Objectives: Renal allograft thrombosis is an important cause of early renal allograft loss. A previous study from our unit showed thrombosis rates in patients who received heparin that were similar to those who did not receive any thromboprophylaxis. This study evaluated the impact of aspirin prophylaxis on renal allograft thrombosis rates in pediatric renal transplant recipients.
Materials and Methods: We conducted a retrospective study of 456 consecutive pediatric allografts from deceased and living related donors over age 22 years in a single center. Routine perioperative heparin was introduced in 1994 and was subsequently changed to aspirin prophylaxis in 2000. Group 1 comprised 126 patients who did not receive any thromboprophylaxis, group 2 comprised 128 patients who received heparin, and group 3 comprised 202 patients who received aspirin therapy. Variables associated with increased risk of renal allograft loss were examined using multivariable logistic regression.
Results: Thrombosis occurred in 11% (14/126) of grafts in group 1, 9% (11/128) of grafts in group 2, and 1% (2/202) of grafts in group 3 (odds ratio for aspirin group = 0.38, 95% confidence interval, 0.22-0.64; P = .02). In patients who received aspirin (group 3), there was only one renal allograft loss secondary to hemorrhage, and no grafts were lost in patients younger than 5 years of age.
Conclusions: After our center introduced a change from heparin to aspirin prophylaxis, the thrombosis rate in pediatric renal allografts fell from 9% to 1%. Although there are a number of possible confounding variables, the introduction of aspirin has led to a reduced rate of renal allograft thrombosis.
Key words : Early graft loss, Heparin, Kidney transplant
Renal allograft thrombosis occurs in up to 12% of pediatric renal transplants1,2 and is a significant cause of early renal allograft loss, particularly in the first 4 weeks postoperatively. Factors associated with increased risk of renal allograft thrombosis include young age of donor and recipient (< 5 years due to vessel size), multiple donor vessels, prolonged cold ischemia time, surgical complications, right-sided donor grafts, and thrombophilic hypercoagulable states.3-11 Children who have received renal allografts are at high risk of thrombosis, and investigations have focused on whether prophylactic treatment can result in decreases in renal allograft thrombosis and loss. However, any benefits of antithrombotic therapies must be balanced against a potentially increased risk of surgical complications, including hemorrhage, blood transfusions, and surgical reexploration.3,6,12
Low-dose aspirin prophylaxis has been reported to be beneficial in terms of reduced early renal allograft thrombosis in adult renal transplant recipients.12,13 However, the role of aspirin in pediatric renal transplant recipients has not yet been reported in the literature.
Our previous retrospective study in our unit examined the effects of heparin prophylaxis on pediatric renal allograft thrombosis and concluded that there was no significant reduction in graft thrombosis in patients who received heparin.3,5 This retrospective study aimed to evaluate the effects of a change to aspirin prophylaxis on the incidence of pediatric renal allograft thrombosis.
Materials and Methods
We studied all recipient and donor data for children who received renal transplants over 22 years. Only first renal allografts were included in our study to prevent reanalyses of children, as this would violate the independent assumption on which our logistic regression analysis is based. Seven patients were excluded because of inadequate donor details (there were no renal allograft losses in these patients).
Data for factors known to increase the risks of renal allograft thrombosis were collected from medical records. These included recipient and donor age and sex, donor status (living or deceased donor), cold ischemia time (CIT), graft vasculature (single or multiple vessels), side of organ donation, site of anastomosis (aorta/inferior vena cava or iliac vessels), and surgical complications. From 2000 onward, patients were screened for inherited and acquired thrombophilic disorders, including protein C and S and antithrombin III deficiency, factor V Leiden mutation, prothrombin mutation (G20210A), MTHFR mutation, and lupus anticoagulant and anticardiolipin antibodies in those with positive personal or family history. Demographic data, time to renal allograft loss, and histopathology of graft loss were also compiled.
The primary outcome measure was renal allograft thrombosis, as defined by histopathologic examination with exclusion of acute rejection. In addition, cases of renal allograft loss secondary to hemorrhage and those requiring surgical reexploration for bleeding were recorded. The day of nephrectomy was recorded as the day of renal allograft loss.
