Non-operative versus operative treatment for blunt pancreatic trauma in children

  • Review
  • Intervention

Authors

  • Michael V Haugaard,

    Corresponding author
    1. Rigshospitalet, Copenhagen University Hospital, Department of Surgery and Transplantation C2122, Copenhagen, Denmark
    • Michael V Haugaard, Department of Surgery and Transplantation C2122, Rigshospitalet, Copenhagen University Hospital, Blegdamsvej 9, Copenhagen, DK-2100 Ø, Denmark. michaelhaugaard@hotmail.com.

    Search for more papers by this author
  • André Wettergren,

    1. Kirurgisk Klinik Hvidovre, Hvidovre, Denmark
    Search for more papers by this author
  • Jens Georg Hillingsø,

    1. Rigshospitalet, Copenhagen University Hospital, Department of Surgery and Transplantation C2122, Copenhagen, Denmark
    Search for more papers by this author
  • Christian Gluud,

    1. Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, The Cochrane Hepato-Biliary Group, Copenhagen, Denmark
    Search for more papers by this author
  • Luit Penninga

    1. Rigshospitalet, Copenhagen University Hospital, Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Copenhagen, Denmark
    Search for more papers by this author

Abstract

Background

Pancreatic trauma in children is a serious condition with high morbidity. Blunt traumatic pancreatic lesions in children can be treated non-operatively or operatively. For less severe, grade I and II, blunt pancreatic trauma a non-operative or conservative approach is usually employed. Currently, the optimal treatment, of whether to perform operative or non-operative treatment of severe, grade III to V, blunt pancreatic injury in children is unclear.

Objectives

To assess the benefits and harms of operative versus non-operative treatment of blunt pancreatic trauma in children.

Search methods

We searched the Cochrane Injuries Group's Specialised Register, Cochrane Central Register of Controlled Trials (Issue 5, 2013), MEDLINE (OvidSP), EMBASE (OvidSP), ISI Web of Science (SCI-EXPANDED and CPCI-S) and ZETOC. In addition, we searched bibliographies of relevant articles, conference proceeding abstracts and clinical trials registries. We conducted the search on the 21 June 2013.

Selection criteria

We planned to select all randomised clinical trials investigating non-operative versus operative treatment of blunt pancreatic trauma in children, irrespective of blinding, publication status or language of publication.

Data collection and analysis

We used relevant search strategies to obtain the titles and abstracts of studies that were relevant for the review. Two review authors independently assessed trial eligibility.

Main results

The search found 83 relevant references. We excluded all of the references and found no randomised clinical trials investigating treatment of blunt pancreatic trauma in children.

Authors' conclusions

This review shows that strategies regarding non-operative versus operative treatment of severe blunt pancreatic trauma in children are not based on randomised clinical trials. We recommend that multi-centre trials evaluating non-operative versus operative treatment of paediatric pancreatic trauma are conducted to establish firm evidence in this field of medicine.

Résumé scientifique

Traitement non opératoire versus opératoire des traumatismes fermés du pancréas chez l'enfant

Contexte

Le traumatisme pancréatique de l'enfant est un état grave avec une morbidité élevée. Les traumatismes fermés du pancréas chez l'enfant peuvent être traités chirurgicalement ou sans opération. Pour les lésions pancréatiques fermées moins graves, de stade I et II, une approche non opératoire ou conservatrice est généralement adoptée. Actuellement, il n'est pas clair quel est le traitement optimal chez l'enfant des traumatismes pancréatiques fermés graves, de stade III à V, le traitement opératoire ou non opératoire.

Objectifs

Évaluer les bénéfices et les inconvénients du traitement opératoire versus non opératoire du traumatisme fermé du pancréas chez l'enfant.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre spécialisé du groupe Cochrane sur les blessures, le registre Cochrane des essais contrôlés (numéro 5, 2013), MEDLINE (OvidSP), EMBASE (OvidSP), ISI Web of Science (SCI-EXPANDED et CPCI-S) et ZETOC. Nous avons également fait des recherches dans les bibliographies des articles pertinents, les résumés d'actes de conférences et les registres d'essais cliniques. Ces recherches ont été effectuées le 21 juin 2013.

Critères de sélection

Nous avions prévu de sélectionner tous les essais cliniques randomisés examinant le traitement non opératoire versus opératoire du traumatisme pancréatique fermé chez l'enfant, indépendamment de la mise en aveugle, du statut ou de la langue de publication.

Recueil et analyse des données

Nous avons utilisé des stratégies de recherche appropriées afin d'obtenir les titres et les résumés des études pertinentes pour la revue. Deux auteurs de la revue ont indépendamment évalué l'éligibilité des essais.

Résultats principaux

La recherche a identifié 83 références pertinentes. Celles-ci ont toutes été exclues, et nous n'avons identifié aucun essai clinique randomisé examinant le traitement du traumatisme pancréatique fermé chez l'enfant.

