The pre-transplant anemic condition is independent of long-term outcome in living-related kidney transplantation

Abstract

The relationship between pre-transplant Hemoglobin (Hb) concentration and long-term outcome of living-related kidney transplantation is far from well addressed. A retrospective cohort study was conducted by reviewing the medical profile of the patients who received living-related kidney transplantations at our center from January 2006 to January 2013. Patients were divided into two groups: high Hb group (≥10 g/dL) and low Hb group (<10 g/dL). Cox regression model was utilized to analyze the effect of pre-transplant hemoglobin concentration on the patient and graft survival. About 422 patients were of Hb level <10 g/dL (78.30 ± 14.18 g/dL), 280 were >10 g/dL (116.2 ± 14.43 g/dL) (p < 0.001). In a follow-up of 35.34 ± 18.12 months, we did not find any difference in serum creatinine between the two groups. Low Hb concentration is not associated with increased risk of developing DGF (HR = 1.186, 95% CI: 0.53–2.654), acute rejection (HR = 1.338, 95% CI: 0.919–1.947), overall infection (HR = 1.263, 95% CI: 0.847–1.885) nor perioperational infection (HR = 1.019, 95% CI: 0.513–2.026). Though we detected a trend that low Hb level group were of higher incidence of patient death and graft failure, the two groups did not differ significantly (2.38% vs. 0.71%, p = 0.096; and 4.04% vs. 2.14%, p = 0.165, respectively). Cox regression model revealed that pre-transplant Hb level <10 g/dL was independent of increased overall mortality (HR = 3.379; 95% CI: 0.706–17.172) and increased death censored allograft failure risk (HR = 1.556; 95% CI: 0.595–4.069). Pre-transplant Hb concentration <10 g/dL is independent of poor long-term outcome of living-related kidney transplantation.

• Graft failure
• hemoglobin concentration
• living-related kidney transplantation
• patient survival
• rejection

Introduction

Anemia is universal in patients with chronic kidney disease (CKD), attributed to several factors including impaired endocrine function in production of erythropoietin, iron, folate or vitamin B12 deficiency, gastrointestinal blood loss, shortened red blood cell survival, and less frequently, by disorders such as hemolytic uremic syndrome or aplastic anemia.1,2 Of them, decreased erythropoietin synthesis is the most important and specific etiology causing CKD-associated anemia. After the kidney transplantation, anemia remains a common complication with a prevalence ranges from 20% to 60% at 1 year and >30% during the follow up.3–6 There is a growing body of literature documented anemia in the post-transplant period, demonstrating that anemic condition associated with a reduced graft and patient survival.7–9 However, little is known about the relationship between pre-transplant hemoglobin (Hb) level and long-term outcome in kidney recipients. Results from retrospective studies with small patient population indicated higher pre-transplant Hb conferred comparable patient and graft outcome, but a better graft function.10,11 Another study with the attempt to correct pre-transplant anemia with recombinant human erythropoietin (rHuEPO) showed that rHuEPO-hyporesponsive patient suffered higher mortality and higher incidence of graft failure at 5 years,12 suggesting an elevation in pre-transplant Hb may improve long-term outcome. Nevertheless, Muirhead et al.13 found that pre-transplantation Hb concentration did not significantly affect DGF, graft loss, or vascular thrombosis in kidney recipients.

Given the uncertainty of whether correcting anemic condition before transplantation will result in better long-term outcome, we reported our single-center experience, seeking to evaluate the usefulness of correcting anemia by reaching a targeting pre-transplant Hb level.

Materials and methods

Patients

Between 1 January 2006 and 30 June 2012, a total of 702 living-related kidney transplants were performed in our center. The demographic and clinical data were obtained by reviewing patients’ profile. As the previous studies revealed that patients with pre-transplant Hb level <10 g/dL experienced comparable short-term outcome but poorer renal function,11,12 patients were divided into two groups according to the pre-transplant Hb concentration: <10 g/dL and ≥10 g/dL. During the follow up period, the Hb concentration and serum creatinine were collected at months 3 and 6 after transplant and every 6 months thereafter.

