|Home | About | Journals | Submit | Contact Us | Français|
Iron deficiency contributes to anemia after transplantation. The magnitude of iron loss from blood loss in the peri-transplantation period has not been quantified.
We prospectively estimated phlebotomy and surgical losses over the first 12-weeks following transplantation in 39 consecutive renal transplant recipients on hemodialysis (HD), peritoneal dialysis (PD), or chronic kidney disease (CKD).
At transplant, ferritin levels were <200 ng/mL in 51% of patients, and iron saturation was ≤ 20% in 44%. CKD patients more commonly had ferritin levels <200 ng/mL than either HD or PD patients (100% vs. 21% vs. 67%, P < 0.0002, respectively). Blood loss was similar among HD, PD and CKD patients (833 ± 194 vs. 861 ± 324 vs. 755 ± 79 ml respectively, P= NS), and not different between deceased and living donor transplant recipients (881 ± 291 vs. 788 ± 162 ml, P= 0.33). Based on a baseline hemoglobin (Hgb) of 11.8 g/dL we estimated an additional 330 mg of additional iron was needed to normalize hemoglobin to 13 g/dL, and 605 mg to increase hemoglobin to 14 g/dL.
Blood and iron losses over the first 12 weeks post-transplant are substantial and may warrant early administration of intravenous iron.
Anemia is common after renal transplantation and is frequently under-treated [1–6]. A European study found 38.6% of 4263 patients who received a kidney transplant within the previous 5 years were anemic (defined as hemoglobin <13 g/dL for men and <12 g/dL for women) . Other studies have also found a high prevalence of anemia among renal transplant recipients, even when graft function is normal [2, 6–9]. Anemia is prevalent among these patients despite the rapid increase in endogenous erythropoietin production following successful graft function.
Anemia in chronic kidney disease (CKD) reduces quality of life and is associated with development and progression of left ventricular hypertrophy (LVH). In kidney transplant patients, anemia has been associated with development of post transplant congestive heart failure [10, 11]. Iron deficiency is a major cause of anemia and is independently associated with post-transplant cardiovascular events, the leading cause of death in kidney transplant recipients .
The prevalence of iron deficiency, defined as the percent hypochromic red blood cells ≥ 2.5%, in 438 renal transplant recipients was 20% . Shibagaki et al found significant anemia in 192 kidney transplant recipients (hemoglobin <11 mg/dL in women and <12 mg/dL in men) was common at 19.3% at 6 months and 19.8% at 12 months. Only 36% of these anemic patients had their iron profile assessed during their study. Among these, half of them were iron deficiency (defined as ferritin <100 µg/dl, or transferrin saturation (TSAT) <20%) . Shah et al studied 1511 renal transplant patients, and found the prevalence of anemia after transplant to be 45.6%. Ferritin levels were measured only in 699 (46%) patients .
Major contributors to iron deficiency anemia in renal transplant recipients include inadequate iron stores at the time of transplant, surgical blood losses, and frequent post transplant phlebotomy. With recovery of renal function, erythropoiesis depletes storage iron. Moore et al found that 60% of renal transplant recipients without iron deficiency at the time of transplant became iron deficient by six-months post transplant . Taken together, these studies support that iron deficiency is an overlooked and correctable contributor to post renal transplant anemia.
To determine the magnitude of blood and iron losses in the first 12 weeks following transplant, we performed a prospective observational study of 39 consecutive renal transplant recipients, assessing baseline iron storage markers, and recording actual blood losses via surgery and phlebotomies over the first 12-weeks, and administering intravenous iron to those found to be iron deficient.
The study was approved by the Washington University Human Research Protection Office. We identified 42 consecutive patients receiving a living or deceased donor renal allograft between January 1 and April 27, 2004 at our center. We excluded three patients from our analysis; one due to primary graft dysfunction, one who received a combined liver-kidney transplant, and one who received intravenous iron supplementation despite a TSAT of 23%, which was not our usual clinical practice.