Before 1994, children undergoing renal transplant did not routinely receive aspirin or heparin prophylaxis. Unfractionated subcutaneous heparin following induction of anesthesia was introduced as routine prophylaxis in 1994, with thrice daily administration thereafter (3000 U/day for children under 15 kg, 4500 U/day for children weighing 15-20 kg, and 7500 U/day for children weighing over 20 kg). Activated partial thrombin time (APTT) was maintained at less than twice the control value. Heparin was discontinued once patients were fully mobile at 7 to 10 days posttransplant. In 2000, oral aspirin (1 mg/kg with maximum dose of 75 mg/day) prophylaxis from induction until at least 1 month posttransplant (unless bleeding problems occurred) was introduced. Aspirin was not discontinued at the time of renal allograft biopsies (as usually they were urgent biopsies without time to reverse effects).
There were 126 patients who did not receive any thrombosis prophylaxis (group 1), 128 patients who received heparin (group 2), and 202 patients who received aspirin (group 3). In all patients, adequate peri- and postoperative perfusion was achieved by maintaining the central venous pressure above 10 cm H2O and ensuring that toe core temperature was less than 2°C.
Immunosuppressive regimens included prednisolone, azathioprine, and cyclosporine until 1997 when cyclosporine was replaced by tacrolimus. Graft nephrectomies were examined histopathologically.
Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 19.0, IBM Corporation, Armonk, NY, USA). Descriptive statistics were produced for previously described risk factors for graft loss (young age of donor and recipient [< 5 y], sex of recipient and donor, donor status, multiple donor vessels, prolonged CIT, side of graft donation, site of anastomosis, and surgical complications). The Pearson chi-square test was calculated to assess whether these risk factors differed significantly among the 3 groups. Each potential risk factor was analyzed by univariate logistic regression to determine whether there was an association with renal allograft loss, and significant risk factors were included in a multivariable analysis using a stepwise backward elimination model. The level of significance was set at P < .05.
Our total cohort included 456 renal transplant recipients (64% males) aged less than 18 years (range, 1-18 y; median of 9 y), of which 62% received deceased-donor renal transplants.
The primary diagnosis of end-stage kidney disease in our cohort was 35% with congenital anomalies of the kidney and urinary tract with renal dysplasia with or without vesicoureteric reflux and with or without posterior urethral valves. We observed significant differences among the treatment groups with regard to patient characteristics (Table 1). The aspirin-treated group (group 3) had fewer recipients under 5 years of age than the other 2 groups: 33 patients (16%) in group 3 versus 39 patients (31%) in group 1 and 35 patients (27%) in group 2 (P = .002). Group 3 also had no donors under 5 years of age compared with 19 (15%) in group 1 and 9 (7%) in group 2 (P < .001). Not unexpectedly, patients treated more recently received proportionally more grafts from living donors (102 in group 3 vs 29 in group 1 and 32 in group 2; P < .001); therefore, there were more left-sided grafts (133 in group 3 vs 52 in group 1 and 55 in group 2; P < .001), fewer renal allografts with CIT of more than 24 hours (6 in group 3 vs 40 in group 1 and 29 in group 2; P < .001), and fewer grafts with multiple vessels (53 in group 3 vs 25 in group 1 and 19 in group 2; P = .007).
Renal allograft loss
There were 29 graft losses (6.4%), with 27 (5.9%) due to renal allograft thrombosis and 2 (0.4%) due to hemorrhage. We observed a significantly lower incidence of renal allograft thrombosis in patients who received aspirin, with 2 cases (1%) in group 3 versus 14 cases (11%) in group 1 and 12 cases (9%) in group 2 (odds ratio [OR] for aspirin group = 0.38, 95% confidence interval [95% CI], 0.22-0.64; P = .02) (Figure 1). There was one renal allograft loss secondary to hemorrhage from the anastomotic site each in group 3 (0.4%) and group 2 (0.7%). We observed no significant difference in the median time to renal allograft loss secondary to thrombosis between the groups, with results of 6 days (interquartile range [IQR], 2-10 d) in group 1, 10 days (IQR, 5-11 d) in group 2, and 13 days (IQR, 12-14 d) in group 3. The APTT was < 60 seconds in all cases of renal allograft loss secondary to thrombosis in those who received heparin (group 2). The APTT was 120 seconds in the single case of renal allograft loss secondary to hemorrhage in the same group.
