in a randomized controlled trial..."

Comment: Probably not done, as study did not report blinding, and compared topical and infiltrated forms of anaesthesia

Incomplete outcome data (attrition bias)
All outcomes
Low risk100 enrolled patients in study and no missing outcome data or exclusions

Hegenbarth 1990

MethodsTwo-centre RCT
Participants467 patients, 18 years or younger, with dermal lacerations located on the face, scalp, extremity and trunk
Interventions1. TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%),
applied for 30 minutes (n= 262) 2. Intra-dermal infiltration with lidocaine 1% (n= 205)
Outcomes

The pain during the suturing process was not directly assessed.
1. Prior to laceration repair, the physician probed the wound with a 26-gauge needle to determine the adequacy of initial anaesthesia (adequate, inadequate or unable to access). The physician administered infiltrated anaesthetic to patients in the TAC group with "inadequate" anaesthesia.
2. The study reported any acute adverse reactions to the anaesthetic.

Results
1. Adequate initial anaesthesia for facial and scalp lacerations (topical TAC = 81% versus infiltrated local anaesthetic = 87%, P=0.005). Adequate initial anaesthesia for the extremity and trunk wound group (topical TAC = 43% versus infiltrated local anaesthetic = 89%, P<0.0001)
2. No acute anaesthetic related adverse effects

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "Randomization of anaesthetic treatment was determined by the final digit of the patients medical record number, with odd numbers receiving lidocaine and even numbers receiving TAC"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "Randomization of anaesthetic treatment was determined by the final digit of the patients medical record number"

"unblinded study"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: "We conducted a prospective, randomized, unblinded study..."

Comment: Probably not done

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk467 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Kendall 1996

MethodsSingle-centre RCT
Participants107 paediatric patients, 3-16 years old, with lacerations less then 4 cm in length, located anywhere on the body except mucous membranes or digits.
Interventions1. Topical AC solution (epinephrine 1:2000, cocaine 4.7%), applied for 10-15 minutes (n= 51)
2. Intra-dermal infiltration with lidocaine 1% (n= 51)
Outcomes

1. Children less than 10 years of age rated pain during both laceration repair and anaesthetic application with the Wong-Baker faces scale. Patients 10 years or older, used a VAS (10 cm) score to rate pain during suturing and anaesthetic administration.
2. Physician-rated VAS (10 cm) pain scale scores.
3. Parent-rated VAS (10 cm) pain scale scores.
4. The parent rated the overall acceptability of the procedure.
5. The study reported any acute adverse effects to the anaesthetic

Results:
(Standard deviations were not reported)
1. Patient-rated pain scores (pooled VAS and Wong-Baker faces scores) (mean score: topical AC = 4.50 versus infiltrated local anaesthetic = 4.40, P=NS).
2. Physician-rated VAS (10 cm) pain scale scores (mean score: topical AC = 2.60 versus infiltrated local anaesthetic = 3.60, P=NS).
3. Parent-rated VAS (10 cm) pain scale scores (mean score: topical AC = 3.10 versus infiltrated local anaesthetic = 3.80, P=NS).
4. The parent's rating of overall acceptability of the procedure (topical AC = 14.5% unacceptable versus infiltrated local anaesthetic = 39% unacceptable, P<0.01).
5. No acute anaesthetic related adverse effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "Children presenting with an appropriate laceration were consecutively assigned to receive either conventional intradermal lignocaine or topical AC preparation"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "consecutively assigned to receive either conventional intradermal lignocaine or topical AC preparation"

"groups could not be blinded"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: "The nature of the trial meant that the two groups could not be blinded"

Comment: Probably not done

Incomplete outcome data (attrition bias)
All outcomes
Low risk120 patients were enrolled but 13 excluded prior to data analysis (8 incomplete data collection, 2 received Steristrips and not sutures and 3 did not attend follow-up). We concluded low risk of bias, because reasons for exclusion unlikely to be related to pain scores during laceration repair.

Krief 2002

MethodsRCT (unclear if single or multi-centre)
Participants41 adult and paediatric patients, 5 to 23 years of age, with simple lacerations < 5cm in length
Interventions1. Topical LET gel (lidocaine, epinephrine, tetracaine), applied for 60 minutes (n=22) 2. EMLA cream (lidocaine 2.5%, prilocaine 2.5%), applied for 60 minutes (n=19)
Outcomes

1. Patient-rated VAS (100 mm) pain scale scores
2. Legal guardian-rated VAS (100mm) pain scale scores (when applicable)

3. Physician-rated VAS (100 mm) pain scale scores

4. Requirement of supplemental lidocaine infiltration

Pain scores were obtained at four points in time: after irrigation, first suture or staple placement, last suture or staple placement and during supplemental infiltration of lidocaine (if applicable).

