Seasonal Initiation of Hemodialysis: A Single-Center Experience

Objective: This study aimed to evaluate the seasonal pattern of hemodialysis initiation in a warm/subtropical climate region (Adana) and its association with the mode of initiation, early complications, and mortality. Materials and Methods: A total of 319 patients who started chronic hemodialysis at Adana City Training and Research Hospital, Department of Nephrology, between 2022 and 2025 were retrospectively analyzed. Chi-square/Fisher’s exact and Kruskal–Wallis tests were used for seasonal comparisons. Multivariable logistic regression models adjusted for age, sex, diabetes, hypertension, and cardiovascular disease (coronary artery and/or valvular heart disease) were fitted for emergent initiation and for any early (30-day) complication, with a likelihood-ratio test of season as a whole; 3-month mortality (13 events) was described by season and analyzed with a season-only Firth model. Catheter dysfunction was analyzed among catheter-exposed patients, followed by sensitivity analyses. Results: A total of 58.3% of patients started dialysis emergently and 41.7% in a planned manner; only 11.3% had an arteriovenous fistula as initial vascular access. No significant seasonal differences were found in initiation type, indications, or the frequency of any early complication (161/314, 51.3%; p = 0.090). In the adjusted complication model, spring versus summer gave OR = 2.33 (95% CI: 1.23–4.41, p = 0.009; Holm-adjusted p = 0.028 across the three season contrasts), but the omnibus test of seasons was not significant (p = 0.069), and spring did not differ significantly from autumn or winter. Among catheter-exposed patients, catheter dysfunction occurred in 31.4% (22/70) in the spring and 16.5% (13/79) in the summer (p = 0.035; four-season omnibus p = 0.191); this contrast was not significant after excluding initiations from February to December 2023 (p = 0.111) or when the season was assigned at the midpoint of the risk window (p = 0.114). Season was not associated with emergent initiation (omnibus p = 0.230) or with 3-month mortality (Firth OR for spring versus summer 1.12, 95% CI: 0.28–4.48). Conclusions: Hemodialysis initiation in Adana was not associated with a winter-peak pattern in initiation counts or blood pressure. Spring initiation was associated with a higher frequency of early complications, mainly catheter dysfunction, but this exploratory single-center finding was not supported by the four-season omnibus tests and was sensitive to analytic choices, and none of the 33 seasonal omnibus tests remained significant after Holm correction. It is hypothesis-generating only and does not support a seasonal practice recommendation; confirmation in larger, multicenter, prospective studies is needed.