Serum Trimethylamine -Oxide Levels as a Predictor of Peripheral Arterial Disease in Kidney Transplant Recipients

Introduction

Kidney transplantation (KT) remains the optimal treatment for end-stage renal disease (ESRD), providing a superior quality of life and survival than that of maintenance dialysis [1,2]. Despite the benefits of KT, patients undergoing KT still experience reduced life expectancy due to ESRD-related metabolic disturbances, with an expected remaining lifetime of 44.0 years, compared with 61.8 years for a 20-year-old individual without ESRD [3]. Cardiovascular disease (CVD) remains the leading cause of morbidity and mortality in KT recipients with a functioning graft, despite the reduction in CVD risk achieved with transplantation [4]. KT recipients experience a spectrum of CVDs, including coronary artery disease (CAD), heart failure, pulmonary hypertension, and valvular heart disease. Multiple risk factors, both uremic and transplant-related, contribute to the development of atherosclerosis, leading to peripheral artery disease (PAD) [5]. PAD, along with CAD, is associated with an increased risk of subsequent cardiovascular events, poorer transplant outcomes, and even death [6]. The 5-year and 10-year cumulative incidences of PAD were 4.2% and 5.9% in KT recipients, and KT recipients with pre-transplant PAD had a much higher likelihood of post-transplant PAD events compared to those without [7]. PAD is a significant complication in KT recipients, associated with reduced patient survival and graft survival [7]. The ankle-brachial index (ABI) is a reliable and cost-effective screening tool for PAD, with a normal cutoff value of 0.9 to 1.4 [8]. Recognizing the factors that predispose patients to CVD and PAD is crucial for optimizing KT outcomes and identifying potential therapeutic targets.

Emerging evidence highlights the role of the gut microbiota and its metabolites, including short-chain fatty acids, bile acids, and trimethylamine N-oxide (TMAO), in the pathogenesis of CVD [9,10]. TMAO is a metabolite of dietary phosphatidylcholine, choline, and L-carnitine produced by the intestinal microbiota, affecting cholesterol metabolism and platelet aggregation and promoting thrombosis and atherosclerosis [11]. Elevated TMAO levels are strongly associated with CVD, major adverse cardiac events, and cardiovascular mortality, including in patients with PAD [12–14]. Additionally, serum TMAO levels have been linked to vascular stiffness in patients with advanced non-dialysis chronic kidney disease (CKD), underscoring its role in vascular complications across the CKD spectrum [15]. TMAO levels also correlate with renal function in CKD and predict graft failure in KT recipients [16,17]. Given these findings, this study seeks to investigate the association between serum TMAO levels and PAD risk in KT recipients, aiming to elucidate its clinical significance and potential as a biomarker for improving KT management.

Material and Methods

PATIENTS:

This cross-sectional study was conducted on 109 KT recipients between December 1, 2021, and June 30, 2022, at a medical center in Hualien, Taiwan. Data were collected from all patients at a single point in time during the study period, regardless of the duration since their transplantation. All patients provided informed consent before participation in the study. From the medical records, data on patients’ baseline characteristics, regular medication use, significant medical history, and immunosuppressive drug use (tacrolimus, mycophenolate mofetil, steroids, rapamycin, and cyclosporine) were collected. In the present study, diabetes mellitus (DM) was defined as a fasting plasma glucose concentration of ≥126 mg/dL or the use of antidiabetic medications. Hypertension was defined as a systolic blood pressure of ≥140 mmHg, diastolic blood pressure of ≥90 mmHg, and the use of antihypertensive agents in the 2 weeks prior to study inclusion. Exclusion criteria included participation refusal, malignancy, acute infection, acute rejection status, myocardial infarction, stroke, amputation, heart failure, treatment with cilostazol or pentoxifylline during blood sampling, and elevated ABI (>1.3). Figure 1 depicts the flow chart of this study. Finally, 98 KT patients were included in this study. The study received approval from the Research Ethics Committee of Hualien Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation (IRB108-219-A), and complied with the general recommendations of the Declaration of Helsinki.

