- Research article
- Open Access
- Open Peer Review
Post renal transplant anemia: severity, causes and their association with graft and patient survival
BMC Nephrologyvolume 20, Article number: 51 (2019)
Post transplantation anemia (PTA) is common among kidney transplant patients. PTA is associated with increased graft loss and in most studies with increased mortality. However, the effect of the severity of anemia on this associations was not thoroughly evaluated.
Patients who underwent kidney transplantation in Rabin Medical Center (RMC) were included in the study. Data were collected during the years 2002–2016. Anemia was defined as hemoglobin (Hb) level less than 12 g/dL in women and less than 13 g/dL in men, in accordance with World Health Organization (WHO) criteria. Severe anemia was defined as hemoglobin lower than 11 g/dL. Primary outcome was a composite of patient and graft survival. We used univariate and multivariate models to evaluate association between severity and specific causes of anemia with the outcomes. As the risk associated with anemia changed over time we analyzed the risk separately for the early and the late period (before and after 1251 days).
Our cohort included 1139 patients, 412 (36.2%) of which had PTA and 134 (11.7%) had severe anemia. On multivariable analysis, severe anemia was highly associated with the primary outcome at the early period (HR 6.26, 95% CI 3.74–10.5, p < 0.001). Anemia due to either AKI & acute rejection (11.9% of patients) or infection (16.7%), were associated with primary outcome at the early period (HR 9.32, 95% CI 5.3–26.41, p < 0.001 and HR 3.99, 95% CI 2.01–7.95, p < 0.001, respectively). There was non-significant trend for association between anemia due to Nutritional deficiencies (29.1%) and this outcome (HR 3.07, 95% CI 0.93–10.17, p = 0.067).
PTA is associated with graft loss and mortality especially during the first three years. Anemia severity affects this association. An anemia workup is recommended for PTA.
Early PTA is commonly defined as anemia during the first six months after the transplantation [5, 7]. It is usually due to iron deficiency [8, 9]. Early PTA is also influenced by the slowly increasing levels of the newly graft-produced-erythropoietin [8, 10,11,12].
Late PTA, seen in 23–36% of patients, is usually defined as anemia that occurs more than six months after transplantation, and can appear as late as up to eight years [5, 7, 13]. It is influenced by reduced allograft function, but also by iron deficiency, immunosuppressive medications, infections and other factors [2, 5, 14].
PTA has been shown to be negatively associated with long-term outcomes: higher rates of all-cause mortality [6, 15,16,17,18], graft failure [4, 19, 20], congestive heart failure  and a decline in estimated glomerular filtration rate (eGFR) [6, 13]. Surprisingly, severity level was not thoroughly studied, and only one of the above studies classified patients by hemoglobin levels . Many factors have been associated with PTA including: iron deficiency, worsening renal function, recurrent transplantations, use of immunosuppressive medications, use of drugs acting on the renin angiotensin axis, infections, donor’s age and abnormally low levels of erythropoietin [11, 16, 22]. However, only a few studies showed an association between anemia etiology (mostly iron deficiency [23, 24]) with long-term prognosis.
Data are scarce regarding the association between anemia severity with patient and graft outcome, as well as the distribution of specific etiologies for anemia and their association with outcomes.
In this study, we investigated the association between anemia severity and etiology and long-term prognosis.
Study design and patients
Single center retrospective cohort study, using the Rabin Medical Center (RMC) kidney transplant registry. Inclusion criteria: all patients who underwent transplantation between the years 2002–2016, who had a functioning graft after six months, underwent a baseline anemia work-up at six months post-transplant, and a follow-up period of at least 12 months (i.e. 18 months from transplant). Exclusion criteria were age < 18 years, lack of documented clinic visits during the first 18 months and lack of a documented creatinine level after six months.
Anemia was defined as hemoglobin (Hb) level less than 12 g/dL in women and less than 13 g/dL in men, in accordance with World Health Organization (WHO) criteria . An episode of anemia was defined by at least two consecutives measurements of low Hb followed by two normal consecutive Hb values. Hb level was evaluated by routine complete blood count done at the Beilinson hospital laboratory using automated cell counter)Siemens, ADVIA 2120i).
The definition of anemia severity varies between different studies. We chose to define severe anemia using the commonly used criteria of Hb < 11 g/dL [3, 18, 26, 27]. For evaluation of anemia severity we used the mean value of all hemoglobin levels documented during the anemic episode.
Causes of anemia
We classified anemia according to 5 main causes:
Acute rejection and/or acute kidney injury (AKI)
Infection (acute or chronic)
Nutritional deficiencies (iron, folic acid and/or vitamin B12)
Miscellaneous causes: hematological (hemoglobinopathy, hemolysis), neoplastic (lymphoma, plasma cell dyscrasia), bleeding (Gastro-intestinal, genitourinary or peri-procedural), hypothyroidism, chronic disease.
