The investigators performed the study retrospectively on a cohort and Başkent University Medical and Health Sciences Research Board approved the study with project number KA 22/494. In 2017, clinical use of SGLT-2 inhibitors began in Turkey, with the introduction of first dapagliflozin and then empagliflozin. No other new member of the SGLT-2 family became available in Turkey during the time we conducted the study. We recruited the study group from patients diagnosed with type 2 DM who applied to the Endocrinology and Metabolic Diseases Outpatient Clinics of Başkent University, Turkey, between 1 January 2017 and 30 September 2022. Patients were required to be on SGLT-2 inhibitors (empagliflozin or dapagliflozin) continuously for at least 12 months and be over 18 years of age. We did not randomize the patient recruitment; we collected data from all patients who met the criteria during the study period. The total number of patients with type 2 diabetes evaluated in this way and registered in the database was 83,295. After the elimination of duplicate enrolments, the remaining number of patients with type 2 diabetes was 25,112. Of these diabetics, 3004 individuals who started treatment with an SGLT-2 inhibitor and used it for the prescribed period formed the study population.
Exclusion criteria for this study population were discontinued use of SGLT-2 inhibitors, failure to attend the 3-month and/or 12-month follow-up visits, using drugs that reduce serum uric acid levels, using drugs that reduce uric acid production (allopurinol and febuxostat), using uricolytic drugs (pegloticase), uricosuric drugs (probenecid, sulphinpyrazone, fenofibrate, and losartan), drugs that increase the serum uric acid level (diuretics, cyclosporine, tacrolimus, levodopa, pyrazinamide, and ethambutol) and being treated for acute gout.
On the other hand, the inclusion criteria were: To have applied to Başkent University Endocrinology and Metabolic Diseases Outpatient Clinic between 1 January 2017 and 30 September 2022, to have received a SGLT-2 inhibitor drug use report or prescription and to have used this drug for at least 12 months, to have come for the control examination visit in the 3rd and 12th month after starting the drug. It was also required that serum glycated hemoglobin (HbA1C), fasting plasma glucose (FBG), creatinine, glomerular filtration rate (GFR), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides, uric acid and urine glucose levels were measured and recorded completely at each visit. After applying the inclusion and exclusion criteria, the remaining 410 patients from the study population constituted the sample of the study.
The parameters evaluated in the study were age, gender, type of SGLT-2 inhibitor active substance used (empagliflozin or dapagliflozin), serum HbA1C, FBG, creatinine, GFR, HDL-C, LDL-C, triglyceride, uric acid, and urine glucose levels. The independent variable of the study was SGLT-2 inhibitor treatment. The modulatory variable of the study is GFR. The dependent variable of the study is the SUA level. GFR, which is the regulatory variable of the study, was analyzed in three groups following the Kidney Disease Improving Global Outcomes (KDIGO) guideline [15]. These were categorized into three main groups GFR ≥ 90 ml/min/1.73m2 (normal or high), GFR = 60–89 ml/min/1.73m2 (mildly decreased), and GFR = 30–59 ml/min/1.73m2 (moderately decreased).
We used data scanned from the hospital's electronic medical information system as the data source for the laboratory values measured before receiving the SGLT-2 inhibitor and the laboratory values measured at month 3 and month 12 of SGLT-2 inhibitor treatment in patients who met the inclusion criteria. To prevent bias, the investigators did not interfere with the file information in any way during the data collection and patient evaluation process.
We present means, standard deviations, medians, minimums and maximums in descriptive statistics for continuous data; numbers and percentages for discrete data. To examine the conformity of continuous data to normal distribution, we used Kolmogorov Smirnov test. Missing data and marginal values were not intervened. As the study was retrospective and based on routine follow-up data, there were no cases of exclusion or loss of data in the analysis process among the patients included in the study group. We started the study cohort retrospectively with 410 patients and completed it with the same number of patients.
To compare serum HbA1C, FBG, creatinine, HDL-C, LDL-C, triglycerides, uric acid and urine glucose measurements of patients before SGLT-2 inhibitor treatment, 3 months on treatment and 12 months on treatment we used the Friedman test. We also applied the Friedman's multiple comparison test to analyze which measurements were different. We grouped the patients according to their GFR levels before SGLT-2 inhibitor treatment, and changes in SUA levels between the groups before SGLT-2 inhibitor treatment, after 3 months, and after 12 months of treatment were analyzed using the Mann-Whitney U test. IBM SPSS version 20 (Statistical Package for Social Sciences; v.20; Chicago, IL, USA) program was used in all evaluations and p < 0.05 was accepted as the limit of statistical significance at 95% confidence interval.