Abstract

1. Introduction

Cardiovascular diseases (CVD) pose a significant challenge in the aftermath of kidney transplantation, exerting a profound influence on both graft outcomes and patient survival. Despite notable advancements in immunosuppressive therapy, which have reduced acute rejection rates and improved short-term transplant outcomes, the trajectory of long-term allograft survival has remained relatively stagnant over the past two decades [1,2]. A critical factor contributing to this enduring challenge is the persistent burden of post-transplant cardiovascular disease (PTCVD), which becomes increasingly prevalent in kidney transplant recipients (KTRs) who experience extended lifespans. A substantial body of evidence consistently indicates a significantly higher incidence of cardiovascular disease among kidney transplant recipients compared to the general population [3,4]. Beyond the cardiovascular risks associated with chronic kidney disease and dialysis, kidney transplantation introduces unique risk factors, including immunosuppressive medication-induced post-transplant hyperglycemia, dyslipidemia, and hypertension. Additionally, the effects of suboptimal allograft function, such as chronic inflammation, oxidative stress, endothelial dysfunction, and chronic volume overload, as well as anemia, mineral bone disease, left ventricular hypertrophy, and homocysteinemia further heighten the risk of PTCVD [5,6,7]. Emerging evidence also suggests a role of adaptive cellular and humoral immunity in the development of CVD, with the expansion of pro-inflammatory and anti-apoptotic cytotoxic T-cells being associated with PTCVD [6]. Post-transplant DNA viremia such as infection with cytomegalovirus has also been shown to increase cardiovascular mortality by upregulating cytotoxic T-cells [6,7].

Cardiovascular disease accounts for one-third of all post-transplantation deaths, with the risk increasing significantly for individuals aged 50 years and above compared to the general population [8].

With the introduction of newer and more potent immunosuppressive agents, the interplay between traditional, chronic kidney disease (CKD)-related and transplant-specific risk factors in the genesis of cardiovascular complications warrants careful examination. Furthermore, despite evidence supporting cardiovascular risk modification through interventions such as blood pressure control, lipid modification, renin-angiotensin system blockage, and antiplatelet treatment, these strategies are often inadequately implemented in transplant recipients [4].

Against the backdrop of contemporary immunosuppressive strategies, this study aims to refine our understanding of PTCVD dynamics in KTRs without pre-existing CVD. Through the evaluation of a sizable cohort of KTRs spanning two decades, our aim is to unravel the nuanced interplay of factors contributing to the development of PTCVD and delineate its associations with graft outcomes and recipient mortality.

In the evolving landscape of kidney transplantation, where the optimization of long-term outcomes has become imperative, insights derived from this study may guide clinicians in tailoring interventions to mitigate the burden of PTCVD, thereby advancing the holistic care of kidney transplant recipients.

In this context, we aimed to assess the cumulative incidence of CVD after transplantation and determine the risk factors of PTCVD in recipients devoid of a pre-existing diagnosis of CVD. Second, we assessed the impact of PTCVD on long-term graft survival and recipient survival.

2. Materials and Methods

3. Results

A total of 1114 KTRs who underwent transplantation between January 2000 and December 2020 were evaluated; 365 recipients who did not meet the inclusion criteria, including 214 (19%) patients with prior CVD, were excluded from the final analysis, leaving 749 recipients for complete analysis. The flow chart of patient selection is shown in Figure 1, and the baseline characteristics of the participants are summarized in Table 1. The cohort was followed up for a total of 6017 person-years, and the median follow-up was 8.1 (IQR 4.5–13.0) years. The mean age at transplantation was 45 ± 15 years; 40% were female, and 81% were of White ethnicity. Pre-emptive transplantation was performed in 32% of recipients, while those without pre-emptive transplants had a median duration on dialysis of 26 months. The most common etiology of chronic kidney disease (CKD) was glomerulonephritis (GN). Approximately 30% of recipients received an allograft from a living donor. Most (90%) recipients received tacrolimus-based maintenance immunosuppression, and 46% had more than six months of corticosteroid therapy. The median baseline eGFR was 51 (IQR 41–64 mL/min/1.73m².

Open in a separate window

Figure 1

Flowchart of subject selection for the study.

After transplantation, 83 (11%) recipients experienced PTCVD, 69 (9%) recipients experienced graft loss, and 115 (15%) recipients died.

