Abstract
OBJECTIVE: Using American College of Rheumatology (ACR) and Systemic Lupus International Collaborating Clinics (SLICC) criteria for systemic lupus erythematosus (SLE) classification as gold standards, we determined sensitivity, specificity, positive and negative predictive values (PPV and NPV) of having SLE denoted as the primary cause of end-stage renal disease (ESRD) in the US Renal Data System (USRDS). METHODS: ESRD patients were identified by International Classification of Diseases, Ninth Revision codes in electronic medical records of 1 large tertiary care center, Montefiore Hospital, from 2006 to 2012. Clinical data were extracted and reviewed to establish SLE diagnosis. Data were linked by social security number, name, and date of birth to the USRDS, where primary causes of ESRD were ascertained. RESULTS: Of 7,396 ESRD patients at Montefiore, 97 met ACR/SLICC SLE criteria, and 86 had SLE by record only. Among the 97 SLE patients, the attributed causes of ESRD in the USRDS were 77 SLE and 12 with other causes (unspecified glomerulonephritis, hypertension, scleroderma), and 8 missing. Sensitivity, specificity, PPV, and NPV for SLE in the USRDS were 79%, 99.9%, 93%, and 99.7%, respectively. Of the 60 patients with biopsy-proven lupus nephritis, 44 (73%) had SLE as primary ESRD cause in the USRDS. Attribution of the primary ESRD causes among SLE patients with ACR/SLICC criteria differed by race, ethnicity, and transplant status. CONCLUSION: The diagnosis of SLE as the primary cause of ESRD in the USRDS has good sensitivity, and excellent specificity, PPV, and NPV. Nationwide access to medical records and biopsy reports may significantly improve sensitivity of SLE diagnosis.