Anemia: A Continuing Problem Following Kidney Transplantation

American Journal of Transplantation 2003; 3: 1426–1433 Blackwell Munksgaard C Blackwell Munksgaard 2003 Copyright  ISSN 1600-6135 doi: 10.1046/j.1600-6135.2003.00224.x Anemia: A Continuing Problem Following Kidney Transplantation T. Christian H. Mixa , Waqar Kazmia , Samina Khana , Robin Ruthazerb , Richard Rohrerc , Brian J. G. Pereiraa and Annamaria T. Kausza, ∗ Divisions of a Nephrology and b Clinical Care Research, Department of Medicine, and c Transplantation, Department of Surgery, Tufts-New England Medical Center, Boston, MA ∗Corresponding author: Annamaria T. Kausz, [email protected] Cardiovascular disease is a leading cause of death among kidney transplant recipients. Anemia, a risk factor for cardiovascular complications among patients with chronic kidney disease, has not been well characterized in kidney transplant recipients. We performed a retrospective cohort study of the prevalence of and factors associated with anemia among 240 patients who underwent kidney transplantation at our institution. The mean hematocrit (Hct) rose from 33% at 1 month after transplantation to 40% at 12 months after transplantation. The proportion of patients with Hct < 36% was 76% at transplantation and 21% and 36%, 1 year and 4 years after transplantation, respectively. Six months after transplantation, women had higher likelihood (OR = 3.61) of Hct < 36%, while higher Hct at 3 months (OR = 0.67 for 1% higher Hct) and diabetes (OR = 0.14) were associated with a lower likelihood of Hct < 36%. Similar associations were seen 12 months after transplantation. Even among patients with Hct < 30%, only 36% had iron studies, 46% received iron supplementation and 40% received recombinant human erythropoietin. Awareness of factors associated with a lower Hct may prompt better anemia screening and management, potentially improving cardiovascular outcomes among kidney transplant recipients. Key words: Anemia, chronic kidney disease, erythropoietin, glomerular filtration rate, kidney, transplant Received 14 March 2003, revised 12 May and accepted for publication 5 June 2003 Introduction The glomerular filtration rate (GFR) among kidney transplant recipients is consistently less than normal physiologic ranges for men and women. Yet, it was only the recent in1426 clusion of kidney transplant recipients in the National Kidney Foundation (NKF) Kidney Disease Outcomes Quality Initiative (K/DOQI) Guidelines for Chronic Kidney Disease (CKD) (1) that brought into focus the fact that transplant recipients are also patients with CKD. Evidence is emerging to suggest that complications of CKD are common among these patients (2–4), and that the management of these complications is less than optimal (5). We have recently shown that adult patients returning to dialysis after a failed kidney transplant have levels of hematocrit (Hct), serum albumin and recombinant human erythropoietin (rHuEPO) use before initiation of dialysis (5) that are not significantly different from those seen in the general incident end-stage renal disease (ESRD) population (6). These indices of poor care are manifest despite presumed regular medical care from nephrologists and transplant physicians. Complications of CKD adversely affect morbidity and mortality in patients with CKD. Anemia has been associated with cardiovascular complications among nontransplant patients with CKD, including left ventricular (LV) hypertrophy and LV growth (7,8). Lower Hct and hemoglobin have been associated with higher risk of hospitalization in patients with CKD (9,10), and higher risk of mortality in patients with decreased GFRs and decreased LV function (11). Associations between anemia, CHF, LV hypertrophy and mortality have also been identified in kidney transplant recipients (12,13). The importance of these findings may be underscored by the fact that greater than 42% of kidney transplant recipients die with a functioning graft (4), and the majority of these deaths (42%) result from cardiovascular disease. Anemia is a common and early complication of CKD. Among patients who initiated dialysis in the United States between 1995 and 1997, 51% had a Hct less than 28%, and 67% had a Hct less than 30% (6). A study of patients with CKD seen in nephrology clinics demonstrated that 45% of the patients with serum creatinine levels less than 2 mg/dL had a Hct less than 36% (14). There are, however, limited data regarding the prevalence of anemia among kidney transplant recipients (15,16). We undertook this retrospective longitudinal cohort study to characterize changes in Hct following kidney transplantation and to identify factors associated with those changes. Improved understanding of the natural history of anemia among kidney transplant recipients and risk factors associated with anemia may enable better timing Anemia in Kidney Transplant Recipients and targeting of anemia-directed therapies, with the goal of ameliorating cardiovascular disease in this high-risk population. Methods coexisting disease, to 3 = uncontrolled condition that causes moderate-tosevere disease manifestations during medical care. Each patient was also assigned a global IDS score, which represents the maximum score assigned among the 20 domains. Ischemic heart disease, nonischemic heart disease and HTN were recorded separately as present