ANTHROPOMETRIC ANALYSIS AND BIOCHEMICAL INVESTIGATIONS:

Height and weight were measured 3 times with the patient in casual clothing and stocking feet. The height, which was checked with patients standing erect, was the distance from the sole of the foot to the top of the head (H910, Nagata Scale Co., Ltd, Tainan City, Taiwan). Body mass index (BMI) was calculated as body weight (kg) divided by height squared (m2). Following an overnight fast of approximately 8 h, 5 mL of blood sample was collected from each participant and centrifuged at 3000× g for 10 min. Then, the serum levels of fasting glucose, total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), blood urea nitrogen, creatinine, calcium, and phosphorus were measured using Siemens Advia 1800 autoanalyzer (Siemens Advia 1800, Siemens Healthcare GmbH, Erlangen, Germany). Furthermore, the serum levels of intact parathyroid hormone (iPTH) were assessed using a commercial enzyme immunoassay or enzyme-linked immunosorbent assay kit (catalog no. NM59041; IBL International, Hamburg, Germany) [18]. The intra- and inter-assay coefficients of variation were 3.6% and 2.8% for iPTH, respectively. To calculate the estimated glomerular filtration rate (eGFR), we used the 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.

MEASUREMENT OF AERUM TMAO LEVELS BY HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY-MASS SPECTROMETRY:

TMAO levels were quantified in serum using high-performance liquid chromatography coupled with mass spectrometry. A 100-μL serum sample was transferred to a 1.5-mL Eppendorf tube, mixed with 100 μL of sodium phosphate dibasic heptahydrate solution (50 mM), and then combined with 100 μL of deuterium-labeled internal standard solution (300 ng/mL d9-TMAO) in methanol. The mixture was gently vortexed. We used the Simplified Liquid Extraction method (Phenomenex, Novum), using 1.5 mL of ethyl acetate as the eluent. The samples were dried under nitrogen and reconstituted in 100 μL of methanol for analysis. Serum TMAO levels were measured using the Waters e2695 Separations Module high-pressure liquid chromatography (HPLC) system with a single quadrupole mass spectrometer (ACQUITY QDa, Waters Corporation, Milford, MA, USA). A Phenomenex Luna C18(2) HPLC column (particle size, 5 μm; dimension, 250×4.60 mm; pore size, 100 Å) was used, with a temperature of 40°C, flow rate of 0.8 mL/min, and injection volume of 30 μL. The mobile phase consisted of component A (water with 0.1% formic acid) and component B (methanol with 0.1% formic acid) in a gradient elution profile: initially 95% A and 5% B were combined for 1 min, then B was gradually increased to 70% over 12 min, held at 70% for 2 min, decreased to 50% over 1 min, and maintained for 2 min to re-equilibrate. This modified approach was used for HPLC mass spectrometry [19]. Quantification was conducted in positive-ion mode (TMAO) using electrospray ionization. The parameters were as follows: desolvation temperature, 600°C; capillary voltage, 800 V; and cone voltage, 15.0 V. The mass scan range was 50 to 450 m/z for positive-ion modes and 100 to 350 m/z for negative-ion modes, with TMAO detected at 76.0 m/z and d9-TMAO at 85.1 m/z. The retention times for TMAO and d9-TMAO were 2.54 min. Data were collected and analyzed using Empower 3.0 software (New York, NY, USA). Calibration curves for TMAO were constructed using isotope-labeled d9-TMAO standards, to account for endogenous TMAO levels.