Immunosuppressive regimen

Induction therapy by rabbit anti-human thymocyte immunoglobulin (1.0–1.5 mg kg⁻¹ d⁻¹ for 3–5 days) or anti-CD25 monoclonal antibody (1.0–1.5 mg kg⁻¹ right before the surgery and 10–14 days thereafter) was instituted according to the pre-transplantation evaluation of recipients. Maintenance immunosuppression therapy consisted of a calcineurin inhibitor [CNI; cyclosporine A (CsA) or tacrolimus (TAC)] in combination with mycophenolate mofetil (MMF) and prednisone. CsA or TAC was initiated at the 2nd day post-transplant with a fixed dosage of CsA 100 mg or TAC 2 mg, twice a day. MMF was given 1 day before the transplant with a dosage of 2.0 g d⁻¹. The dosage was adjusted according to the blood drug concentration, with a trough level for TAC of 5–8 ng mL⁻¹, CsA of 75–150 ng mL⁻¹ and MMF area under the curve 30–60 mg h/L⁻¹. Methylprednisolone was given on the day of surgery of 7 mg kg⁻¹ intravenously for 3 days, and then given as prednisone of 60 mg d⁻¹ and tapering 10 mg d⁻¹, maintained with 5–10 mg d⁻¹ thereafter.

Primary outcomes of the study were patient and graft survival, secondary outcomes were the incidence of DGF, the rejection free interval during the post-transplant period and the infection. Death censored graft survival is death with a functioning graft. DGF was defined as need for dialysis after transplantation. Rejection was diagnosed on the basis of an increase in serum creatinine, confirmed by examination of a biopsy sample in needed and was treated primarily with bolus doses of methylprednisolone and, in cases of refractory rejection, with antithymocyte globulin. Infection was defined as any infection the recipients get, including wound infection, pulmonary, urinary tract and skin infection. For further analysis, we also analyze the incidence of perioperational infection (<1 month after surgery).

Statistical analysis

Demographic characteristics were compared using the chi-squared test for category variables and means ± SE for continuous variables. Comparisons were made between groups in both primary and secondary outcomes. Quantitative variables were compared by Student’s t test or the Wilcoxon test, when appropriate. Logistic regression was applied to estimate the relationship between DGF, infection and pre-transplant Hb. The first acute rejection episode, patient and kidney graft survivals were performed according to the Kaplan–Meier method supplemented by hazard ratio (HR) calculations based on Cox proportional models. Statistical significance was indicated by a two-tailed p < 0.05 for all analysis. Data analysis was performed with SPSS 17.0 software (SPSS Inc., Chicago, IL).

Results

About 422 patients (60%) were of Hb level <10 g/dL (mean Hb, 78.30 ± 14.18 g/dL), 280 (40%) were >10 g/dL (116.2 ± 14.43 g/dL, Table 1). There is no difference in donor-related characteristics. Other than Hb level, all demographic data regarding recipients are of no significant discrepancy, except the higher Hb level group receiving dialysis of a longer period, lower rate of erythropoiesis-stimulating agent use and iron supplementation. Both groups are comparable in HLA mismatch, panel reactive antibody (PRA) screening and induction therapy. Majority use TAC and MMF as their maintenance immunosuppression regime, 62.32% in low Hb group and 71.07% in high Hb group. During the follow up, 47 recipients receiving CsA as primary immunosuppressant in low Hb group converted to TAC, and 15 in the high Hb group. Almost half conversion was because of acute rejection or creeping serum creatinine and the left was attributed to complications, such as hairy or gingival hyperplasia.

Table 1. Demographic data of included patients.

