Impacto da cetoacidose recorrente no risco de mortalidade de pacientes com diabetes mellitus do tipo 1: estudo de uma coorte brasileira

Abstract

Context: Recurrent Diabetic Ketoacidosis (rDKA) is an already recognized condition that presents some peculiarities, such as associated psychosocial illnesses, non-compliance to insulin treatment and which courses with an increase in mortality.(1, 6, 8-11) Objective: To study the impact of rDKA mortality and its associated risk factors Patients and Methods: Through the analysis of the electronic database of our institution (PEP Unifesp) we accessed clinical and laboratorial data of patients hospitalized with ICD-10 E.10.1 and E14.1 that were in the fourth character for diabetic ketoacidosis of the period from January 2007 to January 2018. Of these, only those with type 1 diabetes mellitus (DM1) were included for analysis, and those with another form of diabetes, even insulin-dependent and DKA, were excluded. In total, we analyzed and managed to follow-up 231 patients, who were included in this retrospective cohort. For medium-term analysis (1823 days), follow-up was carried out through new medical appointments, phone calls and virtual contacts. For mortality analysis, we subcategorized patients with DKA into 4 groups: I) DKA at diagnosis of DM1/ DM1 Newly Diagnosed II) 1 episode of DKA during follow-up after diagnosis of DM1 III) 2-5 episodes of DKA IV) >5 episodes of DKA during follow-up. Results: A mortality rate of 16.02% (37/231) was found after hospital discharge, with age of death of 38.7 years [IQR 28.43-48.91 years], duration of diabetes of 19.2 years [IQR 13.67-23.80]. In addition, we found the number of hospitalizations (p<0.001), duration of diabetes (p<0.001), microangiopathic complications (Retinopathy, Nephropathy and Neuropathy) (p<0.001), mood disorders (p=0.001), use of antidepressants (p<0.001), statin use (p<0.001) were risk factors for the outcome of death in patients with CAD. Analyzing the CAD subgroups, it was found in the survival curve of Kaplan-Meier, there was a notable difference between those with more than two episodes in the projected period. With a projection of 1926 days (approx. 5 years) the probabilities of death were HR 7.78%; 4.58%; 24.40%; 26.63% in group I, II, III and IV, respectively. Conclusion: Patients with CADr (those with more than 2 episodes of CAD in the period analyzed) are about 4-5 times more likely to die than those who had a single episode (Group I and II). CADr is a entity that requires specific care and follow-up that should include personalized medicine and psychosocial care for individuals with DM1.