ANKLE-BRACHIAL INDEX MEASUREMENTS:

The ABI was measured in all patients using an oscillometric method (VaSera VS-1000; Fukuda Denshi, Tokyo, Japan). Blood pressure was measured 3 times at each site – brachial artery (both arms) and ankle (dorsalis pedis and posterior tibial arteries) – while the patient was in a supine position. The highest systolic blood pressure from the dorsalis pedis or posterior tibial artery in each ankle was used for the measurement. The ABI was calculated as the highest ankle systolic blood pressure ratio to the highest brachial systolic blood pressure. Real-time electrocardiography was recorded for over 15 min to ensure accurate assessment. For each patient, the recorded ABI value was averaged over 3 consecutive measurements. The calibration process strictly adhered to the operator’s manual of the VaSera VS-1000 device. To optimize the accuracy of the study results, the cuffs and air lines used in the study were all validated. After each measurement of ABI, the quality control indicator based on the variability of ABI was displayed. An ABI value of less than 0.9 in either leg (left or right) was used as the diagnostic threshold for PAD, consistent with established guidelines [20].

STATISTICAL ANALYSIS:

To detect a correlation coefficient of approximately 0.3 between serum TMAO levels and ABI, with an alpha level of 0.05 and a power of 80%, it was determined that a minimum of 88 patients should be included in the study.

The normality of the data distribution was determined using the Kolmogorov-Smirnov test. Normally distributed continuous variables presented as mean±standard deviation were compared using the t test. Conversely, non-normally distributed variables (eg, triglycerides, fasting glucose, blood urea nitrogen, creatinine, iPTH, and TMAO), which are expressed as median (interquartile range), were compared using the Mann-Whitney U test. In analyzing the categorical variables, we used the chi-square test. Multivariate logistic regression analysis tested the association of TMAO with PAD in the KT population. The relationship between TMAO concentrations and clinical variables was assessed using Spearman’s rank correlation coefficient. To determine the optimal cutoff of serum TMAO level by the Youden index with sensitivity and specificity to predict PAD in KT recipients, we examined the area under the receiver operating characteristic (ROC) curve (MedCalc Software for Windows version 22.019, MedCalc Software Ltd., Ostend, Belgium). Data were analyzed using the SPSS Statistics 19.0 software, with a P value ≤0.05 considered significant.

Results

BASELINE CHARACTERISTICS:

The study included 98 KT recipients for a median duration of 90 months after transplant. Among them, 49.0% were women (n=48), 31.6% (n=31) had DM, and 43.9% (n=43) had a history of hypertension. Sixteen patients (16.3%) received kidneys from living donors. The cohort was divided into 2 groups according to ABI values: the low-ABI group (ABI<0.9, n=22) and the normal-ABI group (n=76). Table 1 shows the baseline characteristics of the 2 groups.

The low-ABI group had significantly higher rates of DM (P=0.035) and elevated serum TMAO levels (P<0.001) than the normal-ABI group. However, sex, age, BMI, KT duration, blood pressure, and immunosuppressive agent use showed no significant differences between the 2 groups.

SERUM TMAO LEVELS AND PAD:

The mean serum TMAO levels were significantly higher in the low-ABI group than in the normal-ABI group (P<0.001). Multivariate logistic regression analysis adjusted for potential confounders (eg, DM, hypertension, BMI, and KT duration) revealed that each 1 μg/mL increase in serum TMAO levels was associated with an 11.8% increased risk of PAD (adjusted OR [aOR]: 1.118, 95% CI: 1.050–1.191, P<0.001). However, further adjustment for systolic and diastolic blood pressure, total cholesterol, triglycerides, fasting glucose, eGFR, and iPTH (Model 2) showed that for each 1 μg/mL increase in TMAO levels, the risk for PAD increased by 15.4% (aOR: 1.154, 95% CI: 1.062–1.255, P=0.001) (Table 2).

CORRELATIONS BETWEEN TMAO AND ABI/EGFR:

Spearman correlation analysis showed that left and right ABI values were negatively correlated with serum log-transformed TMAO (log-TMAO) levels (r=−0.476 and −0.532, respectively; both P<0.001). Serum TMAO levels were also negatively correlated with eGFR (r=−0.281, P=0.005) (Table 3).

