Sex-stratified long-term outcomes after living donor hepatectomy: A nationwide matched cohort study

Background: Living donor hepatectomy exposes a healthy person to major surgery, and roughly one-third of Korean living liver donors are females.Previous nationwide analyses have treated sex as a variable to match on and adjust for rather than as a potential effect modifier.We examined whether sex modifies the long-term consequences of donor hepatectomy.Methods: We linked 12,371 adults who donated part of the liver between 2002 and 2018 in the mandatory Korean national donor registry to National Health Insurance Service records and matched each donor to 10 healthy controls by sex and 5-year age band (n=123,710).All-cause mortality was the primary outcome.Sex-stratified Cox models were fitted, and effect modification was judged by a formal donation-by-sex interaction test rather than by significance within either sex.Crude absolute rate differences per 1,000 person-years were calculated as descriptive measures of clinical burden.Results: Over 97,767 donor person-years, 89 donors died (68 males and 21 females).The adjusted hazard ratio for death was 1.73 (95% confidence interval, 1.28-2.34) in males and 1.58 (0.86-2.90) in females, with no significant donation-by-sex interaction (p=0.79).Hepatic failure, depression, cancer, diabetes, and hypertension were more frequent among donors of both sexes, and none showed a significant interaction by sex.Crude absolute rate differences nonetheless differed in pattern: excess mortality was greater in males (0.52 vs. 0.35 per 1,000 person-years), whereas excess depression (6.72 vs. 3.91), hypertension (2.03 vs. 1.22), and diabetes (1.83 vs. 1.47) were greater in females.Conclusion: We found no clear evidence that sex substantially modified the relative association between liver donation and long-term outcomes; however, the smaller number of events among females limited precision, and modest differences cannot be excluded.Sex-stratified absolute rate differences should be interpreted as descriptive and hypothesis-generating.