Operative caries management in adults and children

  • Review
  • Intervention

Authors


Abstract

Background

The management of dental caries has traditionally involved removal of all soft demineralised dentine before a filling is placed. However, the benefits of complete caries removal have been questioned because of concerns about the possible adverse effects of removing all soft dentine from the tooth. Three groups of studies have also challenged the doctrine of complete caries removal by sealing caries into teeth using three different techniques. The first technique removes caries in stages over two visits some months apart, allowing the dental pulp time to lay down reparative dentine (the stepwise excavation technique). The second removes part of the dentinal caries and seals the residual caries into the tooth permanently (partial caries removal) and the third technique removes no dentinal caries prior to sealing or restoring (no dentinal caries removal). This is an update of a Cochrane review first published in 2006.

Objectives

To assess the effects of stepwise, partial or no dentinal caries removal compared with complete caries removal for the management of dentinal caries in previously unrestored primary and permanent teeth.

Search methods

The following electronic databases were searched: the Cochrane Oral Health Group's Trials Register (to 12 December 2012), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2012, Issue 11), MEDLINE via OVID (1946 to 12 December 2012) and EMBASE via OVID (1980 to 12 December 2012). There were no restrictions regarding language or date of publication.

Selection criteria

Parallel group and split-mouth randomised and quasi-randomised controlled trials comparing stepwise, partial or no dentinal caries removal with complete caries removal, in unrestored primary and permanent teeth were included.

Data collection and analysis

Three review authors extracted data independently and in triplicate and assessed risk of bias. Trial authors were contacted where possible for information. We used standard methodological procedures exacted by The Cochrane Collaboration.

Main results

In this updated review, four new trials were included bringing the total to eight trials with 934 participants and 1372 teeth. There were three comparisons: stepwise caries removal compared to complete one stage caries removal (four trials); partial caries removal compared to complete caries removal (three trials) and no dentinal caries removal compared to complete caries removal (two trials). (One three-arm trial compared complete caries removal to both stepwise and partial caries removal.) Four studies investigated primary teeth, three permanent teeth and one included both. All of the trials were assessed at high risk of bias, although the new trials showed evidence of attempts to minimise bias.

Stepwise caries removal resulted in a 56% reduction in incidence of pulp exposure (risk ratio (RR) 0.44, 95% confidence interval (CI) 0.33 to 0.60, P < 0.00001, I2 = 0%) compared to complete caries removal based on moderate quality evidence, with no heterogeneity. In these four studies, the mean incidence of pulp exposure was 34.7% in the complete caries removal group and 15.4% in the stepwise groups. There was also moderate quality evidence of no difference in the outcome of signs and symptoms of pulp disease (RR 0.78, 95% CI 0.39 to 1.58, P = 0.50, I2 = 0%).

Partial caries removal reduced incidence of pulp exposure by 77% compared to complete caries removal (RR 0.23, 95% CI 0.08 to 0.69, P = 0.009, I2 = 0%), also based on moderate quality evidence with no evidence of heterogeneity. In these two studies the mean incidence of pulp exposure was 21.9% in the complete caries removal groups and 5% in the partial caries removal groups. There was insufficient evidence to determine whether or not there was a difference in signs and symptoms of pulp disease (RR 0.27, 95% CI 0.05 to 1.60, P = 0.15, I2 = 0%, low quality evidence), or restoration failure (one study showing no difference and another study showing no failures in either group, very low quality evidence).

No dentinal caries removal was compared to complete caries removal in two very different studies. There was some moderate evidence of no difference between these techniques for the outcome of signs and symptoms of pulp disease and reduced risk of restoration failure favouring no dentinal caries removal, from one study, and no instances of pulp disease or restoration failure in either group from a second quasi-randomised study. Meta-analysis of these two studies was not performed due to substantial clinical differences between the studies.

Authors' conclusions

Stepwise and partial excavation reduced the incidence of pulp exposure in symptomless, vital, carious primary as well as permanent teeth. Therefore these techniques show clinical advantage over complete caries removal in the management of dentinal caries. There was no evidence of a difference in signs or symptoms of pulpal disease between stepwise excavation, and complete caries removal, and insufficient evidence to determine whether or not there was a difference in signs and symptoms of pulp disease between partial caries removal and complete caries removal. When partial caries removal was carried out there was also insufficient evidence to determine whether or not there is a difference in risk of restoration failure. The no dentinal caries removal studies investigating permanent teeth had a similar result with no difference in restoration failure. The other no dentinal caries removal study, which investigated primary teeth, showed a statistically significant difference in restoration failure favouring the intervention.

Due to the short term follow-up in most of the included studies and the high risk of bias, further high quality, long term clinical trials are still required to assess the most effective intervention. However, it should be noted that in studies of this nature, complete elimination of risk of bias may not necessarily be possible. Future research should also investigate patient centred outcomes.

Résumé scientifique

Prise en charge opératoire des caries chez l'adulte et l'enfant

Contexte

La prise en charge des caries dentaires consiste généralement à retirer intégralement la dentine molle déminéralisée avant de placer une obturation. Toutefois, les avantages liés au retrait complet d'une carie ont été remis en cause en raison d'inquiétudes concernant d'éventuels effets indésirables liés au retrait intégral de la dentine molle de la dent. Trois groupes d'études ont également examiné la doctrine consistant à retirer complètement une carie en la colmatant dans la dent à l'aide de trois techniques différentes. La première technique consiste à retirer progressivement la carie en deux visites à plusieurs mois d'intervalle, permettant ainsi à la pulpe dentaire de laisser reposer la dentine de réparation (technique d'excavation par étapes). La seconde technique retire une partie d'une carie dentinale et colmate définitivement la carie résiduelle dans la dent (retrait partiel d'une carie) et la troisième technique ne retire aucune carie dentinale avant de l'avoir colmatée ou restaurée (retrait d'aucune carie dentinale). Ceci est une mise à jour d'une revue Cochrane publiée pour la première fois en 2006.

Objectifs

Comparer les effets d'un retrait par étapes, partiel ou d'aucune carie dentinale par rapport au retrait complet d'une carie dans la prise en charge d'une carie dentinale sur des dents de lait et définitives n'ayant subi aucune restauration antérieure.

Stratégie de recherche documentaire

Les bases de données électroniques suivantes ont fait l'objet de recherches : le registre d'essais du groupe Cochrane sur la santé bucco-dentaire (jusqu'au 12 décembre 2012), le registre Cochrane des essais contrôlés (CENTRAL) (The Cochrane Library, 2012, numéro 11), MEDLINE via OVID (de 1946 au 12 décembre 2012) et EMBASE via OVID (de 1980 au 12 décembre 2012). Il n'y avait aucune restriction concernant la langue ou la date de publication.

Critères de sélection

Des essais contrôlés randomisés et quasi randomisés en groupes parallèles et en bouche fractionnée, comparant un retrait par étapes, partiel ou d'aucune carie dentinale au retrait complet d'une carie sur des dents de lait et définitives non restaurées, ont été inclus.

Recueil et analyse des données

Trois auteurs de la revue ont indépendamment extrait des données en triple et évalué les risques de biais. Les auteurs des essais ont été contactés, lorsque cela était possible, afin d'obtenir des informations complémentaires. Nous avons utilisé des procédures méthodologiques standard extraites par The Cochrane Collaboration.

Résultats principaux

Dans cette revue mise à jour, quatre nouveaux essais ont été inclus, augmentant ainsi le total des essais à huit, avec 934 participants et 1 372 dents. Trois comparaisons ont été effectuées : le retrait par étapes d'une carie et le retrait complet d'une carie en une seule étape (quatre essais) ; le retrait partiel d'une carie et le retrait complet d'une carie (trois essais) ; le retrait d'aucune carie dentinale et le retrait complet d'une carie (deux essais) (un essai divisé en trois bras comparait le retrait complet d'une carie au retrait par étapes et partiel d'une carie). Quatre études ont examiné des dents de lait, trois des dents définitives et une incluait les deux. Tous les essais ont été évalués comme présentant des risques de biais élevés, bien que les nouveaux essais montrent des preuves de tentatives de minimisation des biais.

Le retrait par étapes d'une carie diminuait de 56 % l'incidence d'une exposition de la pulpe (risque relatif (RR) 0,44, intervalle de confiance (IC) à 95 % 0,33 à 0,60, P < 0,00001, I2 = 0 %) par rapport au retrait complet d'une carie et se basait sur des preuves de qualité moyenne et l'absence d'hétérogénéité. Dans ces quatre essais, l'incidence moyenne d'une exposition de la pulpe était de 34,7 % dans le groupe de retrait complet d'une carie et de 15,4 % dans les groupes de retrait par étapes. Il y avait également des preuves de qualité moyenne montrant une absence de différence concernant les résultats liés aux signes et symptômes d'une maladie de la pulpe (RR 0,78, IC à 95 % 0,39 à 1,58, P = 0,50, I2 = 0 %).

Le retrait partiel d'une carie diminuait l'incidence d'une exposition de la pulpe de 77 % par rapport au retrait complet d'une carie (RR 0,23, IC à 95 % 0,08 à 0,69, P = 0,009, I2 = 0 %) et se basait également sur des preuves de qualité moyenne et une absence de preuves d'hétérogénéité. Dans ces deux études, l'incidence moyenne d'une exposition de la pulpe était de 21,9% dans les groupes de retrait complet d'une carie et de 5% dans les groupes de retrait par étapes d'une carie. Il y avait des preuves insuffisantes pour déterminer s'il existait une différence ou pas au niveau des signes et symptômes d'une maladie de la pulpe (RR 0,27, IC à 95 % 0,05 à 1,60, P = 0,15, I2 = 0 %, preuves de qualité médiocre) ou un échec de la restauration (une étude ne montrant aucune différence et une autre étude ne montrant aucun échec dans les groupes, preuves de qualité très médiocre).

Le retrait d'aucune carie dentinale était comparé au retrait complet d'une carie dans deux études très différentes. Il y avait des preuves de qualité moyenne ne montrant aucune différence entre ces techniques concernant les résultats liés aux signes et symptômes d'une maladie de la pulpe et une diminution des risques d'échec de la restauration favorisant le retrait d'aucune carie dentinale, d'après une première étude, et aucun cas de maladie de la pulpe ou d'échec de la restauration dans les groupes d'après une seconde étude quasi randomisée. Une méta-analyse de ces deux études n'a pas été réalisée en raison de différences cliniques significatives entre les études.

Conclusions des auteurs

Une excavation par étapes et partielle réduisait l'incidence d'une exposition de la pulpe sur des dents de lait et définitives cariées, vitales et asymptomatiques. Par conséquent, ces techniques montrent des avantages cliniques concernant le retrait complet d'une carie dans la prise en charge d'une carie dentinale. Il n'y avait aucune preuve d'une différence concernant les signes ou symptômes d'une maladie de la pulpe entre le retrait par étapes et le retrait complet d'une carie. De même, les preuves étaient insuffisantes pour déterminer s'il existait ou pas une différence concernant les signes et symptômes d'une maladie de la pulpe entre le retrait partiel d'une carie et le retrait complet d'une carie. Dans le cas du retrait partiel d'une carie, il y avait également des preuves insuffisantes pour déterminer s'il existait ou pas une différence concernant les risques d'échec d'une restauration. Les études relatives au retrait d'aucune carie dentinale examinant des dents définitives présentaient un résultat similaire et aucune différence au niveau des échecs de restauration. L'autre étude relative au retrait d'aucune carie dentinale, qui examinait des dents de lait, montrait une différence statistiquement significative au niveau des échecs de restauration favorisant l'intervention.

En raison d'un suivi à court terme réalisé dans la majorité des études incluses et des risques de biais élevés, d'autres essais cliniques à long terme et de qualité supérieure seront nécessaires afin d'évaluer l'intervention la plus efficace. Toutefois, notons que dans les études de ce type, l'élimination complète des risques de biais peut ne pas être systématiquement possible. D'autres recherches devront également examiner les résultats axés sur les patients.

Plain language summary

Techniques for managing decay in teeth

Tooth decay (dental caries) is a common problem around the world. It can cost a lot of money to treat and causes infection, pain and the loss of teeth. Tooth decay can be controlled by what are known as 'non-operative' methods which include cleaning plaque from teeth thoroughly, advising a healthy diet and using fluoride to prevent the decay getting worse. If the decay progresses these 'non-operative' techniques may need to be combined with 'operative' management which involves filling and restoring teeth where the holes caused by decay prevent cleaning. The fillings will improve the appearance of the teeth and allow the patient to clean them.  

Traditionally dentists have removed all of the decay with a dental drill or instruments before a filling is placed. However, removal of all the decay has some disadvantages, including damage to the nerve of the tooth, toothache and possibly weakening of the tooth structure. This method is known as one step complete caries removal. 

Despite the large number of fillings and restorations placed worldwide on a daily basis, dentists remain uncertain as to which is the best 'operative management' strategy for tooth decay. This review has been carried out by researchers from the Cochrane Oral Health Group to assess the most effective ways of treating and managing tooth decay (dental caries) when operative methods are used in first and permanent teeth.

The most recent search of relevant studies for this review was carried out on 12th December 2012. Eight studies with 934 patients (1372 teeth) were included.

Three alternative operative caries management interventions were assessed by comparing them with the traditional treatment of removing all the decay in one go (complete caries removal). These interventions were.

- Stepwise excavation - this technique removes caries in stages over two visits some months apart, allowing the dental pulp time to repair itself and lay down dentine.
- Partial caries removal - the dentist removes part of the dentinal caries and seals what is left into the tooth permanently.
- No dentinal caries removal - no dentinal caries is removed before sealing or restoring.

It was found that when the complete caries removal technique was compared with stepwise excavation, the pulp or nerve of the tooth would have been exposed in 347 of every 1000 teeth treated with complete caries removal, whereas when the stepwise excavation technique was used, this would have occurred in only 154 teeth per 1000. 

When the partial caries removal technique was used, it was found that the pulp or nerve would have been exposed in 50 teeth out of 1000 treated. However, when the complete caries removal technique was used this figure would have been 219 teeth per 1000 treated.