ROC CURVE FOR PREDICTING PAD:

The ROC curve for serum TMAO levels predicting PAD in KT recipients was 0.868 (95% CI: 0.784–0.928, P<0.001). The optimal TMAO cutoff point for predicting PAD, determined by the Youden index, was 18.65 μg/L, with sensitivity, 68.18% (95% CI, 0.451–0.853); specificity, 89.47% (95% CI, 0.798–0.950); positive predictive value, 65.21% (95% CI, 0.428–0.828); and negative predictive value, 90.67% (95% CI, 0.811–0.958) (Figure 2). The cutoff point of 18.65 μg/L for serum TMAO provides a clinically valuable tool for identifying PAD in KT recipients, with excellent specificity, moderate sensitivity, and strong predictive values.

Discussion

In this study, we found that KT recipients with low ABI exhibited significantly higher rates of DM and elevated TMAO levels than those with normal ABI. Increased serum TMAO levels were strongly associated with PAD, given that each increase of 1 μg/mL in TMAO raised the odds of PAD by 15.4%. The optimal TMAO cutoff point for predicting PAD was 18.65 μg/L, which demonstrated high specificity and sensitivity. Additionally, TMAO level was negatively correlated with ABI values and eGFR, indicating its potential as a cardiovascular risk marker in KT recipients.

PAD, characterized by the occlusion of peripheral arteries in the upper and lower extremities resulting from atherosclerotic plaques, can be assessed through limb vessel palpation, ABI measurement, and certain symptoms, such as intermittent claudication [8]. An ABI below 0.9 indicates PAD and is associated with a 2-fold to 4-fold increase in mortality [21]. Multiple risk factors, including traditional, uremic, and transplant-related factors, contribute to atherosclerosis, leading to PAD [5]. DM is a known risk factor for PAD; when coexisting with PAD, it further aggravates the risk of developing critical limb ischemia [22]. The prevalence of PAD is 2 to 7 times higher in people with DM [23,24]. Although tight glycemic control can improve microvascular and macrovascular outcomes, its benefit in advanced PAD or critical limb ischemia remains insufficiently established [25]. Patients with DM often experience poor post-ischemic recovery, leading to limb amputation, physical disability, or death [26]. The higher DM rate in our study’s low-ABI group aligns with previous findings. The European Society of Cardiology and the American Heart Association recommend ABI screening for PAD in patients with DM aged over 50 years, those with DM for more than 10 years, or those with additional risk factors, such as smoking, hypertension, and dyslipidemia [21,27]. Early detection and management of PAD in patients with DM are crucial, especially in those with KT, owing to the dual impact of DM and the hyperglycemic effects of specific immunosuppressants.

PAD increases the risk of major adverse cardiac events, including myocardial infarction, ischemic stroke, and mortality. However, PAD often goes undiagnosed or overlooked, compared with CAD; consequently, its atherosclerotic burden and associated mortality risks are underestimated [28]. Recent research has highlighted the role of dietary nutrient metabolites in atherosclerosis. Wang et al found that TMAO was significantly associated with CVD and mortality in 1876 patients [29]. Senthong et al identified TMAO as an important predictor of all-cause mortality in patients with PAD, with a 2.7-fold increased risk of death. High TMAO levels were more common in older individuals with DM and lower eGFR, consistent with other studies, regardless of the presence of CAD or other clinical subgroups [28]. Additionally, TMAO is implicated in the pathogenesis of insulin resistance and DM by inhibiting bile acid protein synthesis and transport, thereby regulating glucose metabolism. TMAO’s elevation of nitroso compounds also contributes to DM [30]. Elevated TMAO levels were reported as a determinant of all-cause mortality in patients with diabetic kidney disease [31,32]. As previously noted, serum TMAO levels have been linked to vascular stiffness in advanced CKD, underscoring their role in vascular complications [15]. This complements our findings, further emphasizing TMAO as a critical biomarker for assessing cardiovascular and renal risks.