Results:
1. Patient self-reported V AS (100 mm) pain scores were not significantly different between the two anaesthetic groups (mean pain scores not provided, P > 0.05)

2. Legal guardian-reported V AS (100 mm) pain scores were not significantly different between the two groups (mean pain scores not provided, P > 0.05)

3. Physician reported V AS (100 mm) pain scores were greater in the EMLA group during irrigation (mean VAS EMLA = 21.4mm versus LET gel = 10.1mm, P=0.3) and during first suture/staple placement (mean VAS EMLA = 41.7mm versus LET gel = 14.0mm, P=0.004)

4. Requirement of supplemental infiltrated anaesthesia: 13/19 patients in the EMLA group required infiltrated lidocaine (68%) compared to 5/22 in the LET group (23%), P=0.005%

NotesTrial published as an abstract only
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "We conducted a double-blind, randomized trial..."

Comment: Unclear, as method of sequence generation not described

Allocation concealment (selection bias)Unclear riskComment: Unclear
Blinding (performance bias and detection bias)
All outcomes
Unclear risk

Quote: "We conducted a double-blind, randomized trial...

Comment:Unclear as reported to be double-blind, but no details provided

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk41 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Kuhn 1996

MethodsSingle-centre RCT
Participants180 adult and paediatric patients, 6 years or older, with lacerations 3-7 cm in length, located on the head (n=114) or extremity (n=66)
Interventions1. Topical MAC solution (bupivacaine 0.5%, epinephrine 1:2000, cocaine 10.0%), applied for at least 10-15 minutes for head lacerations and for 30 minutes for extremity wounds (n=92)
2. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 10.0%), applied for at least 10-15 minutes for head lacerations and for 30 minutes for extremity wounds (n=88)
Outcomes

1. Children less than 12 years of age rated pain during laceration repair with the Wong-Baker faces scale.
2. Patients 12 years or older, used a VAS (10 cm) score to rate pain during suturing.
3. The physician assessed the effectiveness of initial anaesthesia using pinprick.
4. The patients noted their preference for topical anaesthesia in the future.
5. The study reported any acute adverse effects to the anaesthetic

Results:
1. Children under 12 years of age used the Wong-Baker faces (1-9) scale (mean score ± SD: topical MAC = 2.35 ± .50 versus topical TAC = 2.46 ± 2.34, P=0.96).
2. Patients 12 years or older used the VAS (100 mm) pain scale (mean score ± SD: topical MAC = 6.9 ± 10.9 versus topical TAC = 12.0 ± 14.5, P=0.16). 3. Adequacy of initial anaesthesia (topical MAC = 73% versus topical TAC = 74%, P=0.87)
4. The patients preference for topical anaesthesia in the future (topical MAC = 77% versus topical TAC = 81%, P=0.42)
5. No acute anaesthetic related adverse effects

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "The study was a double-blinded, randomized, prospective trial.."

Comment: Unclear as study reported to be randomized but method of sequence generation was not described

Allocation concealment (selection bias)Low risk

Quote: "Solutions of MAC and modified TAC were prepared and placed in syringes marked A or B by a pharmacist not involved in study. All study participants remained blinded throughout the trial"

Comment: Probably done, assuming solutions were visually identical

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Solutions of MAC and modified TAC were prepared and placed in syringes marked A or B by a pharmacist not involved in study. All study participants remained blinded throughout the trial"

Comment: Probably done, assuming solutions were visually identical

Incomplete outcome data (attrition bias)
All outcomes
Low risk191 patients were enrolled but 10 excluded prior to data analysis (5 less then 6 years old, two had wounds greater 5mm deep, 2 were not sutured, and 1 had a digital laceration). We concluded low risk of bias, because reasons for exclusion unlikely to be related to pain scores during laceration repair.

Pryor 1980

MethodsSingle-centre RCT
Participants158 adult and paediatric patients, range from 10 months to 53 years old (mean = 9 years old)
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for minimum of 10 minutes
(n=82).
2. Intra-dermal infiltration with lidocaine (n=76)
Outcomes

1. Patients 10 years of age or older rated anaesthetic efficacy (complete, partial or none) depending on whether they experienced pain during laceration repair.
2. Also, after completion of wound repair, the patient or parent rated the anaesthetic acceptability (excellent, good or poor).

Results:
1. Verbal rating (complete, partial or none) of anaesthetic efficacy (complete: topical TAC = 84% versus infiltrated local anaesthetic = 88%, P=not reported).
2. Anaesthetic acceptability: patients 17 years or younger preferred topical TAC (P<0.005), while there was no difference between the two anaesthetic groups in patients older then 17 years.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "A prospective study of topical TAC and lidocaine infiltration was taken with the last digit of the patients military sponsor's social security number used as the selection variable, Odd numbered patients were anaesthetised with topical TAC; even numbered patients were anaesthetised with lidocaine"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "the last digit of the patients military sponsor's social security number used as the selection variable"

Comment: Probably not done. Anaesthetic agents visually different and no mention of safeguards to prevent concealment of identity

Blinding (performance bias and detection bias)
All outcomes
High risk

Quote: None

Comment: Probably not blinded as the paper did not state whether patients or clinicians were blinded between the topical and infiltrated anaesthetics

Incomplete outcome data (attrition bias)
All outcomes
Low riskA total of 158 patients enrolled and no dropouts or exclusions