There was less nerve damage when part of, or all of the decay, was left behind, for both baby and adult teeth. There was no difference in the number of teeth with toothache with any of the techniques. One of the no dentinal caries removal techniques needed fewer replacement fillings, although there was no difference found when comparing any of the rest of the techniques to complete caries removal.

In the included studies, the fillings were mostly placed by specialist dentists and the teeth were followed up for a relatively short time (1 year). More studies are needed to help answer further questions. Future studies should be carried out by non-specialist dentists to check whether the results would be similar. These studies should follow the patients for a longer time, and check if there are any differences in toothache, further decay and filling replacements. They should also check which techniques patients prefer and if there is a long term difference in cost.

Résumé simplifié

Techniques de prise en charge des caries dentaires

La carie dentaire est un problème répandu à l'échelle mondiale. Son traitement peut se révéler onéreux et provoquer des infections, des douleurs et la perte de dents. Une carie dentaire peut être contrôlée grâce à des méthodes dites « non opératoires » qui consistent en un détartrage minutieux des dents, des conseils pour une alimentation saine et l'utilisation de fluoride pour éviter toute aggravation de la carie. En cas de progression d'une carie, ces techniques « non opératoires » peuvent être associées à une prise en charge « opératoire » qui consiste à combler et à restaurer la dent dans laquelle les trous, dus à la carie, empêchent tout nettoyage. Les obturations vont améliorer l'apparence de la dent et permettre au patient de la nettoyer.  

En général, les dentistes retirent l'intégralité d'une carie à l'aide d'une fraise ou d'instruments dentaires avant de placer une obturation. Toutefois, le retrait complet d'une carie présente certains inconvénients, notamment des lésions au niveau du nerf de la dent, des douleurs dentaires et éventuellement un affaiblissement de la structure dentaire. Cette méthode s'appelle « retrait complet d'une carie en une seule étape ».

Malgré le grand nombre d'obturations et de restaurations réalisées quotidiennement à l'échelle mondiale, les dentistes ignorent quelle est la stratégie la plus efficace de « prise en charge opératoire » d'une carie. La présente revue a été effectuée par des chercheurs du groupe Cochrane sur la santé bucco-dentaire afin d'évaluer la méthode la plus efficace de traitement et de prise en charge d'une carie lors de l'utilisation de méthodes opératoires sur des dents de lait et des dents définitives.

Les dernières recherches d'études pertinentes pour cette revue ont été réalisées le 12 décembre 2012. Huit études, composées de 934 patients (1 372 dents), ont été incluses.

Trois interventions de prise en charge opératoire alternative de caries ont été évaluées en les comparant à un traitement standard consistant à retirer l'intégralité d'une carie en une seule fois (retrait complet d'une carie). Ces interventions étaient les suivantes :

- Excavation par étapes - Cette technique permet de retirer progressivement une carie en deux visites à quelques mois d'intervalle, permettant ainsi à la pulpe dentaire de se régénérer et de laisser la dentine se reposer.
- Retrait partiel d'une carie - Le dentiste retire une partie de la carie dentaire et colmate de façon permanente ce qui reste dans la dent.
- Retrait d'aucune carie dentinaire - Aucune carie dentinaire n'est retirée avant son colmatage ou sa restauration.

Nous avons constaté que lorsque la technique de retrait complet d'une carie était comparée à une excavation par étapes, la pulpe ou le nerf de la dent était exposé(e) dans 347 dents sur 1 000 dont les caries ont été entièrement retirées, alors que si la technique d'excavation par étapes avait été utilisée, ce résultat se serait produit sur seulement 154 dents sur 1 000.

Lors de l'utilisation de la technique de retrait partiel d'une carie, nous avons constaté que la pulpe ou le nerf était exposé(e) dans 50 dents sur 1 000 ayant été traitées. Toutefois, si la technique de retrait complet d'une carie était utilisée, ce chiffre aurait été de 219 dents sur 1 000 traitées.

Il y avait moins de lésions nerveuses lorsque une partie, ou l'ensemble d'une carie, était laissée sur des dents de lait et d'adulte. Il n'y avait aucune différence concernant le nombre de douleurs dentaires avec l'une des techniques utilisées. L'une des techniques consistant à ne retirer aucune carie dentinale nécessitait moins d'obturations de remplacement, bien qu'il n'y ait aucune différence lorsque l'une des autres techniques était comparée au retrait complet d'une carie.

Dans les études incluses, les obturations étaient principalement placées par des dentistes spécialisés et les dents faisaient l'objet d'un suivi pendant une durée relativement courte (1 an). D'autres études seront nécessaires pour répondre à d'autres questions. Ces études devront être réalisées par des dentistes non spécialisés afin de vérifier si les résultats seraient similaires. Elles devront suivre les patients pendant une durée plus longue et vérifier s'il existe des différences au niveau des douleurs dentaires, de l'apparition d'autres caries et des obturations de remplacement. Elles devront également déterminer les techniques que les patients préfèrent et s'il existe une différence à long terme en termes de coûts.

Notes de traduction

Traduit par: French Cochrane Centre 22nd March, 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 pour la France: Minist�re en charge de la Sant�

Laički sažetak

Tehnike uklanjanja zubnog karijesa

Propadanje zuba (dentalni karijes) je problem koji se viđa u cijelom svijetu. Liječenje karijesa može biti skupo te uzrokovati infekciju, bol i gubitak zuba. Propadanje zuba može se liječiti ''ne-operativnim“ metodama koje uključuju temeljito čišćenje zubnih naslaga (plaka), savjetovanje o zdravoj prehrani i uporabu fluorida kako bi se spriječilo pogoršanja kvarova zubi. U slučaju pogoršanja kvara, te ''ne-operativne'' metode mogu zahtijevati kombiniranje s ''operativnim'' metodama koje uključuju ispune i popravak zuba u kojima oštećenja zuba onemogućuju propisnu higijenu. Ispuni će popraviti izgled zuba i omogućiti pacijentu njihovo lakše čišćenje.

Do sada su liječnici dentalne medicine sva oštećenja uklanjali dentalnom bušilicom ili instrumentima prije postavljanja ispuna. No, uklanjanje zubnog karijesa ima neke nedostatke, kao što su oštećenje zubnog živca, zubobolja i moguće slabljenje zubne strukture. Ta je metoda poznata kao potpuno uklanjanje karijesa u jednom koraku. 

Unatoč velikom broju ispuna i popravaka koji se svakodnevno obavljaju diljem svijeta, doktori dentalne medicine i dalje nisu sigurni koja je najbolja ''operativna'' metoda za liječenje pokvarenih zubi. Ovaj sustavni pregled proveli su istraživača iz Cochrane skupine za oralno zdravlje (engl. Cochrane Oral Health Group) kako bi procijenili najučinkovitije načine tretiranja i sanacije zubnoga kvara (zubnog karijesa) pri korištenju operativnih metoda liječenja mliječnih i trajnih zuba.

Zadnje pretraživanje studija za ovaj sustavni pregled provedeno je 12. prosinca 2012. U sustavni pregled uključeno je 8 studija s 934 pacijenta (1.372 zuba).

Analizirane su tri različite intervencije za operativno uklanjanje karijesa i napravljena je usporedba s tradicionalnim tretmanom uklanjanja cjelokupnog kvara u jednom potezu (potpuno uklanjanje karijesa). Analizirani su sljedeći zahvati.

-Postupna ekskavacija – ta tehnika uklanja karijes postupno, u fazama tijekom dva posjeta u razmaku od nekoliko mjeseci, čime se daje mekom tkivu u unutrašnjosti zuba (zubnoj pulpi) vrijeme za oporavak i stvaranje novog sloja tkiva dentina.
-Djelomično uklanjanje karijesa prilikom kojeg doktor dentalne medicine uklanja dio zubnoga karijesa i trajno pečati ono što je ostalo u zubu.
-Nedentinsko uklanjanje karijesa – nedentinski karijes se uklanja prije pečaćenja ili obnove zuba.

Utvrđeno je da bi tehnikom potpunog uklanjanja karijesa s postupnom ekskavacijom pulpa ili zubni živac bili izloženi u 347 od svakih 1000 zuba, dok bi se korištenjem tehnike postupne ekskavacije to dogodilo u samo 154 zuba od 1000. 

Kada je korištena tehnika djelomičnog uklanjanja karijesa, utvrđeno je da bi pulpa ili živac bili izloženi u 50 od 1000 tretiranih zuba. Međutim, kada je korištena tehnika potpunog uklanjanja karijesa ta brojka bi bila 219 od 1000 tretiranih zuba.

Manje oštećenja živaca uočeno je kad je ostavljen dio ili cijeli karijes i kod djece i odraslih Nije bilo razlike u broju zuba sa zuboboljom ni kod jedne tehnike. Jedna od tehnika nedentinskog uklanjanja karijesa je bila povezana s manjim brojem zamjene ispuna (plombi), iako nije pronađena razlika u usporedbi bilo koje od ostalih tehnika s potpunim uklanjanjem karijesa.

U uključenim studijama, ispune su uglavnom postavljali specijalisti dentalne medicine te je stanje zuba praćeno razmjerno kratko vrijeme (1 godina). Potrebno je više studija da bi se dobio odgovor na daljnja pitanja. Buduća istraživanja bi trebala ispitati i postupke koje ne provode liječnici specijalisti dentalne medicine da bi se provjerilo dobivaju li se slični rezultati. Ta istraživanja bi trebala pratiti pacijente kroz dulje vrijeme te provjeriti postoje li ikakve razlike u zubobolji, daljnjem oštećenju zuba i učestalosti zamjene ispuna Također bi trebali provjeriti koje tehnike pacijenti bolje prihvaćaju i postoji li dugoročno razlika u cijeni.

Bilješke prijevoda

Hrvatski Cochrane
Prevela: Laura Vrdoljak
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

Ringkasan bahasa mudah

Teknik untuk mengendali kerosakan gigi

Kerosakan gigi (karies gigi) adalah masalah yang lazim di seluruh dunia. Ia melibatkan kos untuk dirawat dan boleh menyebabkan jangkitan, sakit dan kehilangan gigi. Kerosakan gigi boleh dikawal dengan kaedah 'bukan-operatif' yang melibatkan pembersihan plak gigi dengan teliti, nasihat diet yang sihat dan penggunaan fluorida untuk mencegah kerosakan lebih teruk. Jika proses kerosakan berlarutan teknik 'bukan operatif' ini mungkin perlu digabung dengan pengendalian 'operatif' yang melibatkan tampalan dan restorasi gigi yang berlubang akibat kerosakan yang menghalang daripada dicuci. Tampalan akan memperbaiki rupa bentuk gigi dan membolehkan pesakit membersihkannya. 

Secara tradisi, doktor gigi akan membersihkan kerosakan dengan alat gerudi pergigian sebelum bahan tampalan diletakkan. Walau bagaimanapun, pembuangan bahagian gigi yang rosak mempunyai beberapa keburukan termasuk kerosakan saraf gigi, sakit gigi dan mungkin juga melemahkan struktur gigi. Kaedah ini dipanggil pembuangan karies sempurna dalam satu langkah. 

Walaupun banyak tampalan dan restorasi dibuat di seluruh dunia setiap hari, doktor gigi masih tidak pasti strategi 'pengendalian operatif' terbaik untuk kerosakan gigi. Ulasan yang dibuat oleh penyelidik-penyelidik Kumpulan Kesihatan Oral Cochrane ini bertujuan menilai cara yang paling berkesan untuk merawat dan mengendali kerosakan gigi (karies gigi) apabila kaedah operatif diguna dalam gigi susu dan gigi kekal.

Carian terkini untuk kajian-kajian yang relevan bagi ulasan ini dibuat pada 12hb Disember 2012. Lapan kajian dengan 934 pesakit (1372 gigi) dimasukkan.

Tiga intervensi pengendalian karies alternatif dinilai dengan membandingkan rawatan tradisional membuang semua kerosakan gigi dalam satu langkah (pembuangan karies sempurna) Intervensi tersebut adalah

- Ekskavasi berperingkat - teknik membuang karies berperingkat dalam dua lawatan selang beberapa bulan, membolehkan pulpa gigi sembuh dan membentuk lapisan dentin.
- Pembuangan karies separa - doktor gigi membuang sebahagian dentin berkaries dan memengap dengan kekal selebihnya dentin berkaries dalam gigi.
- Tiada pembuangan dentin berkaries - tiada dentin berkaries dibuang sebelum memengap atau merestorasi.

Apabila teknik pembuangan karies sempurna dibanding dengan ekskavasi berperingkat, 347 dari setiap 1000 pulpa atau saraf gigi mungkin terdedah dengan teknik pembuangan karies sempurna sedangkan bagi penggunaan teknik ekskavasi berperingkat, pendedahan hanya berlaku kepada 154 dari 1000 gigi. 

Apabila teknik pembuangan karies separa diguna, didapati 50 daripada 1000 pulpa atau saraf gigi mungkin terdedah. Walau bagaimanapun, apabila teknik pembuangan karies sempurna diguna, angka ini mungkin berubah menjadi 219 setiap 1000 gigi yang dirawat.

Terdapat kurang kerosakan apabila sebahagian atau semua kerosakan ditinggalkan bagi gigi susu dan gigi kekal. Tiada perbezaan bilangan gigi yang sakit dengan sebarang teknik-teknik tersebut. Salah satu teknik tanpa pembuangan karies memerlukan kurang penggantian tampalan, walaupun tiada sebarang perbezaan didapati apabila membandingkan teknik-teknik lain untuk melengkapkan pembuangan karies

Dalam kajian-kajian tersebut, kebanyakan tampalan diletakkan oleh doktor pakar pergigian dan gigi-gigi tersebut dipantau dengan singkat (1 tahun) Lebih banyak mkajian diperlukan untuk membantu menjawab soalan lanjutan. Sebilangan kecil kajian mungkin boleh dibuat oleh doktor gigi yang bukan pakar untuk menentukan sama ada keputusan adalah sama. Kajian-kajian ini sepatutnya menyusul pesakit untuk jangkamasa lebih lama untuk menentukan jika ada sebarang perbezaan sakit gigi, kerosakan lanjutan dan penggantian tampalan. Ia juga patut menentukan teknik yang dipilih oleh pesakit dan jika ada perbezaan kos dalam jangkamasa panjang.

