What's new?

Kidney transplant recipients (KTRs) are particularly prone to developing adverse clinical outcomes of COVID‑19 in both the short and long term. The mechanisms underlying long‑COVID syndrome remain unclear, yet dysfunction of the small arteries may be at play. We evaluated microvascular structural and functional changes in KTRs 8 weeks after SARS‑CoV‑2 infection by employing cutting‑edge technologies to evaluate eye and skin microcirculation in vivo. Our study identified an abnormal microvascular response to ischemic stimuli linked to SARS‑CoV‑2 infection, suggesting that this part of the vascular tree requires further observation and research.

Introduction

COVID‑19, caused by SARS‑CoV‑2, may affect different organs and present with a wide range of clinical manifestations, both acute and chronic (long‑COVID syndrome).¹ Patients with chronic kidney disease (CKD), especially those receiving renal replacement therapy, are characterized by increased COVID‑19 incidence rates and greater COVID‑19–related mortality when compared with the general population.^2-5 This may be partly explained by common immunosuppressive therapies that alter immune functions, accelerate aging, and exacerbate chronic low‑grade inflammation and endothelial dysfunction, which make kidney transplant recipients (KTRs) particularly vulnerable to communicable diseases and worsen their prognosis.^2-7

Several microvascular‑dependent mechanisms have been postulated as responsible for higher morbidity and mortality rates in patients after COVID‑19.^8,9 They may include vessel rarefaction, endothelial cell apoptosis and endothelial dysfunction, preponderance of vasoconstriction over vasodilation, hypercoagulation, and clot formation, which together promote solid organ hypoperfusion, ischemia, and failure.^2,10-14 This process may be further aggravated by microvascular rarefaction, as recently reported in the deep capillary plexus of the central part of the retina observed in KTRs after COVID‑19.⁴ Importantly, endothelial damage also results in impaired hemostasis and antithrombotic activity, adding to the risk of acute kidney injury in the course of recent SARS‑CoV‑2 infection.^2,15 In contrast to macrovascular complications in CKD, which have been extensively investigated for decades, studies on microcirculation are underrepresented, mostly because of their invasive nature.¹⁴ Therefore, the aim of our investigation was to evaluate the association between microcirculatory changes and SARS‑CoV‑2 infection in KTRs using cutting‑edge, noninvasive, in vivo technologies in 2 KTR subgroups (with and without a history of a recent, clinically overt SARS‑CoV‑2 infection). Moreover, we included all KTRs, irrespective of the presence of signs and symptoms of the so‑called long‑COVID syndrome, as defined by the World Health Organization.¹⁶ In this study, an Rt × 1e retinal camera (Imagine Eyes, Orsay, France) equipped with adaptive optics technology was employed to assess microvascular structure, and intensity of cutaneous nicotinamide adenine dinucleotide hydrogen (NADH) fluorescence was used a marker of microvascular function.^17-19

Patients and methods

Results

The baseline characteristics of the study groups are presented in Table 1. The mean (SD) age of the participants was 47.6 (12.7) years (25 men; 46% of the study group). The COV and non‑COV groups were comparable in terms of age, sex, BMI, blood pressure, eGFR (Chronic Kidney Disease Epidemiology Collaboration), comorbidities, and pharmacological treatment.

Microvascular structural characteristics were not significantly different in the COV and non‑COV groups. However, the functional parameters showed differences in both groups, as presented in Table 2. As compared with the non‑COV group, the COV group had significantly lower median (IQR) IR_max values (7 [5–8.7] vs 9.6 [8.8–11.5]; P = 0.04) and IR index values (3.1 [2.3–4.2] vs 7.2 [5.2–8.2]; P = 0.01). Additionally, in Table 3, we present a multivariable logistic regression model analysis in which COVID‑19 was the only variable independently associated with microvascular deterioration, denoted by a decline in IR. The remaining multivariable regression models showed no association between COVID‑19 and microvascular parameters.

A negative correlation was observed between lymphocyte count and IR (R = –0.65; P = 0.02), while CRP positively corelated with microstructural remodeling, as demonstrated for WT (R = 0.42; P = 0.02) and WLR (R = 0.48; P = 0.01; Table 4).

In the COV group, the relationship between eGFR and NOI was evident (R = 0.39; P = 0.03; Figure 1). No such relationship was observed in the non‑COV controls (P >0.05 for both). In the multivariable logistic regression analysis, eGFR was the sole factor associated with deterioration of microvascular function expressed by low NOI (below the median value; Table 5). In the remaining multivariable regression models, there was no association between eGFR and the microvascular parameters.

Discussion

Our study yielded 3 main findings. First, several weeks after COVID‑19, the patients showed altered microvascular IR, which reflects mitochondrial dysfunction. Second, the renal graft function corresponded to microvascular oscillations represented by NOI, implying a link between cutaneous and renal microcirculation in the KTRs after COVID‑19. Third, the severity of the inflammatory process in the KTRs who contracted COVID‑19 was associated with the worsening of structural (WT, WLR) and functional (IR index) microvascular markers, observed 8 weeks after the infection. Our study also showed the feasibility of a comprehensive, noninvasive, and in vivo assessment of the microcirculation phenotype in the KTRs.

