Endothelin-1 and Accelerated Cardiac Allograft Vasculopathy After Heart Transplantation

Post-heart transplantation (HT) cardiac allograft vasculopathy (CAV) is characterized by diffuse intimal hyperplasia and negative remodeling of the coronary circulation. Current treatment has had negligible impact on worldwide incidence and outcomes, underscoring the need to identify new therapeutic targets.¹

Endothelin-1 (ET-1) is a powerful vasoconstrictive and pro-inflammatory peptide. Animal studies have demonstrated a causal link between ET-1 and CAV.² Initial human studies reported consistent findings, but were limited by less sensitive imaging tools.^3,4 Utilizing volumetric intravascular ultrasound (IVUS), we previously reported that among 90 non-consecutive HT recipients from 2002–2014, higher circulating ET-1 levels after HT were associated with significantly increased risk of accelerated CAV at 1 year and late death.⁵ While these retrospective findings are compelling, prospective validation in a contemporary population is lacking. Furthermore, data regarding the association of 1) change in ET-1 levels over the first year (ΔET-1) with CAV; and 2) ET-1 with plaque characteristics are limited.

This prospective cohort study included consecutive adult (≥18 years) HT recipients at our institution from October 28, 2018 to December 31, 2021. Exclusion criteria included multi-organ transplantation and comorbidities precluding baseline (within 4 months post-HT) cardiac catheterization. The study protocol was approved by the UCLA Institutional Review Board and all patients provided informed consent prior to enrollment. The de-identified data that support the findings of this study are available from the corresponding author upon reasonable request.

The post-HT pharmacologic regimen, cardiac catheterization, and ET-1 assay protocol have been previously described.⁵ Notable changes in the present study included 1) tacrolimus as the default calcineurin inhibitor (100% vs. 58% in our prior study); and 2) Makoto (Infraredx, a Nipro Company, MA, USA) as the preferred IVUS catheter, which quantifies lipid rich plaque using near-infrared spectroscopy (NIRS).

NIRS-IVUS measurements of the left anterior descending artery were performed in a blinded manner by the Stanford Cardiovascular Core Analysis Laboratory. CAV was defined as lumen volume loss at 1 year, which equaled plaque volume gain (intimal hyperplasia) plus vessel volume loss (negative remodeling). The primary endpoint was accelerated CAV, and the component secondary endpoints were accelerated intimal hyperplasia and negative remodeling. Like previous studies, we selected the 75^th percentile cutoff to define accelerated disease.⁵ Early lipid accumulation was defined as an increase in the maximum lipid core burden index over a 4mm segment (maxLCBI_4mm) at 1 year.

A multivariable logistic regression model was fit incorporating baseline ET-1, recipient and donor demographic factors, recipient clinical factors, and transplant factors with a univariable logistic regression p-value <0.20. A separate model was fit for ΔET-1 controlling for baseline ET-1 to adjust for potential confounding.

During the study time period, 170 adult patients received a HT, of whom 60 were excluded (31 multi-organ transplantation; 23 did not undergo baseline catheterization; 6 declined consent). Among the 110 consecutive isolated HT recipients included in the present analysis, the median age was 56.1 (41.4–63.5) years and 79 (72%) were men. The median baseline ET-1 level was 1.25 (1.04–1.57) pg/mL and ΔET-1 level was −0.05 (−0.39–0.38) pg/mL.

Eighty-four patients completed baseline and 1 year catheterization with interpretable IVUS data. Following adjustment, baseline ET-1 (odds ratio [OR]=3.39, 95% confidence interval [CI]: 1.01–11.48; p=0.049) and female sex (OR=3.98, 95% CI: 1.13–14.03); p=0.032) remained significantly associated with accelerated CAV. Baseline ET-1 also demonstrated a trend towards accelerated negative remodeling (OR=2.53, 95% CI: 0.79–8.06; p=0.12). ΔET-1 was not significantly associated with accelerated CAV (OR=0.19, 95% CI: 0.02–1.49; p=0.11).

Among the 66 patient subset with interpretable NIRS data, 15 (23%) developed early lipid accumulation at 1 year; these patients had significantly higher baseline ET-1 levels than their counterparts without early lipid accumulation [1.44 (1.32–1.50) vs. 1.21 (1.03–1.34) pg/mL, p=0.005]. Baseline ET-1 was significantly correlated with ΔmaxLCBI_4mm (r=0.31, p=0.01).

In summary, this prospective, contemporary cohort study of 110 consecutive adult isolated HT recipients validated and built upon our previous work by demonstrating that: 1) higher baseline circulating levels of ET-1—but not ΔET-1—were significantly associated with accelerated CAV at 1 year; and 2) this association may be driven by effects throughout the vessel wall, as suggested by the trend towards negative remodeling and significant correlation with early lipid accumulation. Taken together, these data further implicate ET-1 in the development of CAV (Central Illustration).

Central Illustration. Endothelin-1 and Accelerated Cardiac Allograft Vasculopathy.

Serial NIRS-IVUS of the LAD obtained at baseline (upper panels) and 1 year (lower panels) after heart transplantation. Volumetric IVUS analysis showed significant lumen loss (D: −3.56 mm³/mm) due to intimal hyperplasia (D: 1.69 mm³/mm) with negative remodeling (D: −1.86 mm³/mm). NIRS analysis also revealed significant early lipid accumulation in the proximal LAD segment (arrowhead).

IVUS, intravascular ultrasound; LAD, left anterior descending artery; NIRS, near-infrared spectroscopy

In contrast to prior studies, we did not detect an association between lower ΔET-1 and CAV; this difference likely reflects that unlike earlier analyses, we controlled for baseline ET-1 given its association with CAV and to mitigate any potential statistical biases due to regression to the mean and ceiling/floor effects with serial measurements.⁴ Additional ET-1 data points, in particular pre-HT and immediately post-HT, would shed more light on the longitudinal effect of ET-1 on CAV.

Other notable limitations included the modest sample size, though it was comparable to that of other single-center HT studies, and the absence of microvascular function assessment. Nonetheless, these data further link ET-1 to the pathogenesis of CAV and lay the groundwork to investigate whether ET-1 inhibition attenuates its development.