Pretransplant dialysis treatment and vascular calcification of the iliac artery and abdominal aorta in kidney transplant patients

Patients with chronic kidney disease often suffer from cardiovascular disease, and vascular calcification has been identified as one of the risk factors for cardiovascular disease. We aimed to examine the effect of dialysis treatment before kidney transplantation on graft survival, vascular calcification, and its progression after kidney transplantation. Among the 102 patients who underwent kidney transplant between 2008 and 2017, two patients were excluded for moved and lost to follow-up and primary nonfunction. The clinical characteristics and laboratory data were assessed according to pretransplant treatment modality. Rapid progression of vascular calcification was defined when patients showed an increase in the highest tertile of progression of each iliac artery calcification thickness (IACT) and aortic calcification index (ACI). Cox proportional hazard models did not show any significant association between pretransplant treatment modality and graft survival to the doubling of creatinine from nadir creatinine during the first 3 months after kidney transplantation. At baseline, the IACT was significantly higher in hemodialysis patients than in preemptive kidney transplant patients, whereas the ACI was comparable among the pretransplant treatment modality groups. IACT was independently associated with dialysis vintage. There was no significant association between rapid progression of vascular calcification (IACT and ACI) and dialysis modality. Dialysis modality was an independent factor related to IACT, whereas there was no legacy effect for the progression of vascular calcification after kidney transplantation.


Introduction
It is well known that cardiovascular complications predict the prognosis of patients and that vascular calcification is a risk factor for cardiovascular disease and mortality. Coronary artery calcification in particular correlates with cardiovascular events [1][2][3] and mortality [4,5]. A past study also showed that abdominal aortic calcification is a strong predictor of cardiovascular and cerebrovascular events and death [6]. However, these reports involve the general population, and there are few reports specifically on chronic kidney disease patients. It has not been clarified whether vascular calcification of the aorta and common iliac artery may impact clinical outcomes after kidney transplantation. Moreover, there are few studies comparing the effect of dialysis treatment on vascular calcification, and there is no report about the legacy effect of dialysis treatment performed before kidney transplantation.
The measurement of the calcification of the iliac artery may be convenient and useful in comparison with aortic calcification index (ACI) which is time-consuming, complicated, and not available in clinical practice. Although peripheral arterial disease has been associated with poor graft survival and high mortality after kidney transplantation [7], it remains controversial as to whether iliac artery calcification is associated with graft survival and mortality [8,9].
We conducted this study to clarify the relationship between vascular calcification in this area and pretransplant dialysis treatment or preemptive kidney transplantation, and the development of vascular calcification after kidney transplantation.

Study design
Among the patients who underwent kidney transplant in the decade between 2008 and 2017, two patients were excluded for moved and lost to follow-up (n = 1) and primary nonfunction (n = 1). Consequently, the study population comprised 100 patients, including 61 who had been on hemodialysis, 22 on peritoneal dialysis, and 17 who underwent preemptive kidney transplantation (crosssectional cohort). We analyzed a longitudinal sample of 62 patients who were scanned twice with an interval of at least half a year (longitudinal cohort). Data were collected through history taking, physical examination, and laboratory findings. Study variables were age, sex, co-morbid medical conditions, medications, body weight, height, systolic and diastolic blood pressure, laboratory findings, including creatinine, calcium, phosphate, HbA1c, parathyroid hormone, and low-density lipoprotein (LDL) cholesterol. Demographic information, underlying medical conditions, and medications were documented at the time of hospitalization. Following the previous report, cardiovascular disease (CVD) events were defined as stroke, myocardial infarction, hospitalization for unstable angina, coronary intervention (coronary artery bypass surgery or angioplasty), hospitalization for heart failure, and peripheral artery disease [10,11]. The study complies with the ethical guidelines of the 1975 Declaration of Helsinki, was approved by the appropriate institutional review committee (authorization No. 448). The board waived the need for patient consent, since retrospective data originated from standard care (in which one can opt-out for the use of their data for scientific research).

Assessment of vessel calcification
A previous study reported that maximal thickness was only associated with kidney transplant outcomes for three parameters of iliac artery calcification: (1) maximal thickness, (2) greatest percentage circumference, and (3) percentage length [8]. Regions of interest were placed around all lesions found within the bilateral common iliac arteries. The threshold for a calcific lesion was set at a computed tomographic density of 130 Hounsfield units within an area of ≥ 1 mm 2 (Fig. 1). ACI was obtained as previous reported [12,13]. Briefly, the images were obtained with a 5-mm slice thickness from the site where the renal artery arises to the bifurcation into the common iliac arteries of the aorta. Calcification was present if an area displayed a density of 130 Hounsfield units. The cross-section of the abdominal aorta on each slice was radially divided into 12 segments. Aortic calcification index was calculated as follows: ACI = ([total calcification score on 10 slices]/[12 × 10]) × 100%.