Effects of recipient age, donor age, cold ischemia time, side of graft, and site
of anastomosis on graft loss
Univariate logistic regression analysis showed that other significant variables for renal allograft thrombosis were lower recipient and donor age, prolonged CIT, left-sided graft, and anastomosis to the aorta (Table 2).
The odds of graft thrombosis were increased in recipients aged 5 years or less.
Recipients aged from 6 to 10 years had lower odds of graft thrombosis
(OR of 0.10; 95% CI, 0.03-0.38; P = .001), as did recipients aged from 11 to 16 years (OR of 0.17; 95% CI, 0.04-0.65; P = .01). The odds of graft thrombosis were increased in recipients of grafts from younger donors. Recipients with donors who were 6 to 15 years old had lower odds of graft thrombosis (OR of 0.22; 95% CI, 0.06-0.76; P = .016) than those with donors who were 5 years old or less, as did recipients whose donors were > 16 years old (OR of 0.28; 95% CI, 0.11-0.75; P = .011).
Patients who received grafts with a CIT of > 24 hours had higher odds of renal allograft thrombosis than those with a CIT of < 24 hours (OR of 4.37; 95% CI, 1.77-10.80; P = .001). Left-sided grafts had higher odds of thrombosis than right-sided grafts (OR of 3.49; 95% CI, 1.38-8.84; P = .008). Recipients with grafts anastomosed to the iliac had lower odds of loss of graft due to thrombosis than those with grafts anastomosed to the aorta (OR of 0.24; 95% CI, 0.09-0.63; P = .004).
Recipient sex and donor sex had no significant effect on renal allograft thrombosis (P = .117 and P = .248, respectively). Donor status (deceased or living donation; P = .356) and number of graft vessels (single or multiple; P = .427) also had no significant effect on renal allograft thrombosis.
To adjust for confounding variables, multivariable logistic regression was performed using risk factors found to be significant by univariate analysis. A main effects model of all significant variables (treatment group, recipient age, donor age, CIT, side of graft, and site of anastomosis) showed that, when these confounding variables were factored in, the effect of aspirin on reducing the odds of graft thrombosis was no longer significant (OR of 0.70; 95% CI, 0.13-3.91; P = .687). However, a stepwise backward elimination analysis showed that renal allograft thrombosis in our data can be predicted with a model using treatment group and recipient age only (Table 3). Therefore, although the use of prophylaxis was not an independent predictor of renal allograft thrombosis, the effect of aspirin on reducing the risk of graft loss is significant when considered in conjunction with the age of the recipient.
Thrombophilic risk factors
Results of thrombophilic screens were available for 179 patients transplanted from 2000 onward. Most of these patients (64.8%) were given aspirin, with 62 patients (34.6%) receiving prophylaxis and only 1 patient receiving heparin. Thrombophilic risk factors were identified in 38 patients (21.2%): 12 patients (7%) were homozygous for the MTHFR mutation, 3 patients (2%) were heterozygous for the factor V Leiden mutation, 4 patients (2%) were heterozygous for the prothrombin mutation, 13 patients (7.3%) were positive for anticardiolipin antibodies, 3 patients (1.7%) had protein S deficiency, 2 patients had both factor V Leiden mutation and antiphospholipid antibodies, and 1 patient had factor V Leiden and homozygous MTHFR mutations. Of patients with identified risk factors, 16 (42%) did not receive prophylaxis and 22 (58%) were given aspirin.
There were 5 cases of renal allograft thrombosis within the cohort of patients who had available thrombophilic screening results: 2 in the no prophylaxis group (group 1), 2 in the aspirin group (group 3), and 1 who received heparin (group 2). Interestingly, both patients in the group 1 who developed renal allograft thrombosis were homozygous for the MTHFR mutation, and both patients in group 3 with graft thrombosis were heterozygous. No thrombophilic risk factors were identified in the patient in group 2 who developed renal allograft thrombosis.
