The choice of primary repair or mesh repair for paraesophageal hernia: a decision analysis based on utility scores.

BACKGROUND

Controversy exists on the use of mesh in the repair of paraesophageal hernias (PEH). This debate centers around the type of mesh used, its value in preventing recurrence, its short- and long-term complications, and the consequences of those complications compared with primary repair. Decision analysis is a method to account for the important aspects of a clinical decision. The purpose of this study was to determine whether or not the addition of mesh would be superior in PEH repair.

METHODS

A decision analysis model of the choice between primary repair and mesh repair of a PEH was constructed. The essential features of the decision were the rate of perioperative complications, PEH recurrence rate, reoperation rate after recurrence, rate of symptomatic recurrence, and type of outcome after reoperation. The literature was reviewed to obtain data for the decision analysis and the average rates used in the baseline analysis. A utility score was used as the outcome measure, with a perfect outcome receiving a score of 100 and death 0. Sensitivity analysis was used to determine if changing the rates of recurrence or reoperation changed the dominant treatment.

RESULTS

Using the baseline analysis, mesh repair was slightly superior to primary repair (utility score 99.59 vs 99.12, respectively). However, if recurrence rates were similar, primary repair would be slightly superior; whereas if reoperation rates were similar, mesh repair would be superior. Using sensitivity analysis, there are combinations of recurrence rates and reoperation rates that would make one repair superior to the other. However, these differences are relatively small.

CONCLUSIONS

Depending on what the decision-maker accepts as the recurrence and reoperation rates for these types of repair, either mesh or primary repair may be the treatment of choice. However, the differences between the two are small, and, perhaps, clinically inconsequential.