Resch 1998

MethodsSingle-centre RCT
Participants194 paediatric patients with lacerations of the face and scalp
Interventions1. Topical LAT solution (lidocaine 4%, epinephrine 1:2000, tetracaine 0.5%), applied for 20 minutes (n=103)
2. Topical LAT gel (lidocaine 4%, epinephrine 1:2000, tetracaine 0.5%), applied for 20 minutes (n=91)
Outcomes

1. The physician assessed the adequacy of initial anaesthesia by probing the wound with a 27-gauge needle. If any pain was elicited with probing, then the anaesthetic was considered "inadequate" and infiltrated lidocaine was given.
2. At the conclusion of laceration repair, the physician rated the anaesthetic effectiveness (complete, partial or incomplete) based on painful responses during suturing.
3. The study reported acute adverse reactions directly related to the anaesthetic.

Results:
1. Adequacy of initial anaesthesia (adequate anaesthesia: LET solution = 84% versus LET gel = 82%, P>0.05).
2. Effectiveness of anaesthesia (complete anaesthesia: LET solution = 76% versus LET gel = 85%, P=0.007).
3. There were no acute anaesthetic-related adverse effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote: "A computer-generated random number table was used by a hospital pharmacy personnel to label standard amber vials from 1 to 200"

Comment: Probably done

Allocation concealment (selection bias)Low risk

Quote: "hospital pharmacy personnel to label standard amber vials from 1 to 200"

"it was required that the study medication be applied by a nurse not involved in the suturing"

Comment: Probably done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "To ensure blinding of suture personnel, in the trial, it was required that the study medication be applied by a nurse not involved in the suturing"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk

200 patients enrolled and 3 withdrawn before test of initial anaesthesia because patients uncooperative or complicated laceration that did not meet inclusion criteria. Out of the 197 available for analysis, 3 data sheets inadvertently lost.

We concluded low risk of bias, because plausible effect size among missing outcomes not enough to have a clinically relevant impact on observed effect size.

Schaffer 1985

MethodsSingle-centre RCT
Participants107 paediatric patients 10 years of age or younger, with lacerations on the face (n=84) or scalp (n=23)
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 10 minutes (n=56)
2. Topical TA solution (tetracaine 0.5%, epinephrine 1:2000), applied for 10 minutes (n=51)
Outcomes

1. The physician rated the anaesthetic effectiveness (complete, partial or inadequate) based on the ability of the patient to tolerate manipulation of the wound during repair. The anaesthesia was "complete" if the patient did not cry, complain or wince during suturing. The anaesthesia was "partial" if the patient had some discomfort, but did have an avoidance reaction. "Inadequate" anaesthesia was defined as obvious discomfort with minimal manipulation of the wound.
2. Requirement of rescue lidocaine infiltration
3. The study reported any acute adverse reactions to the anaesthetic.

Results:
1. Physician-rating (complete, partial or inadequate) of anaesthetic effectiveness (complete anaesthesia: topical TA = 47.1% versus topical TAC = 75%, P<0.05).
2. Requirement of rescue lidocaine infiltration (topical TA = 27.5% versus topical TAC = 8.9%, P=0.01).
3. No acute anaesthetic related adverse effects. However, after returning home from the emergency department, 10.7% of children treated with TAC and 7.8% who received topical AC became drowsy or excitable. There was no evidence that the symptoms were causally related to the topical anesthetic and the author concluded that these were not anaesthetic induced adverse effects.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "Patients who received topical anaesthesia were randomized by alternating between A and B solutions"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "...randomized by alternating between A and B solutions"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Neither patients nor treating physicians were informed of the composition of the anaesthetic solutions"

Comment: Probably done assuming topical TAC and TA were visually identical

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk107 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Schilling 1995

MethodsSingle-centre RCT
Participants151 patients, age 1 to 17 years old with facial (69.6%) and scalp (30.4%) lacerations
Interventions1. Topical TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%), applied for 15 minutes (n=73)
2. Topical LET solution (lidocaine 4%, epinephrine 0.1%, tetracaine 0.5%), applied for 15 minutes (n=78)
Outcomes

1. The physician assessed the adequacy of initial anaesthesia by probing wound with a 27-gauge needle.
2. After laceration repair, the physician rated the anaesthetic effectiveness (complete, partial or incomplete). The anaesthesia was "complete" if the patient did not have a painful response to suturing. The anaesthesia was "partial" if the patient had a painful response to suturing, between 15 and 30 minutes after removal of topical solution. The anaesthesia was considered "incomplete" if the patient had a painful response, within 15 minutes after removal of the topical agent.
3. The study reported any acute adverse reactions directly related to the anaesthetic.

Results:
1. Adequacy of initial anaesthesia (topical LET = 74.4% versus topical TAC = 79.5%, P=0.46)
2. Physician-rated anaesthetic effectiveness (complete, partial, incomplete) (complete anaesthesia: topical LAT = 82.4% versus topical TAC = 75.9%, P=0.18).
3. No acute anaesthetic related adverse effects

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "...vials of the anaesthetic solutions were assigned random numbers.."