Conclusions des auteurs

Cette revue montre que les stratégies concernant le traitement non opératoire versus opératoire du traumatisme pancréatique fermé chez l'enfant ne sont pas basées sur des essais cliniques randomisés. Nous recommandons la réalisation d'essais multicentriques évaluant le traitement non opératoire versus opératoire des traumatismes du pancréas pédiatriques afin d'établir des preuves solides dans ce domaine de la médecine.

Plain language summary

Treatment of severe blunt pancreatic lesions in children

Background

Optimal treatment of severe blunt pancreatic trauma in children has long been debated, some doctors advocate operation while others prefer a conservative approach without operation as primary treatment. We conducted this systematic review to assess which strategy should be preferred. Blunt trauma to the pancreas typically occurs after crashes involving a bicycle handlebar, road traffic crashes or other types of injury that cause a blow to the upper abdomen.

Study characteristics

We searched medical databases for randomised (where two groups of children were randomly assigned to treatment or no treatment) clinical trials of children treated for blunt trauma to the pancreas by an operation or no operation. The children were aged 17 years or younger. The search was current to June 2013.

Key results

We found no randomised clinical trials investigating operative treatment compared with non-operative treatment of severe pancreatic injury in children, hence we have no firm evidence to support either operative treatment or non-operative treatment of severe pancreatic lesions in children. Although difficult because of the rarity and the acute nature of these lesions, we recommend that multicentre randomised clinical trials of good quality are conducted.

Résumé simplifié

Traitement des traumatismes fermés graves du pancréas chez les enfants

Contexte

Le traitement optimal des traumatismes fermés graves du pancréas chez l'enfant a longtemps fait débat, certains médecins conseillant la chirurgie tandis que d'autres préfèrent une approche conservatrice sans opération en traitement primaire. Nous avons réalisé cette revue systématique pour évaluer laquelle de ces stratégies devrait être préférée. Un traumatisme fermé (ou contondant) du pancréas se produit typiquement lors d'un accident impliquant un guidon de vélo, un accident de la route ou un autre type de blessure qui provoque un coup à l'abdomen supérieur.

Caractéristiques des études

Nous avons effectué des recherches dans les bases de données médicales pour des essais cliniques randomisés (dans lesquels deux groupes d'enfants ont été aléatoirement assignés à un traitement ou à l'absence de traitement) portant sur des enfants traités pour un traumatisme fermé du pancréas par chirurgie ou sans opération. Les enfants étaient âgés de 17 ans ou moins. Les recherches sont à jour jusqu'à juin 2013.

Résultats principaux

Nous n'avons pas trouvé d'essais cliniques randomisés comparant le traitement chirurgical par rapport au traitement non opératoire des traumatismes fermés graves du pancréas chez les enfants, par conséquent, nous ne disposons d'aucune preuve solide permettant d'appuyer soit l'un, soit l'autre. Bien que difficile à mettre en place en raison de la rareté et du caractère aigu de ces lésions, nous recommandons la réalisation d'essais cliniques randomisés multicentriques de bonne qualité.

Notes de traduction

Traduit par: French Cochrane Centre 15th June, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

Background

Description of the condition

Pancreatic trauma in children is a serious condition that is associated with high morbidity (Kao 2003). Pancreatic pseudocyst formation, pancreatitis and pancreatic fistula formation are common complications of pancreatic trauma, and 19% to 55% of people experience pancreas-related complications after pancreatic trauma (Wales 2001; Stringer 2005; Haugaard 2012). Furthermore, morbidity, including pneumonia and pleural fluid collections, frequently occurs in people with traumatic lesions of the pancreas (Haugaard 2012). Mortality in people with pancreatic trauma is estimated to be 3% to 8% (Bosboom 2006; Haugaard 2012). Isolated pancreatic trauma is seldom lethal, but pancreatic trauma in children can be life threatening in association with other organ injuries, cranial injuries or injuries of major blood vessels in the peripancreatic area (Kao 2003).

Trauma of the pancreas is divided into blunt and penetrating injuries. Blunt pancreatic trauma is by far the most common type of trauma mechanism in children, and it is typically seen after crashes involving a bicycle handlebar, road traffic crashes or other types of crashes that cause a blow to the upper abdomen (de Blaauw 2008). Penetrating injuries of the pancreas are much more seldom in children, and may be a result of gunshot or stabbing injuries (Akhrass 1997).

Pancreatic trauma is the fourth most common type of internal organ trauma in children. It has been reported to occur in 3% to 12% of blunt injuries and in 1.1% of penetrating injuries in children (Juric 2009). Pancreatic injury is normally graded in five categories according to the organ injury scale of the American Association for the Surgery of Trauma (AAST), with pancreatic duct status being the most important factor (Moore 1990). These categories are summarised in Table 1.

