BACKGROUND: Despite the poor prognosis of patients with both cirrhosis and renal failure, most reports on renal function and outcomes of cirrhosis have come from liver transplant registries. The present study aimed to investigate the association between renal function and cirrhosis in a broader population. METHODS: 3,857 patients were enrolled after the exclusion of patients with incomplete data. The most recent demographic data after the latest laboratory measurements obtained at outpatient or inpatient department were collected. RESULTS: In predicting renal function, estimated glomerular filtration rate (eGFR) was found to be dissociated from Child-Pugh points (β = -0.01, p = 0.691), different causes of cirrhosis, and presence of diabetes (β = -0.03, p = 0.112). In terms of predicting in-hospital mortality, the sensitivity (60-82%) and specificity (70-90%) of the model for end-stage liver disease (MELD) score increased with the decrease in eGFR. However, the blood urea nitrogen (BUN)/creatinine ratio was better than the MELD score in patients with normal eGFR. CONCLUSION: In contrast to prerenal causes of kidney injury, the underlying causes of cirrhosis or diabetes had relatively minor effects on renal function in cirrhotic patients. The BUN/creatinine ratio was a better index than the MELD score in predicting in-hospital mortality in cirrhotic patients with normal renal function.
BACKGROUND: Despite the poor prognosis of patients with both cirrhosis and renal failure, most reports on renal function and outcomes of cirrhosis have come from liver transplant registries. The present study aimed to investigate the association between renal function and cirrhosis in a broader population. METHODS: 3,857 patients were enrolled after the exclusion of patients with incomplete data. The most recent demographic data after the latest laboratory measurements obtained at outpatient or inpatient department were collected. RESULTS: In predicting renal function, estimated glomerular filtration rate (eGFR) was found to be dissociated from Child-Pugh points (β = -0.01, p = 0.691), different causes of cirrhosis, and presence of diabetes (β = -0.03, p = 0.112). In terms of predicting in-hospital mortality, the sensitivity (60-82%) and specificity (70-90%) of the model for end-stage liver disease (MELD) score increased with the decrease in eGFR. However, the blood ureanitrogen (BUN)/creatinine ratio was better than the MELD score in patients with normal eGFR. CONCLUSION: In contrast to prerenal causes of kidney injury, the underlying causes of cirrhosis or diabetes had relatively minor effects on renal function in cirrhoticpatients. The BUN/creatinine ratio was a better index than the MELD score in predicting in-hospital mortality in cirrhoticpatients with normal renal function.