if the respective IDS scores were >0. Patient population Analytical methods Subjects for this retrospective longitudinal cohort study included all patients aged 18 years or older who underwent kidney transplantation at Tufts-New England Medical Center between October 1, 1990 and September 30, 1999. Post-transplant data collection was extended through September 30, 2000, to allow a minimum 1-year follow-up period for all patients. Data were obtained from the time of transplantation until the end of the study period, the occurrence of graft failure, transfer of care to a different institution, death or loss to follow up. Descriptive analyses: Patient demographic, clinical and laboratory variables were reported as mean and standard deviation (SD) for continuous variables and proportions for categorical variables. Demographic variables among patients who were lost to follow up or transferred their care were compared with those of the remaining cohort to investigate whether bias was introduced as a result of their loss from the cohort. Data Data were abstracted by physician investigators from hospital charts, nephrology and transplant surgery out-patient clinic charts and Clinical Information Systems (CIS), a computerized patient database that contains laboratory results, hospitalization and out-patient clinic data, and other hospitalbased care data. Data were directly entered into an electronic database designed for this study. Baseline data at the time of kidney transplantation included (1): patient demographic information such as age at transplantation, gender and race (2), clinical information such as primary cause and duration of end-stage renal disease (ESRD), comorbid conditions/index of disease severity (IDS) score, modality of renal replacement therapy before transplantation, history of prior kidney transplant, and (3) transplant-related information such as type of kidney donor, histocompatability and cross match data, viral serology, use of induction immunosuppression with muromonab-CD3 (OKT3), antithymocyte globulin (ATG) or an IL-2 receptor antagonist (IL-2RA), requirement for perioperative blood transfusion and occurrence of delayed graft function. Medication use and laboratory data were recorded when available from patient charts. Longitudinal data were collected monthly for the first 6 months after transplantation and then every 3 months for up to 60 months after transplantation. Data collection at each point included outpatient blood pressure, weight and laboratory data, medication use, and hospitalization data, including occurrence and treatment of acute rejection. The reason for the last follow up was recorded. If a laboratory value was not available coincident with the visit, the value closest to the visit within ±15 days was used during the first 6 months of follow up, and the value closest to the visit within ±45 days was used beyond 6 months after transplantation. Data collection terminated when a study end-point was reached. Data categorization and definitions Kidney function was expressed as GFR, estimated with an equation derived from the Modification of Diet in Renal Disease (MDRD) Study, which includes age, gender, race and serum creatinine (17,18), and has been validated for use in the kidney transplant population (19,20). Serum creatinine levels at T-NEMC were calibrated to Cleveland Clinic levels (where serum creatinine used in the development of the MDRD equation was measured) by subtracting 0.2 mg/dL, the mean difference between the values obtained at T-NEMC and the Cleveland Clinic labs. Episodes of acute rejection were identified from hospitalization records. Confirmation of the diagnosis of acute rejection by kidney transplant biopsy was documented if a biopsy was performed. Co-morbidity was assessed using the Index of Disease Severity (IDS), as described by Greenfield and colleagues (21), and later modified by Athienites and colleagues (22) for use in ESRD patients. The modified IDS assigns one of four severity levels to 20 disease domains; from 0 = absence of American Journal of Transplantation 2003; 3: 1426–1433 Hematocrit values were determined at the following specified time intervals after transplantation: 1, 3, 6, 12, 24, 36, 48 and 60 months. Given the variability in Hct values, the representative Hct for a given time interval was taken as the mean of visit-related Hct values around each time interval as follows: Hct at 1 month after transplantation was taken as the mean of visitrelated Hct values recorded for the first 1.5 months after transplantation, Hct at 3 months after transplantation was taken as the mean of visit-related Hct values recorded for months 1.5 through 4.5 after transplantation, Hct at 6 months after transplantation was taken as the mean of visit-related Hct values recorded for months 4.5 to 9 after transplantation, Hct at 12 months and each subsequent year after transplantation was taken as the mean of visit-related Hct values recorded within the 3-month intervals immediately before and after each year of transplantation. The representative estimated GFR at 3, 6 and 12 months after transplantation was similarly taken as the mean of estimated GFR values determined during the time intervals described earlier. Patients were then assigned to a Hct category (<30%, 30 to <33%, 33 to <36% and ≥36%) within each time interval, reflecting