TRANSPLANT RENAL ARTERY STENOSIS IN 77 PATIENTS-DOES IT HAVE AN IMMUNOLOGICAL CAUSE?¹

Abstract

Transplant renal artery stenosis (TRAS) is a common complication after transplantation and is an important cause of graft dysfunction. Damage from graft rejection, trauma, and atherosclerosis have been implicated as possible causes. We reviewed all 917 patients transplanted in our unit since 1978 to study the prevalence, clinical features, and possible causes of TRAS. Seventy-seven patients with TRAS were identified. The detected incidence was 2.4% before the introduction of color doppler ultrasonography (CDU) and rose to 12.4% after CcU was introduced in 1985, giving an overall incidence of 8.4% during a mean follow-up period of 6.9 years. The TRAS group was compared with a control group of 77 transplanted patients matched for age, year of transplant, sex, and number of previous grafts. Mean ages for the study and control groups were 43.6±15 and 44.8±13.7 yr. A total of 25% of cases of TRAS were diagnosed within the first 8 wk of transplantation and in 60% within the first 30 wk (median=23 wk). All patients were treated with angioplasty, 28 patients had recurrence of TRAS requiring multiple angioplasties (maximum 5) and 1 went on to have surgery. Angioplasty resulted in a significant fall in plasma creatinine. Patient and graft survival were significantly worse in the TRAS group: 69% vs. 83%(P<0.05) and 56% vs. 74%(P<0.05) (TRAS vs. Control), respectively. There was a significantly higher incidence of rejection, especially cellular rejection in the TRAS group, 0.67 vs. 0.35 episodes per patient(P<0.01) (TRAS vs. Control). Recurrence but not occurrence of TRAS was associated with the use of cyclosporine.

Transplant renal artery stenosis (TRAS)^* has a reported incidence of between 1.6% and 16% ⁽¹⁾. It can lead to graft dysfunction and hypertension. Angioplasty is an established form of treatment that has been shown to be effective^(2-5). Damage from graft rejection has been suspected as a possible cause ^(6-9). Other factors suggested include atherosclerotic changes, faulty suture technique, renal artery trauma, perfusion injury, turbulent flow secondary to malpositioning of the kidney, and kinking or compression of the artery⁽¹⁰⁾. Although it has been associated with cellular rejection ⁽⁹⁾, published data have only involved a small number of patients. We therefore retrospectively reviewed all 917 patients transplanted in our unit in the past 16 years to address these points.

MATERIALS AND METHODS

All patients who received a cadaveric renal allograft at this unit between January 1978 and April 1994 with TRAS diagnosed by angiography were identified by examining medical and radiological records. A stenosis of 50% or more was considered significant. Since 1985, color doppler ultrasonography (CDU) had been used at least weekly during the immediate postoperative period, monthly during the first year, and as annual screening thereafter or when clinically indicated. Abnormal Doppler findings (peak systolic velocities of >1.5 m/s or an increase of over 50% in peak systolic velocity in the course of the renal artery) were followed up with angiography ⁽¹¹⁾. This screening method for TRAS has been shown to be effective by our unit with a sensitivity of 100% and specificity of 75% ⁽¹¹⁾ and by others ^{(12, 13)}.

Comparison was made with a case control group matched for age (±10 years), year of transplant, sex, and previous number of grafts. Since patients with failed grafts were not screened for TRAS, in order to compare meaningfully patient and graft survival in the TRAS and control groups, an additional matching criterion was that the control graft should be functioning after transplant when the TRAS was diagnosed in the matched pair. This matching was performed without knowledge of the plasma creatinine in any potential control patient. There were 10 patients with diabetes in both the TRAS patient and control groups. Ischemic heart disease was present in 6 and 5 TRAS and control group patients, respectively. Peripheral vascular disease existed in 3 patients in the TRAS group and in 1 patient in the control group. None of these figures reach statistical significance. All patients in the study and control groups were transplanted by the same surgeon (M.B.). Patients were on prednisolone and azathioprine. Cyclosporine was added to the regimen in 1982.