No obvious reason found: presumably due CKD and/or immunosuppressive drugs. We combined these causes since by definition all renal transplant recipients are considered CKD , and all of them receive immunosuppressive drugs.
Iron deficiency anemia was defined as anemia with a low ferritin and/or transferrin saturation level, based on CKD level, microcytosis and hypochromia, and according to KDOQI  (see Additional file 1 for detailed definitions).
Vitamin B12 deficiency
Total serum vitamin B12 is not a reliable biomarker of vitamin B12 availability [30, 31]. We chose to define B12 deficiency as < 300 pmol/L, as suggested by others for borderline and deficiency [32,33,34,35].
Graft failure was defined as chronic (≥3 months) treatment with hemodialysis, re-transplantation or death with functioning graft.
Acute rejection was defined by renal biopsy showing rejection of Banff score of 1A or higher.
AKI was defined, according to the KDIGO criteria .
Patients’ characteristics were collected at baseline. For each patient, we documented all available episodes of anemia. Diagnosis of an episode of anemia was defined as the first time a decreased level of Hb was documented, according to the WHO criteria .
For each episode of anemia, full laboratory workup was collected. For patients without anemia, laboratory data were collected at six months following transplantation.
Acute infections based on culture, serological results and biopsy-proven acute graft rejections were collected from the electronic chart.
All possible causes of anemia for each episode were reviewed by two researchers (AS and BRZ) and in case of disagreement a third researcher evaluated the case (AG).
The primary outcome was the composite endpoint of graft failure (return to dialysis or re-transplantation) and all-cause mortality at the end of follow up.
Secondary outcomes were death censored graft failure (defined as re-establishment of long-term dialysis therapy, the need for re-transplantation) and all-cause mortality with a functioning graft.
Continuous data are presented as mean ± standard deviation or median and range, and dichotomous data as rate and percentages. Two-sample t-test and Mann–Whitney U-test were used for normally and non-normally distributed data, respectively. Differences in dichotomous variables were assessed by χ2 test. When numbers were small, the Fisher’s exact test was used instead of the χ2 test.
For the survival analysis we used a hierarchical method in order to attribute only one cause for anemia at a given time point. When anemia episode could be attributed to more than one cause and when two or more episodes of anemia due to a different cause occurred during the study period, the cause with higher hierarchy was considered as the anemia cause. Thus, every patient was assigned a single cause for his anemia (the highest-ranked cause in our model). The hierarchical order from lowest to highest was as follows: no identified cause, metabolic deficiencies, hemorrhage/hemolysis/hematologic causes, infection and AKI/rejection.
As not all anemia episodes occurred at six months we used a time dependent covariate model in which anemia was the time dependent covariate. Univariate and multivariate time varying Cox proportional hazard models were used, with the anemia incidence and severity as the time dependent covariate. The proportionality of the hazard was evaluated by adding the interaction for every covariate with time and assessing for the null hypothesis. Since anemia did not satisfy the propensity of the hazard assumption, we used a model with changing hazard. In order to find the point at which the hazard changed, we compared three possible models according to the quartiles of follow up time. We used the Wald statistics to evaluate the model with the best fit. The first quartile (180 to 1251 days) had the best fit according to the Wald statistics and was selected. When the proportionality of the hazard was evaluated for each time period (180 to 1251 days and 1251 days and onwards) the assumption was fully satisfied. As a result we analyzed the hazard separately for the early period (180 to 1251 days) and late period (1252 and onwards). As anemic patients were different in several factors we used extensive multivariate model to adjust for these differences.
The multivariate analysis was done using two models. First, we used a stepwise backwards regression model with p-value of 0.05 for inclusion and 0.15 for exclusion to adjust for the factors significantly associated with the outcome by univariate analysis (Model 1), and then we added to the model variables that might be associated with anemia or the outcome according to the literature (Model 2).
Covariates entered into the model included the following:
Recipient features including age, gender, kidney disease, pre-transplant diabetes mellitus (DM), pre-transplant heart disease, time on dialysis before transplantation, last Panel Reactive Antibodies (PRA) results classified as positive (10% or more) or negative, cytomegalovirus (CMV) serostatus, infection with hepatitis C (HCV) or hepatitis B virus (HBV) and former transplantations. Donor features included donor type (living or deceased), donor age, gender and HLA mismatch between the donor and recipient. Transplantation-related variables included cold ischemia time, induction therapy [none, anti thymocyte globulin (ATG), interleukin (IL)-2 antagonists and others], immunosuppression (tacrolimus, cyclosporine, mTOR inhibitor and antimetabolites and steroids alone), presence of delayed graft function, length of hospital stay (log transformed with the natural base in order to normalize data distribution), eGFR at 6, platelet count white blood cell (WBC), all at 6 months.
