Robotic-assisted versus laparoscopic repair of type II, III and IV hiatal hernias: A retrospective study comparing adverse outcomes

Abstract

Objective

Robotic-assisted surgery (RAS) is continuing to expand in use in surgical specialties, including foregut surgery. The available data on its use in large hiatal hernia (HH) repair are limited and conflicting. This study sought to determine whether there are significant differences in adverse outcomes following HH repair performed with a robotic approach vs. a laparoscopic approach. This study was limited to outcomes in patients with type II, III, and IV HHs, as these hernias are typically more challenging to repair.

Methods

A retrospective analysis was performed from data obtained from TriNetX, a large deidentified clinical database, over a 10-year period. Adult patients who underwent type II, III, or IV HH repair were included in the study. HH with robotic repair was compared to laparoscopic repair. Cohorts were propensity score matched for demographic information and comorbidities. Risk ratios, risk differences (RDs) with 95% confidence intervals (CIs), and t test for each examined adverse outcome were used to estimate the effects of robotic repair vs. laparoscopic repair.

Results

In total, 20,016 patients who met the inclusion criteria were identified; 1,515 patients utilized RAS, and 18,501 used laparoscopy. Prior to matching, there were significant differences in age, sex, comorbidity, and BMI between the two cohorts. After 1:1 propensity score matching, analyses of 1,514 well-matched patient pairs revealed no significant differences in demographics or comorbidities. Patients who underwent robotic repair were more likely to experience major complications, including venous thromboembolism (RD: 0.007, 95% CI: 0.003, 0.011; p = 0.002), critical care (RD: 0.023, 95% CI: 0.007, 0.039; p = 0.004), urinary/renal complications (RD: 0.027, 95% CI: 0.014, 0.041; p < 0.001), and respiratory complications (RD: 0.046, 95% CI: 0.028, 0.064; p < 0.001). RAS was associated with a significantly shorter length of stay (32.4 ± 27.5 h vs. 35.7 ± 50.1 h, p = 0.031), although this finding indicated a reduction in the length of stay of less than 4 hours. No statistically significant differences in risk of esophageal perforation, infection, postprocedural shock, bleeding, mortality, additional emergency room visits, cardiac complications, or wound disruption were found.

Conclusions

Patients who undergo robotic-assisted large HH repair are at increased risk of venous thromboembolism, need critical care, urinary or renal complications and respiratory complications. Due to variations in RAS technique, experience, and surgical volumes, further study of this surgical approach and complication rates is warranted.