Catatan terjemahan

Diterjemahkan oleh Noorliza Mastura Ismail (Kolej Perubatan Melaka-Manipal). Disunting oleh Mohd. Shaharudin Shah Che Hamzah (Universiti Sains Malaysia). Untuk sebarang pertanyaan berkaitan terjemahan ini sila hubungi noorliza.mastura@manipal.edu.my

Summary of findings(Explanation)

Summary of findings for the main comparison. Stepwise excavation versus complete one stage caries removal prior to restoration of carious teeth
  1. 1 Four trials at high risk of bias. Assumed risk is the mean risk in the control groups of these four studies.
    2 Two trials at high risk of bias. Assumed risk is the mean risk in the control groups of these two studies.
    3 Two trials at high risk of bias. Assumed risk is the mean risk in the control groups of these two studies.

Stepwise excavation versus complete one stage caries removal prior to restoration of carious teeth
Patient or population: Patients with carious teeth
Settings: Secondary dental care
Intervention: Stepwise excavation
Comparison: One stage complete caries removal
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
One stage complete caries removalStepwise excavation
Pulp exposure during caries removal347 per 1000154 per 1000
(114 to 208)
RR 0.44
(0.33 to 0.6)
627
(4 studies)
⊕⊕⊕⊝
moderate 1
 
Pulp exposure during caries removal - Primary dentition407 per 1000126 per 1000
(69 to 232)
RR 0.31
(0.17 to 0.57)
173
(2 studies)
⊕⊕⊕⊝
moderate 2
 
Pulp exposure during caries removal - Secondary dentition325 per 1000166 per 1000
(117 to 234)
RR 0.51
(0.36 to 0.72)
454
(3 studies)
⊕⊕⊕⊝
moderate 2
 
Signs or symptoms of pulpal disease (1 year)
Follow-up: mean 1 year
101 per 100079 per 1000
(39 to 159)
RR 0.78
(0.39 to 1.58)
278
(2 studies)
⊕⊕⊕⊝
moderate 3
 
*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.

Summary of findings 2 Partial caries removal compared to complete caries removal for patients with carious teeth

Summary of findings 2. Partial caries removal compared to complete caries removal for patients with carious teeth
  1. 1 Two small trials at high risk of bias. Assumed risk is the mean risk in the control groups of these two studies.
    2 Three small trials with wide confidence interval around the estimate indicating imprecision.

    3 One very small study at high risk of bias.This estimate is very imprecise. Assumed risk is the risk in the control group.

Partial caries removal compared to complete caries removal for patients with carious teeth
Patient or population: Patients with carious teeth
Settings: Secondary dental care
Intervention: Partial caries removal
Comparison: Complete caries removal
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Complete caries removalPartial caries removal
Pulp exposure during caries removal219 per 100050 per 1000
(18 to 151)
RR 0.23
(0.08 to 0.69)
141
(2 studies)
⊕⊕⊕⊝
moderate 1
 
Signs or symptoms of pulpal disease (1 year)
Follow-up: mean 1 year
47 per 100013 per 1000
(2 to 75)
RR 0.27
(0.05 to 1.60)
172
(3 studies)
⊕⊕⊝⊝
low 1,2
 
Failure of restorations62 per 1000

21 per 1000

(1 to 476)

RR 0.33 (0.01 to 7.62)

32

(1 study)

⊕⊝⊝⊝

very low 3

 
*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.

Summary of findings 3 No dentinal caries removal versus complete caries removal prior to restoration

Summary of findings 3. No dentinal caries removal versus complete caries removal prior to restoration
  1. 1 One split-mouth randomised controlled trial, at high risk of bias, conducted in primary care, comparing preformed metal crowns with a variety of restoration methods for primary teeth. Finding has not been independently replicated.

    2 One additional quasi-randomised study at high risk of bias could not be included in the meta-analysis due to clinical heterogeneity. (Conducted in secondary care, different degree of caries, restoration of permanent teeth with composite materials.)

No dentinal caries removal versus complete caries removal prior to restoration
Patient or population: Patients with carious teeth
Settings: General dental practice and secondary dental care
Intervention: No dentinal caries removal
Comparison: Complete caries removal
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Complete caries removalNo dentinal caries removal
Signs or symptoms of pulpal disease (1 year)
Follow-up: mean 1 year
50 per 1000

10 per 1000

(1- 86 per 1000)

RR 0.20
(0.02 to 1.72)
132
(1 study)
⊕⊕⊕⊝
moderate 1
Additional quasi-randomised study showed no events in either group.2
Failure of restorations370 per 1000

30 per 1000

(7 - 96 per 1000)

RR 0.08 (0.02 to 0.26)

132

(1 study)

⊕⊕⊕⊝

moderate 1

Additional quasi-randomised study showed 1 event in both groups.2
*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

Description of the condition

Dental caries is one of the most commonly occurring diseases worldwide and its treatment has high cost implications both in monetary and biological (dental pain/infection and tooth loss) terms. Non-operative measures (plaque and diet control, fluoride application) are considered important treatments to control caries progression. However, operative dentistry (placement of restorations) has a role to play in facilitating plaque control and in restoring tooth form and function. Cavitated lesions that cannot be cleaned are restored to allow the patient to clean effectively.

This operative treatment of dental caries traditionally has involved complete removal of the caries before placing a restoration. This tissue is heavily infected with bacteria and is removed using a slow speed bur or hand excavating instruments (Kidd 1998). Histologically and clinically, dentinal caries has been characterised as having two distinct layers: the outer zone (stains with caries detector dyes) where the dentine is highly demineralised, the collagen denatured and heavily infected with bacteria (often referred to as the infected zone), and the inner zone (does not stain with a caries detector dye) where the dentine is demineralised but the collagen intact and minimally infected (often referred to as the caries affected zone) (Fusayama 1972).

Traditionally all remnants of caries were removed during cavity preparation (Black GV 1908). However, contemporary cavity preparation clears caries from the periphery of the cavity and only removes the outer caries infected zone pulpally.This type of treatment has been accepted and practiced for generations by dentists. However, there are a number of consequences associated with such restorative intervention.

  • Entry to the restorative cycle: Whilst caries removal and restoration give the short term benefit of removing the infected, soft demineralised dentine and restoring tooth form, they cannot be regarded as a definitive solution as the majority of restorations placed by dentists are replacement restorations (Elderton 1990). Individual teeth often undergo re-restoration a number of times due to restorative material failure, new caries adjacent to the restoration and tooth structure failure as a result of weakening.

  • Reduction in remaining dentine thickness (RDT): When a cavity is prepared and a restoration is replaced, the cavity becomes larger and the RDT between the floor of the cavity and the dental pulp often becomes thinner. The RDT has been shown to be the most critical variable in cavity preparation which has an impact on pulpal health (Murray 2003).

  • Pulpal exposure: At its most extreme, caries removal in dentine lesions can lead to exposure of vital pulp tissue. Traditionally, such exposures in symptomless teeth have been managed by placement of a direct pulp cap or with a pulpotomy. These techniques can achieve good success rates when they are used to treat pulps exposed through dental trauma (Cvek 1978). However, following carious exposures (where there is infected dentine and the pulp is more likely to be compromised) outcomes are poor (Al-Hiyasat 2006; Barthel 2000).

  • Dentine is a vital tissue, and local anaesthetic (LA) is required for conventional cavity preparation, as it often involves removal of non-necrotic dentine. Studies involving children have found less reported discomfort and pain when softened demineralised dentine was removed with hand excavation without LA than when LA and burs were used (Rahimtoola 2000; van Bochove 2006).

More conservative approaches often known as minimal intervention techniques for the management of caries have been adopted and are becoming more widely accepted in efforts to address and reduce the adverse consequences of restorative treatment. The interpretation of what minimal intervention means varies, but often consists of what would be regarded as complete caries removal. It is uncertain whether the traditional approach to caries removal is necessary and this dogma has been challenged by a number of procedures and studies (Kidd 2004) where partial or no caries removal was carried out (Thompson 2008).

Description of the intervention

There is a lack of standardisation in the terminology describing the various degrees of caries removal, but those that move away from complete caries removal can be broadly grouped as follows.

  1. Stepwise excavation: Partial caries removal is carried out and provisionally restored, followed a few weeks later by removal of the provisional restoration and any soft caries. A definitive restoration then being placed (Bjørndal 1997).

  2. Partial caries removal and placement of a definitive restoration (Ribeiro 1999).

  3. No dentinal caries removal and use of a restorative material to seal caries into the tooth (Handelman 1991; Mertz-Fairhurst 1998). This group has a more diverse range of procedures which includes placement of traditional fissure sealants in the pits and fissures and novel techniques such as preformed metal crowns (Innes 2007) and those sealing proximal lesions using resin sealants (Martignon 2006; Paris 2007).

How the intervention might work

Sealing infected demineralised dentine into a cavity with a restoration that provides a good peripheral seal, deprives the microorganisms of substrate from the oral cavity. The bacteria reduce in numbers (Handelman 1976) and the caries process arrests. Not only do the bacteria reduce in numbers, but the microbial diversity becomes less complex. Only those microorganisms capable of breaking down pulpal tissue fluid glycoproteins are able to survive (Paddick 2005). A number of other clinical studies discussed by Fejerskov and Kidd 2008 and Thompson 2008 have investigated and provided evidence to support this theory.

The progressive reduction in the number of organisms and a change to a less cariogenic microflora within sealed carious dentine, leads to a gradual reduction in lesion activity and hence lesion progression. This allows time for the pulp-dentine complex to lay down tertiary dentine and peri-tubular dentine leading to tubular sclerosis and reduces the permeability of the remaining dentine. This reduction in pulpal exudate further depletes the nutrient source for the bacteria. In stepwise excavation the provisional restoration is removed after a period of time to allow further caries removal. The tertiary dentine which has now had time to form provides further protection of the dental pulp and reduces the risk of pulpal exposure. Avoidance of carious pulpal exposures is critical to the long term outcome of the tooth, as management of such exposures using a direct pulp capping technique is associated with a poor prognosis for maintaining a vital pulp (Al-Hiyasat 2006; Barthel 2000). Extirpation of the damaged pulp and root canal treatment would then be required.

Partial caries or no dentinal caries removal techniques also have the potential to reduce cavity size and hence preserve tooth structure. However, a consequence of such techniques is that the restoration does not have a sound foundation. The impact of this upon restoration longevity is still debated, although it may not be such an important issue in the long term for primary teeth as they exfoliate.

Why it is important to do this review

Despite the high prevalence of dental caries and the large number of restorations placed worldwide on a daily basis, clinicians remain uncertain as to which is the best operative caries management strategy: complete, partial or no dentinal caries removal. At the present time there is considerable variation in clinical practice and teaching. Therefore there is a need to systematically review the literature on stepwise, partial, no dentinal or complete caries removal prior to definitive restoration. A review of this literature was published in 2006. Since publication, further evidence has become available and there is a need to update the results.

This is an update of the Cochrane review originally entitled Complete or ultraconservative removal of decayed tissue in unfilled teeth previously published in 2006 (Ricketts 2006). It has been re-titled to encompass all studies which remove varying amounts of caries, including those that do not remove any dentinal caries.

Objectives

Primary objectives

To assess the effects of stepwise, partial or no dentinal caries removal compared to complete caries removal in both primary and permanent teeth. 

Specifically, the primary outcomes were: exposure of the dental pulp during caries removal, signs or symptoms of pulpal disease, progression of caries, and restoration failure. The secondary outcomes were: health economic measures, oral health related quality of life, patient/carer and dentist perceptions of treatment, and patient discomfort during treatment.

Secondary objectives

To investigate factors such as depth of lesion, surface(s) affected, extent of caries removal, method of caries removal, restorative material used and effect of primary or secondary teeth.

Methods

Criteria for considering studies for this review

Types of studies

Parallel group and split-mouth randomised controlled trials (RCTs), including quasi-randomised trials, that compared stepwise, partial or no dentinal caries removal with complete caries removal, prior to restoration.

Types of participants

Participants with caries, affecting any tooth surface(s), in unrestored primary and permanent teeth.

Types of interventions

Stepwise, partial or no dentinal caries removal prior to restoration. The control groups involved complete caries removal. To avoid including dental pulps compromised by previous treatment only teeth with no previous restorations were considered.

Types of outcome measures

Primary outcomes
  • Exposure of the dental pulp during caries removal.

  • Signs or symptoms of pulpal disease.

  • Progression of caries.

  • Restoration failure.

Secondary outcomes
  • Health economic measures.

  • Oral health related quality of life.

  • Patient/carer and dentist perceptions of treatment.

  • Patient discomfort during treatment.

Search methods for identification of studies

Electronic searches

For the identification of studies included or considered for this review, we developed detailed search strategies for each database searched. These were based on the search strategy developed for MEDLINE (OVID) but revised appropriately for each database. The search strategy used a combination of controlled vocabulary and free text terms and was linked with the Cochrane Highly Sensitive Search Strategy (CHSSS) for identifying randomised trials (RCTs) in MEDLINE: sensitivity maximising version (2008 revision) as referenced in Chapter 6.4.11.1 and detailed in box 6.4.c of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) (Higgins 2011).

We searched the following electronic databases:

  • The Cochrane Oral Health Group's Trials Register (to 12 December 2012) (Appendix 1)

  • The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2012, Issue 11) (Appendix 2)

  • MEDLINE via OVID (1948 to 12 December 2012) (Appendix 3)

  • EMBASE (1980 to 12 December 2012) (Appendix 4).

Details of the MEDLINE search are provided in Appendix 3. The search of EMBASE was linked to the Cochrane Oral Health Group filter for identifying RCTs in EMBASE (Appendix 4).

Searching other resources

Handsearching for this review was done as part of the Cochrane worldwide handsearching programme (Cochrane Masterlist contains details of journals searched to date). All relevant trials found by this handsearching were retrieved by the search of the Cochrane Oral Health Group's Trials Register.

The search was not limited by language and any non-English published trials considered to be relevant were translated.

First authors of included studies were contacted where possible for unpublished data. Reference lists of included studies were screened for further trials.