Statistical analysis
Patient clinical characteristics and laboratory data are shown in each pretransplant treatment modality (hemodialysis [HD], peritoneal dialysis [PD], and preemptive kidney transplantation [PEKT]). Data for categorical variables are given as number of patients (%) and data for continuous variables are given as mean ± standard deviation or median (interquartile range: IQR) values. Rapid progression was defined when patients showed increasing highest tertile of each ACI and increasing iliac artery calcification thickness (IACT).
Differences between groups were evaluated, when appropriate, using a student's t test and Turkey post-hoc test for multiple comparisons for continuous variables and the chi-squared test for categorical variables. Spearman correlation coefficients were calculated for univariate analysis of the associations among clinical and biochemical variables. Cox proportional hazards models evaluated the associations of pretransplantation treatment modality with graft survival, initially without adjustment, and subsequently adjusting for confounding factors, as follows: (1) the demographic adjusted model adjusted for age and gender; (2) the fully adjusted model adjusted for the demographic variables as well as complications of diabetes mellitus, nadir creatinine (lowest creatinine during the first 3 months after transplantation), dialysis vintage, body mass index (BMI), donorspecific antigen, ABO incompatibility, rejection after transplantation, and baseline IACT or ACI. Multiple stepwise linear regression analysis was performed to assess potential independent predictors of vascular calcification and its rapid progression. These were also evaluated with repeated-measures ANOVA. The adjusted odds ratios were provided with 95% confidence intervals and p values. All statistical tests were twosided, and a p < 0.05 was considered statistically significant. The statistical analyses were performed with the JMP version 13 (SAS institute Inc., Cary, NC, USA).

Study population
The final study population included 100 patients with a mean age of 47 ± 13 years, 69% were male, 18% had diabetes, and 80% received living kidney transplantation. At baseline, calcification of the common iliac artery was present in 62% of patients, and the IACT was 2.4 (IQR, 0-4.6) mm. Similarly, abdominal aortic calcification was present in 66% of patients, and median ACI was 10 (IQR, 0-30). The mean interval of CT from kidney transplantation was 40.3 ± 28.6 months, and the mean duration of follow-up was 68.2 ± 29.1 months. During the follow-up period, four (4%) patients died, eight (8%) had graft loss, and 18 (18%) showed a doubling of creatinine from baseline (lowest creatinine during the first 3 months after transplantation). Baseline characteristics of study subjects according to pretransplant treatment modality are shown in Table 1. Prevalence of male sex, diabetes, history of CVD, and BMI were comparable among the three groups. Dialysis vintage was longer in the HD group than the PD group. There were no significant differences among the three groups in terms of immunological factors, such as ABO incompatibility, donor-specific antigen, and HLA mismatch counts. In the cross-sectional cohort, diastolic blood pressure was higher in the PEKT group than the HD group. The prevalence of living donor transplantation was higher in the PD and PEKT groups than the HD group. In the longitudinal cohort, patients in the PEKT group were significantly younger than the HD group (not significant in cross-sectional cohort).

Association of dialysis modality with vascular calcification, graft survival, and mortality
At baseline, the prevalence of calcification of the common iliac artery was comparable among the three groups (70% in HD group, 55% in PD group and 41% in PEKT group, p = 0.07). Similarly, the prevalence of abdominal aortic calcification was comparable among the three groups (74% in HD group, 55% in PD group and 53% in PEKT group, p = 0.12). The baseline IACT was significantly higher in the HD group than the PEKT group (p = 0.049, Fig. 2A). The baseline ACI was comparable among the three groups (Fig. 2B). Death and RRT were comparable among the three groups (Logrank test, p = 0.34), and the time for doubling of serum creatinine from nadir creatinine during the first 3 months after kidney transplantation was comparable among the three groups (Log-rank test, p = 0.70, Fig. 3).
After adjustment for multiple variables, including ACI and IACT, Cox proportional hazard regression models showed that pretransplant treatment modality was not associated with doubling of creatinine ( Table 2). The results of stepwise multiple regression analyses in terms of the determinants of vascular calcification at kidney transplantation are shown in Table 3. In multivariate regression analyses, age, sex, diabetes, and dialysis vintage  . IACT is significantly higher in the HD group than in the PEKT group, whereas ACI is comparable among the three groups. There are no significant differences in terms of progression of IACT (C) and ACI (D) among the three groups

