Cherisse Baldeo1, Professor Mark Denton2, Patrick O’Kelly3, Abdul Ali1

Author affiliations
1RCSI medical students
2Consultant Nephrologist, Beaumont Hospital, Dublin
3Statistician, Beaumont Hospital, Dublin

Royal College of Surgeons in Ireland Student Medical Journal. 2011;4(1):10-15.

Cytomegalovirus (CMV) seropositivity has been reported to be a major determinant of erythropoietin-stimulating agent (ESA) resistance in patients with chronic kidney disease (CKD). It is hypothesised that prior CMV infection induces significant changes in T-cell subtypes that promote bone marrow resistance to ESA. We examined whether CMV or Epstein-Barr virus (EBV) serostatus influenced haemoglobin (Hb) levels and/or ESA resistance in our population of CKD stage 5 patients. Data on CMV and EBV serology, age, sex and Hb was collected on 1,417 patients presenting for a renal transplant from 2000 to 2009 in Ireland. Patients were split into four groups (CMVneg/EBVneg, CMVpos/EBVneg, CMVneg/EBVpos, and CMVpos/EBVpos) and analysis of variance was performed to examine whether CMV and EBV serostatus influenced Hb levels pre transplantation. Data was then collected on 117 patients currently on the transplant pool from Beaumont and Tallaght Hospitals. CMV and EBV serostatus, Hb, ESA dose and other parameters associated with ESA responsiveness (iron studies, B12/folate, albumin, PTH, diabetes, vascular access, dialysis adequacy) were collected. Multivariate analysis was performed to determine factors associated with increased ESA dosage. CMV positivity was found to have no effect on Hb levels or ESA dosage. Likewise, EBV serostatus had no effect on these parameters. First, analysis of 1,417 patients showed no difference in mean Hb between the various CMV/EBV serostatus groups. Second, analysis of patients currently on the transplant pool also showed that there was no difference in mean Hb between CMV and EBV groups. Analysis of a subpopulation of haemodialysis patients alone showed that CMV positivity was associated with a higher mean Hb and a lower mean ESA dose. Our results contrast with those of a recent report by Betjes et al. linking ESA resistance to CMV seropositivity. No association was found between CMV or EBV serostatus and Hb levels or ESA dosage. CMV seropositivity is not associated with increased ESA requirements in our population. These conflicting results may be due to differences in patient demographics and a lower target Hb level in our study.

Keywords: Haemoglobin, erythropoietin, CMV, EBV, renal disease.

Anaemia associated with chronic kidney disease (CKD) is a common problem that is associated with reduced quality of life and increased morbidity and mortality.1 The main cause is erythropoietin (Epo) deficiency. Epo is secreted by the peritubular capillary cells of the kidney and its production decreases when glomerular filtration rate (GFR) falls below approximately 40ml/min. Other causes of anaemia in CKD are decreased red blood cell (RBC) survival, bone marrow dysfunction in the setting of uraemia, increased RBC loss from uraemic gastritis and blood loss associated with the dialysis procedure. The main treatment is administration of erythropoiesis-stimulating agents (ESAs) and ensuring adequate iron stores. ESAs are similar to the naturally occurring protein Epo, which stimulates RBC formation (erythropoiesis). Available ESAs include Epo, epoetin alfa (Procrit/Epogen), epoetin beta (NeoRecormon) and darbeopoetin (Aranesp).2,3 However, patients differ in their sensitivity to ESA treatment, and in those that are hyporesponsive, a larger dose is required to maintain the haemoglobin (Hb) at an adequate level.4 There is controversy over the optimal Hb target, but it is usually taken to be 11-12g/dL. Of note, complete correction of anaemia to Hb levels higher than these values may be associated with increased cardiovascular risk.5,6

There are a number of factors that cause Epo hyporesponsiveness, including iron deficiency, systemic inflammation, hyperparathyroidism, vitamin B12 or folate deficiency, and co-morbidities such as diabetes, ongoing blood loss and inadequate dialysis.7

