Research

Erythropoietin mimics the acute phase response in critical illness

John Michael Elliot¹ , Tanit Virankabutra², Stephen Jones³, Surasak Tanudsintum⁴, Graham Lipkin⁵, Susan Todd⁶ and Julian Bion⁷

¹Research Fellow, University Department of Anaesthesia and Intensive Care, Queen Elizabeth Hospital, Birmingham, UK

²Research Fellow, University Department of Anaesthesia and Intensive Care, Queen Elizabeth Hospital, Birmingham, UK

³Consultant in Clinical Chemistry, Department of Biochemistry, Queen Elizabeth Hospital, Birmingham, UK

⁴Research Fellow, University Department of Anaesthesia and Intensive Care, Queen Elizabeth Hospital, Birmingham, UK

⁵Consultant in Renal Medicine, Department of Nephrology, Queen Elizabeth Hospital, Birmingham, UK

⁶Medical Statistician, Medical and Pharmaceutical Statistics Research Unit, University of Reading, Reading, UK

⁷Senior Lecturer in Intensive Care Medicine, University Department of Anaesthesia and Intensive Care, Queen Elizabeth Hospital, Birmingham, UK

author email corresponding author email

Critical Care 2003, 7:R35-R40doi:10.1186/cc2185

Published:	24 April 2003

Abstract

Background

In a prospective observational study, we examined the temporal relationships between serum erythropoietin (EPO) levels, haemoglobin concentration and the inflammatory response in critically ill patients with and without acute renal failure (ARF).

Patients and method

Twnety-five critically ill patients, from general and cardiac intensive care units (ICUs) in a university hospital, were studied. Eight had ARF and 17 had normal or mildly impaired renal function. The comparator group included 82 nonhospitalized patients with normal renal function and varying haemoglobin concentrations. In the patients, levels of haemoglobin, serum EPO, C-reactive protein, IL-1β, IL-6, serum iron, ferritin, vitamin B₁₂ and folate were measured, and Coombs test was performed from ICU admission until discharge or death. Concurrent EPO and haemoglobin levels were measured in the comparator group.

Results

EPO levels were initially high in patients with ARF, falling to normal or low levels by day 3. Thereafter, almost all ICU patients demonstrated normal or low EPO levels despite progressive anaemia. IL-6 exhibited a similar initial pattern, but levels remained elevated during the chronic phase of critical illness. IL-1β was undetectable. Critically ill patients could not be distinguished from nonhospitalized anaemic patients on the basis of EPO levels.

Conclusion

EPO levels are markedly elevated in the initial phase of critical illness with ARF. In the chronic phase of critical illness, EPO levels are the same for patients with and those without ARF, and cannot be distinguished from noncritically ill patients with varying haemoglobin concentrations. Exogenous EPO therapy is unlikely to be effective in the first few days of critical illness.