Hospital records were reviewed for demographic data, laboratory data, dialysis status prior to transplant, the administration of blood transfusions or intravenous iron from the time of transplant to 12 weeks following transplant. Blood loss was calculated by adding blood loss during the surgery as stated in the operation report, and the blood volume of all phlebotomies performed. Blood loss from phlebotomy was determined by the fill volume of each phlebotomy tube and accounting for a 10 mL blood volume wasted with clearing of central lines at the time of obtaining blood samples during the hospitalization. Most vials had a 5 ml fill volume, while tacrolimus level samples, type and screens, and cytomegalovirus PCR testing vial volumes were 7 ml each.
Standard testing immediately pre-transplant included a complete blood count (CBC), type and screen, complete metabolic panel (CMP), prothrombin and partial thromboplasmin time, final cross-match, lipid panel, human immunodeficiency virus (HIV), Epstein Barr virus, hepatitis B and hepatitis C virus, creatine phosphokinase, lactate dehydrogenase, gamma glutamyl transpeptidase (GGT), folate, iron battery, ferritin, vitamin B12, phosphorus, uric acid, and quantitative cytomegalovirus IgG (CMV). Patients typically had a CBC and a renal panel checked four more times during the transplant hospitalization, followed by weekly monitoring of calcineurin inhibitor levels, CBC, renal panel, and CMV testing during the time period studied.
We estimated 1.1 mg of iron per ml of packed red blood cells and used an average hematocrit of 35% to calculate the phlebotomy and bleeding-associated iron losses. The estimated 1.1 mg of iron per ml of packed cells was based on 3.3 mg of iron in each gram of hemoglobin or 660 mg of iron in 200 g of hemoglobin . The average volume of packed cells would be 200–250 ml. Thus if there are 3.3 mg iron per 1 g Hgb; if Hgb is 11.5g/dL, then (3.3 × 11.5) = 38 mg iron in 100 ml of blood. Thus the average blood loss ~780 ml blood = 295 mg iron. We estimated additional iron requirements to achieve normalization of hemoglobin of 13 or 14 g/dL from each patient’s pre-transplant hemoglobin.
Our clinical practice was to administer intravenous iron (iron dextran (INFeD, Watson Pharmaceuticals) 1000 mg IV over 6 hours after a 25 mg test dose during the transplant hospitalization when immediate pre-transplant testing suggested iron deficiency (defined by TSAT ≤ 20% or ferritin < 200 ng/mL). Patients without iron deficiency received no iron.
Thirty-seven patients received thymobglobulin 6 mg/kg over 4 days, 1 patient received basiliximab 20 mg × 2 doses, and one patient, a recipient of a 2-haplotype living related allograft, received no induction. All patients received tacrolimus with target trough levels of 7–10 ng/mL for the first month and 3–7 ng/mL thereafter. Thirty-two patients received mycophenolate mofetil with an average maintenance dose of 500 mg bid. Seven patients received azathioprine with an average dose of 150 mg/d. Patients who were seropositive or whose donor was seropositive for cytomegalovirus (CMV) were randomized to receive prophylaxis with 450–900 mg/d of valganciclovir with the dose adjusted for renal function based or preemptive therapy upon detection of a positive CMV-PCR.
No patient was on an angiotensin-converting enzyme inhibitor (ACE-1) or angiotensin-receptor blocker (ARB) after transplantation during the study period.
Frequencies and means were tabulated for demographic variables and descriptive variables of interest. For all analyses, a P < 0.05 was considered statistically significant. Chi-square statistics were calculated for categorical variables of interest. Fisher’s exact test was used for contingency tables where any cell had an expected frequency < 1 or where ≥ 20% of cells with expected frequencies were ≤ 5. One-way analysis of variance (ANOVA) was used to compare means of continuous variables by modality types. Statistical analysis was performed using SAS for Windows software, version 9.1 (SAS Institute, Cary, NC).