There has been recent evidence that prior exposure to cytomegalovirus (CMV) infection resulting in seropositivity to CMV is associated with resistance to Epo treatment in CKD patients.8 CMV is a herpes DNA virus that causes infectious mononucleosis and is commonly acquired in the teenage years. Often, the infection is asymptomatic or associated with a mild flu syndrome. The prevalence of seropositivity increases with the age of the population, such that approximately 50% of 50 year olds are seropositive.9 It does not cause significant clinical problems unless a patient becomes immunosuppressed such as in renal transplant patients. CMV infection in immunocompromised individuals can be life threatening, such that reactivation of the virus can cause CMV retinitis and pneumonitis. Furthermore, it is associated with distinct changes in peripheral T-cell populations. Betjes et al. have shown that CKD patients with positive CMV serology required significantly more Epo to achieve target Hb values (median 12,000 vs. 6,300 units/week; p=0.02). They provide evidence that this Epo-resistant state is due to a CMV-mediated change in peripheral T-cell subsets. Prior CMV infection is associated with expanded CD4+ CD28 null T-cells that are pro-inflammatory and mediate bone marrow resistance to the effects of Epo.10

This study examined if there was an association between CMV serostatus and Epo responsiveness in our CKD population. A retrospective analysis of 1,417 CKD patients presenting for renal transplantation was conducted to determine whether CMV serostatus correlated with Hb values prior to transplantation. Thereafter, a detailed analysis of 117 patients on the transplant waiting list in Beaumont and Tallaght Hospitals was performed to determine factors associated with Epo resistance in a multivariate analysis. In addition, we included serostatus for Epstein-Barr virus (EBV) infection in these patients as we hypothesise that infection with EBV may have a similar effect on Epo requirement. EBV is also a herpes virus like CMV and any findings in this area would be novel.

Hospital data for the 1,417 patients who were transplanted in Ireland over the past 10 years (2000-2009) were collected. This information was obtained from hospitals across Ireland. No confidential patient details such as names were recorded and therefore ethics approval was not required before data collection. The data was then used in a retrospective analysis to determine whether CMV and EBV seropositivity affected Hb levels. The patients in our list were divided into four subgroups according to their serostatus, namely CMV+EBV+, CMV+EBV-, CMV-EBV+, and CMV-EBV-. Analysis of variance (ANOVA) was used to examine the mean Hb level for the various EBV and CMV status groups, and so a comparison was made to determine if the presence of CMV or EBV infection had an effect on Hb levels. The significance of other confounding factors such as age at transplantation, sex of patient and the year of transplant were also determined for these patients. Next, a sub-analysis of the 117 patients on the renal transplant list from Beaumont and Tallaght Hospitals was conducted. The criteria for selection were for the patient to have been on the transplant list from January 2010 and to be a patient at either of these hospitals. These patients in the transplant pool were selected for study, as a patient’s CMV and EBV serostatus are now routinely determined prior to being put on the transplant list. Patient information was collected using case notes, referral letters and hospital databases (Clinical Vision and PIPE for Beaumont Hospital renal patients). The data collected was then categorised and entered into an electronic database.

First, the mean Hb levels for CMV+, CMV-, EBV+ and EBV- patient groups were calculated and differences among these values were compared. Second, the different types of Epo used by these patients and the popularity of each were examined. Analysis of Epo dose required the use of the conversion system from either ESA (Table 1). Darbepoetin alfa (Aranesp) and epoetin beta (NeoRecormon) doses were quoted weekly, while Mircera was given monthly. A standardised scoring system was derived to allow comparison of differences in dosage of the various Epo types (Table 2). The validity of results derived from using this system was verified by standardising Aranesp dose from micrograms into units in the same manner as Betjes et al.7 This was done by multiplying the Aranesp dose by 200. Third, a multivariate analysis was performed using the patient data and these standardised doses to investigate the effect CMV and EBV serostatus has on Epo dose and Hb concentration, while simultaneously looking at the effects of other key variables known to cause Epo hyporesponsiveness. The specific variables included were serum iron, vitamin B12 and folate levels, serum ferritin, transferrin saturation, serum albumin, diabetic status, parathyroid hormone (PTH) levels and the use of an angiotensin II receptor blocker (ARB) or ACE inhibitors (ACEIs).7 This enabled a comparison of the extent of influence each variable had on Epo dose and Hb levels.

Table 1. (Click image to enlarge.)

Table 2. (Click image to enlarge.)

The retrospective analysis showed that there were no significant differences in Hb levels among the various groups. Overall, the CMV-EBV- group reported high mean Hb levels, while the CMV+EBV- had the lowest mean Hb (Table 3). This difference in Hb levels between the seropositive and the seronegative patients was not significant when compared to the other confounders. However, the variables that were found to have significant effects on Hb levels were the gender of the patient and the year of the transplant (Table 4).

