Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Fibrin sealants for the prevention of postoperative pancreatic fistula following pancreatic surgery

Authors' declarations of interest

Version published: 15 February 2012 Version history

https://doi.org/10.1002/14651858.CD009621

This is not the most recent version

view the current version 19 June 2023
Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the safety, effectiveness and potential adverse effects of fibrin sealants for the prevention of POPF following pancreatic surgery.

Background

Fibrin sealant (also known as fibrin glue) is a kind of surgical tissue adhesive that is widely used worldwide in various surgical procedures for bleeding control, incision closure etc (Chow 2010; Spotnitz 2010). It is a product derived from human or animal blood (Carless 2003).

Description of the condition

Pancreatic cancer ranks 13th in terms of most common cancers and 8th as a cause of cancer death from a global viewpoint (Anderson 2006; Dragovich 2011; Lowenfels 2006). Regional differences exist in the incidence (the number of new patients diagnosed per year) (Anderson 2006; Dragovich 2011; Lowenfels 2006). The overall incidence of pancreatic cancer is approximately four to 10 cases per 100,000 persons per year (Anderson 2006; Dragovich 2011; Lowenfels 2006). The most common cause of pancreatic cancer is heavy tobacco usage (Anderson 2006; Dragovich 2011; Lowenfels 2006).

Although the exact incidence of chronic pancreatitis worldwide is unknown, it affects approximately six cases per 100,000 persons per year in the European and probably all western countries (Spanier 2008). The prevalence (the total number of patients at a designated time) of chronic pancreatitis in the United Kingdom, France, Japan and south India is three cases, 26 cases, four cases and 114 to 200 cases per 100,000 persons, respectively (Bornman 2001; Braganza 2011; Garg 2004; Lévy 2006). The most common cause of chronic pancreatitis is alcohol abuse (Braganza 2011; Spanier 2008).

Pancreatic surgery is performed to treat pancreatic and extra‐pancreatic diseases, including pancreatic cancers, chronic pancreatitis, as well as biliary, ampullary (related to the union of the pancreatic duct and the common bile duct), and duodenal malignancy (Connor 2005; Gurusamy 2010; Lillemoe 2004). Although the mortality (the proportion of deaths after operation) of pancreatic surgery has been reduced to less than 5% currently, the overall morbidity (the proportion of patients with any postoperative complications) is still high, ranging from 30% to 60% (Bassi 2005; Connor 2005; Gurusamy 2010). Postoperative pancreatic fistula (POPF) is one of the most frequent and potentially life‐threatening complications (Dong 2011; Gurusamy 2010). It is defined by the International Study Group on Pancreatic Fistula as "a drain output of any measurable volume of fluid on or after postoperative day 3 with an amylase content greater than 3 times the serum amylase activity" (amylase is an important enzyme that is present in pancreatic juice, a high level of amylase in abdominal collections indicates POPF) (Bassi 2005). Its reported incidence varies between 2% and 24% in different studies (Bassi 2005; Connor 2005).

POPF generally originates from the pancreatic stump (cut surface) following pancreatic resection, as well as from the pancreatic‐enteric anastomoses (the surgical connection of pancreatic duct and intestine to form a continuous channel) following pancreaticoduodenectomy (a major surgical operation involving the pancreas, duodenum, and other organs) (Bassi 2005; Hackert 2010). The natural history of POPF is variable in different patients (Case 1960). Many factors have been considered to influence the development of POPF (e.g. age, obesity, cardiovascular diseases, diabetes mellitus, pancreatic texture, and pancreatic duct size) (Ramacciato 2011). It seems that older (e.g. more than 60 years of age), overweight people with cardiovascular diseases, diabetes mellitus, soft pancreatic texture, a small pancreatic duct diameter (e.g. less than 3 mm) are more likely to suffer POPF (Ramacciato 2011, Riall 2008).