Data collection and analysis

Selection of studies

The titles and abstracts of all reports identified through the electronic searches were scanned independently by three review authors (David Ricketts (DR), Nicola Innes (NI) and Thomas Lamont (TL)) for eligibility. For studies appearing to meet the inclusion criteria, or for which there were insufficient data in the title and abstract to make a clear decision, the full report was obtained. These full reports were assessed independently by the same three authors to establish whether the studies met the inclusion criteria or not. Agreements were reached through discussion and where resolution was not possible a fourth review author (Janet Clarkson (JC)) was consulted. Studies rejected at this or subsequent stages were recorded in the Characteristics of excluded studies table and the reason for exclusion recorded.

Data extraction and management

All studies meeting the inclusion criteria underwent a risk of bias assessment and data extraction was carried out using a specially designed data extraction form. Review authors were not masked to the journal title or authors. Data were extracted by three authors (DR, NI and TL) independently and in triplicate. One of the included studies was conducted by an author of this review (NI). For this study the other two review authors extracted the data (DR and TL) in duplicate.

For each trial the following data were recorded:

  • the year of publication, country of origin and source of study funding;

  • details of the participants including demographic characteristics, criteria for inclusion and exclusion and withdrawals;

  • details of the type of intervention, timing and duration; and

  • details of the outcomes reported including methods of assessment.

The primary outcome measures were: exposure of the dental pulp during caries removal, signs or symptoms of pulpal disease, progression of caries and restoration failure. Other reported outcomes were recorded for descriptive purposes.

The number of studies eligible for inclusion were recorded (Characteristics of included studies). Data were analysed using Review Manager (RevMan) software (RevMan 2011) and reported according to Cochrane Collaboration criteria.

Assessment of risk of bias in included studies

Assessment of risk of bias in included studies was conducted by three review authors (DR, NI and TL) using the Cochrane risk of bias assessment tool (Higgins 2011). Any disagreements were resolved by discussion. One of the included studies was conducted by an author of this review (NI). For this study the other two review authors (DR and TL) assessed the risk of bias. Seven domains were assessed for each included study: sequence generation, allocation concealment, masking of participants and personnel, masking of outcome assessment, incomplete outcome data, selective outcome reporting and 'other bias'. Within each domain, a description of what happened, as reported in the study, was recorded along with a judgement of either 'low', 'high' or 'unclear' risk of bias. An overall risk of bias assessment was also made.

After taking into account any additional information provided by the authors of the trials, studies were grouped into the following categories.

  • Low risk of bias (plausible bias unlikely to seriously alter the results).

  • Unclear risk of bias where one or more of the domains was assessed as unclear.

  • High risk of bias (plausible bias that weakens confidence in the results) where one or more domains was assessed as at high risk of bias.

Measures of treatment effect

For dichotomous outcomes, the estimate of effect of an intervention was expressed as risk ratios (RR) together with 95% confidence intervals (CIs). For any possible continuous outcomes, mean differences and standard deviations were used to summarise the data for each group using mean differences and 95% CIs.

Unit of analysis issues

The units of randomisation and analysis in the included trials would ideally all have been at the level of the individual. When split-mouth studies were included and each individual trial participant had one tooth randomised to intervention and another randomised to control, analyses took into account the paired nature of the data. In trials where the unit of randomisation was the tooth and the number of teeth included in the trial was not more than twice the number of participants, the data were treated as if the unit of randomisation was the individual. It was recognised that the resulting 95% confidence intervals produced would appear narrower (i.e. the estimate will seem to be more precise) than they should be, and would therefore be interpreted accordingly.

Dealing with missing data

Attempts were made to retrieve missing data from authors of trials. Although there were no continuous data for this version of the review, methods for estimating missing standard deviations in section 7.7.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) will be used in future updates of this review.The techniques described by Follmann (Follmann 1992) were to be used to estimate the standard error of the difference for split-mouth studies, where the appropriate data were not presented and could not be obtained.

Assessment of heterogeneity

The significance of any discrepancies in the estimates of the treatment effects from the different trials was assessed by means of Cochran's test for heterogeneity and heterogeneity would have been considered significant if P < 0.1.

The I2 statistic, which describes the percentage total variation across studies that is due to heterogeneity rather than chance, was used to quantify heterogeneity with I2 over 50% being considered moderate to high heterogeneity.

Data synthesis

Meta-analyses were carried out only where there were studies of similar comparisons reporting the same outcome measures. Risk ratios were combined for dichotomous data, and mean differences would have been used for continuous data, using random-effects models provided there were more than three studies in the meta-analysis. Data from split-mouth studies were combined with data from parallel group trials using the method outlined by Elbourne (Elbourne 2002), using the generic inverse variance method in RevMan.

Subgroup analysis and investigation of heterogeneity

Clinical heterogeneity was assessed by examining the types of participants and interventions for all outcomes in each study. It was decided not to formulate any hypotheses to be investigated for subgroup analyses since no significant meta-analysis was expected. However, this may be carried out in future updates of this review.

Sensitivity analysis

It was planned to undertake sensitivity analyses to examine the effect of the risk of bias for each study on the overall estimates of effect. If there were insufficient trials to undertake this, the effect of including unpublished literature on the review's findings would also have been examined.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification.

Results of the search

The electronic searches identified 690 titles and abstracts, four additional studies were identified through searching the reference lists from published reviews and from this, we obtained 30 full reports. The titles and abstracts were screened by three review authors (David Ricketts (DR), Nicola Innes (NI) and Thomas Lamont (TL)) to assess whether the inclusion criteria were met. Full reports of papers that appeared to meet the inclusion criteria were obtained and from these, eight trials were included (Bjørndal 2010; Innes 2007; Leksell 1996; Lula 2009; Magnusson 1977; Mertz-Fairhurst 1987; Orhan 2010; Ribeiro 1999). All of these studies were randomised controlled trials.

The eight trials included in the review involved 934 participants with 1372 teeth treated and outcome data for 1191 teeth were available for analysis.

Included studies

Characteristics of the trial, setting and investigators

One study was carried out in America (Mertz-Fairhurst 1987), two in Brazil (Lula 2009; Ribeiro 1999), two in Sweden (Leksell 1996; Magnusson 1977), one in Scotland (Innes 2007), one in Turkey (Orhan 2010) and one multicentre trial in Sweden and Denmark (Bjørndal 2010). The trials were published in eight reports between 1977 and 2010. Funding for these studies was varied and included government and pharmaceutical sources, although for some the funding remained unclear.

Three studies reported no drop-outs and the remaining five studies reported 3% to 19% of teeth lost to follow-up at 1 year.

Two of the studies were of split-mouth design (Innes 2007; Mertz-Fairhurst 1987) and six were parallel group studies (Bjørndal 2010; Leksell 1996; Lula 2009; Magnusson 1977; Orhan 2010; Ribeiro 1999). Most studies compared two different treatments but one study (Orhan 2010) compared three different treatments.

The unit of randomisation varied within the parallel group studies from patient level (Magnusson 1977) to tooth level (Bjørndal 2010; Leksell 1996; Lula 2009; Orhan 2010; Ribeiro 1999). In studies where more than one tooth per participant was included, no cluster analyses were performed.

Three secondary care based studies (Leksell 1996; Magnusson 1977; Orhan 2010) reported the number of operators (n = 1-6) and treatment centres (n = 1-6). Innes 2007 reported 17 operators (general dental practitioners) in 10 primary care dental practices. In Bjørndal 2010 there were six treatment centres and in Mertz-Fairhurst 1987 and Lula 2009 one treatment centre, but none reported the number of operators although all three implied that more than one operator was involved. No information was provided on operator or number of treatment centres for the remaining study (Ribeiro 1999).

Characteristics of the participants

Four of the trials included only primary teeth (Innes 2007; Lula 2009; Magnusson 1977; Ribeiro 1999) where participants' ages ranged from 3 to 11 years. Three trials were carried out on permanent teeth (Bjørndal 2010; Leksell 1996; Mertz-Fairhurst 1987) and the ages of the subjects ranged from 6 to 89 years. One study included both primary and permanent teeth (Orhan 2010) and the patients ranged from age 4 to 15 years.

Characteristics of carious lesions

Information on the depth of lesions included in the studies varied from no information (Ribeiro 1999 (lesion width given only as "at least 2 mm wide (faciolingually)") to lesions extending radiographically up to the outer half of dentine (Mertz-Fairhurst 1987), lesions extending radiographically into the pulpal half of dentine (Innes 2007; Lula 2009) or the pulpal quarter of dentine (Bjørndal 2010; Leksell 1996 (pulpal exposure expected); Magnusson 1977; Orhan 2010). The different methods of reporting caries lesion depth estimation, reflects the current lack of an accepted standardised measure. 

Whilst the study by Mertz-Fairhurst 1987 only investigated occlusal (Class I) lesions, Innes 2007; Lula 2009; Orhan 2010 and Ribeiro 1999 looked at both occlusal and occlusal-approximal (Class II) lesions, and Bjørndal 2010 looked predominantly (96%) at occlusal-approximal (Class II) lesions. The illustrations, examples and descriptions presented by Magnusson 1977 and Leksell 1996 would suggest that only occlusal lesions were investigated, however, this cannot be confirmed from the text.

Characteristics of the intervention

For all studies, the intervention was compared with complete caries removal as the control. Three studies investigated stepwise excavation (Bjørndal 2010; Leksell 1996; Magnusson 1977). One study (Orhan 2010) reported both stepwise excavation and partial caries removal compared to complete caries removal. In the stepwise excavation studies, the time interval between first stage excavation and second stage varied between 4 and 24 weeks. Two further studies compared partial caries removal and all three of these trials (Lula 2009; Orhan 2010; Ribeiro 1999) sealed the caries into the tooth definitively. 

Two studies did not remove any dentinal caries (Innes 2007; Mertz-Fairhurst 1987) prior to placing a crown or a restoration respectively.

What authors termed partial caries removal at the first stage of the stepwise excavation technique varied between studies. Magnusson 1977 and Orhan 2010 removed caries until the operator determined that there was a significant risk of pulpal exposure with further excavation, whereas Bjørndal 2010 appears to be more conservative and described only "removal of the superficial necrotic and demineralized dentin with complete excavation of the peripheral demineralized dentin, avoiding excavation close to the pulp". In both of the studies investigating partial caries removal, with no re-entry (Lula 2009; Ribeiro 1999), only the enamel-dentine junction was cleared of caries, and superficial necrotic dentine was removed from the pulpal and axial walls of the cavity. In the two studies that did not make any attempt to remove dentinal caries, one was in permanent teeth and involved bevelling the enamel at the entrance of pits and fissures prior to restoration with composite resin (Mertz-Fairhurst 1987). The second study investigated primary teeth and restoration was with a preformed metal crown using the Hall Technique (Innes 2007).

In the control group of the Innes study, dentists were asked to carry out their routine management of carious lesions; 76% reported complete caries removal and 24% partial caries removal. The data presented in this review were for the complete caries removal control group and were obtained from personal communication with the first author.

The restorative materials varied throughout the included studies: the provisional restorative materials used in the stepwise excavation technique were zinc oxide-eugenol (Leksell 1996; Magnusson 1977; Orhan 2010) and glass ionomer (Bjørndal 2010), whereas the definitive restorations provided for the other techniques were composite (Lula 2009; Mertz-Fairhurst 1987; Ribeiro 1999) and preformed metal crowns (Innes 2007).

Characteristics of the outcome measures

The primary outcomes were: exposure of the dental pulp during caries removal, signs or symptoms of pulpal disease, progression of caries and restoration failure. Five studies measured exposure of the dental pulp during caries removal (Bjørndal 2010; Leksell 1996; Lula 2009; Magnusson 1977; Orhan 2010). Seven studies reported signs or symptoms of pulpal disease at 1 year follow-up (Bjørndal 2010; Innes 2007; Lula 2009; Magnusson 1977; Mertz-Fairhurst 1987; Orhan 2010; Ribeiro 1999). None of the studies reported upon progression of caries. Restoration failure was presented in six studies (Innes 2007; Leksell 1996; Lula 2009; Mertz-Fairhurst 1987; Orhan 2010; Ribeiro 1999) with this being the major focus for two of the studies (Mertz-Fairhurst 1987; Ribeiro 1999). The study by Innes reported outcome data at 2 years (Innes 2007) and 5-year data are available (Innes 2007). To improve consistency with the other studies, the author provided 1 year follow-up data, including results of the exposure of dental pulp during caries removal.

For the secondary outcomes of health economic measures, oral health related quality of life, patient/carer and dentist perceptions of treatment, and patient discomfort during treatment, only one study (Innes 2007) reported on perceptions of treatment and patient discomfort.

Excluded studies

Fifteen trials were excluded for the following reasons: no control, the atraumatic restorative treatment (ART) technique was compared to conventional caries removal where the ART technique was judged to constitute complete caries removal.

Risk of bias in included studies

A summary of the risk of bias is presented in Figure 1.

Figure 1.

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

Allocation

Sequence generation

Reporting on sequence generation varied between the studies. Two studies provided no information on the method of sequence generation and were assessed at unclear risk of selection bias (Leksell 1996; Ribeiro 1999). Allocation was based on date of birth in Magnusson 1977, so this quasi-randomised study was assessed at high risk of bias. Four studies were assessed at low risk of selection bias based on the reported method of sequence generation: paper based "randomised treatment assignment sheet" (Mertz-Fairhurst 1987) to computerised sequence generation (Bjørndal 2010; Innes 2007; Lula 2009).

In Orhan 2010, teeth were allocated to treatment by the investigator drawing lots. However, in this study randomisation was carried out in one stage for both the control and intervention groups. At this stage the intervention group was divided into two subgroups, stepwise and partial caries removal. This meant that when the operator was partially removing caries, it was known whether there would be a subsequent appointment to remove further caries (the stepwise group). The authors commented that this may have introduced bias because the operator knew there was to be a second caries removal stage, and this might have affected the extent of caries excavation they were carrying out. The authors point out there should have been a second randomisation, following the initial caries removal to divide the group into two, and decide which teeth would be allocated to stepwise or partial caries removal. Orhan 2010 was therefore assessed as having a high risk of selection bias.