Comment: Unclear as study was reported to be randomized, but method of sequence generation not described

Allocation concealment (selection bias)Low risk

Quote: "Both TAC and LET solutions are aqueous and have the same blue tint and viscosity"

"labelled to ensure appropriate blindness of suture personnel"

"A double blind topical application using 3ml of the test solutions was performed in study entry"

Comment: Probably done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Both TAC and LET solutions are aqueous and have the same blue tint and viscosity. Unit-dose, amber vials of the anaesthetic solutions were assigned random numbers; labelled to ensure appropriate blindness of suture personnel; and stored under refrigeration in the ED. A double blind topical application using 3ml of the test solutions was performed in study entry"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
Low risk171 patients were initially enrolled, but data analysis for only 151 subjects. Five patients excluded after consent obtained (1 sedated prior to anaesthetic administration, 2 topical anaesthetics applied inappropriate duration, 2 data sheets lost). 15 additional patients were withdrawn prior to evaluation of anaesthetic effectiveness because subjects were either uncooperative or it was discovered that the wound involved deeper tissue layers then inclusion criteria permitted. We concluded low risk of bias, because reasons for exclusion unlikely to be related to pain scores during laceration repair.

Smith 1996

MethodsSingle-centre RCT
Participants240 patients, 2 to 17 years old, with lacerations 5 cm or less
located on the face (n=134), scalp (n=57) and extremity (n=49)
Interventions1. Bupivanor (BN) solution (0.48% bupivacaine with 1:26,000 norepinephrine), applied for 20 minutes (n=30)
2. Etidonor (EN) solution (0.95% etidocaine with 1:26,000 norepinephrine), applied for 20 minutes (n=30)
3. Mepivanor (MN) solution, (1.90% mepivacaine with 1:26,000 norepinephrine), applied for 20 minutes (n=30)
4. Prilonor (PN) solution, (3.81% prilocaine with 1:26,000 norepinephrine), applied for 20 minutes (n=30)
5. TAC solution (tetracaine 1.00%, epinephrine 1:4000, cocaine 4.0%), applied for 20 minutes (n=60)
6. Infiltrated lidocaine 1% (n=60)
Outcomes

1. Patients 5 years of age or older, reported discomfort using the VAS (100 mm) pain scale score.
2. Observer-reported VAS (100 mm) pain scale scores (suture technicians and research assistants).
3. Observer-reported Likert (1-7) pain scale scores (parents and suture technicians).
4. Oberver-rated (RICDRS) Restrained Infants and Children Disress Rating Scale (0 to 8) (research assistant and suture technician).
5. Suture technician-rated anaesthetic effectiveness scale.

Results (topical BN versus topical EN vs. topical MN vs. topical PN versus topical TAC versus infiltrated local anaesthetic):
(Standard deviations not reported for any of the outcomes)
1. Patient-reported VAS (100 mm) pain scores (mean scores: topical BN = 18.3 versus topical EN = 46.5 versus topical MN = 27.0 versus topical PN = 36.0 versus topical TAC = 12.0 versus infiltrated local anaesthetic = 26.3) (TAC significantly outperformed EN, P<0.05; no significant difference between any other groups).
2a. Suture technician-reported VAS (100 mm) pain scores (mean scores: topical BN = versus topical EN = versus topical MN = versus topical PN = versus topical TAC = versus infiltrated local anaesthetic = (EN significantly outperformed by BN, TAC and infiltrated anaesthetic, P<0.05; no significant difference between any other groups).
2b. Research assistant-reported VAS (100 mm) pain scores (mean scores: topical BN = versus topical EN = versus topical MN = versus topical PN = versus topical TAC = versus infiltrated local anaesthetic = ) (TAC outperformed both EN and PN, P<0.05; infiltrated anaesthetic outperformed both EN and PN, P<0.05; no significant difference between any other groups).
3a. Suture technician-reported Likert (1-7) pain scores (mean scores: topical BN = 2.05 versus topical EN = 2.6 versus topical MN = 2.4 versus topical PN = 2.1 versus topical TAC = 1.55 versus infiltrated local anaesthetic = 1.6 (TAC outperformed both EN and MN, P<0.05; infiltrated anaesthetic outperformed both EN and MN, P<0.05; no significant difference between any other groups)
3b. Parent-reported Likert (1-7) pain scores (mean scores: topical BN = 2.8 versus topical EN = 3.5 versus topical MN = 3.3 versus topical PN = 3.6 versus topical TAC = 2.11 versus infiltrated local anaesthetic = 2.33 (TAC outperformed EN, MN and PN, P<0.05; infiltrated anaesthetic outperformed EN and PN, P<0.05; no significant difference between any other groups).
4a. Suture technician-reported RICDRS (0-8) (mean scores: topical BN = 2.5 versus topical EN = 3.6 versus topical MN = 2.3 versus topical PN = 2.5 versus topical TAC = 1.4 versus infiltrated local anaesthetic = 1.63 (TAC outperformed EN, P<0.05; infiltrated anaesthetic outperformed EN, P<0.05; no significant difference between any other groups).
4b. Research assistant-reported RICDRS (0-8) (mean scores: topical BN = 2.4 versus topical EN = 3.1 versus topical MN = 2.7 versus topical PN = 2.9 versus topical TAC = 1.6 versus infiltrated local anaesthetic = 1.8 (TAC outperformed both EN and PN, P<0.05; infiltrated anaesthetic outperformed EN, P<0.05; no significant difference between any other groups).
5. Anaesthetic effectiveness scale (scores not reported) (TAC outperformed EN and MN, P<0.05; infiltrated anaesthetic outperformed BN, EN, MN, PN, P<0.05; no significant difference between any other groups).