TMAO may have a predictive value, and studies focusing on dietary modifications to reduce the TMAO burden are needed. Dietary choline, betaine, and L-carnitine are metabolized into trimethylamine (TMA) by the gut microbiota, subsequently oxidizing into TMAO by hepatic flavin-containing monooxygenase. Approximately 95% of TMAO is excreted via glomerular filtration, with minimal tubular secretion or reabsorption [33]. Flavin-containing monooxygenase enzyme activity, intestinal dysbiosis, and renal clearance significantly influence circulating TMAO levels [34]. Elevated TMAO induces vascular oxidative stress, inflammatory responses, and endothelial dysfunction [35]. In CKD, high TMAO levels are primarily due to reduced renal clearance and are implicated in fibrosis and nephron loss, ultimately contributing to increased cardiovascular risk [36]. TMAO levels show an inverse correlation with eGFR and increase with CKD severity, rising 2.5- to 30-fold in patients receiving hemodialysis, compared with healthy individuals. These levels temporarily decrease after hemodialysis and normalize following KT [16,33,37,38]. Higher TMAO levels are also associated with worse renal outcomes and long-term mortality, with Tang et al reporting a 1.9-fold higher 5-year mortality risk [36]. Our study identified a weak negative correlation between serum log-TMAO levels and eGFR, reinforcing previous findings that TMAO accumulation reflects declining renal function. However, elevated TMAO likely results from impaired clearance rather than directly causing renal dysfunction. The value of TMAO as a biomarker extends beyond traditional markers like serum creatinine, offering additional insights into metabolic and cardiovascular risks in KT recipients. Monitoring TMAO may enhance risk stratification and management in this population.

TMAO was first investigated in KT recipients by Flores et al to predict graft failure and its dietary determinants [17]. They reported a graft failure rate of 12.9% after a median follow-up of 5.3 years in 448 KT recipients from the Netherlands. Higher plasma TMAO levels were associated with an increased risk of graft failure, independent of traditional risk factors, such as DM, hypertension, BMI, lipid profile, smoking, donor characteristics, and baseline renal function [17]. Another study examined the interaction between immunosuppressive agents (tacrolimus and cyclosporin) and TMAO in 403 KT recipients. They found an association between TMAO levels and cyclosporine, though it was insignificant in multivariate analysis [39]. Our study, however, found no correlation between TMAO levels and immunosuppressive agents. Possible reasons for these differences could include variations in dosing regimens, patient subgroups, or measurement time points. Further research exploring these factors may help clarify the interaction between immunosuppressive drugs and TMAO metabolism and its effect on patient outcomes.

This study’s cross-sectional design limits the ability to infer causality. While TMAO may promote PAD through mechanisms like cholesterol deposition and vascular inflammation, reverse causality, where PAD affects TMAO levels, also warrants consideration. Longitudinal studies are needed to clarify these relationships. Additionally, the small sample size from a single medical center can limit the generalizability of the findings and the power to detect small effects. Third, reliance on self-reported medical history and the absence of dietary data can introduce bias. Future research should address these limitations by employing larger, longitudinal studies and incorporating validated medical history and dietary assessment tools to reduce bias and better understand the role of diet in TMAO levels and PAD risk.

Conclusions

This study identified elevated serum TMAO levels as an independent predictor of PAD in KT recipients, with significant negative correlations with ABI and eGFR. These findings highlight TMAO’s potential as a biomarker for cardiovascular risk and renal function decline, aiding in predicting and managing PAD in KT recipients. However, the study’s cross-sectional design and single-center nature limit the ability to infer causality and can affect generalizability. Further longitudinal and multicenter studies are needed to explore underlying mechanisms and evaluate interventions to reduce TMAO levels, potentially improving patient outcomes.