Table 3. (Click image to enlarge.)

Table 4. (Click image to enlarge.)

The clinical and demographic data for the 117 patients on the transplant pool are outlined in Table 5. Of the 117 patients, 43 were female and 74 were male. Mean Hb was slightly higher in the CMV+ group (11.36g/dL) compared to the CMV- group (11.30g/dL). Aranesp proved to be the most popular Epo used by the patients.

Table 5. (Click image to enlarge.)

In the transplant pool, there were more people with diabetes in the CMV+ group (16.7% compared to 7.8%). In addition, the patients in this study were younger than those in Betjes’ study (mean age 47.3 years in the CMV+ group and 49.8 in the CMVgroup, compared to 57.4 years in Betjes’ haemodialysis patients). Albumin levels were roughly the same in both groups (CMV- 40.16g/dL; CMV+ 40.09g/dL). Serum iron was higher in the CMV- group (15μg/dL), and ferritin was higher in the CMV+ group (434ng/ml). The mean Epo score (Table 2) for the CMV+ group was 0.44 higher than the CMV- group. Of note, analysis of these patients also showed no difference in Hb levels among the various serostatus groups. The EBV patients had the highest mean Hb level of 11.57g/dL, but it was not statistically significant (Figure 1).

FIGURE 1: Relationship between patient serostatus and mean haemoglobin levels showed no difference in haemoglobin levels among the various serostatus groups. (Click to enlarge.)

Separate analysis conducted for patients on haemodialysis (excluding patients on peritoneal dialysis) showed that the mean Hb was higher in the CMV+ group (11.44g/dL) than in the CMV- group (10.55g/dL, p=0.016) (Table 6). Mean Hb for patients not on Epo was higher than for those not on treatment (CMV+ patients: 12.50g/dL vs. 11.15g/dL; CMV- patients: 12.38g/dL vs. 10.77). CMV+ patients in both categories had higher Hb levels than CMVpatients (Table 7).

Table 6. (Click image to enlarge.)

Table 7. (Click image to enlarge.)

Patients not on Epo therapy and those receiving Mircera treatment were excluded, as they were few in number and would lower accuracy. The Epo doses of the remaining patients were then standardised by Betjes’ method and this data was represented on a box plot (Figure 2).10 This showed that the CMV- patients required a higher median Epo dose than CMV+ patients, 8000U/wk and 6000U/wk, respectively. Univariate analysis showed that ferritin was the only variable that had a significant effect on Epo dose (p=0.0253), while the multivariate analysis showed that serum iron had a significant effect on erythropoietin dose (p=0.0441) (Table 8).

FIGURE 2: Box-plot of erythropoietin doses standardised by Betjes’ method. CMV- patients required a higher median erythropoietin dose of 8000U/wk in comparison to the CMV+ patients, who required 6000U/wk. (Click to enlarge.)

Table 8. (Click image to enlarge.)

This study examined 1,417 CKD patients transplanted in Ireland over the past ten years and found that neither their CMV nor EBV serostatus affected their mean Hb levels at the time of transplantation. The sex of the patient and the year of transplantation had an effect on Hb levels. It is well known that women have lower Hb levels, and any effect the year of transplantation might have may in fact represent changes in clinical practice, such as different target Hb in current practice. The lack of an effect of CMV serostatus on Hb is similar to the findings by Betjes et al.10 However, any potential effect of CMV serostatus on Hb values may be masked by the fact that most CKD patients are receiving ESA agents and clinical management involves increasing the dose of ESA to achieve Hb levels within the target range. Betjes et al. concluded that CMV seropositivity affects anaemia management in CKD by significantly decreasing the bone marrow responsiveness to ESA, and showed that patients who are CMV+ require significantly higher doses of ESA agents.10 Moreover, Betjes et al. examined a subpopulation of patients not on ESA therapy and found that CMV+ patients were on average more anaemic. In this study, CMV serostatus did not affect responsiveness to ESA agents in the CKD population and these results were not in agreement with Betjes et al.’s findings. Analysis of patients on haemodialysis shows that the mean Hb is higher in CMV seropositive patients (11.44g/dl) than in the CMV seronegative group (10.55g/dl). However, CMV+ patients received a lower mean dose of ESA. This is a novel finding, as previous studies found that CMV seropositivity leads to higher ESA requirements.10 In the multivariate analysis, the only factor positively associated with ESA dose was serum iron, and it is well known that iron deficiency causes ESA hyporesponsiveness.The marked difference in the results here may possibly be due to differences in patient demographics between study populations. Moreover, the mean Hb in our patient population was lower than that in Betjes et al.’s study. Overall, our patient population received a lower mean dose of ESA (mean 6871.9U/wk for haemodialysis patients) compared to those in Betjes et al.’s study (10,399U/wk). This may be because the target Hb level at our institutions is lower, or perhaps the effect of CMV serostatus on ESA responsiveness only manifests when attempting to achieve a higher Hb level in these patients. The mean age of our population was younger than those studied by Betjes et al. (47.3 years vs. 57.4 years). This may be an important factor because the effect of CMV on T-cell subpopulations (enhanced CD4+, CD28- population) is age dependent. We also had a lower percentage of patients with diabetes in our study, and there was a difference in the percentage of patients with diabetes between groups. This may be important because of the influence of diabetes on anaemia. Of note, renal impairment due to microvascular complications of diabetes could lead to impaired Epo production and, ultimately, anaemia.11 Additionally, the patients on the transplant list were perhaps in better health by virtue of being younger than those with end-stage renal disease included in Betjes et al.’s study.