varying degrees of severity of anemia. The categories were based partly on the Medicare-determined threshold of Hct (<30%) for initiation of rHuEPO in the Medicare-eligible population, and on the K/DOQI target Hct range of 33–36% for patients with CKD (23). The proportion of patients within each Hct category was determined for each time interval after transplant. The 95% confidence interval (CI) for each proportion was determined using the normal approximation to the binomial, bounded at 0% and 100%. Changes in the distribution of Hct across different levels of kidney function were illustrated by first categorizing patients by stage of CKD according to the K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease (1), using the mean GFR at 6 and 12 months after transplantation. The proportion and 95% confidence interval (CI) of patients with Hct < 30%, 30 to <33%, 33 to <36% and ≥36% was then determined within each CDK stage. Glomerular filtration rate and Hct data from clinic visits that coincided with rHuEPO use were excluded from these analyses. The proportion of all patients in the cohort who underwent testing for iron deficiency, and the proportion of all patients who received iron supplementation and/or rHuEPO, was determined. These proportions were also determined among the subsets of patients with a lowest recorded Hct < 33% and <30%, selected as two thresholds likely to prompt investigation of and treatment for anemia. Factors associated with anemia at 6 and 12 months after transplantation We explored three different thresholds for anemia: Hct < 36% (mild anemia), Hct < 33% (moderate anemia) and Hct < 30% (severe anemia) at 6 and 12 months after kidney transplantation. We chose to investigate factors associated with anemia at 6 and 12 months after transplantation because 1427 Mix et al. this appears to be the time of fullest possible recovery of erythropoiesis (24). It was also presumed that the post-transplant course would have stabilized after 6 to 12 months. Univariate associations (p < 0.05) between anemia and clinical factors likely to be associated with anemia were identified with Chi-square testing for categorical variables and t-tests for continuous variables. Odds ratios for these univariate associations were determined with logistic regression. Variables with significant univariate associations with anemia were then incorporated into multivariate logistic regression models utilizing a stepwise selection process. We decided a priori to adjust the model for age, gender and race, regardless of the significance of the univariate association. Race was described as a three-level categorical variable with Caucasian being the referent group. We strived for a target ratio of 10 outcomes per variable in the model. No attempt was made to impute missing data. Given the anticipated strength of the association between Hct in the preceding time interval and anemia, regression models were created that excluded Hct in the preceding time interval to facilitate the identification of other clinical factors associated with anemia. Steps were taken to ensure that values for explanatory variables used in regression models predated Hct values that were used to identify a patient as anemic. As such, models of anemia at 6 months after transplantation used values for explanatory variables recorded before 4.5 months after transplantation. Similarly, models of anemia at 12 months after transplantation used values for explanatory variables recorded before 9 months after transplantation. Odds ratios were reported with 95% CI. All statistical analyses were performed using SAS System for Windows, version 8.02 (SAS Institute Inc., Cary, NC). Table 1: Baseline characteristics of kidney transplant recipients at Tufts-New England Medical Center between 1990 and 1999 compared with the national population Patient characteristic Age at transplantation (years) 18 to <45 45 to <65 ≥65 Female Race Caucasian Black Asian Other Cause of ESRD Glomerulonephritis Diabetes Hypertension Polycystic kidney disease Other Type of kidney donor Cadaveric Living Previous kidney transplant T-NEMC (n = 240) USRDSa (n = 109,127) 45% 48% 7% 38% 52%b 43% 5% 40% 81% 8% 4% 7% 71% 23% 4% 2% 42% 17% 13% 10% 18% 23% 22% 15% 7% 33% 51% 49% 5% 74% 26% 11% T-NEMC = Tufts-New England Medical Center; USRDS = United States Renal Data System; ESRD = end stage renal disease. a Averages of kidney transplant recipients spanning 1988–1998 per USRDS Annual Data Report; recipients age ≥20 years. b Represents the USRDS patient age range of 20 to <45 years. Results Patient population Two hundred and forty patients underwent kidney transplantation between October 1, 1990 and September 30, 1999. Table 1 shows the demographic and clinical characteristics of the cohort and compares it with national averages. The mean age of the recipients at transplantation was 45 years, 38% were females and 81% were Caucasians. The most common cause of kidney failure was glomerulonephritis (42%) followed by diabetes mellitus (17%). Hemodialysis was the most common mode of renal replacement therapy among patients on dialysis at the time of transplant. Twenty-three patients (10%) received a preemptive kidney transplant. Almost half of the kidney transplants came from living donors. Twelve (5%) patients had undergone a prior kidney transplant. Human leukocyte antigen data