Objective: To investigate whether the use of triclosan-containing sutures in paediatric patients unselected by surgery could lead to a clinically relevant reduction in the occurrence of surgical site infections (SSIs).
Design: randomised, double-blind, controlled trial.
Setting: University hospital with a Paediatric Surgery Department in Oulu, Finland.
Study sample: children aged less than 18 years admitted to hospital for any type of elective or emergency surgery and with anticipated use of absorbing sutures. Out of 2124 patients that met the inclusion criteria, 1633 agreed to participate and were randomly allocated. Children undergoing surgical correction of the foreskin or a cleft lip or palate were excluded due to the use of sutures with shorter resorption times that were not available with triclosan. Also excluded were patients from the neonatal and paediatric intensive care units (ICUs) and the oncological ward because their baseline risks for infection were too different and there were too few of these surgeries to be analysed as a subgroup. Losses to followup were similar and of less than 10% in both groups.
Intervention: monofilament or multifilament sutures were used based on the desired absorption times, coated with triclosan in the intervention group (IG) and uncoated in the control group (CG). A statistician that was not involved in the study performed the randomised allocation of patients: 802 patients allocated to the IG and 813 to the CG. Twenty-one and twenty percent of patients in these groups, respectively, did not receive the assigned treatment and were excluded from the per-protocol analysis. Patients, surgeons, staff in the operating room and individuals in charge of evaluating surgical wounds were blinded to the type of suture.
Outcome measures: the primary outcome was the occurrence of a deep or superficial SSI in the 30 days following the operation and according to the criteria of the CDC. In the post-hoc analysis, the authors compared the frequency of wound dehiscence, the number of positive bacterial cultures from surgical site samples, the number of required antimicrobial courses and the number of unplanned visits to a nurse or physician. The authors presented the results of the intention-to-treat and per-protocol analyses.
Main results: there were no differences in the baseline characteristics of the two groups. Six surgeons performed 77% of the interventions, but a total of 69 participated in the study. The intention-to-treat analysis found that SSIs occurred in 3% of the IG (n = 778) compared to 5% of the CG (n = 779) (risk ratio [RR], 0.48; 95% confidence interval [95 CI], 0.29 to 0.81), which resulted in a number needed to treat (NNT) with this type of suture of 36 children (95 CI, 21 to 111) to prevent one SSI. In the per protocol analysis, SSIs occurred in 3% of patients in the IG (n = 636) compared to 6% in the CG (n = 651) (RR: 0.47; 95 CI, 0.27 to 0.81) (NNT*, 32; 95 CI, 19 to 108). Adverse events were similar in both groups (sutures not resorbing in the expected time frame in 6%).
Conclusion: consistent with the findings of previous studies in adults, the use of triclosan-coated absorbable sutures is effective in reducing the occurrence of SSIs in paediatric patients.
Conflicts of interest: explicitly disclosed by the authors.
Funding source: study funded by The Alma and KA Snellman Foundation, past fees from Bioretec Ltd and MSD Finland Ltd received by author outside of this work.