The inconsistency between the results presented by Betjes et al. and those in this study demonstrates that the relationship between CMV serostatus and Epo dose is still unclear. Further large-scale studies should be conducted to definitively determine the association between patient CMV serostatus and required Epo dose, if any. Currently, patients with renal disease may pay thousands annually for Epo treatment, and costs can be as much as double that paid for the alternative treatment of RBC transfusions.12 Careful control and surveillance of influencing factors, such as glucose control in patients with diabetes and prevention of iron deficiency, may help to lower Epo requirements, and thereby decrease the cost for patients.

I would like to thank Dr Mark Denton for all his guidance, as well as Mr Patrick O’Kelly, Abdul Ali, Susan Spencer and the dialysis nurses at Beaumont and Tallaght Hospitals for all their help during this research project.


  1. Weiss G, Goodnough LT. Anaemia of chronic disease. N Engl J Med.2005;352(10):1011-23.
  2. Ashley C, Morlidge C (eds.). Introduction to Renal Therapeutics. London:The Pharmaceutical Press, 2008:45-54.
  3. Eschbach JW, Kelly MR, Haley NR, Abels RI, Adamson JW. Treatment of the anaemia of progressive renal failure with recombinant human erythropoietin. N Engl J Med. 1989;321(3):158-63.
  4. Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW. Correction of the anaemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial. N Engl J Med. 1987;316(2):73-8.
  5. Drueke T. Hyporesponsiveness to recombinant human erythropoietin. Nephrol Dial Transplant. 2001;16(Suppl. 7):25-8.
  6. Courtney AE, Maxwell AP. Critiques of clinical guidelines in nephrology: anaemia. Nephron Clin Pract. 2008;110(2):c115-25.
  7. Singh AK, Szczech L, Tang KL, Barnhart H, Sapp S, Wolfson M et al.; CHOIR Investigators. Correction of anaemia with epoetin alfa in chronic kidney disease. N Engl J Med. 2006;355(20):2085-98.
  8. Kanbay M, Perazella M, Kasapoglu B, Koroglu M, Covic A. Erythropoiesis stimulatory agent-resistant anaemia in dialysis patients: review of causes and management. Blood Purif. 2010;29(1):1-12.
  9. Kumar P, Clark M (eds.). Clinical Medicine (6th ed.). London: Elsevier Saunders, 2005:680.
  10. Betjes MGH, Weimar W, Litjens NHR. CMV seropositivity determines epoetin dose and haemoglobin levels in patients with CKD. J Am Soc Nephrol. 2009;20(12):2661-6.
  11. Dikow R, Schwenger V, Shömig M, Ritz E. How should we manage anaemia in patients with diabetes? Nephrol Dial Transplant. 2002;17(Suppl. 1):67-72.
  12. Fain J, Hilsenrath P, Widness JA, Strauss RG, Mutnick AH. A cost analysis comparing erythropoietin and red cell transfusions in the treatment of anaemia of prematurity. Transfusion. 1995;35(11):936-43.

Back to list of articles.