The overall cumulative incidence of PTCVD at 1, 5, 10 and 20 years was 1.1% (0.51%, 2.0%), 5.4% (3.8%, 7.3%), 14.3% (11.1%, 17.9%), and 22.5% (17.2%, 28.2%), respectively (Figure 2). The median time to CV events was 5.8 (IQR 2.5–6.2) years, and the median CV event-free survival was 7.6 (IQR 4.1–12) years. The incidence rate of CVD per 1000 person-years was 13.7 (i.e., 1.37% per year).

Open in a separate window

Figure 2

Cumulative incidence of PTCVD over the follow-up period.

The PTCVD patients were older at transplantation (50 vs. 44.5 years; p < 0.001) and more likely to have received corticosteroids for greater than 6 months (60.2% vs. 44.2%; p = 0.020). The PTCVD group had a lower baseline eGFR (median eGFR:46 vs. 52 mL/min), a lower average tacrolimus level (median: 4.6 vs. 5.7 ug/L; p = 0.003), and a lower mean hemoglobin level (mean:121.3 vs. 126.7 g/L; p = 0.023). The PTCVD group also exhibited a higher median parathyroid hormone (PTH) level (13.1 vs. 10.0 ng/L; p = 0.009) and a higher median urinary protein-to-creatinine ratio (uPCR) (40.5 vs. 22.7 mg/mmol; p = 0.024).

Conversely, factors such as sex, ethnicity, primary renal disease, number of HLA mismatches, history of acute rejection, DNA viremia, and smoking history were comparable between the PTCVD and non-PTCVD groups.

3.1. Predictors of PTCVD

Table 2 presents both the univariable and multivariable competing risk regression models assessing the risk factors associated with the development of PTCVD. In the univariable competing risk model, considering graft loss and recipient death as competing risks, recipient age, duration of dialysis vintage (per year), and the slope of estimated glomerular filtration rate (eGFR) decline exhibited significant associations with PTCVD. In the multivariable model, the independent predictors of PTCVD included an increase in recipient age (sub-hazard ratio [SHR]: 1.22; 95% CI: 1.01–1.46; p = 0.036) per year, duration of dialysis (SHR: 1.07; 95% CI: 1.00–1.14; p = 0.048) for each year spent on dialysis, and the slope of eGFR decline (SHR: 0.91; 95% CI: 0.86–0.98; p = 0.007) per mL/min/yr increase. Also, a higher baseline eGFR exhibited an increased protective effect (SHR, 0.98; 95% CI; 0.96–1.00; p = 0.032) per 10 mL/min/1.73m² increase.

Table 2

Univariable and multivariable competing risk regression models showing the predictors of time to PTCVD.

	Univariate Competing Risk Regression Model			Multivariate Competing Risk Regression Model
Characteristic	SHR	95% CI	p-Value	SHR	95% CI	p-Value
Recipient age at transplant (per decade)	1.26	1.10–1.45	<0.001	1.22	1.01–1.46	0.036
Female	1.06	0.68–1.63	0.8
Ethnicity
White	1.00	—	—
Asian	1.28	0.69–2.36	0.43
Black	1.70	0.44–6.60	0.44
Other	0.00	0.00–0.00	<0.001
Pre-transplant BMI (per 5 kg/m2)	1.07	1.00–1.15	0.065
Primary renal disease
ADPKD{ XE “ADPKD” }	1.00	—	—
GN{ XE “GN” }	1.48	0.61–3.56	0.39
DKD{ XE “DKD” }	0.38	0.08–1.85	0.23
HKD{ XE “HKD” }	2.15	0.72–6.39	0.17
Reflux/CPN{ XE “CPN” }	1.58	0.63–3.95	0.33
Unknown	0.95	0.35–2.59	0.92
Other	1.46	0.53–4.00	0.46
Pre-emptive transplant	0.65	0.39–1.09	0.11
Dialysis vintage (per year)	1.06	1.00–1.12	0.037	1.07	1.00–1.14	0.048
Pre-transplant diabetes	0.56	0.25–1.28	0.17
Donor type
DD{ XE “DD” }	1.00	—	—
LD{ XE “LD” }	0.78	0.47–1.30	0.34
Total mismatch	1.03	0.88–1.22	0.71
Total ischaemic time	1.01	0.99–1.04	0.34
CNI{ XE “CNI” }
Tacrolimus	1.00	—	—
Cyclosporine	1.57	0.90–2.75	0.11
Antimetabolite
None	1.00	—
MPA{ XE “MPA” }	0.83	0.47–1.46	0.52
Aza{ XE “Aza” }	0.72	0.33–1.56	0.41
Corticosteroid treatment
<2 wks	1.00	—	—
2 w–6 mo	0.85	0.11–6.62	0.88
>6 mo	1.26	0.80–1.97	0.32
Number of immunosuppressive agents	1.16	0.79–1.68	0.45
Recipient CMV-positive	1.53	0.93–2.52	0.09
Donor CMV-positive	1.08	0.66–1.76	0.76
CMV viremia	0.76	0.39–1.50	0.44
EBV viremia	1.01	0.56–1.81	0.98
Polyoma viremia	1.80	0.98–3.29	0.057
Smoking history
Never smoked	1.00	—
Current smoker	1.31	0.72–2.40	0.37
Ex-smoker	1.17	0.63–2.19	0.62
Median tacrolimus level	0.91	0.82–1.00	0.061	0.93	0.85–1.02	0.10
Post-transplant diabetes	1.55	0.93–2.58	0.092
Baseline eGFR (per 10 mL/min increase)	0.91	0.79–1.03	0.14	0.98	0.96–1.00	0.032
Slope of eGFR	0.96	0.92, 1.00	0.073	0.91	0.86, 0.98	0.007
Average uPCR	1.00	1.00–1.00	0.12
Average PTH (per 10 units increase)	1.04	0.98–1.10	0.18