During the study period 1404 adult patients underwent 1420 kidney transplantations. Of them, a total of 265 patients (18.9%) were excluded: 87 patients (6.2%) did not have functioning graft after 180 days, 86 patients (6.1%) had no follow up data, 77 patients (5.5%) had missing baseline data, 15 patients (1.1%) had other organ transplantation (heart, lung, liver and pancreas). Anemic episodes after second transplantation during the study period were excluded since those 16 patients were considered to achieve the study outcome. Therefore, 1139 patients with functioning graft were available for analysis, Additional file 1: Figure S1. Of them, 412 patients (36.2%) had at least one episode of anemia within 6–18 months after transplantation, with a total of 500 events. Distribution of anemia episodes is as follows: 341 (82.8%) had one episode, 59 (14.3%) had two separate episodes, 11 (2.7%) had three separate episodes, and two (0.5%) had 4 episodes.
Table 1 depicts the demographic, baseline and medication characteristics, according to patients who developed anemia compared to those who did not, and according to severity of anemia. Anemic patients were more likely to have diabetes and cardiovascular comorbidities and had an older donor. Approximately 95% of all patients were treated with a similar immunosuppressive regimen consisting of a calcineurin inhibitor and mycophenolic acid. Distribution of the causes of anemia is presented in Table 2. Nutritional deficiencies were the most common. Of them, the most common deficiency was iron deficiency, which was diagnosed in 143 patients (34.7% of anemic patients) and in 157 separate events (31.4%).
During a median follow up of 5.5 years (interquartile range 3.4–8.7 years), a total of 265 deaths or graft loss occurred (23.3% of the entire cohort). One hundred seventy two patients died (15.1%) and graft loss occurred in 129 patients (11.3%).
The effect of anemia, anemia severity and specific causes on the outcomes are presented for the primary composite outcome (Table 3), death-censored graft survival (Table 4), and all-cause mortality (Table 5).
I. Effect of anemia on outcomes
In a univariate analysis, anemia was significantly associated with graft loss or mortality in the early period [hazard ratio (HR) 3.64, 95% confidence interval (CI) 2.34–5.66, p < 0.001] while at the late period the association was weaker (HR 1.44, 95% CI 0.97–2.13, 0.07), Table 3.
Multivariate analysis showed the same significant association for the early period (HR 2.94, 95% CI 1.86–4.66, p < 0.001, model 2), and a nonsignificant association for the later period (HR 1.44, 95% CI 0.97–2.13, 0.074).
Similar results were obtained for the outcome of all-cause mortality during the early period: (HR 3.24, 95% CI 1.73–6.08, < 0.001) and (HR 2.27, 95% CI 1.17–4.4, p = 0.015, model 2) for univariate and multivariate analyses, respectively. For the late period there was no association between anemia and mortality for both univariate and multivariate (Table 4).
Anemia was significantly associated with death censored graft loss during the early period by both univariate (HR 4.06, 95% CI 2.17–7.57, p < 0.001) and multivariate analysis (HR 3.15, 95% CI 1.63–6.08, p = 0.001, model 2). Anemia was also associated with death censored graft loss at the late period only with smaller hazard (HR 2.17, 95% CI 1.33–3.55, p = 0.002) and (HR 1.853, 95% CI 1.09–3.16, p = 0.024, model 2) for univariate and multivariate analysis respectively (Table 5).
II. Effect of Anemia severity on outcomes
In a univariate analysis of the three groups according to anemia severity (no anemia, mild anemia and severe anemia), all anemia groups were significantly associated with graft loss or mortality during the early period. However, the HR in the severe anemia group was higher than in the mild anemia group (HR 1.89, 95% CI 1.07–3.31, p = 0.027, for mild anemia and 8.01, 95% CI 4.92–13.04, p < 0.001, for severe anemia). By a multivariate analysis, severe anemia was still strongly associated with this endpoint at the early period (HR 6.26, 95% CI 3.74–10.5, p < 0.001, model 2). However, for mild anemia there was no significant association (HR 1. 1.42, 95% CI 0.79–2.54, p = 0.244, model 2). Both mild and severe anemia were not significantly associated with mortality and graft loss during the late period (Table 3).