NotesContacted author to request additional study data, author replied but unable to provide the lacking information. High risk of bias for local anaesthetic versus topical anaesthetic, as this comparison not blinded. However,unclear risk of bias in three domains for comparisons of different topical anaesthetics, because appropriately blinded.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "Study patients were assigned to one of six anaesthetic treatment groups using block randomization"

Comment: Unclear, as exact method of selecting the blocks is not reported

Allocation concealment (selection bias)Unclear riskComment: Unclear
Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Comparisons among the five topical preparations were double blinded. Because lidocaine was given as an injection, its identity was not blinded"; "Anesthetics were prepared in advance by Children's Hospital pharmacy, sealed in envelopes labelled with a study identification number, and stored in a locked cabinet in the emergency department"

Comment: Probably blinded between comparisons of different topical agents, but probably not blinded between infiltrated lidocaine and topical anaesthetic

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk240 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Smith 1997a

MethodsSingle-centre RCT
Participants71 patients, 2-16 years old, with lacerations, 5 cm or less in length located on the face (n=43) or scalp (n=28), 5 cm or less in length
Interventions1. Mepivanor (MN) solution, (mepivacaine 2%, norepinephrine 1:100,000), applied for 20 minutes (n=24)
2. TAC solution (tetracaine 1.0%, epinephrine 1:4000, cocaine 4.0%), applied for 20 minutes (n=24)
3. Intra-dermal infiltration with lidocaine 1% (n=23)
Outcomes

1. Observer-reported VAS (100 mm) pain scales scores (suture technicians, research assistants and video-tape reviewers).
2. Observer-reported Lickert (1-7) pain scale scores (parent, suture technicians).
3. Requirement for supplemental lidocaine infiltration

Results (topical MN vs. topical TAC vs. infiltrated local anaesthetic):
1a. Suture technician-reported VAS (100 mm) pain scores (mean score ± SD: topical MN = 7.1 ± 12.5 versus topical TAC = 2.0 ± 2.7 versus infiltrated anaesthetic = 1.8 ± 4.0). (Both topical TAC and infiltrated anaesthetic outperformed topical MN, P=0.003)
1b. Research assistant-reported VAS (100 mm) pain scores (mean score ± SD: topical MN = 14.8 ± 19.5 versus topical TAC = 4.7 ± 8.5 versus infiltrated anaesthetic = 3.0 ± 4.0). (Both topical TAC and infiltrated anaesthetic outperformed topical MN, P=0.0003)
1c. Video-tape reviewer-reported VAS (100 mm) pain scores (mean score ± SD: topical MN = 5.0 ± 12.5 versus topical TAC = 5.25 ± 16.42 versus infiltrated anaesthetic = 2.0 ± 5.9). (No reported difference between groups, P>0.05).
2a. Suture technician-reported Likert (1-7) pain scores (mean score ± SD: topical MN = 2.2 ± 1.4 versus topical TAC = 1.7 ± 0.9 versus infiltrated anaesthetic = 1.6 ± 1.0). (No reported difference between groups, P=0.18).
2b. Parent-reported Likert (1-7) pain scores (mean score ± SD: topical MN = 3.7 ± 1.9 versus topical TAC = 2.4 ± 1.8 versus infiltrated anaesthetic = 2.4 ± 1.6). (Both topical TAC and infiltrated anaesthetic outperformed topical MN, P=0.02)
3. Requirement supplemental lidocaine infiltration (topical MN = 37.5% versus topical TAC = 8.3%, P=0.04).

NotesObtained additional study data by directly contacting author.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "Enrolled patients were assigned to receive one of three anesthetic preparations by block randomization"

Comment: Unclear, as exact method of selecting the blocks not described in study

Allocation concealment (selection bias)Unclear riskComment: unclear
Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "Comparions between topical mepivonor and TAC were blinded to all observers. Since lidocaine was given as an injection, its identity was not blinded to those present for the procedure. However, after the anaesthetic was administered, suturing procedures were videotaped. These videotapes were later reviewed by an observer who was completely blinded to which local anaesthetic the patient had received"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk71 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Smith 1997b

