|| Checking for direct PDF access through Ovid
To the Editor:Odensten et al1 have made an important contribution with their multi-institutional randomized controlled trial (RCT) assessing the effectiveness of prophylactic mesh for the primary prevention of parastomal hernias (PSHs). PSH is a common and morbid complication of colostomy formation and techniques for prevention are needed. Before the publication of this study, 10 RCTs compared the incidence of PSH between patients who did and did not have prophylactic mesh placed at the time of the index operation. All but a single study – which used biologic rather than the more common synthetic mesh and included patients requiring ileostomy formation2 – demonstrate a lower incidence of PSH when prophylactic mesh is placed; meta-analyses estimate mesh to reduce the odds of PSH by 76% [(95% confidence interval (95% CI): 50–88].3 On this basis, the most recent European Hernia Society guidelines on the prevention and treatment of PSH provide a strong recommendation for prophylactic mesh placement and judge the quality of evidence for this recommendation to be high.4The results of this double-blinded study, the largest to date, challenge our understanding of the role of prophylactic mesh for the primary prevention of PSH. Of the 104 patients randomized to mesh who underwent follow-up at 1 year, 30 (29%) patients were clinically judged to have a PSH, as compared with 32 (30%) of the 107 patients randomized to the control group [odds ratio (OR) 0.95; 95% CI: 0.53–1.72]. Similarly, no differences were found in the incidence of radiologically detected PSH. In light of these results, it is important to fully explore the factors that may be responsible for these very different findings.A possible explanation provided by the authors is the lack of blinding in previous studies. Most previous RCTs blinded patients and radiologists to the randomized group, but not clinical assessors of PSH who generally were the surgeons responsible for mesh implantation. Although this is a possible explanation for differences in the incidence of clinically apparent hernias, previous studies that have blinded outcome assessors, as well as trials that have used routine CT evaluated by blinded radiologists, have reported benefits to mesh use.5–9 These findings seem to suggest that blinding alone does not account for the similar rates of PSH seen between control and intervention groups in this trial.Odensten et al1 also suggest that the greater pragmatism of their trial as compared to previous studies may, in part, explain their findings. Their multicenter study includes 4 university, 2 regional, and 2 county hospitals, with a catchment area that includes 15% of the population of Sweden. The authors contend that their sample of patients and practitioners is more reflective of real-world clinical practice than previous trials conducted at single university affiliated institutions with an interest in PSH prevention. However, the demographics of the patients in this study (ie, age, sex, body mass index) are similar to those of previous trials. The inclusion of regional and county hospitals rather than university hospitals might have included surgeons with a wider range of experience – high-volume surgeons may have different rates of PSH or may attain results with mesh use that are not reproducible by lower volume surgeons. However, all surgeons involved in this study were colorectal surgeons with annual case volumes of over 100 major operations, suggesting that all involved surgeons had relatively high volume.Despite similarities in patient characteristics, mesh choice, placement technique, and outcome definitions, Odensten et al1 demonstrate PSH rates in the mesh group that are inconsistent with those found in previous RCTs. We, therefore, appeal to the authors for further hypotheses regarding their discrepant findings.