Table 1. Pancreatic organ injury scale of the American Association for the Surgery of Trauma
Grade Injury Description
IHaematomaMinor contusion without duct injury
 LacerationSuperficial laceration without duct injury
IIHaematomaMajor contusion without duct injury or tissue loss
 LacerationMajor laceration without duct injury or tissue loss
IIILacerationDistal transection or parenchymal injury with duct injury
IVLacerationProximal transection or parenchymal injury involving ampulla
VLacerationMassive disruption of pancreatic head

Diagnosing pancreatic trauma is difficult. The abdominal symptoms such as pain, nausea and vomiting are often unspecific, and are not correlated to the seriousness of the trauma (Bosboom 2006; Haugaard 2012). Serum amylase is increased in 55% to 91% of people with pancreatic lesions (Bass 1988; Graham 2000). Typically, serum amylase raises 2 to 12 hours after injury, and is elevated for up to two to five days after injury (Bass 1988). Grade III to IV pancreatic lesions have a higher initial and maximum amylase value when compared with grade I to II lesions (Haugaard 2012). Elevation of amylase beyond 10 to 30 days after trauma is associated with an increased risk of pseudocyst formation (Bass 1988; Bosboom 2006).

Radiological investigations are important for the diagnosis of pancreatic trauma. Ultrasound examination is often performed in the emergency department and supplies information on free peritoneal fluid and co-existing damage of other organs such as the liver and spleen (Bosboom 2006). Due to its retroperitoneal localisation, the pancreas is usually difficult to visualise with ultrasound. Computer tomography (CT) is used to detect pancreatic damage; however, it is not always possible to detect pancreatic lesions when CT is performed shortly after the accident. Therefore, CT might be repeated after 24 hours when suspicion of pancreatic lesions persist (Bosboom 2006; Haugaard 2012). Magnetic resonance cholangiopancreatography (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP) are considered valuable diagnostic tools for detecting pancreatic duct lesions (Hall 1986). MRCP is non-invasive, while ERCP is invasive and may cause iatrogenic pancreatitis (Bosboom 2006; Haugaard 2012). ERCP gives the possibility of immediate intervention of the pancreatic duct lesion by inserting a stent (Bosboom 2006; Houben 2007).

Description of the intervention

Pancreatic trauma in children can be treated non-operatively or operatively (Paul 2011; Penninga 2011). The non-operative approach often consists of close monitoring of the child's clinical condition; repeated radiological investigations such as CT, ultrasound and MRCP; monitoring of the child's amylase and lipase levels and, optionally, parenteral nutrition (Haugaard 2012). In addition, ERCP with placement of a stent in the damaged pancreatic duct is used in some centres as part of the non-operative approach (Bosboom 2006; Houben 2007). Experience with placement of a stent using ERCP is very limited, and only described in very few cases (Houben 2007; Canty 2001). In adults a fracture of the pancreas involving the pancreatic duct or the common bile duct can be treated with a papillotomy in some cases (Jaik 2008).

Ultrasound-guided drainage of abdominal and peripancreatic fluid collections, and pancreatic pseudocysts, is frequently applied (Bass 1988; Wales 2001).

The operative approach to pancreatic trauma can consist of laparotomy with drainage of the peripancreatic area, distal pancreatectomy with or without preservation of the spleen, Roux-en-Y pancreaticojejunostomy, and, more seldom, pancreaticoduodenectomy (Whipple's procedure). The type of operation depends on the grade and localisation of the pancreatic lesion (Meier 2001). Spleen-preserving surgery should be attempted to avoid the lifelong increased risk of infections after splenectomy (Kertai 2010). In case of trauma where multiple organs are involved and an acute laparotomy is performed, the damage control resuscitation concept must be applied and the pancreatic resection done as part of a staged surgery in a center with extensive hepato-pancreato-biliary (HPB) experience (Seamon 2009; Sutton 2006).

Grade I and II blunt pancreatic injuries in children are treated non-operatively. The optimal treatment of blunt AAST grade III and IV injuries is unclear (Mattix 2007). There might be a trend towards non-operative treatment of blunt severe paediatric pancreatic lesions (de Blaauw 2008). In adults, blunt grade III and IV pancreatic injuries are generally treated operatively. However, in adults the preference for operative treatment is not based on randomised trials.

Differences in morbidity and hospital stay between operative and non-operative treatment have been reported for children with blunt pancreatic lesions (Meier 2001; Wales 2001; Haugaard 2012). Patients who initially are treated non-operatively might require an operation when the clinical condition of the patient worsens (Stringer 2005). When worsening of the clinical condition of the patient occurs, it is frequently due to infection or sepsis (Stringer 2005). It is currently unclear what consequences the delay of surgery has in patients where initially a non-operative approach has been chosen, but who eventually receive an operation due to worsening of the condition (Vane 2009). It has been suggested that surgery is technically more difficult in patients who initially have been treated non-operatively due to intraabdominal adhesions and leakage of pancreatic enzymes in the surgical field (Meier 2001).

Why it is important to do this review

Pancreatic trauma in children is a serious condition. Currently, the optimal treatment of serious paediatric pancreatic injury (AAST grade III to V) is unclear (Penninga 2011). A Cochrane review is warranted to summarise the available evidence and the benefits and harms of non-operative versus operative treatment in children with serious blunt (AAST grade III to V) pancreatic injury. We have not been able to identify any existing systematic reviews evaluating the benefits and harms of non-operative versus operative treatment for serious pancreatic trauma in children or adults.