	Pre-transplant Hb < 10 g/dL	Pre-transplant Hb ≥ 10 g/dL	p
Donor gender (male/female)	139/283	94/186	0.957
Donor age (year)	46.10 ± 10.06	45.47 ± 9.71	0.41
Donor serum creatinine (μmol/L)	102.59 ± 19.58	102.92 ± 21.08	0.832
Recipient gender (male/female)	311/110	217/64	0.326
Recipient age (year)	32.78 ± 8.69	33.01 ± 8.99	0.731
BMI of recipient (kg/m^2)	21.41 ± 2.80	21.37 ± 2.97	0.887
Dialysis period (months)	7.41 ± 10.14	11.36 ± 13.01	<0.001
Recipient pre-transplant  hemoglobin (g/dL)	78.30 ± 14.18	116.2 ± 14.43	<0.001
Erythropoiesis-stimulating  agent (yes/no)	375/47	57/223	0.001
Iron supplement (yes/no)	368/54	46/234	<0.001
Previous transplantation (yes/no)	3/419	5/275	0.189
HLA mismatch
0	33	18
1	22	21
2	90	62
3	218	123	0.104
4	28	34
5	24	19
6	7	3
Panel reactice antibody (%)
0	368	245
1–20%	37	24	0.948
≥20%	16	12
Induction therapy
No	345	231
Anti-CD25 monoclonal antibody	48	35	0.561
ATG	29	14
Cacineurin inhibitor
TAC	246	181
CsA	129	71	0.232
CsA converted to TAC	47	28

Notes: BMI, body mass index; HLA, human leukocyte antigen; ATG, anti-human thymocyte immunoglobulin; TAC, tacrolimus; CsA, cyclosporin A.

Mean follow up period was 35.34 ± 18.12 months (ranged from 1 to 84 months). At 3 months, the Hb concentration of low Hb group reached 12.77 and 13.35 g/dL for high Hb group. Significant difference in Hb concentration between the two groups persists until 54 months later. However, no difference was detected in serum creatinine throughout the follow up period.

About 12 (1.71%) deaths occurred. Among them, nine were because of lung infection, two cardiovascular events and one intestine obstruction. Furthermore, there were 23 graft failures. Though there is a trend that low Hb group is of higher incidence of patient death and graft failure, the two groups did not differ significantly (2.38% vs. 0.71%, p = 0.096; and 4.04% vs. 2.14%, p = 0.165, respectively). Cox regression model found that pre-transplant Hb<10 g/dL was not associated with significantly increased overall mortality (HR = 3.379; 95% CI: 0.706–17.172) (Figure 1) and increased death censored allograft failure (HR = 1.556; 95% CI: 0.595–4.069) (Figure 2).

Figure 1. Cumulative patient survival in Hb < 10 g/dL versus Hb ≥ 10 g/dL group after transplantation.

Figure 2. Death censored renal allograft survival in Hb < 10 g/dL versus Hb ≥ 10 g/dL group after transplantation.

During the follow up, 27 (3.85%) recipients developed DGF, 16 in low Hb group, 11 in high Hb group, with similar incidences (3.8% vs. 3.91%, p = 0.939). Balancing other confounding factors, recipients with lower Hb level were of higher risk in developing DGF, but did not reach the level of significance (HR = 1.186, 95% CI: 0.53–2.654). About 127 (18.09%) recipients developed acute rejection, 84 (19.91%) in low Hb group, 43 (15.36%) in high Hb group (p = 0.15). About 81 rejection episodes (63.78%) occurred in the first month, and the average time since the operation was 6.46 ± 11.07 months. We did not observe any protective effect exerted by higher Hb level in acute rejection avoidance (HR = 1.338, 95% CI: 0.919–1.947) (Figure 3). With respect to overall infection, 97 (23.0%) recipients got infection in low Hb group during the follow up, and 53 (18.9%) in high Hb group (p = 0.308). Additionally, there was 26 perioperational infection in low Hb group and 16 in high Hb group (p = 0.887). No increased risk for low Hb level was found either in overall infection (HR = 1.263, 95% CI: 0.847–1.885) or in perioperational infection (HR = 1.019, 95% CI: 0.513–2.026).

Figure 3. The first acute rejection episodes in Hb < 10 g/dL versus Hb ≥ 10 g/dL group after kidney transplantation.