Rejection episodes were diagnosed on biopsy specimens and treated according to protocol: high-dose steroid for cellular rejection and antithymocyte globulin, antilymphocyte globulin or OKT3 for vascular or steroid-resistant rejections. Student's t test and the chisquare test were used for statistical analysis.

RESULTS

Seventy-seven patients with TRAS were identified during a mean follow-up period of 6.9 years. Mean ages for the study and control groups were 43.6±15 and 44.8±13.7 years, respectively. The incidence between 1978 and 1984 was 2.4% of all transplanted patients and rose to 12.4% during the period between 1985 and 1994 when CDU was used (Fig. 1).

Time distribution. Twenty-five percent of all cases were diagnosed within the first 8 weeks and 60% within the first 30 weeks posttransplant (median=23 weeks) (Figs. 2 and 3).

Outcome. All TRAS patients were treated with angioplasty and 1 went on to have surgical reconstruction. Overall patient survivals were 69% and 83% in the TRAS group and control, respectively (P<0.05). Graft survival were 56% vs. 74% in the TRAS group and control, respectively(P<0.05) Figs. 4 and 5.

Rejection episodes. Rejection episodes were more frequent in the TRAS group than in the control, as shown in Table 1. There were more patients with rejection, especially cellular, in the study group and a nearby two-fold increase in the average number of cellular rejection episodes per patient in the study group compared with the control group. Vascular rejection was also more common in the study group, but the difference did not reach statistical significance.

HLA matching. Mean class I HLA matching in TRAS patients was 0.68±0.77 (out of a possible 4 for A and B loci) and for control patients it was 0.71±0.78 (not significant). Not enough data were obtained for HLA class II matching to draw any conclusions.

Plasma creatinine. Plasma creatinine was significantly higher in the TRAS group before angioplasty compared with the control group at the equivalent time (P<0.01) but fell and became similar to the control group 1 and 6 months after angioplasty (Fig. 6).

Cyclosporine, blood pressure, hemoglobin, platelets, and antihypertensive therapies. The proportion of patients receiving cyclosporine was similar in the TRAS and control groups. There was no significant difference in blood pressure, hemoglobin, and platelets in the study and control patients before or after angioplasty. The numbers of antihypertensive drugs used in the control and TRAS patients before and after angioplasty were also similar (Table 1).

Recurrent TRAS. Twenty-eight patients had recurrent TRAS and required multiple angioplasties, with 1 patient requiring 5. The time between the diagnosis of TRAS and its first recurrence was highly variable, with a mean of 21.8±20.7 wk. Recurrence of TRAS was associated with the use of cyclosporine but not with the number of previous grafts or rejection episodes(Table 2).

DISCUSSION

In our series, the incidence of TRAS was low before introduction of CDU, suggesting that some patients were not diagnosed during that period. Cyclosporine was added to the immunosuppressive regimen in 1982, and this may have contributed to the rise in incidence of TRAS in the mid 1980s. The number of patients on cyclosporine in the TRAS group was similar to that in the control group (Table 1). This may be due to the fact that the matched control patients were chosen from patients who were transplanted in the same year and therefore treated according to the same immunosuppression protocol. Most of our cases of TRAS occurred earlier than previously described⁽⁹⁾. This may be a result of regular Doppler ultrasonography screening in the present study. Graft survival was poorer in the TRAS group; this may at least be partly due to the higher incidence of rejection. Poorer patient survival in the TRAS group may be a reflection of poorer graft survival, as patients had to return to other modes of renal replacement therapy, presumably with higher long-term mortality rates.

TRAS was found to be associated with graft rejection, especially cellular rejection. The fact that the majority of the stenoses were found distal to the surgical anastomosis suggests that technical problems were not a common cause in these patients. The pathological features are arteritis with intimal proliferation, derangement of the internal elastic membrane, and secondary aggregation of platelets with deposition of fibrin ⁽⁷⁾. Fibrinoid material has been reported in TRAS ⁽⁸⁾. A similar picture is seen in cardiac allografts, where the incidence of coronary arteriosclerosis is associated with the severity of acute rejection⁽¹⁴⁾. The pathological appearances of the coronary arteries included intimal fibrosis, fibrofatty plaques, cellular intimal thickening, and concentric foam cell lesions. During vascular rejection of a renal allograft, inflammation occurs within small arteries. Therefore it might be hypothesized that the same process that occurs within coronary arteries and intrarenal arteries during rejection may also occur in the main renal artery, causing damage and scarring and eventually leading to stenosis. However, this has not been our observation or that of Marcia et al. ⁽⁹⁾. There are two possible explanations: (1) vascular rejection may also be a causative factor for TRAS but the treatment for vascular rejection with antithymocyte globulin, antilymphocyte globulin, or OKT3 may protect the renal artery from immunological damage or (2) vascular rejection does not result in injury to the transplanted artery but has some specificity for small vessels.