Objectives

To assess the benefits and harms of operative versus non-operative treatment of blunt pancreatic trauma in children.

Methods

Criteria for considering studies for this review

Types of studies

We planned to include randomised clinical trials, irrespective of blinding, language and publication status.

Types of participants

Children with blunt pancreatic trauma grade I to V, who were 17 years of age or younger.

Types of interventions

Operative treatment versus non-operative interventions or no intervention for pancreatic trauma.

Types of outcome measures

All outcome measures were planned to be assessed at final follow-up.

Primary outcomes
  1. Mortality.

  2. Pancreatic complications (pancreatic pseudocysts, fistulas, pancreatitis, abscesses).

Secondary outcomes
  1. Other complications (infections, wound problems).

  2. Diabetes mellitus.

  3. Other adverse events. Serious complications/adverse events were defined as any untoward medical occurrence that was life threatening; resulted in death, or persistent or significant disability; or any medical event that might have jeopardised the child, or required further intervention to prevent harm (ICH-GCP 1996).

  4. Hospital length of stay among children who survived until hospital discharge.

Search methods for identification of studies

We placed no restrictions on date, language or publication status.

Electronic searches

The Cochrane Injuries Group Trials Search Co-ordinator searched the following:

  1. Cochrane Injuries Group Specialised Register (21 June 2013);

  2. Cochrane Central Register of Controlled Trials (CENTRAL, Issue 5 of 12, 2013);

  3. MEDLINE (OvidSP) (1946 to June week 2 2013);

  4. Embase (OvidSP) (1974 to 20 June 2013);

  5. ISI Web of Science: Science Citation Index Expanded (SCI-EXPANDED) (1970 to June 2013);

  6. ISI Web of Science: Conference Proceedings Citation Index-Science (CPCI-S) (1990 to June 2013);

  7. ZETOC (1993 to June 2013).

Searching other resources

We searched bibliographies of relevant articles. We searched relevant conference proceeding abstracts and the following online trials registers through 5 December 2013:

Data collection and analysis

The Injuries Group's Trials Search Co-ordinator performed the electronic database searches. She collated results using bibliographic software and then forwarded titles and abstracts to the review authors for screening. The review authors carried out searches of all other resources.

Selection of studies

Two review authors (MH and LP) independently assessed the search results for study eligibility. We obtained full-text copies of all studies that met the inclusion criteria. We listed studies that were subsequently excluded in the review with the reason for exclusion (Characteristics of excluded studies). We solved disagreements by discussion or in consultation with a third review author (CG). We had planned to contact the authors of the trials if information about methodology or data was unclear or missing.

Data extraction and management

Two review authors (MH and LP) planned to independently extract data using standard data extraction forms (Moher 2009; Higgins 2011). We were to translate studies reported in non-English language journals before assessment. Where multiple publications from one study existed, we planned to group reports together and mark the publication with the most complete data as the primary reference. Where relevant outcomes were published in other reports, we planned to use these data. We would have resolved disagreements by consultation with all authors. From each trial, we were to extract the following information: first author, country of origin, study design, inclusion and exclusion criteria, number of participants, participant characteristics, injury severity, single- or multi-organ trauma, number of children treated operatively, number of children treated non-operatively, the intervention administered in the two treatment groups (e.g. type of operation, type of non-operative intervention), follow-up period, primary and secondary outcomes, adverse events and participants lost for follow-up. If information was not available in the published trial, we intended to contact authors of the publications in order to assess the trials correctly. We planned to use 'Summary of findings' tables in which the primary outcome measures would be assessed.

Assessment of risk of bias in included studies

The design and reporting of a randomised clinical trial will restrict bias in the comparison of the intervention with the control (Moher 1998; Gluud 2006). According to empirical evidence (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008), we intended to assess the risk of bias in the included trials according to the following criteria:

Sequence generation
  • Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards and throwing dice are adequate if performed by an independent adjudicator.

  • Uncertain risk of bias: the trial was described as randomised, but the method of sequence generation was not specified.

  • High risk of bias: the sequence generation method was not, or may not be, random.

Allocation concealment
  • Low risk of bias: allocation was controlled by a central and independent randomisation unit, sequentially numbered, opaque and sealed envelopes or similar, so that intervention allocations could not have been foreseen in advance of, or during, enrolment.

  • Uncertain risk of bias: the trial was described as randomised but the method used to conceal the allocation was not described, so that intervention allocations may have been foreseen in advance of, or during, enrolment.

  • High risk of bias: if the allocation sequence was known to the investigators who assigned participants.

Blinding of participants and personnel
  • Low risk of bias: the trial was described as blinded for participants and personnel, and the method of blinding was described, so that knowledge of allocation was adequately prevented during the trial.

  • Uncertain risk of bias: the trial was described as blind for participants and personnel, but the method of blinding was not described.

  • High risk of bias, the trial was not blinded for participants and personnel, and so group allocation was known during the trial.

Blinding of outcome assessment
  • Low risk of bias: the trial was described as blinded for outcome assessors from knowledge of which intervention a participant received, and the method of blinding was described, so that knowledge of allocation was adequately prevented during the trial.