MethodsSingle-centre RCT
Participants240 patients, 1-18 years, with lacerations less then or equal to 5 cm in length, located on the face (51%), scalp (30%), extremity (18%) or other site (1%)
Interventions1. Prilophen (PP) solution (prilocaine 3.56%, phenylephrine 0.99%), applied for 20 minutes, (n=60)
2. Tetraphen (TP) solution (tetracaine 1.0%, phenylephrine 5.0%), applied for 20 minutes, (n=60)
3. Tetralidophen (TLP) solution (tetracaine 1.0%, lidocaine 1.0%, phenylephrine 2.5%), applied for 20 minutes, (n=60)
4. TAC solution (tetracaine 1.0%, epinephrine 1:4000, cocaine 4.0%), applied for 20 minutes, (n=60)
Outcomes

1. Patients 5 years or older reported VAS (100 mm) pain scale scores.
2. Oberver-reported VAS (100 mm) pain scale scores (suture technicians, research assistants and parents).
3. Oberver-reported Likert (1-7) pain scale scores (suture technicians, research assistants and parents).
4. Suture technicians-rated the anaesthetic effectiveness (complete, partial or no anaesthesia).

Results (topical PP vs. topical TP vs. topical TLP vs. topical TAC):
1. Patient self-reported VAS (100 mm) pain scores (mean score ± SD: topical PP = 29.0 ± 43.4 versus topical TP = 24.2 ± 37.2 versus
topical TLP = 30.6 ± 40.3 versus topical TAC = 17.6 ± 34.1). (No reported difference between groups, P=0.5)
2a. Suture technician-rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 7.4 ± 16.0 versus topical TP = 5.1 ± 12.6 versus
topical TLP = 6.0 ± 13.5 versus topical TAC = 3.5 ± 11.8). (topical TAC performed significantly better then topical PP, reported P=0.04)
2b. Research assistant- rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 1.6 ± 2.6 versus topical TP = 1.9 ± 4.2 versus
topical TLP = 1.3 ± 1.7 versus topical TAC = 0.9 ± 1.7). (No reported difference between groups, P=0.09).
2c. Parent-rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 20.0 ± 21.7 versus topical TP = 20.2 ± 21.7 versus
topical TLP = 18.2 ± 18.6 versus topical TAC = 14.0 ± 18.6). (No reported difference between groups, P=0.09).
3a. Suture technician-reported Likert (1-7) pain scores (median score: topical PP = 2.0 versus topical TP = 1.0 versus topical TLP = 2.0 versus topical TAC = 1.0). (topical TAC performed significantly better then either topical PP or topical TLP, P=0.01)
3b. Research assistant-reported Likert (1-7) pain scores (median score: topical PP = 2.0 versus topical TP = 1.0 versus topical TLP = 2.0 versus topical TAC = 1.0). (topical TAC performed significantly better then either topical PP or topical TLP, P=0.03)
3c. Parent-reported Likert (1-7) pain scores (median score: topical PP = 2.0 versus topical TP = 2.0 versus topical TLP = 2.0 versus topical TAC = 2.0). (No reported difference between any of the groups, P=0.06)
4. Anaesthetic effectiveness (complete anaesthesia: topical PP = 63% versus topical TP = 67% versus topical TLP = 65% versus topical TAC = 80%). (No reported difference between any of the groups, P =.18).

NotesContacted author to request additional study data, author replied but unable to provide the lacking information.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "Patients were randomly assigned to one of four anesthetic treatment groups.."

Comment: Unclear as study reported to be randomized but method of sequence generation not described

Allocation concealment (selection bias)Unclear riskComment: Unclear
Blinding (performance bias and detection bias)
All outcomes
Unclear risk

Quote: "Using a prospective, randomized, double-blind design..."

"Anesthetic agents were sealed in envelopes labelled with a study identification number and stored in a locked cabinet in the emergency department".

Comment: Probably done, assuming topical solutions visually identical.

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk240 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Smith 1998a

MethodsSingle-centre RCT
Participants180 patients, age 1-18 years old, with lacerations 5cm or less, located on the face (n=76), scalp (n=59), extremity (n=43) or other (n=2)
Interventions1. Prilophen (PP) solution (3.56% prilocaine, 0.10% phenylephrine) applied for 20 minutes (n=60)
2. Bupivaphen (BP) solution (0.67% bupivacaine, 0.10% phenylephrine) applied for 20 minutes (n=60)
3. TAC solution (tetracaine 1.0%, epinephrine 1:4000, cocaine 4.0%) applied for 20 minutes (n=60)
Outcomes

1. Patients 5 years and older self-reported pain using a VAS (100 mm) scale
2. Observer-reported VAS (100 mm) pain scale scores (suture technicians, research assistants and parents)

Results (topical PP vs. topical BP vs. topical TAC):
1. Patient self-reported VAS (100 mm) pain scores (mean score ± SD: topical PP = 21.0 ± 28.0 versus topical BP = 41.0 ± 35.0 versus topical TAC = 18.0 ± 24.0). (No difference reported between groups, P=0.07)
2a. Suture technician-rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 3.8 ± 8.5 versus topical BP = 5.0 ± 9.0 versus topical TAC = 1.5 ± 3.0). (topical TAC outperformed topical BP, P=0.006; no difference between TAC and PP; no difference between BP and PP)
2b. Research assistant- rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 3.0 ± 6.0 versus topical BP = 3.8 ± 4.9 versus topical TAC = 1.4 ± 2.1). (topical TAC outperformed topical BP, P=0.002; no difference between TAC and PP; no difference between BP and PP)
2c. Parent-rated VAS (100 mm) pain scores (mean score ± SD: topical PP = 24.0 ± 24.5 versus topical BP = 29.0 ± 28.0 versus topical TAC = 17.0 ± 20.5). (TAC outperformed BP, P=0.03; no difference between TAC and PP; no difference between BP and PP)