In contrast, only severe anemia was associated with all-cause mortality at the early period (HR 6.36, 95% CI 3.1–13.03, p < 0.001), while for mild anemia the association was not significant (HR 2, 95% CI 0.93–4.31, p = 0.077). The association between severe anemia and mortality remained significant in the multivariate analysis (HR 5.42, 95% CI 2.51–11.68, p < 0.001, model 2). There was no association between mild and severe anemia and mortality at the late period (Table 4).
When death censored graft survival was evaluated, severe anemia was significantly associated with this outcome, at both the early and late periods (HR 9.78, 95% CI 5.0–19.13, p < 0.001, and HR 2.65, 95% CI 1.25–5.64, p = 0.011, for early and late period respectively). In a multivariate analysis, the results were not changed: early period (HR 7.6, 95% CI 3.69–15.65 p < 0.001, model 2), late period (HR 2.61, 95% CI 1.15–5.92, p = 0.021, model 2). Mild anemia was not associated with death censored graft survival during the early period by both univariate and multi variate analysis (Table 5). In contrast, there was an association between mild anemia and death censored graft survival during the late period but, this association became non-significant by multivariate analysis (HR 2, 95% CI 1.14–3.5, p = 0.015, for univariate and 1.66, 95% CI 0.91–3.01, p = 0.096, for multivariate).
III. Effect of specific causes of anemia on outcomes
In a univariate analysis, AKI & acute rejection (HR 12.3, 95% CI 7.2–21.0, p < 0.001), infections (HR 5.02, 95% CI 2.58–9.78, p < 0.001), nutritional deficiencies (HR 3.12, 95% CI 1.10–8.85, p = 0.032) and miscellaneous reasons (HR 2.43, 95% CI 1.22–4.83, p = 0.012) were all associated with graft loss or mortality during the early period. In contrast, anemia of unidentified cause was not significantly associated with graft loss or mortality. In a multivariate analysis the results were similar to those found in the univariate analysis for anemia due to AKI & acute rejection as well as infections. For anemia due to nutritional deficiencies (HR 3.07, 95% CI 0.93–10.17, p = 0.067) and miscellaneous reasons (HR 1.82, 95% CI 0.85–3.48, p = 0.133) the association was non-significant.
When the late period was evaluated by univariate analysis only AKI & acute rejection (HR 2.32, 95% CI 1.13–4.75, p = 0.022) was associated with graft loss and mortality. By multivariate analysis the association with graft loss and mortality at the late period became a non-significant trend (HR 2.06, 95% CI 0.97–4.39, p = 0.061).
When all-cause mortality during the early period was evaluated, only anemia due to AKI and/or acute rejection (HR 6.88, 95% CI 2.83–16.76, p < 0.001 and HR 4.49, 95% CI 1.77–11.34, p = 0.002, for univariate and model 2 respectively) and infection (HR 6.45, 95% CI 2.76–15.91, p < 0.001, HR 4.27, 95% CI 1.77–10.25, p = 0.001, for univariate and model 2 respectively) were associated with increased mortality. In contrast no association was found between any etiology for anemia and mortality at the late period.
Death censored graft loss at the early period was highly associated with AKI & acute rejection (HR 18.18, 95% CI 9.15–32.32, p < 0.001, and HR 12.94, 95% CI 6.14–27.28, p < 0.001, for univariate and model 2 respectively). Anemia due to infections was significantly associated with this outcome only by univariate analysis (HR 3.47, 95% CI 1.15–10.47, p = 0.002, and HR 3.01, 95% CI 0.97–9.33, p = 0.056, for univariate and model 2 respectively) with similar results for anemia due to miscellaneous reasons (HR 2.75, 95% CI 1.07–7.1, p = 0.037, and HR 2.14, 95% CI 0.81–5.65, p = 0.125, for univariate and model 2 respectively).
Death censored graft loss at the late period was associated only with AKI & acute rejection and the association remained significant after multivariate adjustment (HR 3.93, 95% CI 1.68–9.29, p = 0.002, and HR 4.14, 95% CI 6.14–27.28, p = 0.003, for univariate and model 2 respectively). Anemia due to miscellaneous causes was associated with death censored graft loss only by univariate analysis and the association was not significant after multivariate adjustment (HR 2.57, 95% CI 1.2–5.49, p = 0.015, and HR 1.91, 95% CI 0.86–4.23, p = 0.110, for univariate and model 2 respectively).
Both infection and rejection, known causes for anemia, are greatly impacted by immunosuppression. Since immunosuppression may affect the outcomes, we conducted another analysis for the primary outcome including only the 877 patients treated with tacrolimus (with at least 3 available tacrolimus levels during the first six months). We included the mean tacrolimus level as a covariate, as can be seen in Additional file 1: Tables S1-S3. The results were not very different.