  • Uncertain risk of bias: the trial was described as blinded for outcome assessors, but the method of blinding was not described.

  • High risk of bias, the trial was not blinded for outcome assessors, and so group allocation was known during the trial.

Incomplete outcome data
  • Low risk of bias: the numbers and reasons for dropouts and withdrawals in all intervention groups were described, or it was specified that there were no dropouts or withdrawals.

  • Uncertain risk of bias: the report gave the impression that there had been no dropouts or withdrawals, but the number and reasons were not specifically stated.

  • High risk of bias: the number and reasons for dropouts and withdrawals were not described.

Selective outcome reporting
  • Low risk of bias: predefined, or clinically relevant and reasonably expected outcomes were reported on.

  • Uncertain risk of bias: not all predefined, or clinically relevant and reasonably expected outcomes were reported on or were not reported fully, or it is unclear whether data on these outcomes were recorded or not.

  • High risk of bias: one or more predefined, clinically relevant, or other reasonably expected outcomes were not reported; data on these outcomes were likely to have been collected.

Other biases
  • Low risk of bias (the trial appeared to be free of other sources of bias; e.g. conflict of interest bias).

  • Uncertain risk of bias (there was insufficient information to assess whether other sources of bias were present).

  • High risk of bias (it is likely that potential sources of bias were related to specific design used, or other bias risks are present).

Review authors' judgements were planned to be based on the definitions of the above listed domains.

Measures of treatment effect

For dichotomous outcomes, results were planned to be expressed as risk ratios (RR) with 95% confidence intervals (CI). Where continuous scales of measurement were used to assess the outcomes of treatment, we planned to use the mean difference (MD) or the standardised mean difference (SMD) if different scales were used (Thompson 2002).

Dealing with missing data

We intended to handle missing data as follows:

  • contact the investigators to request missing data;

  • perform sensitivity analyses to assess how sensitive our results are to reasonable changes in the assumptions that were made;

  • for incomplete data, to perform 'worst-worst' case scenario analyses, 'best-best' case scenario analyses, 'worst-best' case scenario analyses, and 'best-worst' case scenario analyses.

Assessment of heterogeneity

We planned to assess heterogeneity using a Chi2 test on N-1 degrees of freedom, with an alpha of 0.05 used for statistical significance, and with the I2 test (Higgins 2003).

Assessment of reporting biases

We planned to use a funnel plot to explore bias, if there were at least 10 studies included in an analysis (Egger 1997; Macaskill 2001). Asymmetry in the funnel plot was to be used to assess this collective form of bias. We planned to perform linear regression to determine the funnel plot asymmetry (Egger 1997), if we had a minimum number of 10 trials.

Data synthesis

We planned to pool the data using both the random-effects and the fixed-effect models to ensure robustness of the result. In case of significant differences of the results from the two models, we planned to provide both results. If the difference in the results was not significant, then we planned to present the results using the random-effects model (Higgins 2002), as heterogeneity might be expected between studies due to differences in preferred operation type, availability of non-operative procedures and type of surgeons performing the operations.

Subgroup analysis and investigation of heterogeneity

We planned subgroup analyses for the:

  1. type of operation performed;

  2. type of non-operative intervention performed;

  3. grade of pancreatic lesion.

We planned to perform a test of interaction to evaluate the differences between the estimates, using the method described in Section 9.6.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Sensitivity analysis

  1. Trials with low risk of bias in the domain allocation concealment compared with trials with allocation concealment at high risk of bias;

  2. to assess how sensitive our results are to reasonable changes in the assumptions that were made, as described in the Dealing with missing data section.

Trial sequential analysis

We planned to apply trial sequential analysis as cumulative meta-analyses are at risk of producing random errors because of sparse data, and repetitive testing on accumulating data (Wetterslev 2008; Wetterslev 2009). To minimise random errors, we intended to calculate the required information size (i.e. the number of participants needed in a meta-analysis to detect or reject a certain intervention effect) (Wetterslev 2008). Information size calculation should also account for the diversity present in the meta-analysis. In our meta-analysis, we planned information size to be based on the assumption of a plausible relative risk reduction of 20% or on the relative risk reduction observed in the included trials with low risk of bias (Wetterslev 2008). The underlying assumption of trial sequential analysis is that significance testing may be performed each time a new trial is added to the meta-analysis. We planned to add the trials according to the year of publication and, if more than one trial was published in a year, we planned to add trials alphabetically according to the last name of the first author (TSA manual 2011). On the basis of the outcome proportion in the control group and the mentioned RRs, the risk for type I (5%) and type II (20%) errors, and the diversity, the required information size, and the trial sequential monitoring boundaries were planned to be constructed (Wetterslev 2008). These boundaries should determine the statistical inference one may draw regarding the cumulative meta-analysis that has not reached the required information size. If a trial sequential monitoring boundary is crossed before the required information size is reached in a cumulative meta-analysis, firm evidence may have been established and further trials may be superfluous. In contrast, if the boundaries are not surpassed, it is most probably necessary to continue doing trials in order to detect or reject a certain intervention effect. We had planned to use as default a type I error of 5%, type II error of 20%, and to adjust information size for diversity unless otherwise stated (Wetterslev 2008; Wetterslev 2009). We planned to perform analyses using trial sequential analysis software (TSA manual 2011).