NotesContacted author to request additional study data, author replied but unable to provide the lacking information.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk

Quote: "68 patients were assigned to each of the three anaesthetic treatment groups using block randomization"

Comment: Unclear, as exact method of selecting the blocks not reported

Allocation concealment (selection bias)Unclear riskComment: Unclear
Blinding (performance bias and detection bias)
All outcomes
Unclear risk

Quote: "Using a prospective, randomized, double-blind design..."

"Anesthetics were sealed in envelopes labelled with a study identification number and stored in a locked cabinet in the ED"

Comment: Probably done, assuming solutions visually identical

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk180 patients included in study, but insufficient reporting of attrition or exclusions to permit judgment

Vinci 1996

MethodsSingle-centre RCT
Participants156 patients, 3 to 18 years old with lacerations on the face/scalp (n= 102), extremity (n=47), trunk (n=7)
Interventions1. TAC 1 solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 11.8%) applied for 15-30 minutes, (n=49)
2. TAC 2 solution (tetracaine 1.0%, epinephrine 1:2000, cocaine 4.0%), applied for 15-30 minutes, (n=49)
3. TAC 3 solution (tetracaine 1.0%, cocaine 4.0%), applied for 15-30 minutes, (n=58)
Outcomes

1. Physician-rating of anaesthetic effectiveness (complete, partial or no anaesthesia). The anaesthesia was "complete" if the patient did not move, flinch or grimace during repair. The anaesthesia was "partial" if the patient complained of pain, moved or grimaced. If supplemental lidocaine infiltration was required then there was "no anaesthesia".
2. Requirement for a second application of topical anaesthetic.
3. Requirement for supplemental lidocaine infiltration.
4. The study reported acute adverse effects directly due to the anaesthetic.

Results for TAC 1 (standard formulation) vs. TAC 3 (tetracaine-cocaine):
1. Incidence of complete anaesthesia (topical TAC 1 = 73% versus topical TAC 3 = 28%, P<0.001).
2. Requirement for a second dose of topical anaesthetic (topical TAC 1 = 30% versus topical TAC 3 = 66%, P<0.003).
3. Requirement for supplemental lidocaine infiltration (topical TAC 1 = 6% versus topical TAC 3 = 9%, P=not reported).

Results for TAC 2 (higher concentration tetracaine, lower concentration cocaine) vs. TAC 3 (tetracaine-cocaine):
1. Incidence of complete anaesthesia (topical TAC 2 = 63% versus topical TAC 3 = 28%, P<0.001).
2. Requirement for a second dose of topical anaesthetic (topical TAC 2 = 46% versus topical TAC 3 = 66%, P<0.003).
3. Requirement for supplemental lidocaine infiltration (topical TAC 2 = 2% versus topical TAC 3 = 9%, P=not reported).
4. A single paediatric patient developed an erythematous rash one day after application of standard topical TAC.

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote: "The solutions were batched in lots of 10 doses to limit expiration of the study drugs. The order of batching was generated using a standard table of random numbers"

Comment: Probably done

Allocation concealment (selection bias)High risk

Quote: "The order of batching was generated using a standard table of random numbers"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
Unclear risk

Quote: "...we conducted a randomized, prospective, double-blind, clinical trial comparing three different formulations of cocaine-containing topical anaesthetics"

Unclear: In the introduction section reported to be a double-blind study, but no details provided in methods or any other sections.

Incomplete outcome data (attrition bias)
All outcomes
Low riskA total of 165 patients randomized in study and no missing outcome data or exclusions

White 1986

MethodsSingle-centre RCT
Participants68 adult patients, older then 18 years old with lacerations less then 5cm in length, located on the face (n=22), or non-facial (n=46)
Interventions1. TAC solution (tetracaine 0.5%, epinephrine 1:2000, cocaine 10.0%), applied for 5-10 minutes (n=36)
2. Tetracaine solution (tetracaine 0.5%), applied for 5-10 minutes (n=32)
Outcomes

1. Patient-rated numerical pain scale score (0-10).
2. Requirement of supplemental lidocaine infiltration.

Results:
1. Patient-rated numerical pain scale (0-10) score (mean pain scores: topical tetracaine = 5.6 versus topical TAC = 3.53, P<0.05, standard deviations not reported).
2. Requirement rescue lidocaine infiltration (topical tetracaine = 59% versus topical TAC = 36%, P=not reported).

