The erythropoietin neuroprotective effect: assessment in coronary artery bypass graft surgery (TENPEAKS) – a randomized, double-blind, placebo-controlled, proof-of-concept clinical trial

Introduction

Human recombinant erythropoietin (rh-Epo) may be a potent neuroprotectant [1]. It is approved in Canada to reduce blood transfusions in coronary artery bypass graft (CABG) surgery, which is complicated by neurocognitive dysfunction (NCD) in 50–80% of patients at discharge, 10–36% at 6 weeks and 42% at 5 years [2]. We conducted a randomized, double-blind, placebo-controlled study in 32 first-time CABG patients to investigate the feasibility and safety of three doses of rh-Epo with the ultimate goal of reducing NCD in patients following CABG.

Methods

Patients were randomized to receive placebo or a total of 375 U/kg, 750 U/kg, or 1500 U/kg rh-Epo divided into three doses: the day before, the time of, and the day after surgery. A battery of 10 neuropsychological tests was performed pre-operatively, at discharge and at 2 months. The primary outcomes were feasibility and safety, including recruitment rates, 28-day all-cause mortality, ICU and hospital length of stay (LOS), incidence of pure red cell aplasia (PRCA) and thrombotic complications (stroke, myocardial infarction, deep vein thrombosis and pulmonary embolism). The secondary outcome was the incidence of NCD defined as a 20% reduction in 20% of the tests.

Results

Consent was obtained in 32/101 of eligible patients. All subjects were male with mean age 60 years (range 46–73 years). Hypertension was present in 66%, diabetes in 22%, dyslipidemia in 81%, CHF in 6% and COPD in 6%. The mean pump time was 76 min (SD 23), and the mean cross-clamp time was 56 min (SD 22). The median ICU LOS was 1 day (IQR range 1–2). The median hospital LOS was 6 days (IQR 5–6.5). Mortality, PRCA and thrombotic complications were not observed. NCD was present in 66% of patients at discharge and 16% at 2 months. One patient declined 2-month follow-up. NCD at discharge, by group, was: placebo 75%, 375 U/kg 50%, 750 U/kg 75% and 1500 U/kg 63%. NCD at 2 months, by group, was: placebo 38%, 375 U/kg 13%, 750 U/kg 13% and 1500 U/kg 0%. Patients receiving rh-Epo at any dose had an incidence of NCD of 8.3%, compared with 38% for placebo (P = 0.085). The comorbidity, mean pump time and mean cross-clamp time were not significantly different between the two groups, nor were the ICU and hospital LOS.

Conclusion

This study of rh-Epo as a neuroprotectant in patients undergoing CABG surgery is feasible and safe. rh-Epo was associated with a trend in the reduction of NCD at 2 months. These data support a multicentre study of rh-Epo to reduce NCD in CABG patients.