This is a retrospective, observational study, including consecutive patients collected from the institutional interventional radiology database who underwent endovascular aorto-iliac intervention for the management of late transplant kidney dysfunction in the authors’ institution from January 2005 till August 2022. Patients’ demographic data and pre-and post-interventional laboratory and clinical data were collected from the patients’ electronic medical records. Patients’ cardiovascular risk factors were summarized based on Rutherford’s standards for lower extremity ischemia [12].

The institutional Ethics Committee approved this retrospective study (MP 022096).

An upstream aorto-iliac stenosis was considered as symptomatic if the patient presented with a concomitant serum creatinine decrease of >30% over a time period of less than 6 months and/or if the patient presented with a progressive and resistant increase in arterial blood pressure up to 150/90 mmHg or more over a time period of less than 6 months, despite administration of additional antihypertensive drugs if supported by the patient.

The arterial RI was measured in the upper pole, mid-portion, and lower pole of the transplant kidney; RI calculation was performed according to Naesens et al., as [1-(V_min ÷ V_max) ] with V_min representing the end-diastolic velocity and V_max the peak systolic velocity [13]. An RI in renal transplants is within normal limits between 0.6 and 0.8; <0.6 and >0.8 were defined as significantly low and significantly high RIs, respectively [11, 13]. Further, the intrarenal Doppler waveform was also analyzed: a ‘tardus (slow) et parvus’ (little) waveform reveals a dampened flow and can be used as an indirect sign of a proximal vascular stenosis.

CTA was performed in patients with an acceptable, residual renal function (serum creatinine <1.2 mg/dL). The arterial phase was performed after intravenous injection of 80 mL of iohexol iodized contrast medium (Omnipaque 350, GE Healthcare, Oslo, Norway) at a flow rate of 4 mL/s and when the automated triggering in the proximal abdominal aorta reached 100 Hounsfield units.

Contrast-enhanced MRA was performed after intravenous injection of gadobutrol 0.1 mL/kg (Gadovist, Bayer AG, Berlin, Germany) at a dose of 0.5 mmol/mL and a volume based on the formula patient weight (kg) x 0.2. A care bolus injection was also performed to determine the optimal arterial phases.

After obtaining the patient’s informed consent, the contralateral common femoral artery was punctured under local anesthesia and a 55 or 65 cm long, 6 or 7 French (F) sheath (Flexor, Cook Medical, Bloomington, IN, USA or Arrow sheath, Teleflex, Morrisville, NC, USA) was placed. Angiographic mapping of the aorto-iliac vessels was performed with iohexol iodized contrast medium (Omnipaque 270, GE Healthcare, Oslo, Norway) and/or carbon dioxide (CO2) (Angiodroid, CO2-injector, San Lazzaro di Savena, Italy or CO2-Angioset, Optimed, Ettlingen, Germany). The iliac stenosis was defined as occlusion, high-grade stenosis (>90% stenosis), intermediate grade stenosis (70%–90% stenosis), or low-grade stenosis (<70% stenosis). After an intra-arterial administration of 5000 international units of heparin, the stenosis was retrogradely cannulated and predilated with a 6 mm standard angioplasty balloon (Admiral Xtreme, Medtronic, Minneapolis, MN, USA and Mustang, Boston Scientific Inc, Cork, Ireland); in case of vessel wall recoil, a balloon-expandable stent was placed in the common iliac artery. Depending on the time period, various types of stents were used including Cordis Palmaz-Genesis (Cardinal Healthcare, Miami Lakes, FL, USA), Dynamic, [Biotronik, Berlin, Germany), Express Vascular (Boston Scientific Inc, Natick, MA, USA), Scuba Invatec (Medtronic, Minneapolis, MN, USA); in case of an aortic stenosis, a covered balloon-expandable stent (CP-stent, Numed, Hopkinton, NY, USA) was inserted and, in case of a stenosis at the distal end of the common iliac artery or proximal external iliac artery, a self-expanding nitinol stent (Zilver, Cook Medical, Bloomington, IN, USA) was inserted. Finally, if an additional stent in the transplant renal artery was needed, a balloon-expandable renal stent (Tsunami, Terumo Europe, Leuven, Belgium) was used. After completion of the angiography, the puncture site was closed with a closure device (Angioseal, Terumo Europe, Leuven, Belgium) and the patient was prescribed 80 mg daily aspirin lifelong.