Description of the intervention

Various methods have been suggested for the prevention of POPF, such as modification of anastomotic techniques, application of pancreatic duct stents, and administration of somatostatin (a kind of hormone produced chiefly by the brain that inhibits the secretion of pancreatic juice) or its analogues (Dong 2011; Gurusamy 2010; Schulick 2009; Wente 2007), but one of the most common and convenient interventions during an operation has been the application of fibrin sealants (Fingerhut 2009; Kuroki 2005; Ohwada 1998). Fibrin sealants are administered to seal the pancreatic‐enteric anastomoses, pancreatic stump or main pancreatic duct during pancreatic surgery.

How the intervention might work

Fibrin sealants are commercial products containing two separate components: primarily fibrinogen (a plasma protein that is converted into fibrin during blood clot formation) and thrombin (a protease in blood that facilitates blood clotting by converting fibrinogen to fibrin) (Carless 2003; Mobley 2002; NLM 1990). Mixing of the two components by single‐ or dual‐syringe systems in a liquid form promotes blood clotting and cross‐linking of fibrin (Mobley 2002; NLM 1990). This process mimics the final stages of blood coagulation and forms a stable fibrin clot that provides a sealing barrier (Carless 2003; Spotnitz 2010). This kind of commercial product is termed fibrin glue.

Another common type is a fibrin sealant patch which consists of a felt coated with a mixture of freeze‐dried fibrinogen and thrombin, and which can be applied directly to the tissue surface without additional manipulations (Chirletti 2009; Ochiai 2010; Rickenbacher 2009). Theoretically, they have the potential to reduce the rate of POPF by mechanically sealing the pancreatic‐enteric anastomoses, pancreatic stump or main pancreatic duct (Fingerhut 2009; Schulick 2009). As POPF often develops into various further abdominal complications (e.g. intra‐abdominal abscess, subsequent sepsis and bleeding) and significantly attributes to mortality and morbidity (Kuroki 2005; Schulick 2009), a reduction in the incidence of POPF might reduce perioperative mortality, morbidity and postoperative hospital stay (Fingerhut 2009; Gurusamy 2010; Schulick 2009).

The cost of fibrin sealants varies much, but is relatively high (Siedentop 2001). For example, the commercial fibrin glue (Tissucol) costs approximately $100 US per 1 ml (Lovisetto 2007). In other words, it costs approximately $500 US per patient using 5 ml Tissucol for pancreatic surgery.

Why it is important to do this review

The use of fibrin sealants during pancreatic surgery is controversial. Fibrin sealants may potentially reduce the incidence of POPF, but it is also possible that they may have no benefits and may be associated with some adverse effects (Carless 2003; Fingerhut 2009; Siedentop 2001). In addition, a recent animal study indicated that the fibrin sealants resulted in toxicity of the rat's pancreas. Theoretically, it may also be harmful to humans (Lämsä 2008). However, there has been no Cochrane review assessing the role of fibrin sealants for the prevention of POPF following pancreatic surgery.

Objectives

To assess the safety, effectiveness and potential adverse effects of fibrin sealants for the prevention of POPF following pancreatic surgery.

Methods

Criteria for considering studies for this review

Types of studies

We will include all randomized controlled trials (RCTs), irrespective of blinding, sample size, publication status, language or publication date. We will exclude quasi‐randomized trials (where the allocation is done on the basis of a pseudo‐random sequence, e.g. odd/even hospital number or date of birth, alternation) (Reeves 2011), non‐randomized studies and duplicate publications.

Types of participants

Patients (irrespective of age, sex, and race) who are about to undergo elective pancreatic resections for any pancreatic or extra‐pancreatic disease will be included.

Patients who are about to undergo total pancreatectomy (no cut surface) will be excluded.

Types of interventions

  • Intervention: fibrin sealants (irrespective of glue or patch) applied to pancreatic anastomoses reinforcement, stump closure or main pancreatic duct occlusion.

  • Comparison: placebo or no fibrin sealants treatment.