Dialysis modality as determinant of progression of vascular calcification
After kidney transplantation, the progression rate of IACT and ACI was 0.03 (IQR, 0-0.33) mm per year and 0.61 (IQR, 0-2.36) per year, respectively. Figure 2C, D shows that the progression rate of IACT and ACI was comparable among the three groups by pretransplant treatment modality. The results of stepwise multiple regression analyses of progression of vascular calcification after kidney transplantation are shown in Table 4. In multivariate regression analyses, diabetes and LDL cholesterol were associated with progression of IACT after kidney transplantation (diabetes: On the other hand, no significant association was found between progression of vascular calcification (IACT nor ACI) and dialysis modality, when the progression of calcification was evaluated as continuous variables. Moreover, dialysis modality did not correlate with the rapid progression of IACT (odds ratio = 1.78, confidence interval = 0.24-13.3, HD vs. PD) and ACI (odds ratio = 1.79, confidence interval = 0.31-10.3, HD vs. PD). Significant predictors of rapid progression of IACT were age (odds ratio = 1.13, confidence interval = 1.05-1.25), male sex (odds ratio = 12.4, confidence interval = 1.79-85.8), and presence of diabetes (odds ratio = 9.25, confidence interval = 1.11-76.9), after adjustment with other covariates (age, sex, diabetes, pretransplant treatment modality, dialysis vintage, calcimimetics prescription, systolic and diastolic blood pressure). Similarly, age was significantly associated with rapid progression of ACI after adjustment with other covariates (odds ratio = 1.08, confidence interval = 1.02-1.17).

Discussion/conclusion
The present study showed that long dialysis period was associated with vascular calcification of the common iliac artery, as well as age and the presence of diabetes at kidney transplantation. Whereas, dialysis modality did not contribute to the progression of vascular calcification, unlike sex and diabetes after kidney transplantation. Moreover, HR hazard ratio, CI confidence interval, HD hemodialysis, PD peritoneal dialysis, PEKT pre-emptive kidney transplantation a Adjusted for age and gender b Adjusted for demographic variables and body mass index, donor-specific antigen, ABO incompatible, rejection after transplantation, aortic calcification index, and iliac arteries calcification thickness Table 3 Stepwise multiple regression analyses for iliac artery calcification thickness (IACT) or aortic calcification index (ACI) (n = 100) A recent meta-analysis suggested that there were no differences in survival rates based on the method of dialysis in incident dialysis patients (overall pooled hazard ratio of death for PD versus HD was 1.06 [95% CI 0.99-1.14]) [14]. In terms of coronary artery calcification, PD was not significantly associated with higher progression of vascular calcification during the 3-year observation period, after adjustment for age, sex, and dialysis vintage [15]. The presence of coronary artery calcification in kidney transplant patients was associated with age and dialysis vintage; however, its progression was significantly associated with age, but not with dialysis vintage [16]. Eight of 67 patients in that study were receiving PD before kidney transplantation, but the association with dialysis modality was not described. As far as we know, this is the first report to show that the pretransplant dialysis modality did not influence the vascular calcification and its progression after kidney transplantation. Another report showed that before the initiation of dialysis therapy, vascular calcification was present and associated with mineral bone disorder and uremic toxins in chronic kidney disease patients [12]. Our results showed that PD patients had a shorter history of dialysis and tended to have less vascular calcification, as did patients who underwent transplantation without dialysis. Multivariate analysis that included dialysis vintage, showed no relationship between the dialysis modality and the progression of vascular calcification.
Recent studies have demonstrated that zero coronary artery calcium predicted a low cardiovascular event rate, which was close to 1% in a 10-year risk assessment in a general population free from CVD at baseline [3]. In non-dialyzed chronic kidney disease patients, severe abdominal aorta calcification was associated with mortality [17], and coronary artery calcification was associated with kidney survival [18]. Whereas, although calcification of the abdominal aorta and iliac artery is associated with surgical complexity and of delayed graft function in kidney transplant patients [19], the association between kidney survival and vascular calcification has not been well studied. Our study showed no apparent significant association between vascular calcification and kidney survival.
A major limitation of this study was the relatively short observational period (the mean duration of followup was 68.2 ± 29.1 months). In one multi-center cohort study involving 1614 kidney transplant patients, the rate of CVD events increased markedly after 7.5 years of follow-up, especially in patients with diabetic nephropathy and CVD history [20]. In order to evaluate patient and kidney survival, longer observation periods seem to be necessary, and this is why surrogate markers of progression of vascular calcification and doubling of creatinine were evaluated in this study.
In conclusion, this study suggests that dialysis vintage was an independent variable related to calcification of the common iliac artery, whereas dialysis modality was not a significant predictor of vascular calcification and its progression in this area. Pretransplant dialysis treatment was associated with vascular calcification at transplantation but did not adversely affect subsequent prognosis. Table 4 Stepwise multiple regression analyses for progression of iliac arteries calcification thickness (IACT) or aortic calcification index (ACI) (n = 62)

Analyses of progression of IACT
Analyses of progression of ACI