Background. Erythropoietin (EPO) has been reported to slow the decline of renal function in predialysis chronic kidney disease (CKD) patients. On the contrary, in the recent large-scale randomized controlled trial (RCT), CREATE and CHOIR, which aimed to keep a higher haemoglobin (Hb) level than former trials, the renoprotective effect of EPO was not observed. Today, the renoprotective effect of EPO has become controversial. In order to test the hypothesis that the usage of EPO in predialysis CKD patients may ameliorate the progression of renal disease, we conducted a macro-level observational study dealing with all Japanese predialysis CKD patients.
Methods. Annually since 1982, the Japanese Society for Dialysis Therapy reports the number of patients that have entered maintenance dialysis in each prefecture of Japan. Based on the 2002-2004 data, we calculated the annual incidence of end-stage renal disease (ESRD) in each of the 47 prefectures. The annual amounts paid for EPO by each prefecture, presumably corresponding to the amounts used, corrected for the estimated predialysis CKD patients, were calculated. We examined the relationship between the incidence of new dialysis and the usage of EPO in each prefecture. Furthermore, the usage of EPO was compared with that of antihypertensive agents including angiotensin converting enzyme inhibitor (ACE-I), and that of statin.
Results. There were prefectural differences in the annual incidence of ESRD from 2002 to 2004. We also found prefectural differences in the usage of EPO for the three consecutive years. The usage of EPO in predialysis patients was negatively correlated with the incidence of ESRD on linear and multiple regression analyses. At the same time, the usage of EPO had strong positive correlations with the usage of antihypertensive agents including ACE-I and with that of statin.
Conclusion. Our nationwide epidemiologic study revealed that a higher use of EPO was associated with a decreased incidence of new dialysis in daily clinical practice. In addi- tion, there were strong correlations among the usage of EPO, antihypertensive agents and statin. These data are supportive of, but do not prove, the hypothesis that EPO may be renoprotective, when used in combination with other strategies.