Types of outcome measures

Primary outcomes

  1. Postoperative pancreatic fistula (the proportion of patients with POPF).

  2. Postoperative mortality (the proportion of deaths after operation).

Secondary outcomes

  1. Overall postoperative morbidity (the proportion of patients with any postoperative complications, including the proportion of patients with POPF).

    1. Pancreas‐associated morbidity (e.g. POPF, delayed gastric emptying, postoperative pancreatitis).

    2. Other general postoperative morbidity (e.g. wound infection, pulmonary complications, renal failure).

  2. Reoperation rate.

  3. Number of patients with any adverse effects attributed to fibrin sealants.

    1. Hepatitis transmission.

    2. HIV transmission.

    3. Allergy.

    4. Others.

  4. Quality of life.

  5. Cost effectiveness.

  6. Total hospital stay.

Search methods for identification of studies

We will design the search strategy with the help of Racquel Simpson (Trial Search Coordinator) before searching irrespective of language, year or publication status.

Electronic searches

Electronic searches will be conducted on the following databases.

  • the Cochrane Upper Gastrointestinal and Pancreatic Diseases Group Controlled Trials Register, contained within the Cochrane Central Register of Controlled Trials (CENTRAL) (Appendix 1).

  • MEDLINE via PubMed (Appendix 2).

  • EMBASE via OVIDSP (Appendix 3).

  • Science Citation Index Expanded (Appendix 4).

  • Chinese BioMedicine Database (CBM) (Appendix 5).

Searching other resources

We will search the following databases, which include ongoing trials.

We will also search the references in relevant publications and meeting abstracts (via http://www.asco.org/ASCOv2/Meetings and Science Citation Index Expanded) to explore further relevant clinical trials. We plan to search manufacturer's Internet websites for additional information. We also plan to communicate with the authors of RCTs included for more information in the review, if necessary.

Data collection and analysis

We will conduct the systematic review according to the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011a).

Selection of studies

After completing all searches, we will merge the search results using the software package Endnote X4 (reference management software) and remove duplicate records of the same report. Two independent review authors (Wu HM, Xiong XZ) will scan the title and abstract of every record identified by the search for inclusion. We will retrieve full text for further assessment if the inclusion criteria are unclear from the abstract. We plan to detect duplicate publication by identifying common authors, centers, details of the interventions, numbers of participants and baseline data (Higgins 2011b). We intend to correspond with the authors of the RCTs to confirm whether the trial results had been duplicated, if necessary. We will exclude papers not meeting the inclusion criteria and list the reasons for the exclusion. A third review author (Cheng Y) will resolve any discrepancy between the two authors by discussion, and if required, by consultation with the review group's editors.

Data extraction and management

Two review authors (Wu HM, Xiong XZ) will independently extract the data, check and enter the data into an electronic data collection form (Microsoft Word) (Figure 1). We will resolve any discrepancy between the two review authors by consensus.

Figure 1
Electronic data collection form (Microsoft Word)

Electronic data collection form (Microsoft Word)

Assessment of risk of bias in included studies

Two review authors (Xiong XZ, Peng S) will independently assess the methodological quality of the included trials. We plan to use the quality checklist recommended by the Cochrane Handbook for Systematic Reviews of Intervention, version 5.1.0 (Higgins 2011c). We will assess the risk of bias of the trials based on the following domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other sources of bias (Gurusamy 2009; Higgins 2011c). In all cases, an answer "Low risk" indicates a low risk of bias, an answer "High risk" indicates high risk of bias and an answer "Unclear risk" indicates that insufficient detail is reported of what happened in the study (Higgins 2011c). We will resolve any disagreements by discussion and referral to a third author (Wu HM) for adjudication. We intend to present the results of risk of bias by two figures (a 'Risk of bias graph' figure and a 'Risk of bias summary' figure) generated using Review Manager 5 (RevMan 2011).

