Agreement between Estimated and Measured Renal Function in an Everyday Clinical Outpatient Setting of Human Immunodeficiency Virus-Infected Individuals

Introduction: Estimated renal function (eRF) has been widely implemented as a screening tool in handling human immunodeficiency virus (HIV)-infected individuals. Our primary objective was to investigate the agreement between measured renal function (mRF) and eRF in HIV-infected individuals in an everyday clinical setting. Methods: A single-center study at the HIV-outpatient clinic at Copenhagen University Hospital, Rigshospitalet. Study period from January 1, 2004-June 1, 2015. We included all HIV-infected individuals who had an mRF performed and compared this with eRF assessed with 9 different serum-creatinine-based equations and the eRF reported by the Department of Clinical Biochemistry. We evaluated performance characteristics of the different eRFs, with concordance correlation coefficient, total deviation index, coverage probability, relative accuracy, and Bland Altman plots. We also evaluated whether exposure to (1) rilpivirine, cobicistat, or dolutegravir (RLP/COB/DTG), (2) protease inhibitors (PIs), or (3) tenofovir disoproxil fumarate (TDF) had an impact on agreement. Furthermore, we compared inter- A nd intra-individual differences between mRF and eRF. Results: Ninety-eight individuals had an mRF performed during the study period. We found that the agreement between mRF and eRF was poor irrespective of the eRF equation. Exposure to RLP/COB/DTG and PIs was not associated with different agreement. Exposure to TDF was associated with statistically significant better agreement for 3 of the evaluated equations. Conclusion: Irrespective of calculation methods, the agreement between mRF and eRF is poor. Surprisingly TDF exposure was associated with a better agreement compared with TDF-unexposed individuals.