Results

Description of studies

Results of the search

The search identified 442 references (Figure 1). After removal of duplicates, 396 references remained. After removal of 313 irrelevant references based on the title or abstract, or both, we retrieved the full-text of 83 articles reporting on paediatric pancreatic trauma. We decided to study all these 83 articles and their reference lists to get a complete overview of the topic, even though the title or abstract of the articles already reported that the type of study was not a randomised clinical trial. After reading all 83 articles, we excluded them all as there were no randomised clinical trials.

Figure 1.

Study flow diagram.

Included studies

We found no randomised clinical trials comparing non-operative versus operative treatment in children with pancreatic trauma.

Excluded studies

We excluded all retrieved trials, as they were not randomised, or not relevant to the topic of this review.

Risk of bias in included studies

We included no trials in this review.

Effects of interventions

We included no trials in this review.

Discussion

Summary of main results

There is currently no consensus whether blunt pancreatic lesions in children should be treated operatively or non-operatively. There is generally consensus that mild blunt pancreatic trauma without lesion of the pancreatic duct (grade I to II) should be treated non-operatively (Haugaard 2012). In contrast, treatment of severe blunt pancreatic trauma with lesion of the pancreatic duct (grade III to V) is controversial, and treatment can either be non-operative or operative.

Quality of the evidence

We found no published or pending randomised clinical trials that compared non-operative versus operative treatment of pancreatic trauma in children for inclusion in this review. Therefore, all available evidence is based on non-randomised studies.

We detected a number of non-randomised studies during the search that reported on non-operative and operative treatment in children with pancreatic lesions. Most of the studies were retrospective case series. We have summarised the results of these including complications seen in the different studies associated with both non-operative and operative treatment (Table 2; Table 3; Table 4).

Table 2. Treatment of blunt pancreatic trauma in children, grade I to II
  1. NOM: non-operative management; OM: operative management.

    'Success of NOM' was defined as primary NOM where no secondary surgery was needed.

Primary author Year, country Number of children (n), grade Treatment Mortality Pancreas-related complications Other complications Success of NOM
Juric 20092009, Croatia7, grade I-IINOM0%43% (Grade ll, pseudocysts)43% (2 sepsis, 1 pleural effusion)100%
Table 3. Treatment of blunt pancreatic trauma in children, grade III to V
  1. '-' indicates that it was not possible to extract data from the given study.

    NOM: non-operative management; OM: operative management; TPN: total parenteral nutrition.

    'Success of NOM' was defined as primary NOM where no secondary surgery was needed.

Primary author Year, country Number of children (n), grade Treatment Mortality Pancreas-related complications Other complications Success of NOM
Canty 20012001, USA18, grade III-IV9 OM vs. 7 NOM, 2 deaths11% (severe head trauma)

OM 0%

NOM 100%

-29%
Keller 19971997, USA154, grade I-V

Grade I-II: 26 OM vs. 97 NOM

Grade III-V: 16 OM vs. 15 NOM

8%

Grade I-II: 10%

Grade III-V: 32%

--
Wales 20012001, Canada9, grade III-V9 NOM0%44%33% (subphrenical abscess, sepsis occurred twice)100%
Table 4. Treatment of blunt pancreatic trauma in children, grade I to V or not specified
  1. '-' indicates that it was not possible to extract data from the given study.

    NOM: non-operative management; OM: operative management; TPN: total parenteral nutrition.

    'Success of NOM' was defined as primary NOM where no secondary surgery was needed.

Primary author Year, country Number of children (n), grade Treatment Mortality Pancreas-related complications Other complications Success of NOM
Wood 20102010, USA44, grade I-IV14 OM vs. 29 NOM, 1 death4% (multi-trauma)OM 21% vs. NOM 73%OM 57% vs. NOM 20%-
Fabbro 20012001, Italy92 OM vs. 7 NOM0%OM 50% vs. NOM 71%-86%
Bass 19911991, South Africa408 OM vs. 32 NOM0%NOM 38%OM 13% (respiratory stop, abscess)100%
Bass 19881988, Canada2611 OM vs. 15 NOM0%OM 18% vs. NOM 53%-93%
de Blaauw 20082008, The Netherlands34, grade I-IV3 OM vs. 31 NOM0%OM 66% vs. NOM 45%NOM 13% (sepsis)90%
Firstenberg 19991999, USA12, grade I-V7 OM vs. 5 NOM17% (multi-organ injury)42%--
Gorenstein 19871987, Canada218 OM vs. 13 NOM10% (aorta laceration, head trauma)OM 25% vs. NOM 77%-77%
Graham 20002000, Scotland161 OM vs. 15 NOM0%NOM 67%-80%
Holland 19991999, Australia1414 NOM0%NOM 57%-79%
Nadler 19991999, USA5127 OM vs. 24 NOM---79%
Kouchi 19991999, Japan201 OM vs. 19 NOM5% (acidosis after TPN)NOM 50%-84%
Mattix 20072007, USA173, grade I-V43 OM vs. 130 NOM---74%
Shilyansky 19981998, Canada28, grade I-V28 NOM0%36%4% (pelvic abscess)100%
Stringer 20052005, UK9, grade I-IV1 OM vs. 8 NOM0%OM 0% vs. NOM 100%OM 100% (sepsis)38%
Takishima 19961996, Japan81 OM vs. 7 NOM0%OM 100% vs. NOM 14.3% -100%