Measures of treatment effect

We will perform the meta‐analyses using the software package Review Manager 5 (RevMan 2011). For dichotomous outcomes, we will calculate the risk ratio (RR) with 95% confidence interval (CI) (Deeks 2011). For continuous outcomes, we will calculate the mean difference (MD) with 95% CI (Deeks 2011). For continuous outcomes with different measurement scales in different RCTs, we will calculate the standardized mean differences (SMD) with 95% CI (Deeks 2011).

Unit of analysis issues

The unit of analysis is each patient. We do not anticipate cluster‐randomized trials, cross‐over trials and studies with multiple treatment groups.

Dealing with missing data

We will contact the original investigators to request further information in case of missing data. If there is no reply, we will perform the analysis on an 'intention‐to‐treat' (ITT) principle, if applicable (Newell 1992). Otherwise, we will use only the available data in the analysis.

Assessment of heterogeneity

We plan to describe the heterogeneity by using the Chi2 test (Deeks 2011). A P value less than 0.10 is considered to be significant heterogeneity (Deeks 2011). We also plan to use the I2 statistic to measure the quantity of heterogeneity. In case of significant heterogeneity, we will perform the meta‐analysis and interpret the result cautiously. We will explore the clinical heterogeneity by comparing the characteristics of participants, interventions, controls, outcome measures and study designs in the included studies. We plan to undertake the following approaches for explanation and solution:

  1. Check again that the data are correct.

  2. Change the effect measure.

  3. Re‐analyse the data using the random‐effects model.

  4. Perform sensitivity analysis by excluding potentially biased trials.

  5. Perform subgroup analysis or meta‐regression.

  6. Present all trials and provide a narrative discussion (Deeks 2011).

Assessment of reporting biases

If meta‐analysis is possible, we plan to use funnel plots to assess reporting biases (Sterne 2011). Visual asymmetry in funnel plots will be used to determine the reporting biases (Sterne 2011). We also plan to perform linear regression to determine the funnel plots asymmetry (Egger 1997). We will not perform funnel plots if the number of trials included is less than 10 (Sterne 2011).

Data synthesis

We will perform the meta‐analyses using Review Manager 5 software provided by The Cochrane Collaboration (RevMan 2011). Two review authors (Peng S, Cheng NS) will independently enter all data into RevMan. Any disagreement will be resolved by consensus. For all analyses, we will employ the random‐effects model. We will perform the meta‐analysis if there are sufficient trials of similar comparisons reporting the same outcomes. Otherwise, we will describe the results of the included RCTs and provide a narrative discussion (Dong 2011).

Subgroup analysis and investigation of heterogeneity

If there is a significant heterogeneity among the RCTs, we plan to perform the following subgroup analyses.

  1. RCTs with low risk of bias versus RCTs with high risk of bias.

  2. The type of fibrin sealants (glue and patch).

  3. Different sealing locations (pancreatic‐enteric anastomoses, pancreatic stump and main pancreatic duct).

  4. Different etiologies (pancreatic cancer, chronic pancreatitis and others).

  5. The type of operation (proximal, distal and central pancreatectomy).

  6. High‐risk patients (e.g. fatty pancreas, soft pancreas, small pancreatic duct) versus low‐risk patients.

Sensitivity analysis

We will perform sensitivity analyses to see whether conclusions are robust to decisions made during the review process following the method as follows:

  1. Changing between a fixed‐effect model and a random‐effects model.

  2. Changing statistics among risk ratio (RR), risk differences (RD) and odds ratios (OR) for dichotomous outcomes.

  3. Changing statistics between mean difference (MD) and standardized mean differences (SMD) for continuous outcomes.

  4. Excluding RCTs with low quality.

  5. Excluding non‐English studies.

If the results do not change, we will consider them to have low sensitivity. If the results change, we will consider them to have high sensitivity.

Electronic data collection form (Microsoft Word)
Figures and Tables -
Figure 1

Electronic data collection form (Microsoft Word)

Navigate to figure in ProtocolOpen in new tab