The average age was 51 ± 15 years old (range: 19 to 72). Three were black, two were Asian and 34 were Caucasian. Nineteen of the 39 patients were female. Etiologies of renal failure were: diabetes (n=7), hypertension (n=9), diabetes and hypertension (n=5), polycystic kidney disease (n=5), IgA nephropathy (n=2), focal segmental glomerulosclerosis (n=2), reflux nephropathy (n=1), chronic interstitial nephropathy (n=1), thin basement membrane disease (n=1), glomerulonephritis (n=2) and unknown (n=4). Sixteen patients received deceased donor allografts, twelve received living unrelated allografts, and eleven received living related allografts. Prior to transplantation, nineteen patients were on hemodialysis, twelve were on peritoneal dialysis, and eight had chronic kidney disease (eGFR< 20 mL/min) (CKD) (Table 1). Anemia parameters at the time of transplantation are shown in Table 2.
At the time of transplant, the average hemoglobin was 11.8 ± 1.7g/dL (range 8.8 to 15.9 g/dL). The mean hemoglobin 6 weeks post-transplant was 11.5 ± 1.7g/dL (range 8.6 to 17.2 g/dL), and 12 weeks post-transplant was 12.2 ± 1.8 g/dL (range: 7.7 to 17.9 g/dL). The prevalence of anemia (defined by hemoglobin < 13 g/dL in men and <12 g/dL in women) was 67% at the time of transplant, 77% 6 weeks post-transplant, and 62% 12 weeks post-transplant. It varied slightly based on renal replacement therapy prior to transplant (Figure 1). There were no significant differences in hemoglobin level between men and women at the three time points. The mean pre-transplant hemoglobin was also similar in those patients on HD, PD or CKD (12.03 vs. 11.77 vs. 11.45 g/dL, P=NS among groups), and between those receiving a deceased donor kidney compared to a living donor kidney (11.8 ± 1.3 g/dL vs. 11.8 ± 2 g/dL).
Immediately prior to transplantation, the mean serum ferritin was 326 ± 309 ng/mL (range 6 to 962 ng/mL). Men tended to have a higher ferritin level than women, although statistically non-significant (415 ± 356 ng/mL vs. 223 ± 224 ng/mL, p=0.07). The average ferritin was much lower in CKD patients (43.9 ng/mL) as compared to HD (518.8 ng/dL, P<0.001) or PD (209.3 ng/mL P=0.04). The overall prevalence of iron deficiency prior to transplant, defined as ferritin <200 ng/mL in HD and PD patients, ferritin < 100 ng/ml in CKD patients, was 49% in all 39 patients, and 88% among the CKD patients (Figure 2).
Immediately prior to transplant, the mean TSAT was 25.4 ± 10.6% (range 6 to 60%). Overall 44% of patients at the time of transplant had a TSAT<20% and there were no significant differences based on TSAT among different renal replacement modalities (HD 26.6% vs. PD 25.3% vs. CKD 21.1% P=NS (Figure 2). Men had a similar TSAT as women (24 ± 8.8% vs. 27 ± 12.6%; p=0.24) and did not differ by renal replacement therapy modality.
The average blood loss during the transplant surgery was 179 ± 133 ml (Range: 43 ml to 600 ml). The average blood loss over the 12 week period (including surgical blood loss and phlebotomy blood loss) just after transplantation was 823 ± 226 ml. The total blood loss ranged from 612 ml to 1854 ml (in a patient with a prolonged hospital course post-transplant). The total blood loss corresponds to an average iron loss of 317 ± 87 mg during the 12 weeks immediately following the transplant, with a range of 239 to 714 mg. Blood losses were similar among HD, PD and CKD patients (833 ± 194, 861 ± 324, and 755 ± 79 mL, respectively, P= NS). Blood losses were not significantly different between deceased donor transplant and living donor transplant recipients (881 ± 291 vs. 788 ± 162 ml, P= 0.33). Blood losses were similar between the group that received intravenous iron and the group that did not receive intravenous iron (780 ± 269 vs. 851 ± 178 ml, P= 0.38).
