#2823 Association between nephrotoxic drugs and the development of chronic kidney disease based on french real-world data

Abstract Background and Aims Chronic kidney disease (CKD) is primarily associated with cardiovascular conditions such as hypertension (HTN) and diabetes. However, the impact of nephrotoxic drugs on the development of CKD remains difficult to assess due to multiple factors, including the slow progression of the disease, polypharmacy, particularly in elderly patients, and a lack of data. Given the constant increase in CKD prevalence and the growing trend of polypharmacy, our objective is to explore the iatrogenic impact of nephrotoxic drugs on the onset of CKD using real-world data. Method This retrospective study is based on the linkage of hospital data alias PMSI (Program for the Medicalization of Information Systems) and the French National Health Data System (SNDS). The SNDS provides information on outpatient drug consumption, while hospital data include biological test results. These two datasets were matched using common information from the PMSI, such as ICD-10 (International Classification of Diseases) diagnoses and medical procedures. We included non-dialysis adult patients hospitalized between 2015 and 2018 who had at least one creatinine measurement. Patients were categorized based on their renal status: healthy patients (without CKD or acute kidney injury) and patients who developed CKD (normal GFR at least once followed by an ICD-10 diagnosis of CKD or three months of clearance < 60 mL/min/1.73m²). Age and sex were identified, along with other comorbidities influencing CKD, such as HTN, diabetes, dyslipidemia, and obesity, identified using ICD-10 codes and the Anatomical Therapeutic Chemical (ATC) classification of outpatient medications. The nephrotoxic drugs studied included drugs known to induce CKD, such as lithium, antiretrovirals (ARVs), fluindione, tyrosine kinase inhibitors (TKIs), triptans, tolvaptan, proton pump inhibitors (PPIs), and non-steroidal anti-inflammatory drugs (NSAIDs). Drug exposure was assessed in a binary manner (exposed/unexposed) over the year preceding the first clearance measurement for healthy patients and over the year preceding the onset of CKD for patients who developed chronic kidney disease. Comorbidities by renal status were analyzed using logistic regression. Subsequently, the impact of each nephrotoxic drug was assessed using logistic regression adjusted for significant comorbidities, with an alpha risk of 5%. Results The study included 244,937 healthy patients (mean age: 63.5 years, 55% female) and 238,676 patients who developed CKD (mean age: 71.6 years, 40% female). According to the preliminary results, comorbidities were more frequent among patients who developed CKD, including HTN (82%), dyslipidemia (48%), diabetes (25%), and obesity (7%). Age was a significant risk factor for CKD development, with a notably higher risk in patients over 75 years old (OR = 10.31 [95% CI: 9.6-11.1]). The total number of nephrotoxic drugs was a significant factor in renal function decline (OR = 1.1 [95% CI: 1.0–1.1]). Among the studied drugs, those with a significant impact on CKD development included ARV 2 (OR = 5.4 [95% CI: 3.9–7.4]), ARV 1 (OR = 3.2 [95% CI: 2.5–4.1]), TKIs (OR = 2.6 [95% CI: 2.2–3.0]), fluindione (OR = 2.2 [95% CI: 2.1–2.3]), lithium (OR = 1.9 [95% CI: 1.6–2.4]), and PPIs (OR = 1.2 [95% CI: 1.1–1.2]). However, in this study, certain drugs such as NSAIDs and triptans were not significantly associated with CKD development. The complete results of the statistical models are presented in Table 1. Conclusion Thanks to full visibility of the patient pathway from outpatient care to hospital settings, we conclude that the significant factors contributing to CKD development include comorbidities such as age, HTN, diabetes, dyslipidemia, and obesity. The nephrotoxic drugs with a significant impact are ARVs, TKIs, lithium, fluindione, and PPIs, which are widely prescribed on a population level. Ongoing studies aim to investigate the relationship between CKD development and drug exposure quantified using the cumulative Defined Daily Dose (DDD), which represents the assumed average maintenance dose per year for a drug used in its primary indication.