Proponents of operative surgical intervention argue that it reduces the risk of secondary complications, especially the formation of pancreatic pseudocysts and that delay in surgical intervention leads to greater morbidity (Smith 1988; McGahren 1995; Meier 2001; Snajdauf 2007). Operative treatment can either be explorative laparotomy with drainage or more extensive surgery. If the distal part of the pancreatic duct is damaged, some surgeons advocate a spleen-sparing distal pancreatectomy (Jobst 1999; Kolar 2005). A spleen-removing distal pancreatectomy is performed in adults, but is avoided in children because of the increased risk of infections (Kertai 2010). In trauma involving the proximal pancreatic duct or the pancreatic head (or both), a Roux-en-Y pancreaticojejunostomy with preservation of the pancreatic tail is performed (Stringer 2005). If both the pancreatic head and duodenum are severely damaged, a radical pancreaticoduodenectomy, Whipple's operation, can be performed (Meier 2001).

In contrast, other doctors propose a non-operative approach, arguing to avoid surgery if possible and instead observe the patient closely, and to only perform an operation if the patient's clinical condition deteriorates or the patient shows signs of haemodynamic instability (Kouchi 1999). As a part of the non-operative treatment, some centres keep the patient fasting, use total parenteral nutrition (TPN) and place a nasogastric tube with suction to alleviate pain and nausea (Alanen 2000). Moreover, these doctors argue that complications seen with non-operative treatment, such as formation of pancreatic pseudocysts, often are treated successfully by non-surgical treatment options such as ultrasound-guided percutaneous external drainage (PED) (Haller 1994; Shilyansky 1998; Lucaya 1998; Holland 1999; Wales 2001), or endoscopic ultrasound-guided drainage (EUS). Other studies recommend the insertion of a stent by ERCP in cases where the pancreatic duct is completely or partially lacerated thereby avoiding open surgical intervention (Rescorla 1995; Canty 2001). Experience with this procedure is limited, and there might be an increased risk of developing strictures of the pancreatic duct, besides the fact that the child has to undergo general anaesthesia (de Blaauw 2008).

The non-randomised studies showed that mortality in children with blunt isolated pancreatic trauma is rare. When mortality occurs, it is associated with other complications such as multi-organ injury, injury of large blood vessels, severe head trauma and sepsis. Pancreas-related morbidity, especially formation of pancreatic pseudocysts, in patients treated non-operatively might be higher than in patients treated operatively. In contrast, morbidity outside the pancreas, such as sepsis, pleural effusion and formation of non-pancreatic abscesses, might be lower when undergoing non-operative treatment compared with operative treatment (Haugaard 2012). Long-term complications after both operative and non-operative treatment occur seldomly, and chronic endocrine and exocrine dysfunction is rarely seen, even in patients with severe pancreatic lesions (Wales 2001).

Penetrating pancreatic lesions are predominantly caused by gunshot wounds, and are associated with a high risk of multiple organ damage and injury of the large blood vessels in the abdominal and thoracic region. Therefore, all penetrating pancreatic lesions are generally treated surgically by an exploratory laparotomy (Firstenberg 1999). However, penetrating pancreatic lesions are not the focus of this review, and consensus exists that penetrating pancreatic lesions are treated surgically.

Operative treatment in all patients with pancreatic trauma is generally indicated if the patient's clinical status worsens, shows signs of peritoneal irritation or becomes haemodynamically unstable.

Traumatic pancreatic lesions in children are rare, and the acute nature of the condition makes it difficult to conduct large randomised trials in this field of medicine. This review found no randomised trials despite searching all relevant databases. This makes it impossible to conclude anything definitive about whether operative or non-operative treatment should be applied in children with severe blunt pancreatic lesions (grade III to V), because such a conclusion would be associated with high risk of bias.

Authors' conclusions

Implications for practice

This review showed that strategies regarding non-operative versus operative treatment of pancreatic trauma in children are not based on randomised trials. The decision to institute either non-operative or operative treatment seems to be made out of general practice in any particular institution.

Implications for research

Lack of randomised clinical trials might be due to the low incidence of blunt pancreatic lesions, the paediatric population and the acute nature of the condition. Performing a randomised clinical trial would be difficult, but not impossible. These trials should be multicentre trials, and should be performed with low risk of systematic error and with low risk of random error, and follow the CONSORT (Consolidated Standards of Reporting Trials) guidelines. Conduct of such multicentre trials evaluating non-operative versus operative treatment of blunt paediatric pancreatic trauma can establish firm evidence.

