Gentamicin-Collagen Implants to Reduce Surgical Infection
Gentamicin-Collagen Implants to Reduce Surgical Infection
The search terms used were gentamicin, collagen; sponge, Collatamp-G, Septocoll, Gentacoll, Syntacoll, and fleece. These search terms were combined with the Cochrane Highly Sensitive Search Strategy to identify all RCTs. The search was performed independently by 2 authors (W.K.C. and S.S.) according to the validated methods of the PRISMA statement (Cochrane Collaboration: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). The databases examined were PubMed, MEDLINE (1948–2011), EMBASE (1980–2011), Cochrane Central Register of Controlled Trials, and SCOPUS. Unpublished trials and conference proceedings were searched through the System for Information on Grey Literature in Europe, the National Research Register (UK), Web of Science, and trial registries. There were no language restrictions applied.
RCTs that evaluated the efficacy of gentamicin-collagen implant in SSI reduction were included. Exclusion criteria were in vitro experiments, animal studies, and nonrandomized studies. Studies were considered for inclusion independently by 2 authors (W.K.C. and S.S.) and any disagreements were resolved by consensus after consultation with the senior authors (A.D.M. and A.G.H.). The quality of included studies was assessed with the methods specified by the Cochrane Collaboration for assessing risk of bias and the Jadad scoring system.
This was performed by 2 authors (S.S. and W.C.) who were masked to the journal and article title and study authors. The risk of bias was assessed using the Cochrane Collaboration tool for assessing risk of bias and displayed using summary figures generated by RevMan 5.0 (Review Manager Version 5.0: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). Assessment of the methodological quality of the included trials was performed according the Jadad scoring system which awards 1 point each for explicit description of randomization, appropriate methods of randomization, withdrawals or dropouts from the trial, double-blinding, and robust methods to achieve double-blinding.
The outcome of interest was the incidence of SSI. The outcomes were counted as per event and assumed as defined if reported. The 2 authors (W.K.C. and S.S.) independently extracted data from eligible articles. Missing information including information from unpublished trials was sought from corresponding authors. The authors were contacted twice over a period of 1 month and if there was no response, the study was excluded. A subset analysis by incision type (clean, clean-contaminated, contaminated, and dirty) was planned a priori. Post hoc analysis was performed according to types of surgery.
The data analysis was performed using Revman 5.0 (Review Manager 5.0: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). Analyses were on an intention-to-treat basis and a random effects model was used to estimate the pooled odds ratio (OR) with the corresponding 95% confidence interval (CI). Forest plots were constructed, with the pooled effect and its 95% CI represented by a diamond that did not cross the vertical line of no effect (OR = 1) and P < 0.05 considered to be statistically significant. The number needed to treat (NNT) was calculated from the pooled OR and patient expected event rate. Heterogeneity was assessed using the I statistic and the χ test for heterogeneity was performed, with P < 0.1 considered as statistically significant heterogeneity. I values below 25% were defined as low heterogeneity, values between 25% and 50% were considered to represent moderate heterogeneity, and values greater than 50% were defined as representing high heterogeneity. Publication bias was assessed using the funnel plot.
Methods
Study Identification
The search terms used were gentamicin, collagen; sponge, Collatamp-G, Septocoll, Gentacoll, Syntacoll, and fleece. These search terms were combined with the Cochrane Highly Sensitive Search Strategy to identify all RCTs. The search was performed independently by 2 authors (W.K.C. and S.S.) according to the validated methods of the PRISMA statement (Cochrane Collaboration: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). The databases examined were PubMed, MEDLINE (1948–2011), EMBASE (1980–2011), Cochrane Central Register of Controlled Trials, and SCOPUS. Unpublished trials and conference proceedings were searched through the System for Information on Grey Literature in Europe, the National Research Register (UK), Web of Science, and trial registries. There were no language restrictions applied.
Study Selection
RCTs that evaluated the efficacy of gentamicin-collagen implant in SSI reduction were included. Exclusion criteria were in vitro experiments, animal studies, and nonrandomized studies. Studies were considered for inclusion independently by 2 authors (W.K.C. and S.S.) and any disagreements were resolved by consensus after consultation with the senior authors (A.D.M. and A.G.H.). The quality of included studies was assessed with the methods specified by the Cochrane Collaboration for assessing risk of bias and the Jadad scoring system.
Validity Assessment
This was performed by 2 authors (S.S. and W.C.) who were masked to the journal and article title and study authors. The risk of bias was assessed using the Cochrane Collaboration tool for assessing risk of bias and displayed using summary figures generated by RevMan 5.0 (Review Manager Version 5.0: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). Assessment of the methodological quality of the included trials was performed according the Jadad scoring system which awards 1 point each for explicit description of randomization, appropriate methods of randomization, withdrawals or dropouts from the trial, double-blinding, and robust methods to achieve double-blinding.
Data Abstraction
The outcome of interest was the incidence of SSI. The outcomes were counted as per event and assumed as defined if reported. The 2 authors (W.K.C. and S.S.) independently extracted data from eligible articles. Missing information including information from unpublished trials was sought from corresponding authors. The authors were contacted twice over a period of 1 month and if there was no response, the study was excluded. A subset analysis by incision type (clean, clean-contaminated, contaminated, and dirty) was planned a priori. Post hoc analysis was performed according to types of surgery.
Statistical Analysis
The data analysis was performed using Revman 5.0 (Review Manager 5.0: The Nordic Cochrane Centre, Copenhagen; The Cochrane Collaboration, 2008). Analyses were on an intention-to-treat basis and a random effects model was used to estimate the pooled odds ratio (OR) with the corresponding 95% confidence interval (CI). Forest plots were constructed, with the pooled effect and its 95% CI represented by a diamond that did not cross the vertical line of no effect (OR = 1) and P < 0.05 considered to be statistically significant. The number needed to treat (NNT) was calculated from the pooled OR and patient expected event rate. Heterogeneity was assessed using the I statistic and the χ test for heterogeneity was performed, with P < 0.1 considered as statistically significant heterogeneity. I values below 25% were defined as low heterogeneity, values between 25% and 50% were considered to represent moderate heterogeneity, and values greater than 50% were defined as representing high heterogeneity. Publication bias was assessed using the funnel plot.
Source...