RESUMEN
BACKGROUND: The role of Ramadan fasting (RF) as a predictive factor of urinary tract infection (UTI) recurrence was controversially discussed in the literature. AIM: The present study aimed to identify the prognosis factors of recurrent UTIs. METHODS: Data were retrospectively collected from patients with UTI diagnosed at the infectious diseases department and its affiliated outpatient department in Hedi Chaker University Hospital, Sfax, Tunisia, between 2010 and 2017. Kaplan-Meier method was used to generate recurrence-free survival (RFS) curves for first episode of UTI diagnosed in two groups: during and outside Ramadan, which were compared using Log-rank test for statistical inference. Univariate and multivariate Cox proportional hazards regression models were used to identify UTI recurrence factors (sex, age, age group, season, comorbidities, first UTI episode diagnosed during Ramadan, UTI requiring hospitalization, length of hospital stay, nosocomial UTI, clinical presentation). RESULTS: During the follow up, among the 867 patients with UTI, 105 (12.1%) developed a recurrent UTI one. The RFS median [95% confidence interval] was 60 [40 to 82] days. Survival curves showed that patients with UTI diagnosed during Ramadan had shorter RFS compared with those diagnosed outside of Ramadan (32 vs. 60 days, respectively, p=0.002). RF (hazard ratio = 2.96; p = 0.033) and diabetes mellitus (hazard ratio =1.6; p = 0.033) were independently associated with UTI recurrence in multivariate Cox regression analysis. CONCLUSION: Recurrent UTI was a prevalent and challenging condition among patients with UTI. Diabetes mellitus and RF had a prognosis value for recurrence in UTI.