The full multivariable model with all of the covariables is shown in Additional file 1: Tables S4-S12.
In this cohort study of 1139 kidney transplant recipients we demonstrated that the severity of anemia is strongly associated with graft failure and mortality. Furthermore, we identified various causes of late PTA, which were associated with prognosis.
PTA has previously been shown to be associated with the composite outcome of all-cause mortality and graft loss [6, 17, 18, 27, 37,38,39], as well as the separate components: death-censored graft survival [4, 19, 20] and all-cause mortality [6, 15,16,17,18]. However, the association between PTA and increased mortality is inconsistent, as several studies failed to show association with all-cause mortality, despite an association with graft loss [4, 19, 40,41,42,43]. Our study shows an association between anemia and the primary outcome of graft failure and mortality, as well as with mortality alone.
All-cause mortality was significantly associated in our study only with severe anemia. These results are in accordance with the findings by Heinze et al.  who found a negative correlation between mortality and hemoglobin level in patients with late PTA. The results of our study may explain the results of the studies that showed no association of late PTA with mortality [4, 19, 40,41,42,43]. Most patients in these studies had mild anemia with mean hemoglobin above 11 g/dl. Death censored graft loss was significantly associated with late PTA by univariate and multivariate analyses in both anemic groups. However, this association was weaker in the mild anemia group. Huang et al.  showed that only the more severe levels of anemia (defined as hemoglobin < 11 g/L and < 10 g/L for men and women, respectively) were associated with lower 3 and 5-year graft survival rates, but did not find an association with all-cause mortality.
Many studies described various causes for PTA. However, most studies did not address the importance of diagnosing the exact etiology for anemia. Nevertheless, two studies showed that iron deficiency  and the percentage of hypochromic RBC’s  are predictors of high rates of all-cause mortality. To the best of our knowledge, our study is the first to thoroughly examine the association between specific etiologies of anemia and prognosis. We demonstrated that the specific etiologies of anemia are associated with prognosis: anemia due to AKI, acute rejection, infection or nutritional deficiencies is associated with higher risk of death or graft loss, while anemia in which no specific cause was found is not.
Folic acid deficiency was diagnosed in 10% of anemic patients, a relatively low rate as compared to 23–41% in other studies [2, 45, 46]. This may be due to the widespread folic acid supplementation of food to prevent neural birth defects.
Our study has several strengths: first, our cohort is large, with over 1000 transplant recipients, and about 400 patients with anemia. In addition, there is an adequate follow up time of over 5.5 years, with separate analyses of the first 3.4 years and a later period. Second, in contrast to other studies, our study was not a cross-sectional, and the presence of anemia was evaluated over a time span of 12 months, and incorporated clinical data derived from routine clinic visits, hospitalizations, and laboratory workup. Anemia severity was evaluated by using the mean Hb value for the whole episode which is probably better than using an arbitrary single value. Furthermore, since we conducted separate analyses for two time periods after transplant, an “early” period of about 3.4 years (1251 days) and a later period, we could show that the association between anemia at any time point between 6 and 18 months after transplantation and graft failure and mortality was seen only during the earlier period. We show for the first time that the risk for mortality associated with anemia changes over time. Although the study is retrospective, and one cannot assume causality between anemia and outcomes, this temporal association in time may elucidate the possible effect that anemia has on outcomes.
Several limitations merit consideration. First, due to the retrospective design of the study, we can only show an association between severity of anemia and outcomes, rather than causality. Some anemia events had multiple causes. Although we used a hierarchical strategy in which every event was assigned a single cause, the effect of each cause cannot be evaluated. In addition, our study had a relatively large group of patients with anemia that we considered as an unknown cause. It is most likely that these patients had anemia due to chronic renal failure (as the KDIGO recommendations define all patients post transplant as CKD patients [28, 47]) and/or due to immunosuppressive therapy. Although CKD (eGFR< 60 ml/min/1.73 m2, CKD-3-5) is an established cause for anemia, we chose not to include renal function as a separate cause for anemia, as reduced renal function is a well-known risk factor for graft loss. However, eGFR was included in our multivariate models so any effect of renal function on graft survival was adjusted for. Immunosuppressive drugs are another established cause of anemia [48,49,50] that was not included in our study, as the vast majority of our cohort (≈95%) was treated by the same protocol of calcineurin inhibitor with mycophenolic acid, the effect of medication could not be properly evaluated in this population. Notably, since this is a retrospective study, there were a few differences between the anemic and non-anemic groups, mainly, anemic patients had an older donor, diabetes and cardiovascular comorbidity. Nevertheless, anemia was indeed shown to be an independent factor for graft failure and death. Another limitation is the lack of patients of African American, Hispanic or Asian origin in our population, so generalization of our results to these population should be done with caution.