Allocation concealment

The concealment of allocation was unclear for four studies (Leksell 1996; Mertz-Fairhurst 1987; Orhan 2010; Ribeiro 1999). Bjørndal 2010; Innes 2007 and Lula 2009 allocated at the time of treatment via a central randomisation sequence, which was concealed from the operators. Magnusson 1977 allocated on the basis of date of birth which was not concealed from the operators and this trial was therefore judged to have a high risk of selection bias.

Blinding

Masking of participants and personnel (performance bias)

Steps were taken in three studies to mask participants to which procedures they had received, or which tooth had received which treatment (Bjørndal 2010; Lula 2009; Ribeiro 1999) by using the same dental material for both the intervention and control groups, and standardising the number of surgical stages for each procedure from the patient perspective. These studies were assessed as having a low risk of bias for masking of participants due to the steps taken in the study design and management which reduced the potential risk of bias.

In the remaining five included studies, masking of participants was not reported and these were assessed as at high risk of performance bias. For two studies, patients could not be masked as different materials were used for the intervention and control groups (Innes 2007; Mertz-Fairhurst 1987). In the remaining three studies, patients had to attend one or two appointments depending on whether they had received the control or intervention (Leksell 1996; Magnusson 1977; Orhan 2010).

As with most dental restorative intervention studies, masking of the operator was not possible in any of the studies included in this review; the intervention and control procedures involved different treatment protocols. All the included studies were therefore assessed as being at high risk of performance bias with regards to masking of personnel providing the trial interventions.

Masking of outcome assessment (detection bias)

For the five studies which investigated the primary outcome of pulp exposure (Bjørndal 2010; Leksell 1996; Lula 2009; Magnusson 1977;Orhan 2010), this outcome was assessed at the time of caries removal by the operator, who knew which treatment they were carrying out. These studies were therefore assessed as having a high risk of detection bias.

Innes 2007 and Mertz-Fairhurst 1987 were also assessed as having a high risk of bias for outcome assessment, as they used different restorative materials and therefore the outcome assessor could not be masked. Ribeiro 1999 was assessed as having a low risk of bias as the same restorative materials and review protocol were used for both intervention and control groups and adequate masking was therefore likely to have been in place.

Where radiographs and microbiology were used to assess outcome measures, the outcome assessors were masked to the intervention. However, where partial caries removal was compared to complete caries removal, the radiologists may have been able to determine which intervention group the tooth was in, which may have introduced an element of bias.

Incomplete outcome data

Leksell 1996 and Mertz-Fairhurst 1987 reported 40% and 19% drop-outs at 1 year respectively; neither provided adequate explanation for their drop-outs and were therefore assessed as having a high risk of attrition bias. Whilst Bjørndal 2010 also failed to provide adequate explanations of the drop-outs from this study, it was considered unlikely to introduce a risk of attrition bias due to the small numbers involved (7%), and therefore risk of attrition bias is assessed as low.

The remaining five studies were also assessed as having a low risk of attrition bias: three reported no drop-outs (Magnusson 1977; Orhan 2010; Ribeiro 1999) and the remaining two had few drop-outs and provided adequate information on those lost to follow-up (Innes 2007; Lula 2009).

Selective reporting

All of the included studies reported in full the outcomes listed in the methods sections. There was no suggestion of selective outcome reporting and therefore all studies were assessed as being at low risk of reporting bias.

Other potential sources of bias

Bjørndal 2010, Leksell 1996 and Orhan 2010 excluded teeth from further outcome assessment once exposure of the dental pulp had occurred. Excluding these teeth from further outcome assessment is a potential source of bias, resulting in potential under-reporting of the other outcomes. This may have a bearing on the complete caries removal groups where there was a higher incidence of pulpal exposure. These trials were therefore assessed as having a high risk of other bias.

There was no evidence of other sources of bias in the remaining studies.

Overall risk of bias

Although the authors of this review acknowledge that the included studies would be classified at high risk of bias based on the assessments above, this has to be taken in the context of the pragmatic nature of dental clinical trials. To design and conduct a study free of all biases would however be impossible for the primary outcome of exposure of the dental pulp, as the outcome assessor would know the allocated caries removal technique after cavity preparation was completed, by virtue of the status of the remaining dentine. Whilst the overall risk of bias of the included studies may be regarded as high, the four new studies included in this update (Bjørndal 2010; Innes 2007; Lula 2009; Orhan 2010) have demonstrated improved methodologies in an attempt to address these shortcomings.

Effects of interventions

See: Summary of findings for the main comparison Stepwise excavation versus complete one stage caries removal prior to restoration of carious teeth; Summary of findings 2 Partial caries removal compared to complete caries removal for patients with carious teeth; Summary of findings 3 No dentinal caries removal versus complete caries removal prior to restoration

See Summary of findings for the main comparison; Summary of findings 2 and Summary of findings 3.

Primary outcomes

Most of the included studies compared complete caries removal prior to restoration with either stepwise, partial or no removal of dentinal caries. Therefore data for stepwise excavation, partial caries removal and no dentinal caries removal were analysed as separate groups. The studies reported outcome data for the primary outcomes of exposure of the dental pulp during caries removal, signs or symptoms of pulpal disease and restoration failure. No studies reported on the primary outcome of progression of caries.

Comparison 1: Stepwise excavation versus complete caries removal

There were four trials comparing stepwise excavation with complete caries removal (Bjørndal 2010; Leksell 1996; Magnusson 1977; Orhan 2010). Orhan 2010 included both primary and permanent teeth in their investigation, Bjørndal 2010 and Leksell 1996 permanent teeth only and Magnusson 1977 primary teeth only.

Exposure of the dental pulp during caries removal (Analysis 1.1)

All four studies reported outcome data for dental pulp exposure. Exposures of the dental pulp at any stage of caries removal (either initial or in the case of stepwise, on re-entry) have been included in the meta-analysis. Data were provided for both primary and permanent teeth. Stepwise excavation was associated with a reduced risk of pulp exposure risk ratio (RR) 0.44 (95% confidence interval (CI) 0.33 to 0.60, P < 0.00001, I2 = 0%) (Analysis 1.1), a 56% reduction compared to complete caries removal, with no evidence of heterogeneity.

For primary teeth, the risk ratio for pulpal exposure during stepwise excavation was RR 0.31 (95% CI 0.17 to 0.57, P = 0.0002, I2 = 0%) and for permanent teeth the RR was 0.51 (95% CI 0.36 to 0.72, P = 0.0002, I2 = 0%) a 69% and 49% reduction in risk respectively compared with complete caries removal. However, the quality of the evidence is graded as moderate due to the risk of bias (Summary of findings for the main comparison).

A breakdown of when these exposures occurred is shown in the table below.

Pulp exposures in stepwise groups
 First stage of caries removalSecond stage of caries removal
 Pulp exposuresTreated%Pulp exposuresTreated%
Bjørndal 201041432.82113915
Leksell 19960640105717.5
Magnusson 1977055085514.5
Orhan 201004904498.2
TOTAL43111.34330014.3
Signs or symptoms of pulpal disease (Analysis 1.2)

Signs or symptoms of pulpal disease were reported by Bjørndal 2010; Leksell 1996 and Orhan 2010. Two trials have been included in the meta-analysis, one included primary teeth and permanent teeth (Orhan 2010) and the other (Bjørndal 2010) included permanent teeth only. From the meta-analysis, there is no evidence of a difference in the development of signs or symptoms of pulpal disease (RR 0.78 (95% CI 0.39 to 1.58, P = 0.50, I2 = 0%) between stepwise and complete caries removal (Analysis 1.2). In another study with 80 participants which evaluated this comparison (Leksell 1996), there were no signs or symptoms of pulpal disease in either group. There is moderate quality evidence of no difference in the outcome of signs and symptoms of pulpal disease (Summary of findings for the main comparison).

In the Magnusson 1977 study, pulp necrosis was reported as occurring in three teeth in the complete caries removal arm and one in the stepwise excavation arm. However, it is not clear from the publication whether this was a finding at excavation or on review, and therefore these data were not included in the analyses on signs and symptoms of pulpal pathology.

Comparison 2: Partial caries removal versus complete caries removal

Three trials compared partial caries removal with complete caries removal (Lula 2009; Orhan 2010 and Ribeiro 1999). Orhan 2010 included both primary and permanent teeth in their investigation, whilst Lula 2009 and Ribeiro 1999 only included primary teeth.

Exposure of the dental pulp during caries removal (Analysis 2.1)

Two studies provided data for meta-analysis (Lula 2009; Orhan 2010) for this outcome. For primary teeth, there was a reduction in risk favouring partial caries removal (RR 0.24 (95% CI 0.06 to 0.90; P = 0.03, I2 = 0%). Only one study provided data for permanent teeth, and this study showed no difference between partial and complete caries removal (Analysis 2.1).

Overall there was a reduction in risk of exposure of dental pulp favouring partial caries removal (RR 0.23 (95% CI 0.08 to 0.69, P = 0.009, I2 = 0%), a 77% reduction in risk of pulp exposure compared to complete caries removal (Analysis 2.1). This is based on moderate quality evidence (Summary of findings 2).

Signs or symptoms of pulpal disease (Analysis 2.2)

All three trials provided data for the outcome of signs or symptoms of pulpal disease. There was no difference in pulpal disease between partial and complete caries removal (RR 0.27, 95% CI 0.05 to 1.60, P = 0.15, I2 = 0% (Analysis 2.2). However, this is based on three small trials at high risk of bias, and there is considerable imprecision in the estimate, assessed as low quality evidence (Summary of findings 2).

Failure of restorations (Analysis 2.3)

Lula 2009 and Ribeiro 1999 also reported the outcome of restoration failure. There was only one failure of restoration recorded, which occurred in the complete caries removal group in Lula 2009, and no restoration failures were reported in either group in the study by Ribeiro 1999. These small studies with a combined total of 80 participants provide insufficient evidence to determine whether or not there is a difference in restoration failure between partial and complete caries removal (Summary of findings 2).

Comparison 3: No dentinal caries removal versus complete caries removal

Two trials compared no dentinal caries removal with complete caries removal (Innes 2007 and Mertz-Fairhurst 1987). They were both split-mouth designs, one with primary teeth where preformed metal crowns were used to seal the occlusal and approximal caries (Innes 2007), and one with permanent teeth which used composite materials to seal occlusal caries (Mertz-Fairhurst 1987). Both studies reported the outcomes of signs or symptoms of pulpal disease and restoration failure.

Exposure of the dental pulp during caries removal could not occur in the intervention groups where no dentinal caries removal was carried out. It was therefore inappropriate to report this outcome in these studies, but we note that no exposure of the dental pulp occurred in either complete caries removal group in either study.

Note that in the Innes 2007 study only the control teeth that actually underwent complete caries removal were included in the analysis throughout the review. These data were provided following communication with the author (Additional Table 1).

Table 1. One year outcome data from Innes 2007
Signs/Symptoms
 No caries removal
Complete caries removal yesno
 yes05
 no194
Restoration failure
  No caries removal
  yesno
Complete caries removalyes037
 no360
Signs or symptoms of pulpal disease (Analysis 3.1)

After 1 year of follow-up, there was no difference in signs or symptoms of pulpal disease in the primary teeth in the study by Innes 2007 (Analysis 3.1; Additional Table 1) (data supplied by author Additional Table 1).

In the study by Mertz-Fairhurst 1987, there were no signs or symptoms of pulpal disease reported for either group.

Failure of restorations (Analysis 3.2)

For primary teeth, Innes 2007 showed a marked reduction in risk of restoration failure favouring no dentinal caries removal (Analysis 3.2) (data supplied by the author Additional Table 1).

The Mertz-Fairhurst 1987 study with permanent teeth showed no difference between the groups in restoration failure.

It was not considered appropriate to combine these studies in any analysis because of their clinical heterogeneity (i.e. primary/permanent teeth, settings of studies, different surfaces of initial caries, different restoration types).

Secondary outcomes

Innes 2007 was the only study that reported on secondary outcomes of patient, carer and dentist perception of treatment and patient discomfort during treatment. In this study 77% of the children, 83% of the carers and 81% of the dentists preferred the no dentinal caries removal and restoration with preformed metal crowns compared to complete caries removal and conventional restoration. In relation to pain as assessed by the dentist, 89% of children were assessed as experiencing "no pain, discomfort" to "mild, not significant" during the intervention, compared to 78% in the control group.

Discussion

Summary of main results

This review compares different caries removal techniques with complete caries removal. Eight trials met the inclusion criteria, 934 patients with 1372 teeth were recruited into the trials and outcome data for 1191 teeth were available for analysis. The trials were divided into three different comparisons, according to the extent of caries removal. Three primary outcomes were reported, with none of the trials providing data for the progression of caries at 1 year.

Comparison 1: Stepwise excavation versus complete caries removal

Four trials reported stepwise excavation versus complete caries removal. The analyses show that there was a 56% reduction in risk of exposure of the dental pulp during stepwise excavation when compared with complete caries removal risk ratio (RR) 0.44 (95% confidence interval (CI) 0.33 to 0.60, P < 0.00001, I2 = 0%). This is a large effect based on moderate quality evidence (Summary of findings for the main comparison).

There was no evidence of a difference in signs or symptoms of pulpal disease when comparing stepwise excavation to complete caries removal. However, in the three studies included (Bjørndal 2010; Leksell 1996; Orhan 2010), once exposure had occurred, the teeth were excluded from the study. As these teeth were not evaluated at follow-up, the effect estimate may be biased towards the control which had a greater incidence of exposure of the dental pulp. This may have resulted in an underestimation of the potential benefit of the intervention.

The studies in this review demonstrate that partial caries removal and sealing into the tooth (first stage stepwise) leads to the systematic and progressive arrest of the carious lesion. This allows time for pulp dentine complex reactions, reducing the risk of pulpal exposure when cavities are re-entered at the second visit of stepwise to remove the remaining demineralised tissue. Indeed, with increasing evidence from studies presenting microbiological data and the randomised controlled trials in this systematic review, some clinicians and researchers question whether re-entry is required in the stepwise excavation procedure. With this in mind, when the first stage of stepwise excavation caries removal was carried out there was only four (1.3%) exposures of the dental pulp compared to 43 (14.3%) exposures at second stage of stepwise.

