Introduction: Previous studies have demonstrated an association between low serum creatinine levels and adverse outcomes in patients undergoing maintenance hemodialysis. However, little is known regarding whether long-term changes in serum creatinine predict outcomes independently and incrementally over a single point evaluation.
Methods: Serum creatinine data at index (June 2013) and for the 18 months prior to the index blood sampling (between January 2012 and June 2013) were evaluated in 346 hemodialysis patients. Patients were followed from the index blood sampling for primary (all-cause mortality) and secondary (cardiovascular death) endpoints.
Findings: During a median follow-up of 5.7 years, there were 82 all-cause and 25 cardiovascular deaths. Compared to patients who survived, those who died displayed a greater time-dependent reduction in creatinine levels during the 18 months prior to the index assessment, coupled with a greater decrease in predialysis body weight (interaction p = 0.007). Patients who displayed creatinine decline over the prior 18 months (∆ creatinine<0 mg/dL) had higher all-cause mortality than those who maintained creatinine levels (∆ creatinine≥0 mg/dL). After adjustment for clinical factors and baseline creatinine index, antecedent creatinine decrease was independently associated with an increased risk of all-cause mortality, with an incremental prognostic value over baseline creatinine index alone. A reduction in creatinine levels was also associated with cardiovascular death independent of the baseline creatinine index.
Discussion: A long-term antecedent decrease in serum creatinine levels is independently associated with clinical outcomes in hemodialysis patients, with an incremental prognostic value over baseline creatinine index alone. Our data suggest that serial creatinine measurements are a useful prognosticator in practice.
Keywords: biomarker; hemodialysis; prognostication; repeated measurements.
© 2019 International Society for Hemodialysis.