Histopathologic findings at graft nephrectomy
On histopathologic examination, thrombus was identified in the renal vein in 8 grafts (6 in group 1, 2 in group 2) due to presumed renal venous thrombosis. There was no evidence of acute rejection in the nephrectomy specimens. A thrombus was seen in the renal artery in 7 grafts (1 in group 1, 5 in group 2, 1 in group 3) due to presumed renal artery thrombosis. Thrombus was seen in both the renal artery and the vein in 2 grafts, one each from groups 1 and 3. In 10 graft nephrectomies, 6 had intrarenal thrombosis and 4 were necrotic on examination.
There were 2 renal allograft losses secondary to hemorrhage: 1 (1%) from a patient who received heparin (group 2) and 1 (1%) from a patient who received aspirin (group 3). Surgical reexploration for bleeding complications was necessary in 5 patients in group 2 (18%) and 2 patients in group 3 (1%). Among those who received heparin (group 2), 3 had bleeding from the site of anastomosis and 2 had clot retention at the ureteric stent. One patient had a prolonged APTT of > 120 seconds; in the other patients, the APTT was within the target range. In the patients receiving aspirin (group 3), 1 had bleeding from the site of anastomosis and 1 patient had an unidentified bleeding point.
There were no cases of adverse reactions (such as Reye’s syndrome) or complications (such as gastrointestinal bleeding) related to aspirin therapy reported among patients in our study.
This is the first report in the published literature of the effects of low-dose aspirin on the rate of graft thrombosis in pediatric renal transplant recipients. This study confirmed that renal allograft thrombosis remains a significant cause of early renal allograft loss in pediatric renal transplant patients.1,4,5 In our study of patients seen over 22 years at our center, the overall rate of renal allograft thrombosis was 6%, a rate comparable to that reported in adults.14 Moreover, this study demonstrated that low-dose aspirin prophylaxis is associated with a significant decrease in the rate of pediatric renal allograft thrombosis to 1%. Analysis of group characteristics did reveal potential confounding variables among the 3 treatment groups.
Before 2000, we did not screen for thrombophilia and may have missed relevant cases that resulted in thrombosis, with transplants performed by different surgical teams during that era. A change in practice over time may have led to decreased incidence of risk factors for thrombosis in the aspirin-treated group, including older donors and recipients and more living donations and therefore more left-sided procedures and single vessel grafts with shorter mean CIT. However, because the introduction of aspirin did result in a significant reduction in graft thrombosis, from 11% in patients receiving no prophylaxis and 9% in those receiving heparin to 1% in those receiving aspirin, we suggest that the role of aspirin in this patient population deserves further analysis.
In the adult renal transplant population, previous retrospective studies have reported a significant reduction in renal venous thrombosis from 6% to 1%13 and in graft thrombosis from 5% to 0%15 in patients receiving aspirin.
Our data concurred with previous reports that prolonged CIT and young donor and recipient age significantly increase the incidence of renal allograft thrombosis.6,7 In addition, in our study, left-sided graft and anastomosis to the aorta were both associated with an increased risk of graft thrombosis. Our data were unable to show that aspirin prophylaxis significantly decreased the rate of thrombosis independent of all other risk factors. However, we were able to accurately predict graft thrombosis using a model involving only aspirin prophylaxis and recipient age.
The incidence of renal allograft thrombosis in patients under 2 years of age has been reported to be as high as 30%.14 Of note, in our study, there were no graft losses in renal transplant recipients under 5 years of age who received aspirin.
The beneficial effects of thromboprophylaxis must be considered in balance with the risk of hemorrhagic complications. Other groups have published conflicting data concerning the bleeding events in pediatric renal transplant recipients who receive heparin treatment. One study showed that anticoagulation was associated with hemorrhage,6 whereas further separate studies found no significant increased risk of bleeding.12-17 To date, there are no published data on the incidence of bleeding complications in pediatric renal transplant recipients who received aspirin. In our study, there were 9 cases of complications resulting from hemorrhage, with 6 recipients who received heparin and 3 recipients who received aspirin. Of these, there were 2 incidences of renal allograft loss secondary to hemorrhage: 1 in the heparin-treated group (group 2) and 1 in the aspirin-treated group (group 3). There were no recorded incidences of bleeding complications following renal allograft biopsies.
