Prediction of Mortality in Patients Undergoing Maintenance Hemodialysis by Charlson Comorbidity Index Using ICD-10 Database

Background/Aims: Many patients with end-stage renal disease have additional comorbidities that are important to clinical study and impact the patient's outcome. The Charlson Comorbidity Index (CCI) is a popular tool and a strong predictor of outcome in end-stage renal disease patients. We obtained comorbidity data from the hospital discharge database using the International Classification of Disease, 10th revision (ICD-10) and analyzed the mortality rate in incident patients undergoing maintenance hemodialysis (HD). Methods: We evaluated the medical records of a total of 456 patients on HD (58 +/- 14 years of age, 56% males). We calculated CCI scores at the start of HD with information from the hospital discharge summary according to the ICD-10 code. We then analyzed patient mortality according to these CCI scores. Results: The percentages of patients that had diabetes with end-organ damage (51.1%), congestive heart failure (9.9%), coronary artery disease (8.1%) and stroke (6.8%) were identified. CCI scores were 5.09 +/- 2.01 (range 2-11). Four comorbidity groups were established by quartile ranking of the CCI scores: low, moderate, high and very high. The mortality rates were: 0.83, 7.70, 14.09 and 18.69 deaths/100 patient-years, respectively (p = 0.001). Compared with the low comorbidity group, the hazard ratios for mortality were 9.22 (95% CI 3.29-25.84) for the moderate group, 16.77 (95% CI 5.97-47.11) for the high group, and 22.37 (95% CI 8.08-61.93) for the very high group. Conclusions: The CCI scores using the ICD-10 database information were significant predictors of mortality in incident patients undergoing maintenance HD. Copyright (C) 2010 S. Karger AG, Basel