Open in a separate window

Competing risk regression showing factors associated with time to PTCVD, with death as the competing risk; SHR, sub-hazard ratio; Variables with p-values less than 0.20 in univariable analysis were included in the multivariable model. Continuous variables missing for more than 10% of the population were excluded from multivariable analyses. Significant variables at p < 0.20 included age at transplantation, ethnicity, pre-transplant BMI, pre-emptive transplant, duration of RRT, type of CNI, recipient CMV status, polyoma viremia, median tacrolimus level, baseline eGFR, mean UPCR, and median parathyroid hormone level. A stepwise backward elimination method was employed in multivariable analysis, retaining variables with p-values less than or equal to 0.10. ADPKD, adult polycystic kidney disease; DKD, diabetic kidney disease; HKD, hypertensive kidney disease; CMV; cytomegalovirus; CI confidence interval; CPN, chronic pyelonephritis; DD, deceased donor; LD, living donor; CNI, calcineurin inhibitor; MPA, mycophenolic acid; Aza, azathioprine; ISA, immunosuppressive agents; NODAT, new-onset diabetes after transplant; PTH; parathyroid hormone; uPCR, urine protein/creatinine ratio.

We conducted a sensitivity analysis to test the robustness of the competing risk regression analysis. In this analysis, we conducted a Cox regression analysis as documented in the Section 2. The results are shown in Supplementary Tables S1 and S2. The outcome of this sensitivity analysis did not substantially differ from that when using our initial method.

3.2. Graft and Recipient Survival

The median survival time for the cohort was >20 years (IQR; 13–20 years). The median survival for those with PTCVD was 15 years (IQR; 9–20 years), whereas those without PTCVD had a median survival of >20 years (IQR; 14–20 years).

Over the observation period, 69 recipients (9.2%) lost their graft—13 (15.7%) from the PTCVD group and 56 (8.4%) from the No-PTCVD group. 132 (17.6%) recipients died with a functioning graft—31 (37.3%) in the PTCVD group vs. 101 (15.2%) in the No-PTCVD group.

The Kaplan–Meier survival curves did not show any difference in death-censored graft survival between the PTCVD and No-PTCVD groups, while all-cause graft survival and recipient survival was significantly lower in those with PTCVD (Figure 3).

Open in a separate window

Figure 3

Kaplan–Meier survival curves of death censored graft survival (A) and all-cause graft survival (B) and recipient survival (C).

The adjusted multivariable Cox proportional hazard model showed no significant difference in the DCGL between the two groups, but there was a notable 71% increase in the relative hazard of all-cause graft loss in the PTCVD group (adjusted hazard ratio [aHR], 1.71; 95% CI, 1.13–2.57; p < 0.001) (Figure 4B). Similarly, the relative hazard of recipient death was 97% higher in the PTCVD group (aHR, 1.97; 95% CI, 1.24–3.13; p = 0.004) (Figure 4C).