Comparison 2: Partial caries removal versus complete caries removal

Comparison 2 included three trials (Lula 2009; Orhan 2010 and Ribeiro 1999) which compared partial caries removal to complete caries removal. Partial caries removal resulted in a 77% reduction in the risk of exposure of the dental pulp during caries removal when compared with complete caries removal (RR 0.23, 95% CI 0.08 to 0.69, P = 0.009, I2 = 0%). This large effect is also based on moderate quality evidence (Summary of findings 2).

All three trials reported signs and symptoms of pulpal disease, and two provided data for meta-analysis, which showed no evidence of a difference (RR 0.27, 95% CI 0.05 to 1.60; P = 0.15, I2 = 0%). The small sample size was reflected in the wide confidence intervals, with Lula 2009 reporting one case of pulpal necrosis in the complete removal group (Summary of findings 2).

There was insufficient evidence to determine whether there was a difference with regards to restoration failure. In Lula 2009 there was one restoration failure in the complete caries removal group and there were no restoration failures in either group in the study by Ribeiro 1999 (Summary of findings 2).

Comparison 3: No dentinal caries removal versus complete caries removal

Comparison 3 included two studies (Innes 2007; Mertz-Fairhurst 1987) which compared no dentinal caries removal with complete caries removal. Exposure of the dental pulp during caries removal was not possible in the intervention groups where no dentinal caries removal was carried out. However, it is interesting to note that no exposures of the dental pulp occurred in the control groups of these studies either. This result might be expected in the Mertz-Fairhurst 1987 study as the included teeth had caries radiographically confined to the outer half of dentine. However in the Innes 2007 study, almost half (42%) of the included teeth demonstrated caries radiographically into the inner half of dentine and some exposures would therefore have been expected in the control group. There are at least two possible explanations for this lack of pulp exposures during caries removal.  Firstly, the participating dentists may have modified their management due to their very involvement in a clinical trial, the so called 'Hawthorne effect' (Fernald 2012). Secondly, although these dentists were assumed to be representative of general practitioners, they volunteered to take part in this trial of a highly conservative approach to caries removal and may have had a general tendency to avoid radical caries removal where there was a risk of exposing the dental pulp.

Innes 2007 reported 3% restoration failure rate in the intervention group, whereas the control group reached 37%. However, the restorations placed in the control group in this study of primary teeth were mainly multi-surface glass ionomer restorations, which have been shown to have poor survival clinically (Chadwick 2007). This failure rate in the control arm was much higher than has been reported from secondary care/specialist practice studies of complete caries removal and restoration. This pragmatic trial presents results achieved in a primary care setting where the majority of dental care is usually provided. Innes 2007 was the only trial that reported on any of the secondary outcomes, namely patient, carer and dentists perceptions of treatment and patient discomfort (assessed by the dentist). In this study the intervention group was preferred by the patient, carer and dentist, and resulted in less discomfort during treatment (Summary of findings 3).

This review only considers the 1-year follow-up data, however, the 5-year data of this trial, have recently been published (Innes 2007) and also show low restoration failure rates in the intervention group (Hall crowns) (Summary of findings 3).

Overall completeness and applicability of evidence

Completeness

Although eight studies have been included in this review, there are only a small number available for each intervention group. There were no data available for the primary outcome of progression of caries, and only one study (Innes 2007) reported upon any of the secondary outcomes.

Applicability

Seven of the studies were carried out in a secondary care environment by specialist dentists, even though the majority of dental treatment is provided in primary care. This poses two questions: are the patients treated in primary care similar to those seen in secondary care, and is the care provided and outcomes achieved in both settings comparable. Some issues related to these questions may include aspects of cost, time allocation, appointments required, funding source (e.g. dental insurance company). Therefore, clinicians will need to assess whether the findings are applicable to the patients in their clinical setting.

Most of the included studies investigated teeth in children and adolescents. Therefore the applicability to patients of all ages has to be considered. Only two of the included studies involved older patients (Bjørndal 2010; Mertz-Fairhurst 1987).

Early work in this field concentrated on occlusal caries where a good seal could readily be obtained to sound enamel, but there was skepticism as to whether the techniques could be applied to approximal lesions. More recent research has included approximal lesions in addition to occlusal lesions (Innes 2007; Lula 2009; Orhan 2010; Ribeiro 1999) and one study has looked almost exclusively at approximal lesions (Bjørndal 2010). Unfortunately, subgroup analyses based on surface could not be carried out as the studies did not always report the outcomes separately.

Quality of the evidence

The quality of evidence provided from a number of the studies included in this review is poor due to their high risk of bias. Ideally in a clinical trial the patient, the operator and the assessor should all be masked as to any treatment/intervention. However, this conventional masking was not always possible in these studies. The operator knew whether caries removal was partial or complete and certainly knew whether an exposure was present. The operator was often the assessor and it must be assumed that this also precluded conventional masking. However, a primary outcome was symptoms of pulpitis or pulp necrosis and it seems unlikely that this would have been affected by a lack of assessor masking. In the Mertz-Fairhurst 1987 and Innes 2007 studies, different restorative materials were used in the intervention and control groups, therefore masking was not possible.

The more recent studies have improved methodologies, with respect to sequence generation, allocation concealment and attempts to mask patients to the treatment received. Using the same restorative materials and re-entering the teeth where complete caries removal had taken place, Bjørndal 2010 may have ensured patients did not know which treatment arm was being carried out. As re-entry took place in Lula 2009, in order to take a microbiological sample, the patients may also have been unaware of which treatment was being carried out.

Although eight studies have been included, they have explored three alternative strategies resulting in a small number of studies in individual analyses. In addition to this, a number of the studies had small sample sizes resulting in wide confidence intervals and lower level of certitude.

One of the sources of heterogeneity within the studies was the depth of the lesions. Variation in lesion depth may have an impact on whether the pulp is exposed during caries removal. The depth of the lesions in the studies varied in radiographic extent, from Mertz-Fairhurst 1987 where all lesions were confined to the outer half of dentine to Bjørndal 2010 and Orhan 2010 where all lesions were over 75% of the way through dentine. Only Bjørndal 2010 included teeth with pre-treatment pain and there were no radiographic signs of periapical or periradicular pathology in any of the studies. The medicaments and materials used in the studies varied widely from zinc oxide-eugenol through standard restorative materials to preformed metal crowns. There were insufficient data to draw conclusions about the respective benefits of the various materials used in the intervention groups.

Potential biases in the review process

A potential source of bias in this review process was that one of the review authors (Nicola Innes) carried out one of the included studies. To decrease the risk of bias in the review process, this author was not involved in the assessment of risk of bias, data extraction, data analysis and interpretation for this study.

Agreements and disagreements with other studies or reviews

This systematic review which has only randomised controlled trials has found similar encouraging results to two review papers (Hayashi 2011; Thompson 2008) that have also included clinical trials.

Authors' conclusions

Implications for practice

The results from the included studies should be interpreted in light of the fact that a number of the studies suffer from a high risk of bias and had a short follow-up period. Whilst bearing this in mind the current evidence demonstrates that in symptomless, vital, carious primary or permanent teeth, stepwise and partial excavation reduced the risk of pulp exposure. Therefore these techniques show clinical advantage over complete caries removal in the management of dentinal caries. Whilst there is insufficient evidence to determine whether it is necessary to re-enter and excavate further in the stepwise excavation technique, the studies that did not re-enter, reported no adverse consequences (Innes 2007; Lula 2009; Mertz-Fairhurst 1987; Ribeiro 1999). In primary teeth, the use of Hall Technique crowns would make re-entry difficult and the study by Innes 2007 demonstrated that this may not be necessary: the restoration having adequate inherent retention in the longer term.

Implications for research

This review found no evidence that incomplete caries removal is harmful. In fact, the reverse is true as complete caries removal is more likely to result in carious exposure of the pulp. The need to re-enter the cavity and excavate further in the stepwise excavation technique must continue to be questioned and should be a priority in future research. This could be investigated by high quality, well conducted randomised controlled trials involving long term follow-up of symptoms of pulpitis and pulp necrosis, radiographic evidence of caries progression, microbiological evidence of caries progression or arrest and longevity of restorations. This review only included trials investigating stepwise, partial or no dentinal caries removal versus complete caries removal; future reviews should consider trials which compare these different techniques.

It should be noted that the only long term clinical trial in permanent teeth (Mertz-Fairhurst 1987) involved occlusal caries where caries had only extended to half way through dentine. There is a need to apply the methodology to deeper lesions, lesions affecting multiple tooth surfaces, lesions in compromised teeth that have had a history of restoration and re-restoration and to permanent as well as primary teeth. A major difficulty in this work will be finding stable populations and it will be essential that such studies include multiple practitioners from primary care.

There is also scope for investigation of bonding of materials to soft, wet and infected dentine.

Future research should also investigate patient centred outcomes including oral health related quality of life and patient perception of treatment. Health economic measures should be used to determine the cost of treatment and patient willingness to pay.

Acknowledgements

We wish to thank Sue Furness for her valued contribution and methodological advice and Anne Littlewood (Cochrane Oral Health Group) for her assistance with literature searching. We also wish to thank the referees and Cochrane editors for their valuable comments.

Data and analyses

Download statistical data

Comparison 1. Stepwise excavation versus complete caries removal
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pulp exposure during caries removal4627Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.33, 0.60]
1.1 Primary dentition2173Risk Ratio (M-H, Fixed, 95% CI)0.31 [0.17, 0.57]
1.2 Secondary dentition3454Risk Ratio (M-H, Fixed, 95% CI)0.51 [0.36, 0.72]
2 Signs or symptoms of pulpal disease (1 year)2312Risk Ratio (M-H, Fixed, 95% CI)0.78 [0.39, 1.58]
Analysis 1.1.

Comparison 1 Stepwise excavation versus complete caries removal, Outcome 1 Pulp exposure during caries removal.

Analysis 1.2.

Comparison 1 Stepwise excavation versus complete caries removal, Outcome 2 Signs or symptoms of pulpal disease (1 year).

Comparison 2. Partial caries removal versus complete caries removal
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pulp exposure during caries removal2141Risk Ratio (M-H, Fixed, 95% CI)0.23 [0.08, 0.69]
1.1 Primary dentition298Risk Ratio (M-H, Fixed, 95% CI)0.24 [0.06, 0.90]
1.2 Permanent dentition143Risk Ratio (M-H, Fixed, 95% CI)0.21 [0.03, 1.60]
2 Signs or symptoms of pulpal disease (1 year)3172Risk Ratio (M-H, Fixed, 95% CI)0.27 [0.05, 1.60]
3 Failure of restorations132Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 7.62]
Analysis 2.1.

Comparison 2 Partial caries removal versus complete caries removal, Outcome 1 Pulp exposure during caries removal.

Analysis 2.2.

Comparison 2 Partial caries removal versus complete caries removal, Outcome 2 Signs or symptoms of pulpal disease (1 year).

Analysis 2.3.

Comparison 2 Partial caries removal versus complete caries removal, Outcome 3 Failure of restorations.

Comparison 3. No dentinal caries removal versus complete caries removal
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Signs or symptoms of pulpal disease (1 year)1 Risk Ratio (Fixed, 95% CI)Totals not selected
1.1 Primary dentition1 Risk Ratio (Fixed, 95% CI)0.0 [0.0, 0.0]
2 Failure of restorations2 Risk Ratio (Fixed, 95% CI)Totals not selected
2.1 Primary dentition1 Risk Ratio (Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 Secondary dentition1 Risk Ratio (Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 3.1.

Comparison 3 No dentinal caries removal versus complete caries removal, Outcome 1 Signs or symptoms of pulpal disease (1 year).

Analysis 3.2.

Comparison 3 No dentinal caries removal versus complete caries removal, Outcome 2 Failure of restorations.

Appendices

Appendix 1. Cochrane Oral Health Group's Trials Register search strategy

((caries or carious or decay*) and (restor* or fill*) and (ultraconservative or "stepwise excavation*" or "atraumatic resto*" or ART or "atraumatic therap*" or "atraumatic technique*" or "atraumatic treat*" or "minima* invas*" or "fissure seal*" or "dental seal*" or "resin cement*" or "resin seal*" or "glass ionomer*"))

Appendix 2. Cochrane Central Register of Controlled Trials (CENTRAL) search strategy

#1 Explode DENTAL CARIES
#2 ((teeth or tooth or dental*) and (caries or carious or decay* or lesion*))
#3 DENTAL RESTORATION PERMANENT
#4 DENTAL RESTORATION TEMPORARY
#5 (restor* or fill*)
#6 (ultraconservative or stepwise excavation* or (atraumatic* near restor*) or (atraumatic* near technique*) or (atraumatic near therapy) or (atraumatic* near treat*) or (minimal next invasion) or (minimum next invasion) or (minim* next invasive))
#7 art:ti
#8 art:ab
#9 PIT AND FISSURE SEALANTS
#10 ((fissure near seal*) or (dental near seal*))
#11 Explode GLASS IONOMER CEMENTS
#12 RESIN CEMENTS
#13 (resin near cement*)
#14 (resin near seal*)
#15 ((glass next ionomer*) or cermet*)
#16 (#1 or #2)
#17 (#3 or #4 or #5)
#18 (#11 or #12 or #13 or #14 or #15)
#19 ((dental near seal*) or (fissure near seal*) or (teeth near seal*) or (tooth near seal*))
#20 (#18 and #19)
#21 (#6 or #7 or #8 or #9 or #10 or #20)
#22 (#16 and #17 and #21)

Appendix 3. MEDLINE (OVID) search strategy

1.exp DENTAL CARIES/
2. ((tooth or tooth or dental$) and (caries or carious or decay$ or lesion$))
3. or/1-2
4. DENTAL RESTORATION PERMANENT/
5. DENTAL RESTORATION TEMPORARY/
6. (restor$ or fill$)
7. (ultraconservative or stepwise excavation$ or (atraumatic$ adj6 restor$) or (atraumatic$ adj6 technique$) or (atraumatic$ adj6 therapy) or (atraumatic$ adj6 treat$) or minimal invasion or minimum invasion or minim$ invasive)
8. ART.ab,ti.
9. "PIT AND FISSURE SEALANTS"/
10. ((fissure adj6 seal$) or (dental adj6 seal$))
11. exp GLASS IONOMER CEMENTS/
12. RESIN CEMENTS/
13. (resin adj6 cement$)
14. (resin adj6 seal$)
15. (glass ionomer$ or cemet$)
16. or/11-15
17. ((dental adj6 seal$) or (fissure$ adj6 seal$) or (teeth adj6 seal$) or (tooth adj6 seal$))
18. 16 and 17
19. 4 or 5 or 6
20. 7 or 8 or 9 or 10 or 18
21. 3 and 19 and 20

The above subject search was linked to the Cochrane Highly Sensitive Search Strategy (CHSSS) for identifying randomised trials in MEDLINE: sensitivity maximising version (2008 revision) as referenced in Chapter 6.4.11.1 and detailed in box 6.4.c of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) (Higgins 2011).