Our findings should encourage physician to thoroughly evaluate patients with anemia. All patients with severe anemia are at risk for a graft failure and mortality, and even for mild anemia, rejection and infections should be excluded and patients should be evaluated for metabolic deficiencies and treated accordingly.
Our study shows an association between severe anemia and outcomes. Therefore, although not assessed in this study, it seems reasonable that an attempt should be made to treat anemia. The management of PTA usually starts with diagnosing and treating all reversible underlying causes. However, there is no specific recommendation for treatment of kidney transplant recipients in KDIGO guidelines . Several issues regarding specific treatment are unclear. Most importantly, the target Hb level for treatment. Second, the type of treatment, erythropoiesis stimulating agents (ESA) and/or iron.
Data on treatment of late PTA are limited, but show a beneficial effect for raising the Hb level. These studies tried to assess the optimal Hb level with ESA administration. A randomized trial by Choukroun et al.  demonstrated that higher Hb level was associated with graft survival. Normalizing PTA (to a Hb level of 13–15 g/dl) with epoetin beta compared with partial correction of Hb to 10.5 to 11.5 g/dL, was associated with reduction in the rate of decline of eGFR and progression to ESRD, and improved death-censored graft survival . Heinze et al. , in a cross sectional retrospective cohort study found that anemia (Hb < 12.5 g/dl) was significantly associated with mortality, in both patients with or without (ESA). In patients without ESA, a spontaneous rise in Hb was associated with decreased mortality at any level. In patients treated with ESA, improvement of anemia up to 12.5 g/dL was associated with decreased mortality as well. Further increments in Hb level led to a tendency of increased rate of mortality, which became significant at Hb level above 14 g/dL .
In contrast to the abundance of data from RCTs demonstrating the efficacy of iron in the CKD population, data regarding treatment with iron in PTA are scarce. Our group reported in a retrospective cohort study of 81 patients that intravenous iron administration after transplantation increases hemoglobin level and delays eGFR decline, both of which were more pronounced in lower hemoglobin levels (mean 9.4 ± 1.2 g/dL) . Yet, a small RCT of 104 patients comparing oral with a single dose of intravenous iron after transplantation did not show a difference in time to anemia correction . However, this trial was conducted in the post-operative period, and does not address late PTA.
Based on our findings of improved prognosis with a lesser degree of anemia and the finding in literature of response to treatment [13, 15], it might be prudent to raise the Hb level. However, the optimal correction level and the best treatment modality are unclear.
In conclusion, we found an association between anemia and mortality and graft failure which is related to the severity and causes of anemia. Anemia workup is highly warranted in order to find the specific underlying causes. Future research should focus on the target Hb level, and the appropriate use of ESA and iron.
Acute kidney injury
Chronic kidney disease
Chronic Kidney Disease Epidemiology Collaboration
Estimated glomerular filtration rate
End stage renal disease
Kidney Disease: Improving Global Outcomes
Mean corpuscular volume
Post transplantation anemia
Randomized controlled trial
World Health Organization
Shibagaki Y, Shetty A. Anemia is common after kidney transplantation, especially among African Americans. Nephrol Dial Transplant. 2004;19(9):2368–73.
Turkowski-Duhem A, Kamar N, Cointault O, et al. Predictive factors of anemia within the first year post renal transplant. Transplantation. 2005;80(7):903–9.
Vanrenterghem Y, Ponticelli C, Morales JM, et al. Prevalence and management of anemia in renal transplant recipients: a European survey. Am J Transplant. 2003;3(7):835–45.
Winkelmayer WC, Chandraker A, Alan Brookhart M, Kramar R, Sunder-Plassmann G. A prospective study of anemia and long-term outcomes in kidney transplant recipients. Nephrol Dial Transplant. 2006;21(12):3559–66.
Yorgin PD, Belson A, Sanchez J, et al. Unexpectedly high prevalence of posttransplant anemia in pediatric and young adult renal transplant recipients. Am J Kidney Dis. 2002;40(6):1306–18.
Gafter-Gvili A, Ayalon-Dangur I, Cooper L, et al. Posttransplantation anemia in kidney transplant recipients: a retrospective cohort study. Medicine. 2017;96(32):e7735.
Hricik DE. Anemia after kidney transplantation--is the incidence increasing? Am J Transplant. 2003;3(7):771–2.
Beshara S, Birgegard G, Goch J, Wahlberg J, Wikstrom B, Danielson BG. Assessment of erythropoiesis following renal transplantation. Eur J Haematol. 1997;58(3):167–73.