Open in a separate window

Figure 4

Effect of PTCVD on graft and recipient outcomes adjusted for confounding factors. (A) showing no difference in death-censored graft loss between those with PTCVD and those without. (B) shows a significantly higher all-cause graft loss associated with PTCVD. (C) shows a significantly higher recipient death amongst those who experience PT CVD. AR, acute rejection; DD, deceased donor; LD, living donor.

4. Discussion

KTRs navigate a distinctive array of health challenges, particularly in relation to the prevention and management of CVD. As highlighted by previous research, KTRs already contend with a considerable burden of CVD risk factors [15,16,17]. While KTRs generally present a lower CVD risk than individuals with end-stage kidney disease (ESKD) awaiting transplantation, their cardiovascular risk remains significantly elevated—it is up to 50 times greater than that of their age-matched counterparts in the general population [17,18].

Our study’s findings offer crucial insights into the prevalence, predictors, and outcomes of PTCVD.

The cumulative incidence of PTCVD, as revealed in our study, aligns with previous reports, acknowledging potential variations arising from differences in CVD definition and cohort characteristics [19,20,21]. Notably, our cohort’s exclusion of recipients with a pre-transplant CVD history distinguishes it from some earlier studies.

The study identified age at transplantation, duration on dialysis, baseline eGFR, and the rate of post-transplant eGFR decline as independent predictors of PTCVD. These findings collectively underscore the enduring influence of CKD-related factors in predisposing recipients to PTCVD. Intriguingly, traditional CVD risk factors, such as smoking, gender, and diabetes, did not exhibit significant associations with PTCVD in our study. This implies that PTCVD may be more strongly linked to factors stemming from impaired kidney function rather than traditional cardiovascular risk factors.

Furthermore, the absence of a significant impact of the immunosuppressive regimen type on PTCVD risk is somewhat reassuring. The reasons for this are not entirely clear, and the lack of recorded data on steroid doses in the cohort adds to the complexity. Despite this limitation, it is likely that recipients were not exposed to high doses of maintenance corticosteroids. In addition, the study’s findings indicate that median tacrolimus levels were not notably elevated at 5.5 ± 2.7 mmol/L.

Immunosuppressive agents exert direct effects on the cardiovascular system. These include left ventricular hypertrophy (LVH), myocardial fibrosis, arrhythmia, hypertension, dyslipidemia, and coronary atherosclerosis. CNIs have been associated with cardiac hypertrophy. CNIs promote the transcription of genes, which increase left ventricular mass through CNI-induced increases in fibrosis and collagen deposition [22]. Furthermore, CNI and corticosteroids are associated with an increased risk of hypertension, hyperlipidemia, and vascular remodeling. On the other hand, mTOR inhibitors have been suggested to reduce cardiac hypertrophy by attenuating cardiac remodeling and reducing cardiac fibroblast proliferation and collagen secretion. There is currently no evidence to suggest that antiproliferative agents have any direct effects on cardiac myocytes [22].

The considerable impact of PTCVD on recipient survival, marked by an 85% increase in the relative risk of death, is consistent with findings in the existing literature. However, the absence of a significant difference in death-censored graft loss between recipients with and without PTCVD in our study deviates from some prior reports [23]. However, this is corroborated by supporting evidence in other studies [20]. This inconsistency could be attributed to the observed tendency of recipients with PTCVD to succumb to cardiovascular events before experiencing graft failure.

The findings of this study emphasize the critical role of risk assessment, early detection, and preventive interventions in mitigating post-transplant cardiovascular disease (PTCVD) risk among kidney transplant recipients. Pre-transplant CVD screening was intended to detect at-risk patients and prevent PTCVD onset by facilitating risk stratification, early detection, and proactive intervention [24,25]. However, there is a lack of evidence supporting the effectiveness of pre-transplant screening in conferring survival benefits during the post-transplant period [24,25]. Current evidence suggests that some screening procedures may not offer clear benefits and may even pose potential risks to patients [24,25]. Non-invasive testing such as myocardial perfusion scans, stress echocardiograms, and computed tomography coronary angiograms are popular screening methods for the initial screening for CVD, but the accuracy of non-invasive screening remains a subject of debate. This is mainly due to the differences in the pathophysiology of coronary artery disease in patients with CKD compared to non-CKD patients [26]. Moreover, a positive non-invasive test for early post-transplant cardiovascular disease may not necessarily imply good discrimination for prognostically significant CVD [27]. While coronary angiography remains the best predictor of cardiovascular adverse events in transplant recipients, its use is associated with an increased risk of complications, including those related to iodinated contrast media, as well as complications due to the invasive nature of the procedure [26]. Therefore, there is a pressing need for further research to understand better the role and effectiveness of pre-transplant CVD screening in kidney transplant recipients.