Patients were followed up by the attending nephrologist; the study follow-up period ended in January 2024. Primary outcome data included changes in RI, in serum creatinine levels, in systolic and diastolic blood pressure, in number of administered antihypertensive drugs, and in daily dose intensity (DDI) for antihypertensive drugs, measured according to Min et al. [14]. Secondary endpoints included transplant kidney and overall patient survival.

Duplex-ultrasound with measurements of the RI were performed 1 week, 1 month, 6 months, and yearly after the index endovascular procedure. Laboratory analysis and clinical evaluation were performed at each follow-up visit; for comparative analysis, serum creatinine levels were collected before and 1 and 3 days after intervention as well as 1 week and 2, 3, and 9 months after the index endovascular procedure. Blood pressure measurement and number of prescribed antihypertensive drugs were collected at 1 week, 1 and 6 months, and 1 year after the index endovascular procedure.

Descriptive statistics were used to characterize the demographic data of the study population.

Linear mixed models for longitudinal measurements were used to analyze the effect of endovascular intervention on blood pressure, serum creatinine level, the number of used antihypertensive drugs, and the intrarenal RI of the transplant kidney. A random intercept was modelled to account for data clustering. Results are reported as mean with 95% confidence intervals and mean differences to compare follow-up measurements (after endovascular intervention) with measurements before the endovascular intervention.

Kaplan Meier estimates were used to estimate the kidney failure rate over time, accounting for death as a competing event. Analyses have been performed using SAS software (version 9.4 of the SAS System for Windows, Cary, NY, USA).

In this study, 26 consecutive patients (14 male and 12 female patients) with a mean age of 53.4 ± 12.7 years were included. One patient treated 3 months after kidney transplantation and presenting with two focal external iliac artery stenoses, proximal and distal to the origin of the transplant renal artery and related to a vascular clamp injury during transplantation, was excluded for further analysis. In the same study period, no institutional patient with identical radiological and clinical characteristics and symptoms was managed with conservative or open surgical techniques; in addition, an extensive search of transplant patients’ medical records, followed up in the authors’ institution before 2005, could be identified, making a comparative outcome analysis impossible. Fourteen out of the 26 included patients (54%) were transplanted between January 2000 and September 2022; in the same time period, 2286 kidney transplantations were performed in the authors’ institution. Patients’ baseline demographics and clinical data are summarized in Table 1.

The kidney transplantation of the included study patients was performed between June 1982 and April 2021. The mean and median time interval between kidney transplantation and endovascular aorto-iliac intervention was 109 ± 89 and 103 months, respectively (range 3–301 months). Details on the initial kidney transplantation and on patients’ clinical and radiological work-up are summarized in Tables 1, 2 respectively.

Angiographic findings are summarized in Table 3 (Figures 1–3). All aorto-iliac stenoses were of atherosclerotic origin.

Overall, technical success of the endovascular procedure was 100% and no procedure-related adverse events were reported.

In the transplant renal upper pole, interpolar region, and lower pole, a significant rise in intrarenal RI of 0.097, 0.095, and 0.130 was found, as summarized in Table 4 and Figure 4.

Serum creatinine levels before versus up to 1 year after the endovascular procedure are summarized in Figure 5, showing a significant and sustained decrease in values over time.

Analysis of the blood pressure before versus after the procedure revealed a significant drop in systolic pressure 1 week and 3 months after versus before the procedure; no significant drop in diastolic blood pressure over time was found, as demonstrated in Figure 6.

Last, no difference in number of prescribed antihypertensive drugs before versus after the procedure was demonstrated, as summarized in Figure 7A; adversely, the mean antihypertensive daily drug dose before the stent procedure was 2.82 (2.13; 3.52, 95% confidence interval) versus 2.18 (1.49; 2.87, 95% confidence interval) 6 months after the stent procedure, resulting in a mean change of daily drug dose of −0.64 (−1.05; −0.24, 95% confidence interval; P = 0.0030) (Figure 7B). In the follow-up period after the endovascular procedure, one patient presented with a progressive transplant kidney failure and dialysis was started up 3 years after a successful CIA stenting, resulting in a transplant kidney survival of 95.0% (78.6; 99.7) and 95.0% (78.6; 99.7) at 5 and 10 years of follow-up respectively (Figure 8). In addition, 17 patients (65%) died during follow-up, resulting in a mean survival of 4.1 (2.0; 10.7) years with an estimated overall survival of 36.0% (15.6; 57.0) and 28.8% (10.3; 50.7) at 5 and 10 years of follow-up respectively (Figure 9).