Aspirin inhibits cyclooxygenase, thereby reducing production of thromboxane A in circulating platelets, a potent platelet aggregator and vasoconstrictor. Cyclooxygenase inhibition also leads to reduced production of prostacyclin (a vasodilator) by vascular endothelium. Low-dose aspirin affects circulating platelets to a greater degree than vascular endothelium and thereby reduces platelet aggregation and thrombosis.6 In addition, aspirin inhibits eicosanoid production, which has been reported to lead to an attenuation of renal ischemic-reperfusion injury and allograft nephropathy in animal models.18-20 Of note, 1 study reported a trend toward a lower rate of chronic allograft nephropathy with aspirin treatment, although this did not reach statistical significance.15 Theoretically, therefore, aspirin has the potential to reduce not only early renal allograft loss, as shown in our study, but also late renal allograft loss.15 We suggest that further studies concerning long-term renal allograft survival in pediatric renal transplant recipients receiving aspirin are required.
Renal allograft thrombosis is thought to be more commonly due to renal vein rather than renal artery thrombosis.14-16 This fact was reflected in our study in that, of the grafts lost in those who did not receive prophylaxis, 6 were due to renal vein thrombosis and only 1 was due to renal artery thrombosis. Interestingly, of the thrombosed allografts in recipients who received heparin, proportionally more showed evidence of renal artery thrombosis (5 cases of renal artery pathology vs only 2 cases of renal vein thrombosis). In the 2 cases of graft thrombosis in recipients who received aspirin, 1 showed pathologic evidence of renal artery thrombosis and the other showed both renal artery and renal vein involvement with thrombus.
Recent reports have highlighted the importance of hypercoagulable disorders, such as protein C and S and anti-thrombin III deficiency, antiphospholipid antibodies, and factor V Leiden, prothrombin, and MTHFR mutations, as risk factors for renal allograft thrombosis in children.21,22 In our study, from the cohort of patients who had available thrombophilic screening results, both patients who developed thrombosis without prophylaxis were homozygous for the MTHFR mutation and both patients on aspirin prophylaxis who developed graft thrombosis were heterozygous. In 2006, Kranz and associates21 reported no cases of renal allograft thrombosis in 18 pediatric renal transplant recipients with hypercoagulable disorders treated with low-molecular-weight heparin followed by aspirin, with 1 serious bleeding complication. The optimal thromboprophylaxis regimen for such patients remains to be established.
Our retrospective single-center study demonstrated a significant reduction in the rate of renal allograft thrombosis following aspirin prophylaxis in pediatric renal transplant recipients. Although the comparison cohorts were historical groups and we acknowledge the potential confounding differences in group characteristics, the magnitude of the effect of aspirin in reducing the rate of renal allograft thrombosis is such that further investigations regarding its role in this patient population are clearly imperative.
Volume : 18
Issue : 2
Pages : 157 - 163
DOI : 10.6002/ect.2018.0358
From the 1Department of Renal Transplantation, Royal Free Hospital NHS Trust, London; the 2Regional Paediatric Nephro-Urology Service, Southampton University Hospitals NHS Trust, Southampton; the 3Department of Paediatric Nephrology, Gazi University; the 4Department of Paediatric Nephrology and 5Department of Haematology, Great Ormond Street Hospital for Children NHS Foundation Trust, London; the 6University College London, Great Ormond Street Institute of Child Health, London; and the 7Department of Transplantation, Guy’s and St Thomas’NHS Foundation Trust, London, United Kingdom
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare. This research was performed at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health, which was made possible by support by the NIHR Great Ormond Street Hospital Biomedical Research Centre and the NIHR Biomedical Research Centre at Guy’s and St Thomas’s NHS Foundation Trust and King’s College London. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health.
Corresponding author: Ammar Al Midani, Renal unit Royal Free NHS Foundation Trust, London NW3 2QG, United Kingdom
Phone: +44 203 3055628
Table 1. Characteristics of Pediatric Renal Transplant Recipients by Treatment Group
Table 2. Univariate Logistic Regression Analysis (Outcome Is Renal Allograft Thrombosis)
Table 3. Multivariate Logistic Regression Analysis (Outcome Is Graft Thrombosis)
Figure 1. Rate of Renal Allograft Thrombosis in Children Receiving No Thromboprophylaxis, Heparin Therapy, or Aspirin Therapy