were available for 228 patients in the cohort. Of these, 124 (54%) had 0–3 human leukocyte antigen (HLA) mismatches with their donor and 104 (46%) had 4–6 HLA mismatches. Subjects were evenly distributed across the four possible donor– recipient combinations of CMV serologic status, with 26%, 25%, 25% and 23% in donor (D)+/recipient (R)–, D+/R +, D–/R– and D–/R+ categories, respectively. Induction therapy was administered to 38 (16%) patients. Ninety-one percent of the subjects were taking cyclosporine at 1 month 1428 after transplantation and 9% were taking tacrolimus. Seventy percent of the subjects were taking azathioprine at 1 month after transplantation, 25% were taking mycophenolate mofetil and 5% had no record of receiving an antimetabolite. Follow up The mean ± SD and median (range) duration of followup of patients in the study was 41 ± 30 months and 35 (1–118 months) months, respectively. Seventeen patients received kidney transplants but no kidney transplant follow up: 13 patients had either primary nonfunction or an immediate postoperative complication (e.g. vascular rejection) that resulted in transplant nephrectomy within the first month, three patients died during the immediate postoperative period and one died shortly after discharge with graft function. One hundred and forty-six (64%) patients were followed through the end of the study period with graft function. Of the remaining patients in the cohort, 24 (11%) returned to dialysis, 13 (6%) died, 20 (9%) transferred care out of our institution and 20 (9%) were lost to follow up. Acute rejection occurred in 77 patients (32% of the cohort). The baseline demographic characteristics of the 40 patients who either transferred care out of our institution or were lost to follow up were not significantly different from those of the remainder of the cohort (data not shown). American Journal of Transplantation 2003; 3: 1426–1433 Anemia in Kidney Transplant Recipients 100% Proportion of patients 90% 80% Hematocrit 70% ≥ 36% 33% − < 36% 30% − < 33% < 30% 60% 50% 40% 30% 20% 10% 0% (Mean Hct) Figure 1: Changes in hematocrit levels after kidney transplantation. [n] 0 1 3 6 12 24 36 48 60 (33%) (33%) (37%) (39%) (40%) (39%) (38%) (38%) (38%) [240] [223] [221] [211] [196] [150] [106] [78] [54] Months after transplantation Prevalence of anemia Hematocrit levels were available for 94% of patients at 1-month follow-up and for 95–100% of patients at subsequent follow-up intervals. The mean Hct and the distribution of Hct levels at each time point during follow up are shown in Figure 1. At the time of transplant surgery, 22% (95% CI 16%, 28%) of the cohort had Hct < 30%. The mean Hct rose from its nadir of 33% (95% CI 27%, 39%), both at the time of transplantation and at 1 month after transplantation, to a peak of 40% (95% CI 34%, 46%) 12 months after transplantation, and declined thereafter. During the first year after transplantation, the proportion of patients with Hct < 36% steadily decreased from 76% (95% CI 70%, 82%) at transplantation to 21% (95% CI 15%, 27%) at 12 months after transplantation. The proportion of patients with Hct < 33% declined over a shorter time interval, from 48% (95% CI 42%, 54%) at transplantation to 7% (95% CI 3%, 11%) at 6 months after transplantation. Factors associated with anemia at 6 months after transplantation Female gender was independently associated with a higher likelihood of Hct < 36% at 6 months after transplantation, as shown in Table 2. Factors independently associated with lower likelihood of Hct < 36% included diabetes as the cause of native kidney failure and higher Hct at 3 months after transplantation. When Hct at 3 months after transplantation was omitted from the model, CMV D+/R– disease and female gender remained as factors independently associated with greater likelihood of anemia. Beyond 1 year after transplantation, the proportion of patients with Hct < 36% increased from 21% (95% CI 15%, 27%) at 12 months to 36% (95% CI 26%, 46%) at 4 years after transplantation. Likewise, the proportion with Hct < 33% increased from 7% (95% CI 3%, 11%) at 6 months after transplantation to 20% (95% CI 10%, 30%) at 4 years after transplantation. Factors associated with anemia at 12 months after transplantation Higher Hct and GFR at 6 months after kidney transplantation were independently associated with lower likelihood of Hct < 36% at 12 months after transplantation, as shown in Table 2. The omission of Hct at 6 months after transplantation from the model allowed female gender to emerge as independently associated with higher likelihood of Hct < 36%. Six and 12 months after kidney transplantation, the proportion of patients with some degree of anemia progressively increased with increasing stages of CKD, as shown in Figure 2. Among patients with GFR ≥ 90 mL/min/1.73 m2 , 11% (95% CI 0%, 25%) and 7% (95% CI 0%, 16%) had some degree of anemia at 6 and 12 months after transplantation, respectively, whereas among patients with GFR < 30 mL/min/1.73 m2 , 60% (95% CI 35%, 95%) and 76% (95% CI 53%, 99%) had some degree of anemia at 6 and 12 months after transplantation, respectively. American Journal of Transplantation 2003; 3: 1426–1433 Several factors were associated with higher likelihood of Hct < 33% at 6 months after transplantation in the univariate analysis. The limited number of outcomes (15 patients with