1. randomized controlled trial.pt.
2. controlled clinical trial.pt.
3. randomized.ab.
4. placebo.ab.
5. drug therapy.fs.
6. randomly.ab.
7. trial.ab.
8. groups.ab.
9. or/1-8
10. exp animals/ not humans.sh.
11. 9 not 10

Appendix 4. EMBASE (OVID) search strategy

1. exp DENTAL CARIES/
2. ((tooth or tooth or dental$) and (caries or carious or decay$ or lesion$)).mp.
3. or/1-2
4. TOOTH FILLING/
5. (restor$ or fill$)
6. (ultraconservative or stepwise excavation$ or (atraumatic$ adj6 restor$) or (atraumatic$ adj6 technique$) or (atraumatic$ adj6 therapy) or (atraumatic$ adj6 treat$) or minimal invasion or minimum invasion or minim$ invasive).mp.
7. ART.ab,ti.
8. FISSURE SEALANT/
9. ((fissure adj6 seal$) or (dental adj6 seal$))
10. GLASS IONOMER CEMENT/
11. RESIN CEMENT/
12. (resin adj6 cement$)
13. (resin adj6 seal$)
14. (glass ionomer$ or cemet$)
15. ((dental adj6 seal$) or (fissure$ adj6 seal$) or (teeth adj6 seal$) or (tooth adj6 seal$))
16. (10 or 11 or 12 or 13 or 14) and 15
17. 4 or 5
18. 6 or 7 or 8 or 9 or 16
19. 3 and 17 and 18

The above subject search was linked to the Cochrane Oral Health Group filter for EMBASE via OVID:

1. random$.ti,ab.
2. factorial$.ti,ab.
3. (crossover$ or cross over$ or cross-over$).ti,ab.
4. placebo$.ti,ab.
5. (doubl$ adj blind$).ti,ab.
6. (singl$ adj blind$).ti,ab.
7. assign$.ti,ab.
8. allocat$.ti,ab.
9. volunteer$.ti,ab.
10. CROSSOVER PROCEDURE.sh.
11. DOUBLE-BLIND PROCEDURE.sh.
12. RANDOMIZED CONTROLLED TRIAL.sh.
13. SINGLE BLIND PROCEDURE.sh.
14. or/1-13
15. ANIMAL/ or NONHUMAN/ or ANIMAL EXPERIMENT/
16. HUMAN/
17. 16 and 15
18. 15 not 17
19. 14 not 18

What's new

DateEventDescription
12 December 2012New search has been performedSearches updated. Four additional trials were included in this review update.
12 December 2012New citation required but conclusions have not changed- Title changed from Complete or ultraconservative removal of decayed tissue in unfilled teeth to Operative caries management in adults and children to encompass all studies which remove varying amounts of caries, including those that do not remove any dentinal caries.
- New review author.
Compared to the previous version of the review, the overall clinical conclusions are unchanged but based on a larger body of evidence.

Contributions of authors

Development of protocol: David Ricketts (DR), Edwina Kidd (EK), Nicola Innes (NI) and Jan Clarkson (JC).
Identification of studies: Thomas Lamont (TL), DR, NI, EK.
Data extraction: TL, DR, NI.
Appraising risk of bias: TL, DR, NI.
Writing to authors of papers for additional information: TL.
Entering data into RevMan: TL.
Analysis and interpretation of data: DR, TL, NI, EK and JC.
Writing the review: DR, NI, TL, EK, JC.
Methodological support: JC.

Declarations of interest

The author of one of the included studies was also an author of this review. To avoid any conflict of interest this author did not take part in any of the data extraction, risk of bias assessment or analyses relating to this study.

Sources of support

Internal sources

  • University of Dundee Dental School, UK.

  • Guy's, King's and St Thomas Dental School, UK.

External sources

  • Cochrane Oral Health Group Global Alliance, UK.

    All reviews in the Cochrane Oral Health Group are supported by Global Alliance member organisations (British Orthodontic Society, UK; British Society of Paediatric Dentistry, UK; National Center for Dental Hygiene Research & Practice, USA and New York University College of Dentistry, USA) providing funding for the editorial process (http://ohg.cochrane.org).

  • National Institute for Health Research (NIHR), UK.

    All reviews in the Cochrane Oral Health Group are supported by NIHR Systematic Reviews Programme infrastructure funding.

Differences between protocol and review

Change of title from Complete or ultraconservative removal of decayed tissue in unfilled teeth to Operative caries management in adults and children to encompass all studies which remove varying amounts of caries, including those that do not remove any dentinal caries.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Bjørndal 2010

Methods

Design: Centrally randomised, patient masked multicentre trial with two parallel groups.

Centres: Two Danish centres and four Swedish centres.

Setting: Secondary care.

Multiple operators implied as multicentre but number not given.

One trial with a nested trial. The main trial compared stepwise versus complete excavation. Pulp exposures were entered into a second (nested) trial which compared direct pulp capping versus partial pulpotomy.

Participants

Patients

Inclusion criteria: ≥ 18 years of age; one primary caries lesion involving 75% or more dentine; and presence of a well-defined radiodense zone between pulp and carious lesion. Only one tooth treated per patient. 58 patients met the inclusion criteria for the second trial and had to have participated in the first trial as pulp exposure was the result of caries excavation.

Teeth/Lesions

Permanent teeth (one per patient). Over 95% of lesions approximal. 7% drop-outs at 1 year.

Number randomised: 314.

Number analysed: 292 (143/149).

Interventions

Comparison: Stepwise excavation versus complete caries removal

Stepwise excavation (n = 156).

Complete caries removal (n = 158).

Both groups treated similarly thereafter; temporised with calcium hydroxide base material (Dycal) and glass ionomer (Ketac Molar); re-entered at 8-12 weeks; and definitively restored with calcium hydroxide base (Dycal) and composite (Herculite XRV). In the second (nested) trial, following caries excavation direct pulp capping with calcium hydroxide (Dycal) was compared with 1.5 mm pulpotomy with calcium hydroxide cement (Dycal) and provisional and definitive restorations as above.

Outcomes

Pulp exposure.

Unexposed pulp vitality without periapical radiolucency at 1 year.

Signs or symptoms of pulpal disease.

FundingDanish Agency for Science Technology and Innovation, and the Danish Regions.
Notes

Study included patients who reported "mild-to-moderate pretreatment pain" with the characteristics of reversible pulpitis "patients who reported pain, the pain was provoked and confirmed by stimulation with cold or compressed air". Patients with symptoms of irreversible pulpitis were excluded.

Final analysis excluded patients with pulp exposure (i.e. those who entered the nested trial).

Sample size calculation reported.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Computer generated, stratified for pain (yes or no), age (18-49 yr or > 50 yr), and centre in blocks of six".
Allocation concealment (selection bias)Low risk"..achieved through central telephone randomization".
Blinding of participants and personnel (performance bias)
Patient
Low risk"Patients were unaware of the treatment assignment, and all were seen in at least two treatment visits".
Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskMasking not possible - the outcome assessor could not be masked for the primary outcome of pulp exposure as this was assessed during treatment, by the operator.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNumbers of drop-outs given 7% – No clear explanation of cause, but low risk of bias due to the small numbers involved.
Selective reporting (reporting bias)Low riskStudy protocol is available and outcomes reported in pre-specified way.
Other biasHigh riskBy eliminating teeth with pulpal exposures the study design introduced a source of potential bias.

Innes 2007

Methods

Design: Randomised controlled split-mouth trial.

Centres: 10 NHS dental practices.

Setting: Primary care (general dental practice) based in Tayside, Scotland.

Operators: 17 general dental practitioners as operators at 10 NHS dental practices.

Participants

Patients

132 patients (264 teeth) aged 3-10 years. Pairs of teeth matched clinically and radiographically. 132 control and intervention restorations placed with 124 children reviewed. 103 teeth in the control arm received complete caries removal. Only the teeth receiving complete caries removal and their corresponding intervention teeth have been included in this review.

Teeth/Lesions

Carious primary molars. 68% approximal lesions and 42% > half way into dentine radiographically where available.

Number randomised: 132 patients (264 teeth).

Number treated: 128 children (256 teeth).

Number analysed: 124 patients (124/124 teeth) 6% drop-outs at 1 year.

Interventions

Comparison: No caries removal versus complete caries removal

No caries removal (n = 132 teeth). Preformed metal crown (PMC) placed by the Hall technique with no caries removal.

The control arm of this trial involved the restorations of the operator's choice (n = 132) - glass ionomer (69%), amalgam (8%), compomer (5%), composite (11%), PMC (1% - with tooth preparation), fissure sealant (2%), and no restoration provided (3%). The majority of these teeth (n = 103), received complete caries removal, whilst some received partial caries removal (n = 29); only those tooth pairs where the control tooth received complete caries removal have been included in this review.

Outcomes

Signs or symptoms of pulpal disease.

Restoration longevity.

Operator, patient and parent/carer preference.

FundingResearch Training Fellowship grant from the Chief Scientist Office of Scottish Executive and financial support from 3M/ESPE and EastRen. "The Sponsors of this trial had no role in its design; the collection, analysis or interpretation of data; or dissemination of results".
NotesStudy showing 23 months data, however 1 year and other relevant data obtained from author. Further publications give data up to 5 years.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Computer generated randomisation for sequence and side were held centrally".
Allocation concealment (selection bias)Low risk"..(randomisation) accessed by telephone to a distant coordinator prior to treatment".
Blinding of participants and personnel (performance bias)
Patient
High risk

"The tooth on one side was restored using the Hall technique and the contralateral tooth with the restorative technique the GDP would normally use". 

Comment: Patient aware of different treatments.

Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskMasking not possible - different restorations for each group therefore the outcome assessor could not be masked.
Incomplete outcome data (attrition bias)
All outcomes
Low riskCONSORT flow diagram provides information on drop-outs and accounts for all patients. Numbers lost and reasons similar in both groups.
Selective reporting (reporting bias)Low riskAll primary outcomes reported upon.
Other biasLow riskThe study appears to be free from other sources of bias. Conflicts of interest declared.

Leksell 1996

Methods

Design: Randomised, parallel group study.

Centres: Three sites in Sweden.

Setting: Secondary care.

Operators: Six operators at three sites.

Participants

Participants

Children and young adults aged 6-16 years. 116 patients with 134 teeth (posterior permanent molars) enrolled; six patients with seven teeth did not return to complete the second stage of stepwise which left 127 teeth for evaluation of the initial stages of treatment (57 stepwise and 70 complete).

80 teeth had a minimum of 1-year follow-up.

Teeth/Lesions

"Permanent posterior teeth were selected if the radiographs revealed carious lesions to such a depth that pulp exposure could be expected if direct complete excavation was chosen".

Number randomised: 116 patients (134 teeth).

Number analysed: 110 patients (127 teeth), six patients (seven molars) lost from stepwise group and no losses from complete group.

Interventions

Comparison: Stepwise verus complete caries removal

Stepwise (n = 64 teeth). Calcium hydroxide (Calasept) was placed as a base and temporised with zinc oxide eugenol (ZOE) cement. Re-entry at 8 to 24 weeks.

Complete caries removal (n = 70 teeth).

The definitive restorations of both groups involved calcium hydroxide (Calasept), ZOE cement and glass ionomer cement (Baseline, 29 teeth; Fuji, 27 teeth), resin-based material (P-50 resin bonded ceramic - which covered the glass ionomer cement) or amalgam (59 teeth).

Outcomes

Pulp exposure.

Symptoms or symptoms of pulpal disease.

FundingUnclear.
NotesTeeth exposed not included at 1-year follow-up.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Randomly selected for either stepwise or direct complete excavation".
Allocation concealment (selection bias)Unclear riskNo indication of when operator was informed of treatment group.
Blinding of participants and personnel (performance bias)
Patient
High riskMasking not possible - different number of appointments.
Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskMasking not possible - the outcome assessor could not be masked for the primary outcome of pulp exposure as this was assessed during treatment, by the operator.
Incomplete outcome data (attrition bias)
All outcomes
High risk

"Twelve months was the minimal follow-up period. Eighty teeth fulfilled this requirement, 40 treated by stepwise and 40 by direct complete excavation".

Six patients in the stepwise group did not return for subsequent appointments.

Selective reporting (reporting bias)Low riskPublished report includes expected outcomes.
Other biasHigh riskBy eliminating teeth with pulpal exposures the study design introduced a source of potential bias.

Lula 2009

Methods

Design: Randomised, parallel group study conducted in Brazil.
Centres: Single centre in Brazil.

Setting: Secondary care.

Operators: Multiple operators implied (caries detector dye applied to reduce examiner subjectivity used) but number not given.

Participants

Participants

26 patients aged 5-8 years, available for follow-up. Inclusion criteria: "healthy child with at least one active primary carious lesion, on a primary tooth, which extends into the inner half of dentine. No radiographic signs of pulp and/or periapical abnormalities". Reasons for withdrawal noted (two per group).

Teeth/Lesions

16 previously unrestored primary teeth per group. Carious lesions in the inner half of dentine, with the buccal-lingual opening measuring at least 2 mm and could involve occlusal or occluso-proximal surfaces.

Number randomised: 30 patients (36 teeth).

Excluded: four patients (four teeth) - reasons- related to complete caries removal and loss of tooth and contact with patient in partial removal group.

Number analysed: 26 patients (32 teeth).

Interventions

Comparison: Partial versus complete caries removal

Partial caries removal (n = 15 patients and 18 teeth).