We estimated an overall need of 330 mg of additional iron to normalize hemoglobin from 11.8 g/dL to 13g/dL, and 605 mg to increase hemoglobin to 14 g/dL. Therefore the average iron requirement to raise hemoglobin to 13 or 14 g/dL and maintain the pre-transplant iron status over the first 12 weeks would be 647 mg (317 mg + 330 mg) and 922 mg (317 mg + 605 mg), respectively.
Nineteen iron deficient patients received intravenous (IV) iron supplementation. No adverse reactions to the IV iron infusion were observed. Overall, 37% of patients receiving IV iron achieved a hemoglobin level greater than 12 g/dL at 6 weeks and 58% at 12 weeks post-transplant. For patients who did not receive IV iron, 30% achieved a hemoglobin level greater than 12 g/dL at 6 weeks and 50% at 12 weeks. However, there was no statistical difference when comparing the proportion of patients who had >12 g/dL hemoglobin level at 6 weeks (p=0.90) or at 12 weeks (0.86) among those who did or did not receive intravenous iron.
At the time of transplant, mean hemoglobin did not differ between patients who received IV iron compared to those did not receive iron (11.7 ± 2.2 g/dL vs. 12.0 ± 1.2 g/dL respectively, P=0.62). Mean hemoglobin concentrations are the same at 6 weeks post-transplant, (11.9 ± 1.7 vs. 11.0 ± 1.6 g/dL, P=NS), and at 12 weeks post-transplant (12.3 ± 1.8 g/dL vs. 12.2 ± 1.8 g/dL, P=0.21) (Figure 3). Hemoglobin increased at 6 weeks in the IV iron group and decreased in the no iron group (Figure 3).
Delayed Graft Function (DGF) did not vary by iron deficiency. Of the 25 with iron deficiency 3 (12%) had DGF and of the 14 non-iron deficient, 2 (14%) had DGF, P= 1.0. Delayed graft function occurred in one (6.7%) of those who were anemic at 6 weeks compared to 4 (16.7%) of those who were not anemic at 6 weeks, P=0.63. There was no difference in the incidence of DGF or slow graft function among those who remained anemic at 6 weeks compared to those who were not anemic as assessed by the MDRD equation (Table 3). However, serum creatinine concentrations were statistically lower among those who remained anemic at 2, 4, and 6 weeks compared to those who were not anemic at these time-points see Table 3 .
Administration of azathioprine or mycophenolate mofetil did not affect anemia. Of those who received azathioprine 14.3% were anemic at 6 weeks post-transplant compared to 40.0% of those who received MMF, P=0.36. This relationship was also not statistically significant at 12 weeks post-transplant, P=0.66.
None of those who received valganciclovir were anemic at 6 weeks post-transplant compared to 41.7% of those who did not receive valganciclovir at 6 weeks post-transplant, P=0. 0.27. This was also not significant at 12 weeks, P= 0.56.
Consistent with previous studies, we found a high prevalence of anemia (67%) and iron deficiency (44%) in patients presenting for transplant at our center. One contribution to the high incidence of anemia following transplant despite return of endogenous erythropoietin production may be insufficient iron. Previous studies have found that iron status is frequently not evaluated in the transplant population [3, 4, 9]. Even when greater attention is focused on anemia before and after kidney transplantation, iron deficiency remains common in the early months post-transplant. Iron deficiency can result in iron-restricted erythropoiesis, as has been observed in dialysis patients .
Our novel report quantifies the magnitude and sources of blood and iron losses in the post-transplant period. We found significant blood loss in the immediate peri-transplant period, largely through standard phlebotomy for allograft function and immunosuppressant drug level monitoring. An additional significant amount of iron is required to increase hemoglobin from the range of 10–12 g/dL common in CKD patients prior to transplant, to the normal range expected in a patient with a well functioning graft. In addition, pre-menopausal women may have return of menses and thus lose iron. In all, the average post-transplant patient has an iron requirement approaching one gram.
