Renal Artery Perfusion Evaluation Before Transplantation via a 3-Dimensional Image Analysis System

Patients

A total of 50 consecutive cases of living donor kidney transplantation with multiple renal artery performed at Tokyo Medical University Hachioji Medical Center from May 2005 to January 2017 were included. Renal transplantations from deceased donors were not included. A retrospective review of the morphologic findings of the kidney graft and the medical records of both the recipients and donors was conducted.

Surgical procedure

Recipient's surgery

The following 4 different reconstruction methods were used: the conjoined method, which created a common orifice between 2 arteries (Fig. 1a); the side-to-end method, wherein the side artery was anastomosed to the main artery wall (Fig. 1b); the individual method, wherein the main and side arteries were anastomosed individually (Fig. 1c); and the ligation-alone method, wherein the side artery was ligated. A combination of any of these methods was classified as another group. The reconstruction method was selected by each surgeon based on the condition of the renal arteries. All main and side arteries, including the reconstructed arteries of the kidney graft, were side-to-end anastomosed to the external iliac artery of the recipient.

View Figure

• Download as Powerpoint
• Download Figure

3D image construction

All living kidney transplant donors underwent a dynamic enhanced computed tomography (CT) scan before surgery. Images were obtained with a 64-multidetector row CT scanner (LightSpeed VCT, GE Healthcare, Milwaukee, Wisconsin) with 1.25-mm thickness. A contrast medium was administered (dose, 1.4–1.6 mg/kg) through a peripheral venous catheter. Scanning in the early, late, and urinary phases was performed at approximately 30, 45, and 150 seconds, respectively.

All CT images were retrospectively analyzed with a 3D image analysis system (Synapse Vincent, Fuji Medical Systems Inc, Tokyo, Japan). The volume of the kidney, number of renal arteries, diameter of each artery, and perfusion area were assessed in detail.

In 3D image analysis, the following steps were performed:

1. Kidney extraction: The kidney was semiautomatically extracted using the region growing method with minor manual modifications (Fig. 2a).

2. Extraction of vessels: Renal arteries, veins, and ureters were extracted from the images (Fig. 2b) and visualized after the renal parenchyma was made transparent (Fig. 2c).

3. Evaluation of each artery construction and its perfusion area: The volumetric analysis was used to calculate the perfusion area (Fig. 2d).

View Figure

• Download as Powerpoint
• Download Figure

In a sample case, the side artery was ligated because its perfusion was less than 10%. After reperfusion of the graft during surgery, a partially ischemic area in the graft was detected, which was similar to that found in the 3D simulation before surgery (Fig. 2e).

Statistical analysis

Differences were determined using bivariate tests, including Pearson χ² and Fisher exact tests, for nominal variables. Student t-test was used to compare quantitative variables, whereas the Mann-Whitney U test was used to compare continuous or ordinal data. The Kruskal-Wallis test was performed to compare the different outcomes among the reconstruction methods. All statistical analyses were performed using the SPSS software version 24.0 (SPSS Inc, Chicago, Illinois). Statistical significance was set at P < 0.05.

Informed consent to use all medical records, including CT scanning images, was obtained from all patients. The Institutional Review Board of Tokyo Medical University Hachioji Medical Center approved this cross-sectional retrospective study (approval number: H-260).

Patient characteristics

Donors had a mean age of 52.5 years. The number of renal arteries identified was 2 in 44 cases (88.0%), 3 in 5 cases (10.0%), and 4 in 1 case (2.0%). The preoperative and postoperative information for each reconstruction method is shown in Table 1. The number of cases was 16 (31.6%) for the conjoined method, 9 (16.4%) for the side-to-end method, 13 (27.6%) for the individual method, 10 (20.2%) for the ligation alone, and 2 (4.1%) for the combined method. The conjoined + side-to-end case had 4 arteries before reconstruction (Fig. 3).

Table 1 Preoperative and postoperative information for each reconstructive method used for kidney grafting ^a

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Surgical information

Warm ischemic time was not statistically different among the reconstruction methods (P = 0.858). However, a statistically significant difference in cold ischemic time and total ischemic time between the methods (P = 0.005 and 0.013, respectively) was found. The mean cold ischemic time and total ischemic time of the individual anastomosis and ligation-alone groups were shorter than those of the other groups, whereas the combination group had longer cold ischemic time and total ischemic time.

Arterial diameter and perfusion volume

Table 2 shows the diameters and perfusion areas of all main and side renal arteries. The mean volume and proportion of the perfusion area of main arteries were 131 mL and 79.7%, respectively. The side arteries were classified according to the anastomosis method. A statistically significant difference in the diameter between the side arteries was noted (P = 0.003).

Table 2 Diameter and supply region of each renal artery ^a

View Fullscreen

A statistically significant difference in the perfusion area (both volume and proportion) between the anastomosis methods was observed (both P < 0.001). Perfusion was lower in the ligated arteries than in the other reconstructed arteries.

Postoperative course

The patency of the reconstructed arteries was monitored using ultrasound when necessary, and the patients had no reconstructed artery complications. Postoperative graft evaluation on the basis of 24-hour creatinine clearance was performed after 14 days, and the serum creatinine level was determined 1 month after the transplantation; no statistically significant difference (P = 0.396 and 0.586, respectively) was observed.