Discussion

The overall prevalence of CKD-associated anemia is ∼50%, and positively correlated to the advanced disease stage. It is estimated that three-quarters of CKD patients starting dialysis suffer from anemia.14 CKD-associated anemia increases morbidity and mortality from cardiovascular complications, including angina, left ventricular hypertrophy and worsening heart failure.15 In addition, anemia is an independent predictor of death in stable coronary artery disease patients with CKD.16 In patients received kidney transplantation, anemic conditions prevail, even years later.3–7 Post-transplant anemia has been proved to associate with greater risk of allograft loss,8,9,17 reduced patient survival.5,7,18 Furthermore, correction of post-transplant anemia will improve the long-term outcome by reducing progression of chronic allograft nephropathy.19

In the present study, pre-transplant Hb concentration is independent of poor allograft and patient survival, an observation in agreement with the findings of Na,11 who divided recipients into two groups: high Hb group (≥100 g/L) and low Hb group (<100 g/L), and followed up for 1 year, revealing similar incidence of acute rejection (7.9% vs. 11.4%, p > 0.05); DGF (11/68 vs. 7/56), comparable patients survival (98.7% vs. 96.8%) and graft survival (98.7% vs. 96.8%).11 Another two studies corroborated the aforementioned observations.12,20 Additionally, the latter two studies also showed that the significant difference in Hb concentration between the two groups persist at 1-year post-transplant. This is in accordance to our results, and comparable Hb level were not attained until 54 months after transplantation.

A report by Molnar et al.,7 who, in a prospective cohort study, found that anemia at baseline significantly predicted mortality and graft failure over 4-year follow-up. Another retrospective studies demonstrated that post-transplant anemia at 12 mo (defined as Hb < 12 g/dL) was associated with decreased patient and graft survival.5 Gheith et al.9 found that anemic condition (Hb < 13 g/dL in males and <12 g/dL in females) at 6 months post-transplant was associated with poorer graft survival, with 93.9% versus 84.8% at 10 years. Moreover, recipients of post-transplant anemia at 3 months (Hb <11 g/dL) were found of significant higher risk of overall mortality (HR = 3.18) and graft loss (HR = 2.67).21 Based on these results, we can extrapolate that only the anemic duration the allograft experienced would greatly affect the fate of the allograft, rather than the pre-transplant anemia. If the anemic condition can be reversed promptly, as our results indicated that both groups got anemic condition corrected at 3 months, the negative affect it exerted might be negligible.

While anemia in CKD can result from multiple mechanisms, decreased erythropoietin synthesis is the predominant one, attributing to tubulointerstitial fibrosis, which compromises renal erythropoietin synthetic capacity. After transplantation, with restoration of kidney filtration function and erythropoietin synthesis, most of the anemic condition could be corrected. As is shown in our results, the recipients of low Hb group get an Hb increment of 5 and 2 g/dL in high Hb group at 3 months. However, there is 28 patients (6.65%) in low Hb group did not reach a Hb concentration of >10 g/dL at 3 months. Even all receiving erythropoiesis-stimulating agents and iron supplementation, indicating a hyporesponsiveness of erythropoietin stimulation or the impaired bone marrow hematopoietic function. As suggested by Campise et al.,12 rHuEPO-hyporesponsive patients suffered a significant higher incidence of graft failure comparing to rHuEPO responders (50% vs. 41.7%), and a higher risk of graft failure (41.7% vs. 32%). That these population whose Hb concentration do not improve months after the restoration of kidney function need further investigation, and alternative treatment other than erythropoiesis-stimulating agents may required.

Our study has several limitations, and the results should be interpreted with caution. One important limitation is in the intrinsic design of the study. Our data, in fact, were collected retrospectively, similar with other studies in this area. A second important limitation is that, the Hb cut off value is arbitrary, because no consensus has reached on the management of pre-transplant anemia. Further investigation on this subject is needed to determine the optimal pre-transplantation Hb concentration.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.

This project was supported by the National Natural Science Foundation of China (Grant No. 30872579), [email protected]