We documented that KTRs had significantly lower IR values in the FMSF‑Post Occlusive Reactive Hyperemia (PORH) test several weeks after SARS‑CoV‑2 infection, suggesting deterioration of the mitochondrial function and impaired balance of the mitochondrial oxidation and reduction processes. It is important to emphasize that the FMSF‑PORH test allows for both in vivo and noninvasive measurements of NADH fluorescence, providing insights into the nearly real‑time function of the mitochondria. Several potential mechanisms could partly explain our observations. NAD, when shifting from the oxidized form (NAD+) to the reduced form (NADH), serves as an electron acceptor or donor in several reactions, maintaining intracellular redox equilibrium.²⁹ Under increased oxidative stress, for example, during inflammatory / infectious diseases, it is postulated that NAD additionally protects against oxidative injury. Mitochondrial DNA, when damaged by reactive oxygen species, is repaired by a NAD+-dependent enzyme, that is, poly ADP‑ribose polymerase 1 (PARP).^29,30 NAD supplies ADP‑ribose units to PARP and is then catabolized.²⁹ Therefore, excessive oxidative stress during SARS‑CoV‑2 infection and concomitant hyperactivity of PARP lead to the consumption of cellular NAD+ reserves, causing redox imbalance, disrupting mitochondrial energy production, and leading to cell death.³⁰ However, it was somewhat surprising that such abnormalities were documented approximately 8 weeks after overt COVID‑19. The lower values of IR in the KTRs after COVID‑19 observed in our study are consistent with the recent reports of Jedrzejewska et al,³¹ who showed decreased IR during recovery soon after mild COVID‑19. The relationship between SARS‑CoV‑2 infection and microvascular alterations was also observed in the retinal area visualized by optical coherence tomography‑angiography.³² Kazantzis et al³² reported long‑term microvascular impairment in patients who recovered from COVID‑19, describing reduced vessel density in the deep capillary plexus and enlarged foveal avascular zone, as compared with the control group. Consistent retinal changes suggesting macular ischemia due to COVID‑19 were also found in other studies.^33,34 Castellino et al³⁴ additionally observed that the degree of inflammation was linked to the decreased vessel density of the superficial capillary plexus 1 month after the acute phase of COVID‑19. Brantl et al,³⁵ apart from lower vessel density in the surrounding quadrants of the fovea in hospitalized patients, did not observe areas of nonperfusion, microaneurysms, or other microvascular anomalies, indicating that further observation is needed.

From a broader perspective, reduced IR values have only recently been documented in chronic diseases accompanied by oxidative stress, that is, coronary artery disease or hypercholesterolemia.^25,36 Interestingly, deterioration of microcirculatory function in the KTRs was detected 8 weeks after acute COVID‑19, which was not the case in the KTRs with no evident history of COVID‑19.

Given that alterations in small‑resistance arteries occur in different vascular districts,^11,12,37 microvascular abnormalities found in the skin may correspond to the ones seen in other organs. Importantly, as documented by Park and Shiffrin,³⁸ microcirculation abnormalities may precede clinical evidence for dysfunction of body organs and systems. We found that the KTRs 8 weeks after acute COVID‑19 showed lower endothelial and neurogenic oscillations (NOI). To date, the presence of impaired microvascular oscillations has been reported in type 1 diabetes, cardiovascular and neurodegenerative diseases, and in post‑COVID syndrome.^28,39-41 Taken together, our results support noninvasive cutaneous microcirculation phenotyping to identify patients at a higher risk for vascular abnormalities associated with COVID‑19. The suggestion of an extended monitoring in the post‑COVID period is even more justified considering our patients in whom an evident relationship between NOI and eGFR was documented.

Moreover, the correlation analyses indicated a relationship between inflammatory markers and structural (WT, WLR) and functional microvascular parameters (IR index) in the KTRs who had COVID‑19. The clinical value of this finding is not clear; however, studies on wall thickening in the smallest arteries showed detrimental effects of inflammatory processes on NO bioavailability. This, in turn, promotes vasoconstriction and may cause smooth muscle cell proliferation.^42,43 Increased WLR constitutes a well‑established hallmark of microvascular remodeling and a surrogate measure of end‑organ damage in several chronic diseases, including heart failure, hypertension, and CKD.^24,42-46 Reduced NO bioavailability may facilitate tissue hypoperfusion, ischemia, and organ damage.^10,42,44 The relationship between vascular remodeling and inflammatory markers reported in our study is consistent with previous reports.^42,43,47 For instance, De Ciuceis et al⁴⁸ demonstrated associations between specific lymphocyte subpopulations and microvascular structural alterations, that is, increased WLR of subcutaneous small‑resistance arteries.