Efforts should be directed toward developing safer and more accurate screening methods tailored to the unique characteristics and needs of CKD patients and transplant recipients. Biomarker development in this area, including the utility of established cardiac biomarkers such as cardiac troponin, NTProBNP, and placental growth factor (PLGF) before and after transplantation, holds promise. Furthermore, the identification of newer biomarker may contribute to the accuracy of imaging and clinical assessments, thereby enhancing the effectiveness of both pre-and post-transplant screening strategies. Enhancing risk assessment and management strategies for this vulnerable patient population is essential for improving long-term transplant outcomes and patient well-being.

Despite its contributions, our study has limitations inherent to its retrospective design, which may introduce potential biases such as selection bias, confounding, misclassification, and incomplete data. While we assumed no pre-transplant CVD in our evaluated cohort based on recorded data and robust pre-transplant screening, may not be entirely accurate due to undiagnosed cases of pre-transplant CVD.

A major limitation was that certain traditional cardiovascular risk factors, such as hypertension and dyslipidemia, were not included in the predictive model. The complex nature of blood pressure fluctuations pre- and post-transplantation, the non-protocolled measurement of lipid profiles only in certain patients in the post-transplant period, and the variable use of statin therapy posed challenges in accurately quantifying their effects on CVD risk. Due to the retrospective nature of this real-world study, including the above factors would have introduced a potential for selection bias, as patients with CVD risk factors or symptoms were more likely to have a lipid profile assay than those without. Similarly, patients at higher risk of CVD were more likely to have statins prescribed.

Lastly, it is worth noting that a considerable number of recipients underwent modifications to their immunosuppression regimen throughout the follow-up period. These alterations might have been prompted by complications such as infections, acute rejection, and medication side effects. Consequently, accurately delineating the impact of their immunosuppressive regimen on PTCVD over the follow-up duration posed a considerable challenge. Despite these limitations, our study aimed to uncover novel post-transplant CVD risk predictors beyond established factors. We acknowledge the need for future research to address the limitations identified in our study. Prospective studies capturing comprehensive data on established risk factors and treatment variables will be essential to elucidate further the complex relationship between these factors and post-transplant CVD risk.

5. Conclusions

In conclusion, our study provides valuable insights into the predictors and outcomes of PTCVD, emphasizing the influence of chronic kidney disease-related factors, such as duration on dialysis, baseline eGFR, and the rate of eGFR decline, in the development of PTCVD. The substantial impact of PTCVD on recipient survival underscores the necessity for a nuanced risk management approach that incorporates strategies to mitigate factors associated with kidney function deterioration. Overall, our findings further enrich the understanding of the complex dynamics surrounding cardiovascular health in kidney transplant recipients.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jcm13102734/s1, Table S1: Summary of variable selection process through backward elimination using AIC. Table S2: Summary of predictors of time to CVD retained in the multivariable Cox regression model after backward elimination based on variable AIC.

Funding Statement

This research received no external funding.

Author Contributions

Conceptualization: C.A.C., R.M. and P.A.K.; Methodology: C.A.C., A.R. and P.A.K.: Data Curation C.A.C.: Data Analysis: C.A.C. Writing—original draft preparation: C.A.C.: Writing—review and editing: C.A.C., A.R. and P.A.K.; Supervision: P.A.K., A.R. and R.M. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study protocol for this study is approved by and registered with the Northern Care Alliance Research and Innovation Department in the United Kingdom on 4 May 2021 (reference number: S21HIP03).

Informed Consent Statement

Patient consent was not required as this study was based on publicly available data. The need for informed consent was waived by the Greater Manchester South Research Ethics Committee in the United Kingdom.

Data Availability Statement

The data presented in this study are available on request from the corresponding author (Chukwuma Austin Chukwu).

Conflicts of Interest

P.A.K. reports the following conflicts of interest: Speaker and advisory board Fees: AstraZeneca, Napp, Bayer, Novartis, Vifor Pharma, Pharmacosmos, Boehringer Ingelheim, Clinical Consultancy: Bayer, Astellas, Otsuka, Unicyte. All other authors declare no competing conflict of interest.

Footnotes

References