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk

Quote: "Prior to delivery to the emergency department, the TAC and tetracaine solutions were assigned odd or even numbers"; "Randomization was achieved by matching the vials to the odd or even numbers at the end of the hospital number"

Comment: Probably not done

Allocation concealment (selection bias)High risk

Quote: "Randomization was achieved by matching the vials to the odd or even numbers at the end of the hospital number"

Comment: Probably not done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: " Only the pharmacist preparing the solutions knew which vials contained tetracaine and which contained TAC"

Comment: Probably done, assuming visually identical solutions

Incomplete outcome data (attrition bias)
All outcomes
Unclear risk68 patients participated in the study. It is not clear whether this was the same number that were randomized or whether there was any withdrawn.

Zempsky 1997

MethodsSingle-centre RCT
Participants32 patients, 5 to 18 years old, with lacerations less then 5 cm long, located on the extremity (n=32)
Interventions1. EMLA cream (lidocaine 2.5%, prilocaine 2.5%), applied for maximum 60 minutes (n=16)
2. TAC solution (formulation not reported by study), applied maximum of 30 minutes (n=16).
Outcomes

1. Patient-rated VAS (100 mm) pain scores.
2. Observer-rated VAS (100 mm) pain scores by the suturing physician and parent.
3. Requirement for supplemental lidocaine infiltration.

Results:
1. Patient-rated VAS (100 mm) pain scores (mean score ± SD: EMLA = 46.0 ± 26.0 versus topical TAC = 40.0 ± 25.0, P=0.50).
2. Parent-rated VAS (100 mm) pain scores (mean score ± SD: EMLA = 42.0 ± 15.0 versus topical TAC = 43.0 ± 25.0, P=1.0) and physician-rated VAS (100 mm) pain scores (mean score ± SD: EMLA = 30.0 ± 16.0 versus topical TAC = 26.0 ± 14.0, P=0.45).
3. Requirement for supplemental lidocaine infiltration (EMLA = 15% versus topical TAC = 55%, P=0.03).

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk

Quote: "...the patient was randomized into one of the two study groups by a table of random number"

Comment: Probably done

Allocation concealment (selection bias)High risk

Quote: "...the patient was randomized into one of the two study groups by a table of random number"

Comment: Porbably not done

Blinding (performance bias and detection bias)
All outcomes
Low risk

Quote: "The sutures, who were blinded to the patients' assignments, were not investigators in the study and were not allowed to see the patient until the anaesthetic had been removed and the wound irrigated"

Comment: Probably done

Incomplete outcome data (attrition bias)
All outcomes
Low riskA total of 32 patients enrolled and no dropouts or exclusions

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Adler 1998This study compared topical lidocaine-epinephrine-tetracaine (LET) only with placebo. There was no comparison with infiltrated local anaesthetics or other topical anaesthetics.
Adriansson 2004Topical xylocaine was not the primary anaesthetic for repair of the dermal injury. Instead the topical anaesthetic was only a pre-treatment prior to infiltration with local anaesthetic.
Bartfield 1995The topical agent was not the primary anaesthetic for repair of the dermal injury. The topical agent was only a pre-treatment prior to infiltration with local anaesthetic.
Bartfield 1996The topical agent was not the primary anaesthetic for repair of the dermal injury. The topical agent was only a pre-treatment prior to infiltration with local anaesthetic.
Bass 1990Not a randomized controlled trial. There are no controls and all patients received topical lignocaine-adrenaline-cocaine.
Bonadio 1988aNot a randomized controlled trial. There are no controls and all patients received topical TAC.
Bonadio 1988bNot a randomized controlled trial. There are no controls and all patients received topical TAC.
Bonadio 1992Not a randomized controlled trial. There are no controls and all patients received TAC gel.
Bonadio 1996The study evaluated patients with lacerations located on mucous membranes.
Chale 2006Compared local anaesthetic with digital anaesthesia. All laceration were pre-treated with topical anaesthetic but this was only too reduce pain from local anaesthetic infiltration. Topical anaesthesia was not used to reduce pain from repair of the lacerations
Chipont 2001Not a randomized controlled trial. There are no controls and all patients received topical LAT.
Liebelt 1997Not a randomized controlled trial. Instead this is a review article.
Little 2004Outcomes not relevant to this review.
Peirluisi 1989Not a randomized controlled trial, this is a retrospective study. Also the outcomes were not relevant to this review.
Priestley 2003Outcomes not relevant to this review.
Singer 2000The topical anaesthetic was only a pre-treatment prior to infiltration with local anaesthetic. Also, some of the wound closure was performed with adhesives.
Singer 2001The topical anaesthetic was only a pre-treatment prior to infiltration with local anaesthetic. Also, some of the wound closure was performed with adhesives.
Smith 1990Some patients (12) were sedated with chloral hydrate.
Smith 1998bThe study evaluated patients with lacerations located on mucous membranes.
Smith 1998cThe study evaluated patients with lacerations located on mucous membranes.
Spivey 1987Outcomes were not relevant to this review.
Stewart 1998The topical agent was not the primary anaesthetic for repair of the dermal injury. The topical agent was only a pre-treatment prior to lidocaine infiltration.
White 2004Not a randomized controlled trial. There are no controls and all patients received LAT gel.
Yamamoto 1997Outcomes were not relevant to this review.

Ancillary