Previous work has shown maximal oral iron absorption to be 3 mg per day in states of iron deficiency . Since the iron deficit was estimated at ~600 −900 mg, it would take at least 60 to 90 days of maximal iron absorption to provide this deficit. Nevertheless, oral iron can cause gastrointestinal side effects leading to poor adherence and in anemic CKD patients is not particularly effective in increasing hemoglobin, TSAT, or ferritin and it may bind to immunosuppressive agents, although this is controversial [18, 19].
Patients may benefit from administration of intravenous iron, as we employed in our clinical practice during the transplant hospitalization. We noted no problems related to administration of 1g of INFeD in the immediate post-operative period. Use of the higher molecular weight iron dextran, Dexferrum, should be avoided due to an unacceptably high rate of severe reactions . Ferric gluconate or iron sucrose can also be administered intravenously as smaller doses each hospital day.
In our non-randomized study, post-operative administration of intravenous iron to patients considered iron deficient was associated with a higher hemoglobin at 6 weeks compared to those not considered iron deficient and not given iron (11.9 ± 1.7 vs. 11.0 ± 1.6 g/dl). Similar hemoglobin levels were present at 12 weeks post transplant. While this trend at 6 weeks may be due to an erythropoietic effect of the intravenous iron, it may also reflect the limited ability to identify iron deficiency by serum ferritin and TSAT testing, or variation due to other factors which can influence anemia, such as immunosuppressant medications and infections. Alternatively, inflammatory cytokines, such as interleukin-6 or tumor necrosis factor, which were not measured but are elevated in ESRD and transplant patients, may have impaired the erythropoietic-response.
A limitation of our study is that it was not a randomized study to investigate the impact of intravenous iron on the recovery of anemia in those who were or were not identified as having iron deficiency at the time of presentation for renal transplantation. A better study would have been to randomize patients who were iron deficient to receive intravenous iron or not. It was our clinical experience that patients who were treated with iron recovered from their anemia more quickly and more completely than those who were not treated with intravenous iron that led to this observational study. The observation that patients who were iron deficient and treated with intravenous iron had numerically higher hemoglobin concentrations at 6 weeks tends to support our clinical impression. Another relevant study would be to administer intravenous iron to all patients regardless of identified iron deficiency. Although this would go beyond a clear clinical indication, it is a study worth pursing in the future.
There were no differences among induction or maintenance immunosuppression among those who were or not anemic at 6 weeks. Most of our patients received thymoglobulin and none received sirolimus. Thus, we were unable to investigate adequately the use of other commonly immunosuppressive maintenance regimens on the influence of anemia.
Although we did not investigate the use of ACEI or ARB at the time of presentation for transplantation, it is unlikely that use of these drugs would have influenced our observations given that the ½ life of these drugs is 6–12 hours.
An important concern with iron supplementation is the development of post-transplant erythrocytosis. None or our patients developed erythrocytosis. Furthermore, in previous studies, discontinuation of iron was usually followed by a decrease in hemoglobin four weeks later [4, 7].
This is the first study to document that despite increased attention to iron deficiency among CKD and ESRD patients that iron deficiency is common in patients presenting for renal transplantation. This study documents the magnitude of blood and iron loss with surgery and routine phlebotomy for post transplant monitoring contribute to anemia. Finally, intravenous replacement of iron among those who were iron deficient at the time of transplant was associated with equalization of hemoglobin concentrations by 12 weeks compared to those who were not iron deficient and did not receive iron. Anemia, however, persisted in both groups. Anemia predisposes to left ventricular hypertrophy. Left ventricular hypertrophy is a major risk factor for cardiovascular mortality and which the vast majority of ESRD patients have. Cardiovascular death is the leading cause of graft loss among renal transplant recipients and whether correction of anemia will decrease this requires further study. The optimal hemoglobin concentration in renal transplant recipients remains to be determined.
In conclusion, routine assessment of iron stores in the post-transplant period is indicated. Given the large iron losses and large iron needs following renal transplantation, further studies assessing the risks and benefits of peri-operative intravenous iron therapy are warranted.
This study is supported in part by P30 DK079333 (DCB).