Teruel JL, Lamas S, Vila T, et al. Serum ferritin levels after renal transplantation: a prospective study. Nephron. 1989;51(4):462–5.
Kahng K, Kang C, Kwak J. Changes in hemoglobin levels after renal transplantation. Transplant Proc. 1998;30(7):3023–4.
Miles AM, Markell MS, Daskalakis P, et al. Anemia following renal transplantation: erythropoietin response and iron deficiency. Clin Transpl. 1997;11(4):313–5.
Sun CH, Ward HJ, Paul WL, Koyle MA, Yanagawa N, Lee DB. Serum erythropoietin levels after renal transplantation. N Engl J Med. 1989;321(3):151–7.
Choukroun G, Kamar N, Dussol B, et al. Correction of postkidney transplant anemia reduces progression of allograft nephropathy. J Am Soc Nephrol. 2012;23(2):360–8.
Saito S, Fujiwara T, Sakagami K, Matsuno T, Tanaka N. Anemia following renal transplantation. Transplant Proc. 1998;30(7):3025–6.
Heinze G, Kainz A, Horl WH, Oberbauer R. Mortality in renal transplant recipients given erythropoietins to increase hemoglobin concentration: cohort study. BMJ. 2009;339:b4018.
Heinze G, Mitterbauer C, Regele H, et al. Angiotensin-converting enzyme inhibitor or angiotensin II type 1 receptor antagonist therapy is associated with prolonged patient and graft survival after renal transplantation. J Am Soc Nephrol. 2006;17(3):889–99.
Imoagene-Oyedeji AE, Rosas SE, Doyle AM, Goral S, Bloom RD. Posttransplantation anemia at 12 months in kidney recipients treated with mycophenolate mofetil: risk factors and implications for mortality. J Am Soc Nephrol. 2006;17(11):3240–7.
Molnar MZ, Czira M, Ambrus C, et al. Anemia is associated with mortality in kidney-transplanted patients--a prospective cohort study. Am J Transplant. 2007;7(4):818–24.
Huang Z, Song T, Fu L, et al. Post-renal transplantation anemia at 12 months: prevalence, risk factors, and impact on clinical outcomes. Int Urol Nephrol. 2015;47(9):1577–85.
de Andrade LG, Abrao JM, Carvalho MF. Anemia at one year is an independent risk factor of graft survival. Int Urol Nephrol. 2012;44(1):263–8.
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(4):1084–90.
Heidenreich S, Tepel M, Fahrenkamp A, Rahn KH. Prognostic value of serum erythropoietin levels in late acute rejection of renal transplants. Am J Kidney Dis. 1995;25(5):775–80.
Eisenga MF, Minovic I, Berger SP, et al. Iron deficiency, anemia, and mortality in renal transplant recipients. Transpl Int. 2016;29(11):1176–83.
Winkelmayer WC, Lorenz M, Kramar R, Horl WH, Sunder-Plassmann G. Percentage of hypochromic red blood cells is an independent risk factor for mortality in kidney transplant recipients. Am J Transplant. 2004;4(12):2075–81.
World Health Organization Technical report sereirs No 405, Nutritional Anemias: Report of a WHO Scientific Group Geneva, Switzerland1968.
Afzali B, Al-Khoury S, Shah N, Mikhail A, Covic A, Goldsmith D. Anemia after renal transplantation. Am J Kidney Dis. 2006;48(4):519–36.
Chhabra D, Grafals M, Skaro AI, Parker M, Gallon L. Impact of anemia after renal transplantation on patient and graft survival and on rate of acute rejection. Clin J Am Soc Nephrol. 2008;3(4):1168–74.
Kidney Disease. Improving global outcomes (KDIGO) CKD work group, KDIGO clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl. 2013;3:1–150.
KDOQI Clinical Practice Guideline and Clinical Practice Recommendations for Anemia in Chronic Kidney Disease. Update of hemoglobin target. Am J Kidney Dis. 2007;50(3):471–530.
Carmel R. Current concepts in cobalamin deficiency. Annu Rev Med. 2000;51:357–75.
Green R. Vitamin B12 deficiency from the perspective of a practicing hematologist. Blood. 2017;129(19):2603–11.
Clarke R, Refsum H, Birks J, et al. Screening for vitamin B-12 and folate deficiency in older persons. Am J Clin Nutr. 2003;77(5):1241–7.
de Benoist B. Conclusions of a WHO technical consultation on folate and vitamin B12 deficiencies. Food Nutr Bull. 2008;29(2 Suppl):S238–44.
Gentil MA, Perez-Valdivia MA, Lopez-Mendoza M, et al. Factor deficiency in the anemia of renal transplant patients with grade III-IV chronic kidney disease: baseline results of the ARES study. Transplant Proc. 2008;40(9):2922–4.