Hct < 33%), however, precluded a multivariate analysis. Five patients had Hct < 30% at 6 months after transplantation. Several factors were associated with higher likelihood of Hct < 33% at 6 months after transplantation in the univariate analysis. The limited number of outcomes (18 patients with Hct < 33%) again precluded a multivariate analysis. Three patients had Hct < 30% at 12 months after transplantation. Other factors that were tested, but showed no significant association with anemia at either 6 or 12 months after 1429 Mix et al. 6 months after transplantation 12 months after transplantation 100% 90% 80% 70% Hematocrit 60% ≥ 36% 33% − < 36% 50% 30% − < 33% 40% < 30% 30% 20% 10% 0% (Mean Hct) [n] <30 30-<60 60-<90 (34%) [15] (39%) [85] (40%) [88] >/=90 (41%) [18] <30 30-<60 (35%) [13] 60-<90 (40%) [82] (42%) [76] >/=90 (41%) [28] GFR (mL/min/1.73 m2) Figure 2: Hematocrit levels at different stages of chronic kidney disease. Analysis includes 206 patients at the 6-month interval after kidney transplantation and 199 patients at the 12-month interval after kidney transplantation. Table 2: Factors associated with anemia after kidney transplantation Hct < 33% (n = 15/209) Hct < 36% (n = 52/209) 6 months after transplantation Univariate OR (95% CI) Univariate OR (95% CI) Multivariate w/Hcta OR (95% CI) Multivariate w/o Hctb OR (95% CI) Female CMV D+/R– vs. other DM vs. no DM HLA mismatche Induction therapy ATG vs. none Hct at 3 monthsf GFR at 3 monthsg ACEI in previous 4.5 months 7.09 (1.93, 25.96) 2.35 (1.19, 4.63) 0.27 (0.08, 0.93) 0.74 (0.56, 0.99) 7.18 (1.61, 31.97) 0.73 (0.62, 0.86) 0.62 (0.44, 0.86) 4.48 (1.53, 13.12) 3.07 (1.61, 5.86) 2.96 (1.39, 6.29) 0.14 (0.03, 0.62) 3.61 (1.60, 8.15) 4.03 (2.00, 8.09) 0.72 (0.65, 0.81) 0.83 (0.72,.97) 0.67 (0.59, 0.77) Not applicable Hct < 33% (n = 18/200) Hct < 36% (n = 43/200) 12 months after transplantation Univariate OR (95% CI) Univariate OR (95% CI) Multivariate w/Hctc OR (95% CI) Multivariate w/o Hctd OR (95% CI) Age at transplanth Female Hct at 6 monthsf GFR at 6 monthsg ACEI in pprevious 9 months 0.62 (0.42, 0.91) 2.63 (1.32, 5.23) 0.63 (0.51, 0.76) 0.52 (0.37, 0.73) 3.00 (1.12, 8.02) 3.89 (1.78, 8.50) 0.64 (0.55, 0.74) 0.70 (0.58, 0.84) 2.19 (1.07, 4.49) 0.65 (0.55, 0.77) 0.77 (0.62, 0.97) Not applicable 0.66 (0.54, 0.80) Hct = hematocrit; CMV D+/R– = cytomegalovirus donor positive/recipient negative versus other serologic combinations; DM = diabetes]mellitus; HLA = human leukocyte antigen; ATG = antithymocyte globulin; ACEI = angiotensin II converting enzyme inhibitor. a Models adjusted for age, gender, race, and include a variable for Hct at 3 months. b Models adjusted for age, gender, race, and exclude a variable for Hct at 3 months. c Models adjusted for age, gender, race, and include a variable for Hct at 6 months. d Models adjusted for age, gender, race, and exclude a variable for Hct at 6 months. e Odds per one additional antigen mismatch. f Odds per 1% point higher Hct. g Odds per 10 mL/min/1.73m2 higher GFR. h Odds per 10 years older at transplantation. 1430 American Journal of Transplantation 2003; 3: 1426–1433 Anemia in Kidney Transplant Recipients transplantation, included type of kidney donor (living vs. cadaveric), induction therapy, baseline immunosuppression, delay of graft function, acute rejection, type of treatment for acute rejection, use of rHuEPO, ACEI and/or ARB drugs, receipt of blood transfusion, smoking history, and burden of comorbid disease. over the next 5 months. While the prevalence of anemia appeared to be lowered by 6 and 12 months after transplant, 25% and 21% of patients remained anemic at these times, respectively. These results are consistent with what is known about the resumption of erythropoiesis in kidney transplant patients (24,27). Diagnosis and management of anemia Twelve percent of the cohort had evaluation for iron deficiency at any time during follow up, 18% received iron supplementation and 10% received rHuEPO. Among patients with the lowest Hct < 33%, 26% had iron studies, 36% received iron supplementation and 25% received rHuEPO. Among patients with the lowest Hct < 30%, 36% had iron studies, 46% received iron supplementation and 40% received rHuEPO. We identified several factors associated with anemia at 6 and 12 months after kidney transplantation. Women had 3–4-fold higher likelihood than men of Hct < 36% at 6 and 12 months after transplantation. One explanation may be the return of menses in age-appropriate females after successful transplantation, which may in turn potentiate anemia through iron deficiency. Studies of reproductive function following kidney transplantation have demonstrated that menstruation generally resumes in ageappropriate females within the first 6 months after transplantation (28,29), and may be seen as early as 6 weeks after transplantation (30). Furthermore, iron deficiency has been reported to be common in both male and female kidney transplant recipients. In a prospective cohort of kidney transplant recipients, 50% were found to have evidence of iron deficiency 2 weeks after transplantation (31), and patients who did not receive iron supplementation remained iron deficient and anemic at 6 months after transplantation. A recent cross-sectional analysis of 432 kidney transplant recipients demonstrated that iron deficiency was present in 20% (15). Alternatively, greater