It may be argued that because of the link between TRAS and raised plasma creatinine, cellular rejection was overdiagnosed in TRAS patients because of the increased biopsy rate in response to the raised creatinine. We think that this is unlikely: only rejection episodes before the diagnosis of TRAS were included in our study, and, since its introduction, all patients undergoing renal biopsies had CDU screening immediately before biopsy. CDU has been shown by our unit to have virtually 100% sensitivity. Accordingly patients who suffered rejection episodes before the diagnosis of TRAS were unlikely to have been biopsied for TRAS-induced raised creatinine. In view of this, the higher incidence of cellular rejection in our TRAS patients is likely to be a true phenomenon and not due to increased biopsy rate and over-diagnosis of cellular rejection.

We have demonstrated in this series that the recurrence but not the occurrence of TRAS is associated with the use of cyclosporine. This association must be interpreted with caution as the cyclosporine era also corresponds to the use of Doppler ultrasonography, which may have improved the detection rate. Cyclosporine may be associated with vascular damage within the graft, and the same process may take place in the transplant renal artery, which has already been damaged by immunological or other causes, leading to a higher incidence of recurrence.

The mean blood pressure did not fall significantly after angioplasty in our series, while other investigators have demonstrated a reduction^{(3, 5)}. There was no significant difference in the number of antihypertensive agents used before and one month after angioplasty. This may be due to good blood pressure control in this series of patients before angioplasty, and perhaps a follow-up period longer than one month is required to detect its beneficial effect on blood pressure.

Creatinine levels fell at 1 and 6 months from levels significantly higher than control to levels similar to the control. Although we have not compared patients treated with angioplasty with patients who did not have treatment for their TRAS, our data at least provide circumstantial evidence supporting the therapeutic use of angioplasty.

The present study demonstrates that TRAS is a common and treatable condition and suggests that its etiology has an immunological component. However, the cause of TRAS is likely to be multifactorial and other precipitating causes must also be important, as some of our patients with TRAS did not have any rejection episodes, as judged by a rise in plasma creatinine and conventional renal biopsy results. Patients most at risk were those who had been transplanted within the previous 30 weeks and had had episodes of cellular rejection. In our unit the routine use of Doppler ultrasonography has been of considerable benefit in detection of this condition. The diagnosis of TRAS should be considered, investigated, and treated with angioplasty if there is a deterioration in graft function, resistant hypertension, or murmur over the graft.

Figure 1: The incidence of TRAS as a percentage of total transplants performed in that year. Color Doppler ultrasonography (CDU) was introduced in 1985.

Figure 2: The number of patients diagnosed with TRAS in relation to time after transplant.

Figure 3: The distribution of time of diagnosis of TRAS in patients diagnosed with TRAS in the first year.

Figure 4: Kaplan-Meier survival graphs for patient survival in patients with TRAS and control patients. The numbers next to curves represent the numbers of patients in the TRAS and control groups at 5, 10, and 16 years, respectively (^* P<0.05, ⁺ P<0.01).

Figure 5: Kaplan-Meier survival graphs for graft survival in patients with TRAS and control patients. The numbers next to curves represent the number of patients in the TRAS and control groups at 5, 10, and 16 years, respectively(^* P<0.05, ⁺ P<0.01,^P<0.001).

Figure 6: Plasma creatinine (μmol/L) in TRAS and control patients before and after angioplasty. (^* P<0.05 TRAS vs. control before angioplasty; ⁺ P<0.05 TRAS before vs. TRAS after angioplasty). Error bars represent standard error of the mean.