Mirkazemi C, Peterson GM, Tenni PC, Jackson SL. Vitamin B12 deficiency in Australian residential aged care facilities. J Nutr Health Aging. 2012;16(3):277–80.
Kidney Disease. Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl. 2012;(2):1–138.
Garrigue V, Szwarc I, Giral M, et al. Influence of anemia on patient and graft survival after renal transplantation: results from the French DIVAT cohort. Transplantation. 2014;97(2):168–75.
Ichimaru N, Obi Y, Nakazawa S, et al. Post-transplant Anemia has strong influences on renal and patient outcomes in living kidney transplant patients. Transplant Proc. 2016;48(3):878–83.
Majernikova M, Rosenberger J, Prihodova L, et al. Posttransplant Anemia as a prognostic factor of mortality in kidney-transplant recipients. Biomed Res Int. 2017;2017:6987240.
Freiberg M, Chiurchiu C, Capra R, et al. Associated factors and clinical implications of post transplant renal anemia. Medicina (B Aires). 2013;73(2):136–40.
Gheith O, Wafa E, Hassan N, et al. Does posttransplant anemia at 6 months affect long-term outcome of live-donor kidney transplantation? A single-center experience. Clin Exp Nephrol. 2009;13(4):361–6.
Kamar N, Rostaing L, Ignace S, Villar E. Impact of post-transplant anemia on patient and graft survival rates after kidney transplantation: a meta-analysis. Clin Transpl. 2012;26(3):461–9.
Schjelderup P, Dahle DO, Holdaas H, et al. Anemia is a predictor of graft loss but not cardiovascular events and all-cause mortality in renal transplant recipients: follow-up data from the ALERT study. Clin Transpl. 2013;27(6):E636–43.
Shah N, Al-Khoury S, Afzali B, et al. Posttransplantation anemia in adult renal allograft recipients: prevalence and predictors. Transplantation. 2006;81(8):1112–8.
Karakus S, Kanbay M, Koseoglu HK, Colak T, Haberal M. Causes of anemia in renal transplant recipients. Transplant Proc. 2004;36(1):164–5.
Mahmud SN, Aziz R, Ahmed E, et al. Anemia characteristics after renal transplantation. Transplant Proc. 2002;34(6):2428.
Levey AS, Eckardt K-U, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO). Kidney Int. 2005;67(6):2089–100.
Augustine JJ, Knauss TC, Schulak JA, Bodziak KA, Siegel C, Hricik DE. Comparative effects of sirolimus and mycophenolate mofetil on erythropoiesis in kidney transplant patients. Am J Transplant. 2004;4(12):2001–6.
Dobrolet NC, Webber SA, Blatt J, et al. Hematologic abnormalities in children and young adults receiving tacrolimus-based immunosuppression following cardiothoracic transplantation. Pediatr Transplant. 2001;5(2):125–31.
Suzuki S, Osaka Y, Nakai I, et al. Pure red cell aplasia induced by FK506. Transplantation. 1996;61(5):831–2.
Rozen-Zvi B, Gafter-Gvili A, Zingerman B, et al. Intravenous iron supplementation after kidney transplantation. Clin Transpl. 2012;26(4):608–14.
Mudge DW, Tan KS, Miles R, et al. A randomized controlled trial of intravenous or oral iron for posttransplant anemia in kidney transplantation. Transplantation. 2012;93(8):822–6.
Availability of data and materials
The datasets generated and/or analysed during the current study are not publicly available due to privacy reasons, but are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
Our study was approved by the Rabin Medical Center Helsinki committee, reference number 0462–17-RMC. Consent was not required because this is a retrospective study.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Iron deficiency definitions. Figure S1. Flow of patients. Table S1. Composite outcome, with tacrolimus level. Table S2. Death censored graft failure, with tacrolimus level. Table S3. All-cause mortality, with tacrolimus level. Table S4. Composite outcome, presence of anemia, without tacrolimus level. Table S5. Composite outcome, severity of anemia, without tacrolimus level. Table S6. Composite outcome, causes of anemia, without tacrolimus level. Table S7. Death censored graft failure, presence of anemia, without tacrolimus level. Table S8. Death censored graft failure, Severity of anemia, without tacrolimus level. Table S9. Death censored graft failure, causes of anemia, without tacrolimus level. Table S10. All-cause mortality, presence of anemia, without tacrolimus level. Table S11. All-cause mortality, severity of anemia, without tacrolimus level. Table S12. All-cause mortality, causes of anemia, without tacrolimus level. (DOCX 59 kb)