odds of Hct < 36% observed in women in our cohort may simply reflect physiologic differences in Hct for women and men (32). Discussion The findings of this study demonstrate that anemia was common following kidney transplantation. At least 70% of kidney transplant recipients had some degree of anemia in the first 6 months after transplantation, and up to one-third were anemic between 1 and 5 years after transplantation. There were transplant- and nontransplantrelated factors associated with anemia at 6 and 12 months after transplantation. The investigation and treatment of anemia were infrequent. The current study provides further insight into the problem of anemia in kidney transplant recipients and highlights the need for greater efforts to investigate and treat anemia in this population of patients with CKD. The importance of anemia among patients with CKD has been well documented. Higher hematocrit and hemoglobin levels are associated with decreased risk of hospitalization among patients with earlier stages of CKD (9,10). Anemia is also a risk factor for death in individuals with decreased kidney function and concomitant decreased left ventricular function (11). Associations between anemia and abnormal LV morphology among patients with chronic kidney disease are well established (7,8). Furthermore, treatment of anemia with rHuEPO has been associated with regression of LV hypertrophy in patients with CKD (25,26). Studies of the relationship between anemia and cardiovascular disease in the transplant population have demonstrated associations between anemia and LV hypertrophy (13) and anemia and the development of CHF (12). The high prevalence of cardiovascular mortality among kidney transplant recipients with graft function, reported to account for 36% of all deaths with graft function (4), heightens the concern that anemia may have significant adverse consequences for this group of patients. We found the prevalence of anemia was largely unchanged at 1 month after transplantation, as compared with the peri-transplant period. Erythropoiesis gradually recovered American Journal of Transplantation 2003; 3: 1426–1433 Cytomegalovirus D+/R– serology was associated with an almost threefold higher likelihood of mild anemia (Hct < 36%) at 6 months after transplantation, in a model that excluded Hct at 3 months after transplantation. Cytomegalovirus D+/R– serology could be a surrogate marker for CMV infection, as this is the serologic group at highest risk. Both CMV infection and prophylaxis against CMV infection have established bone marrow suppressive effects (33,34). We also noted 86% lower odds of Hct < 36% at 6 months after transplantation among patients with diabetes in the model that included Hct at 3 months after transplantation. A potential explanation would be more frequent exposure to physicians as a consequence of the need for diabetes care, in addition to kidney transplant care, as well as more rigorous evaluation and management of anemia given the higher risk of cardiovascular disease in this population. Lower Hct at 3 and 6 months after transplantation was consistently associated with higher likelihood of anemia at 6 and 12 months after transplantation, respectively, indicating that patients who are anemic at 3 and 6 months after transplantation are likely to remain anemic, at least over the next 3 to 6 months. This may suggest an opportunity to improve the approach to anemia management in these patients. 1431 Mix et al. Higher GFR at 3 and 6 months after transplantation was consistently associated with lower likelihood of anemia at 6 and 12 months after transplantation, respectively. This confirmed our expectation that better kidney function in kidney transplant recipients would be associated with more effective erythropoiesis. In addition, the prevalence of anemia at 6 and 12 months after transplantation appeared to increase among patients with a more advanced stage of CKD. These findings are corroborated by a recent analysis of data from the Third National Health and Nutrition Examination Survey (NHANES) (35), which demonstrated a higher prevalence of anemia among individuals with lower estimated GFRs. In the NHANES cohort, the mean Hct at each level of CKD was similar to what we observed in this kidney transplant cohort, suggesting a similarity in the relationship between the GFR and Hct among patients with CKD after kidney transplant and patients with nontransplant-related CKD. It is known that patients with failed kidney transplants return to dialysis with levels of Hct and rHuEPO use that are not significantly different from the general incident ESRD population, in spite of presumed regular medical care (5). It would be reasonable to expect that management of complications of CKD, during the time leading up to kidney transplantation, would be optimized to limit peri-transplant morbidity. Anemia in our cohort, however, was surprisingly common at the time of transplantation: nearly one-quarter of recipients had Hct < 30% and nearly half had Hct < 33%. It is conceivable that better attention to management of anemia before kidney transplantation could minimize not only the risk of adverse cardiovascular events but also the use of perioperative blood transfusions. Anemia management after transplantation also appeared to merit improvement. Anemia-targeted diagnostic