Complete caries removal (n = 15 patients and 18 teeth).

Microbiological samples were taken following complete and partial excavations. Both groups treated similarly thereafter; calcium hydroxide (Dycal) was used as a base prior to restoration with composite (Filtek Z 250). "Within 3-6 months both groups were re-entered for microbiological samples with no further attempt at excavation. Teeth in the phase of exfoliation were extracted and the remaining teeth were protected and restored as above".

Outcomes

Pulp exposure.

Signs or symptoms of pulpal disease.

Microflora levels following complete caries removal and partial caries removal - subsequent levels 3-6 months later.

Radiographic signs of caries progression.

FundingConselho Nacional de Desenvolvimento Cientifico e Tecnologico/Edital Universal and Fundacao de Amparo A Pesquisa do Estado do Maranhao.
NotesReported that sample size based on previous pilot study.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"..the sample was randomized using a sequence of random numbers generated in an electronic spreadsheet by a person who did not belong to the research group".
Allocation concealment (selection bias)Low risk"..this information was passed on to the examiner (E.C.O.L) only at the time of treatment".
Blinding of participants and personnel (performance bias)
Patient
Low risk

"All teeth were protected calcium hydroxide cement, etched with 37% phosphoric acid for 15s, and restored with an adhesive system and resin composite" and "Within 3-6 months after treatment, the teeth were submitted to clinical and radiographic examination to determine signs and symptoms of pulp vitality".

Comment: All teeth were restored with the same materials and all teeth were re-entered therefore it is likely the patients were masked regarding their treatment.

Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High risk

Masking not possible - the outcome assessor could not be masked for the primary outcome of pulp exposure as this was assessed during treatment, by the operator.

"Within 3-6 months after treatment, the teeth were submitted to clinical and radiographic examination to determine signs and symptoms of pulp vitality. Next the restorative materials were removed in the two groups and a new dentine sample was collected".

Comment: Although efforts were made to maintain blinding of the outcome assessor by restoring each group with the same materials and carrying out the same review protocol, overall there is a high risk of bias which is introduced during pulp exposure assessment.

Incomplete outcome data (attrition bias)
All outcomes
Low risk

Two excluded from partial removal group due to exfoliation of tooth and loss of contact with patient. Two excluded from complete removal group due to necrosis and loss of restoration.

Reasons for drop-outs different and related to treatment in complete removal group.

Selective reporting (reporting bias)Low riskPublished report includes expected outcomes.
Other biasLow riskThe study appears to be free of other sources of bias.

Magnusson 1977

Methods

Design: Quasi-randomised, parallel group study conducted in Sweden.

Centres: One.

Setting: Secondary care.

Number operators: Four operators at single site.

Participants

Participants

62 children aged 5 to 10 years with carious molars, with no signs/symptoms of irreversible pulpitis or periradicular involvement.

Teeth/Lesions

110 primary molars. Illustrations, diagrams and descriptions would suggest that only occlusal lesions were investigated although the text does not confirm this.

Number randomised: 62 children (110 molars).

Number analysed: 62 children (110 molars).

Interventions

Comparison: Stepwise versus complete caries removal

Stepwise (n = 55 teeth).

Complete (n = 55 teeth).

In the stepwise group re-entry was carried out at 4 to 6 weeks. Teeth were temporised with calcium hydroxide (Calxyl or Calasept), and intermediate layer of 'Dropsin' and sealed with ZOE cement. Restoration material for the control group was not clearly stated.

Outcomes

Pulp exposure.

Signs or symptoms of pulpal disease (outcomes available for period between excavations stages).

Histological examination of three teeth.

FundingUnclear.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)High risk"If the patient was born on an even date, the excavation was continued on one session until all softened dentine had been removed regardless of the risk of exposing the pulp" "If the child was born on an odd day of the month, the molar was excavated step by step".
Allocation concealment (selection bias)High risk

"When two observers agreed that this situation was present the further procedure was determined by the date of birth of the child".

Comment: Operators could have known the child's date of birth prior to recruitment to the trial.

Blinding of participants and personnel (performance bias)
Patient
High risk

"It was decided at random whether to continue the excavation in one sitting until all carious tissue was removed, or to cover the pulpal floor with calcium hydroxide, seal the cavity for 4 to 6 weeks and then removed the softened dentin".

Comment: Patient would be aware of different number of appointments.

Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskMasking not possible - the outcome assessor could not be masked for the primary outcome of pulp exposure as this was assessed during treatment, by the operator.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised participants included in analysis. No missing outcome data.
Selective reporting (reporting bias)Low riskPublished report includes expected outcomes.
Other biasLow riskThe study appears to be free of other sources of bias.

Mertz-Fairhurst 1987

Methods

Design: Randomised, split-mouth study.

Centres: One in USA.

Setting: Secondary care, School of Dentistry, Medical College of Georgia USA.

Number operators: Multiple operators implied but number not reported.

Participants

Participants

753 adult and child patients (aged 8-52 years) screened. 123 adult and child patients with 312 teeth included. Each participant required at least one pair of class lesions extending into dentine.

Teeth/Lesions

Occlusal lesions extending up to half way into dentine were included. Approximal lesions were excluded from the study. 252 teeth available for follow-up.

Number randomised: 123 patients (156 pairs of teeth).

Number analysed: 126 pairs of teeth.

Interventions

Comparison: No dentinal caries removal versus complete removal

No dentinal caries removal (n = 156) involved a bevel to sound enamel with no deliberate effort to remove dentinal caries and definitively restored with composite (Miradapt).

Complete removal (n = 156). Complete caries removal group restored with amalgam or sealed amalgam.

Outcomes

Restoration longevity.

Other outcomes reported: integrity of sealant.

FundingSupported by NIH grant no DE06112.
Notes

Signs or symptoms not reported. We assume no symptoms.

Duration of 1 year chosen for consistency with other studies. Further publications available giving data up to 10 years.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"According to a randomized treatment assignment sheet".
Allocation concealment (selection bias)Unclear riskNo indication of when operator was informed of treatment group.
Blinding of participants and personnel (performance bias)
Patient
High risk

"..sealed composite restorations were to be compared with traditional cavity outline class 1 amalgam restorations and with ultraconservative sealed amalgam restoration".

Comment: Patient aware of different restorative materials.

Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskOutcome assessor would know treatment group due to different restorative materials.
Incomplete outcome data (attrition bias)
All outcomes
High risk30 pairs of teeth (19%) excluded from analysis at 1 year. No clear explanation provided for drop-outs.
Selective reporting (reporting bias)Low riskPublished report includes expected outcomes.
Other biasLow riskThe study appears to be free of other sources of bias.

Orhan 2010

Methods

Design: Randomised controlled trial. Unit of randomisation = tooth.

Centres: One in Ankara University, Turkey.

Setting: Secondary care.

Number operators: One.

Participants

Participants

Children aged 4-15 years. No symptoms of irreversible pulpitis or signs of pulpal or periradicular pathology. Pulp vitality was confirmed by a cold stimulation tester (Chloraethyl) and/or electric pulp tester (Digitest).

Teeth/Lesions

94 mandibular second primary molars and 60 mandibular permanent first molars. Previously unrestored. Approximal and occlusal lesions included. Caries extended radiographically through at least 75% of the entire dentine thickness.

Number randomised: 123 patients (154 teeth).

Number analysed: 154 teeth.

Interventions

Comparison: Stepwise versus partial versus complete caries removal

Stepwise (n = 49 teeth) two-visit 'indirect pulp treatment' (IPT). Caries was removed until the operator thought pulp exposure would occur with further excavation.

Partial (n = 50 teeth) one-visit 'indirect pulp treatment'. Caries was removed until the operator thought pulp exposure would occur with further excavation.

Complete (n = 55 teeth) direct complete excavation. All carious dentine was removed till hard dentine was reached or pulp exposure occurred.

Two-visit IPT after the initial excavation, calcium hydroxide base was placed and provisionally restored with ZOE cement- re-entry was at 3 months.

In all groups, following the final excavation, glass ionomer (Ionofil) was placed as a cavity base.

Definitive restorations involved compomer (Dyract Extra) in the primary teeth and composite (Grandio Voco) in the permanent teeth.

Deciduous teeth were definitively restored with compomer (Dyract Extra) and all permanent teeth were definitively restored with composite (Grandio Voco).

Outcomes

Pulp exposure.

Signs or symptoms of pulpal disease (only including unexposed teeth).

Bacteriological growth from dentine samples.

FundingThe Scientific and Technological Research Council of Turkey.
Notes

Bias noted by author – randomisation should have occurred in two steps: 1 - DCE or IPT, 2 - 1-visit IPT or 2-visit IPT. The authors declared no conflicts of interest. Approximal lesions included. Lots drawn for randomisation.

Restoration longevity not an explicit outcome and not mentioned in methodology. Althought failure of restorations is included in the result text it is not clear whether these are complete data.

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

"..lots were drawn by the investigator who was blinded to the treatments", "the randomization, however, should have been performed in 2 steps" and "In our opinion, this bias was the limitation of this study".

Comment: Authors note risk of bias.

Allocation concealment (selection bias)Unclear riskTiming of randomisation and concealment unclear.
Blinding of participants and personnel (performance bias)
Patient
High riskPatient would be aware of different number of treatment appointments required.
Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator aware of caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
High riskMasking not possible - the outcome assessor could not be masked for the primary outcome of pulp exposure as this was assessed during treatment, by the operator.
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo drop-outs. Failures explained.
Selective reporting (reporting bias)Low riskPublished report includes expected outcomes.
Other biasHigh riskBy eliminating teeth with pulpal exposures the study design introduced a source of potential bias.

Ribeiro 1999

Methods

Design: Randomised, parallel study conducted in Brazil. Unit of randomisation = tooth.

Centres: Number of centres not reported.

Setting: Secondary care.

Number operators: Number of operators not reported.

Participants

Participants

38 eligible patients (aged 7-11 years) with 48 carious primary molars. All teeth available for follow-up.

Teeth/Lesions

Equal numbers of Class I and Class II lesions (n = 12) in both treatment groups. Primary molars with carious lesions involving dentine both clinically and radiographically. No pulpal exposures expected during cavity preparation.

Number randomised: 38 patients (48 teeth).

Number analysed: 38 patients (48 teeth).

Interventions

Comparison: Partial versus complete caries removal

Partial (n = 24 teeth). Thorough removal of carious dentine from the dentino-enamel junction, while the carious dentine from the pulpal and axial walls was left. Visibly moist and soft carious dentine was intentionally left in the cavity and definitive restoration placed. No re-entry was carried out.

Complete (n = 24 teeth). A caries detecting solution was used. "All identified irreversibly infected dentin was removed with a No 2 low-speed round bur". Preparation removed all stained dentine.

All teeth definitively restored with composite (Z100).

Outcomes

Retention of restoration.

Signs or symptoms of pulpal disease.

Other outcomes reported: marginal integrity, radiographs of residual caries, histology of adhesive bond.

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

"The teeth were randomly assigned to 1 of 2 clinical procedures".

Comment: Randomisation method not explained.

Allocation concealment (selection bias)Unclear riskNo indication of when operator was informed of treatment group.
Blinding of participants and personnel (performance bias)
Patient
Low riskMasking not confirmed although both groups used the same restorative materials and review protocol. Probably masked, outcome not likely to be affected.
Blinding of participants and personnel (performance bias)
Operator
High riskMasking not possible - operator knew caries removal technique.
Blinding of outcome assessment (detection bias)
Primary outcomes
Low riskAll teeth were restored with the same materials – although not confirmed likely to have been masked.
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll randomised teeth included in analysis.
Selective reporting (reporting bias)Low riskAll expected outcomes reported.
Other biasLow riskThe study appears to be free of other sources of bias.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Azrak 2004Compares complete caries removal with chemomechanical caries removal (Carisolv) which aimed to completely remove carious tissue.
Casagrande 2009Not comparing two methods of caries removal. Compares restorative indirect pulp cap materials.
Florio 2001Enamel caries only. Compares fissure sealants with fluoride application, no caries removal/control.
Foley 2003Not comparing two methods of caries removal. All minimal preparation.
Handelman 1981No stepwise, partial or no dentinal caries removal group.
Kreulen 1997Does not compare complete caries removal with stepwise, partial or no dentinal caries removal.
Marchi 2006Not comparing two methods of caries removal. Compares restorative indirect pulp cap materials.
Mertz-Fairhurst 1979Compares sealed caries with open cavities, no complete caries removal group.
Mertz-Fairhurst 19862-year report data in included studies/analysis.
Peters 2001Complete caries removal using air abrasion. Different materials used to restore.
Rock 1974Does not compare complete with stepwise excavation, partial or no dentinal caries removal.
Straffon 1988Does not compare complete with stepwise excavation, partial or no dentinal caries removal. Compares fissure sealant on caries free teeth with complete caries removal of carious teeth.
Walls 1988Conventional caries removal in both groups.
Weerheijm 1992Does not compare complete with stepwise excavation, partial or no dentinal caries removal.
Weerheijm 1993Does not compare complete with stepwise excavation, partial or no dentinal caries removal.
Weerheijm 1999Does not compare complete with stepwise excavation, partial or no dentinal caries removal. Investigates effect of resin-modified glass ionomer cement and amalgam on bacteria.
Welbury 1990Conventional caries removal in both groups.
Williams 1981Not comparing two methods of caries removal. Compares fissure sealant materials.
Zanata 2000Does not compare complete versus stepwise excavation, partial or no dentinal caries removal.

Characteristics of studies awaiting assessment [ordered by study ID]

Borges 2012

MethodsRandomised controlled trial with split-mouth design.
Participants30 children between 5 and 9 years old.
InterventionsFissure sealants versus complete caries removal and restoration.
OutcomesProgression of caries and restoration failure.
Notes 

Phonghanyudh 2012

MethodsRandomised controlled trial, two secondary care centres.
Participants276 children, aged 6-11.
InterventionsGroup 1, partial soft caries removal at enamel-dentine junction; Group 2, complete soft caries removal; and Group 3, conventional caries removal.
OutcomesClinical and radiographic signs of irreversible pulpitis.
Notes 

Ancillary