Acknowledgements

We would like to thank Emma Sydenham, Karen Blackhall and Deirdre Beecher from the Cochrane Injuries Group for their support in preparing the protocol and the review.

Data and analyses

Download statistical data

This review has no analyses.

Appendices

Appendix 1. Search strategy

Cochrane Injuries Group Specialised Register
pancrea* and (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*)

Cochrane Central Register of Controlled Trials
#1 MeSH descriptor Pancreas explode all trees
#2 MeSH descriptor Pancreatic Fistula explode all trees
#3 MeSH descriptor Pancreatic Pseudocyst explode all trees
#4 MeSH descriptor Pancreatitis explode all trees
#5 pancrea*
#6 (#1 OR #2 OR #3 OR #4 OR #5)
#7 MeSH descriptor Wounds and Injuries explode all trees
#8 trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*
#9 (#7 OR #8)
#10 (#6 AND #9)
#11 MeSH descriptor Pancreas explode all trees with qualifier: IN
#12 pancrea* near5 (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*)
#13 (#10 OR #11 OR #12)
#14 child* or infan* or toddler* or pre?school* or preschool* or young* or pediat* or paediat* or minor* or boy* or girl*
#15 (#13 AND #14)

MEDLINE (OvidSP)
1. exp Pancreas/
2. exp Pancreatic Fistula/
3. exp Pancreatic Pseudocyst/
4. exp Pancreatitis/
5. pancrea*.ab,ti.
6. or/1-5
7. exp "Wounds and Injuries"/
8. (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*).ab,ti.
9. 7 or 8
10. 6 and 9
11. exp Pancreas/in
12. (pancrea* adj5 (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*)).mp.
13. 10 or 11 or 12
14. exp Child/
15. exp Pediatrics/
16. (child* or infan* or toddler* or pre?school* or preschool* or young* or pediat* or paediat* or minor* or boy* or girl*).ab,ti.
17. 14 or 15 or 16
18. 13 and 17
19. randomi?ed.ab,ti.
20. randomized controlled trial.pt.
21. controlled clinical trial.pt.
22. placebo.ab.
23. clinical trials as topic.sh.
24. randomly.ab.
25. trial.ti.
26. 19 or 20 or 21 or 22 or 23 or 24 or 25
27. (animals not (humans and animals)).sh.
28. 26 not 27
29. 18 and 28

Embase (OvidSP)
1. exp Pancreas/
2. exp Pancreas Fistula/
3. exp Pancreas Pseudocyst/
4. exp Pancreatitis/
5. pancrea*.ab,ti.
6. or/1-5
7. exp Injury/
8. (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*).ab,ti.
9. 7 or 8
10. 6 and 9
11. (pancrea* adj5 (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*)).mp.
12. exp PANCREAS INJURY/
13. 10 or 11 or 12
14. exp child/
15. exp pediatrics/
16. (child* or infan* or toddler* or pre?school* or preschool* or young* or pediat* or paediat* or minor* or boy* or girl*).ab,ti.
17. 14 or 15 or 16
18. 13 and 17
19. exp Randomized Controlled Trial/
20. exp controlled clinical trial/
21. randomi?ed.ab,ti.
22. placebo.ab.
23. *Clinical Trial/
24. randomly.ab.
25. trial.ti.
26. 19 or 20 or 21 or 22 or 23 or 24 or 25
27. exp animal/ not (exp human/ and exp animal/)
28. 26 not 27
29. 18 and 28

ISI Web of Science: Science Citation Index Expanded (SCI-EXPANDED) and Conference Proceedings Citation Index-Science (CPCI-S)
#1 Topic=(Pancrea* same (trauma* or wound* or injur* or fall* or ruptur* or accident* or stab* or gun*)) AND Topic=(child* or infan* or toddler* or pre school* or pre-school or preschool* or young* or pediat* or paediat* or minor* or boy* or girl*)
#2 TS=(randomised OR randomized OR randomly OR random order OR random sequence OR random allocation OR randomly allocated OR at random OR randomized controlled trial) OR TS=(controlled clinical trial OR controlled trial OR clinical trial OR placebo)
#3 TS=((singl* OR doubl* OR trebl* OR tripl*) SAME (blind* OR mask*))
#4 #2 or #3
#5 #1 and #4

ZETOC
#1 All fields: Pancrea* child*
#2 Title: trauma* or injur* or rupture*
#3 #1 and #2

Contributions of authors

All authors have contributed to the review, and accepted the final version of the review.

Declarations of interest

None known.

Sources of support

Internal sources

  • Rigshospitalet Research Council, Rigshospitalet, Copenhagen, Denmark.

    Grant to LP

External sources

  • No sources of support supplied

Differences between protocol and review

The following outcome measure has been moved from secondary outcome to primary outcome:

Pancreatic complications (pancreatic pseudocysts, fistulas, pancreatitis, abscesses).

We changed the objective to all blunt pancreatic trauma in children, instead of only grade III to V.