and therapeutic interventions were used infrequently. Few patients in our cohort had iron studies, or received iron supplementation or rHuEPO, although the proportions of patients undergoing investigation and treatment of anemia did increase among patients with the lowest Hct levels. Earlier studies raising concerns that rHuEPO use may hasten the decline of the GFR may have limited its use, although these concerns have since been dispelled (36–38). Lastly, the era of our investigation precedes recent heightened attention to treating the complications of chronic kidney disease, resulting from the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (K/DOQI) (1). This study has limitations inherent in retrospective data. The small sample size of this study limits its ability to detect associations between anemia and various clinical factors. Larger studies are therefore needed to confirm risk factors for anemia identified by this study, and to provide further insight into other risk factors for anemia. Differences in patient demographics with respect to national data may limit the ability to generalize our results. While our cohort had a similar proportion of female recipients to national kidney transplant statistics, there was a higher proportion of Cau1432 casian and fewer Black recipients (39). A higher proportion of living donors and fewer patients with prior transplants in our cohort than in the national statistics may have resulted in better graft outcomes and thus a lower prevalence of anemia than that observed in the national transplant population. In addition, the cohort was comprised of patients receiving care at a single university hospital, and the approach to transplant care is not uniform across transplant centers. Finally, the number of patients in the study is relatively small, limiting its power. Despite these potential concerns, the analysis provides insight into trends in anemia after kidney transplantation, factors associated with anemia, the relationship between Hct and the GFR and details regarding the diagnosis and management of anemia. In conclusion, anemia was common after kidney transplantation and persisted in a substantial proportion of kidney transplant recipients. Female gender, lower GFR and lower Hct appear to be important predictors of anemia at 6 and 12 months after transplantation. The investigation and treatment of anemia was suboptimal during the timeframe of the study. This highlights a lack of attentiveness to risk factors that have potential for causing increased morbidity. Increased awareness, timely diagnosis and treatment of anemia after kidney transplantation could be an important strategy to improve cardiovascular outcomes in this highrisk population. Further studies are required to assess the impact of treatment on cardiovascular outcomes among kidney transplant recipients. Acknowledgments We acknowledge the contributions of Debbie Chabot, RN, and Drs Richard Freeman, Michael Angelis and Jeffrey Cooper of the Division of Transplantation, who provided us with access to the transplant surgery clinic records of patients in the study. We also acknowledge the nephrology research assistants Gowri Raman and Aarti Kalra, who assisted in data abstraction and database entry. This project was supported in part by Amgen, Inc., Thousand Oaks, CA, and grants from the National Institutes of Health (1F32DK59704-01 [Dr Mix], F32DK59704-01 [Dr Kazmi], and 1K08DK02745 [Dr Kausz]). Dr. Pereira serves on Amgen’s Aranesp (Medical Advisory Board). References 1. K/DOQI Clinical Practice Guidelines for Chronic Kidney Disease. Evaluation, Classification and Stratification. Am J Kidney Dis 2002; 39: S1–S266. 2. Kasiske BL, Guijarro C, Massy ZA, Wiederkehr MR, Ma JZ. Cardiovascular disease after renal transplantation. J Am Soc Nephrol 1996; 7: 158–165. 3. Guijarro C, Massy ZA, Wiederkehr MR, Ma JZ, Kasiske BL. Serum albumin and mortality after renal transplantation. Am J Kidney Dis 1996; 27: 117–123. 4. Ojo AO, Hanson JA, Wolfe RA, Leichtman AB, Agodoa LY, Port FK. Long-term survival in renal transplant recipients with graft function. Kidney Int 2000; 57: 307–313. American Journal of Transplantation 2003; 3: 1426–1433 Anemia in Kidney Transplant Recipients 5. Gill JS, Abichandani R, Khan S, Kausz AT, Pereira BJ. Opportunities to improve the care of patients with kidney transplant failure. Kidney Int 2002; 61: 2193–2200. 6. Obrador GT, Ruthazer R, Arora P, Kausz AT, Pereira BJG. Prevalence of and factors associated with sub-optimal care prior to initiation of dialysis in the United States. J Am Soc Nephrol 1999; 10: 1793–1800. 7. Levin A, Singer J, Thompson CR, Ross H, Lewis M. Prevalent left ventricular hypertrophy in the predialysis population: Identifying opportunities for intervention. Am J Kidney Dis 1996; 27: 347– 354. 8. Levin A, Thompson CR, Ethier J et al. Left ventricular mass index increase in early renal disease: impact of decline in hemoglobin. Am J Kidney Dis 1999; 34: 125–134. 9. Khan S, Kazmi W, Abichandani R, Tighiouart H, Pereira B, Kausz A. Health care utilization among patients with chronic kidney disease. Kidney Int 2002; 62: 229–236. 10. Holland DC, Lam M. Predictors of hospitalization and death among pre-dialysis patients: a retrospective cohort study. Nephrol Dial Transplant 2000; 15 (5): 650–658. 11. Al-Ahmad A, Rand W, Manjunath G, Konstam M, Salem D, Levey AS. Reduced kidney function and anemia as risk factors for mortality in patients with left ventricular dysfunction. J Am Coll Cardiology 2001; 38: 955–962. 12. Rigatto C, Parfrey P, Foley R, Negrijn C, Tribula C, Jeffery J. Congestive heart failure in renal transplant recipients: risk factors, outcomes, and relationship with ischemic heart disease. J Am Soc Nephrol 2002; 13: 1084–1090. 13. Rigatto C, Foley R, Jeffery J, Negrijn C, Tribula C, Parfrey P. Electrocardiographic left ventricular hypertrophy in renal transplant recipients: prognostic value and impact of blood pressure and anemia. J Am Soc Nephrol 2003; 14: 462–468. 14. Kazmi WH, Kausz AT, Khan S et al. Anemia – An early complication of chronic renal insufficiency. Am J Kidney Dis 2001; 38: 803–812. 15. Lorenz M, Kletzmayr J, Perschl A, Furrer A, Horl WH, SunderPlassmann G. Anemia and iron deficiencies among long-term renal transplant recipients. J Am Soc Nephrol 2002; 13: 794–797. 16. Yorgin PD, Scandling JD, Belson A, Sanchez J, Alexander SR, Andreoni K. Late post-transplant anemia in adult renal transplant recipients. An under-recognized problem? Am J Transplant 2002; 2: 429–435. 17. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999; 130: 461–470. 18. Manjunath G, Sarnak MJ, Levey AS. Prediction equations to estimate glomerular filtration rate: An update. Curr Opin Nephrol Hypertens 2001; 10: 785–792. 19. Stoves J, Lindley EJ, Barnfield MC, Burniston MT, Newstead CG. MDRD equation estimates of glomerular filtration rate in potential living kidney donors and renal transplant recipients with impaired graft function. Nephrol Dial Transplant 2002; 17: 2036–2037. 20. Manotham K, Booranalertpaisarn V, Eiam-Ong S, Chusil S, Praditpornsilpa K, Tungsanga K. Accurately simple estimation of glomerular filtration rate in kidney transplant patients. Transplant Proc 2002; 34: 1148–1151. 21. Greenfield S, Nelson E. Recent developments and future issues in the use of health status assessment measures in clinical setting. Med Care 1992; 30: 23–41. American Journal of Transplantation 2003; 3: 1426–1433 22. Athienites N, Miskulin D, Fernandez G et al. Comorbidity assessment in hemodialysis and peritoneal dialysis using the Index of Coexisitent Disease. Sem Dialysis 2000; 13: 320–326. 23. NKF-K/DOQI Clinical Practice Guidelines for Anemia of Chronic Kidney Disease: Update 2000. Am J Kidney Dis 2001; 37 (Suppl. 1): S182–S236. 24. Brown J, Lappin T, Elder G, Taylor T, Bridges J, McGeown M. The initiation of erythropoiesis following renal transplantation. Nephrol Dial Transplant 1989; 4: 1076–1079. 25. Hayashi T, Suzuki A, Shoji T et al. Cardiovascular effect of normalizing the hematocrit level during erythropoietin therapy in predialysis patients with chronic renal failure. Am J Kidney Dis 2000; 35: 250–256. 26. Portoles J, Torralbo A, Martin P, Rodrigo J, Herrero J, Barrientos A. Cardiovascular effects of recombinant human erythropoietin in predialysis patients. Am J Kidney Dis 1997; 29: 541–548. 27. Sun CH, Ward HJ, Paul WL, Koyle MA, Yanagawa N, Lee DB. Serum erythropoietin levels after renal transplantation. N Engl J Med 1989; 321: 151–157. 28. Merkatz IR, Schwartz GH, David DS, Stenzel KH, Riggio RR, Whitsell JC. Resumption of female reproductive function following renal transplantation. JAMA 1971; 216: 1749–1754. 29. Kim JH, Chun CJ, Kang CM, Kwak JY. Kidney transplantation and menstrual changes. Transplant Proc 1998; 30: 3057–3059. 30. Abramovici H, Brandes JM, Better OS, Peretz A, Paldi E. Menstrual cycle and reproductive potential after kidney transplantation. Report of 2 patients. Obstet Gynecol 1971; 37: 121–125. 31. Moore LW, Smith SO, Winsett RP, Acchiardo SR, Gaber AO. Factors affecting erythropoietin production and correction of anemia in kidney transplant recipients. Clin Transplant 1994; 8: 358–364. 32. Pan WH, Habicht JP. The non-iron-deficiency-related difference in hemoglobin concentration distribution between blacks and whites and between men and women. Am J Epidemiol 1991; 134: 1410–1416. 33. Buhles WC Jr, Mastre BJ, Tinker AJ, Strand V, Koretz SH. Ganciclovir treatment of life- or sight-threatening cytomegalovirus infection: Experience in 314 immunocompromised patients. Rev Infect Dis 1988; 10 (Suppl. 3): S495–S506. 34. Farrugia E, Schwab TR. Management and prevention of cytomegalovirus infection after renal transplantation. Mayo Clin Proc 1992; 67: 879–890. 35. Astor BC, Muntner P, Levin A, Eustace JA, Coresh J. Association of kidney function with anemia: The Third National Health and Nutrition Examination Survey (1988–94). Arch Intern Med 2002; 162: 1401–1408. 36. Lim VS, DeGowin RL, Zavala D et al. Recombinant human erythropoietin treatment in predialysis patients: a double-blind, placebocontrolled trial. Ann Intern Med 1989; 110: 108–114. 37. The US Recombinant Human Erythropoietin Predialysis Group. Double-blind, placebo controlled study of the therapeutic use of recombinant human erythropoietin for anemia associated with chronic renal failure in predialysis patients. Am J Kidney Dis 1991; 18: 50–59. 38. Roth D, Smith R, Schulman G et al. Effects of recombinant human erythropoietin on renal function in chronic renal failure predialysis patients. Am J Kidney Dis 1994; 24: 777–784. 39. USRDS. US Renal Data System. 2001 Annual Data Report. Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Disease, 2001. 1433