Evaluation of Cardiometabolic Risk in Patients with Non-Functioning Adrenal Adenomas Using the Systematic Coronary Risk Evaluation 2 (SCORE2) and the Systematic Coronary Risk Evaluation 2-Older Persons (SCORE2-OP) Algorithms

Introduction

An adrenal incidentaloma (AI) is a randomly discovered mass of greater than 1 cm in size [1] which can be categorized as functional (hormone-secreting) and silent (benign or malignant) lesions [2]. Non-functional adenomas (NFA) are typically detected during imaging procedures and are classified as an AI following an endocrine exclusion procedure, with no secretory capacity [1,3,4]. The rate of NFA is estimated to be between 71% and 84% of all AI cases [1,5–7]. It has been reported that non-functioning adrenal incidentalomas were the most frequent tumor type identified in both clinical (75%) and surgical (69%) studies [8]. Although NFA are considered non-functional, they have been deemed clinically relevant [9]. Numerous instances of hypertension, dyslipidemia, insulin resistance (IR), impaired glucose tolerance, metabolic syndrome (MetS), obesity, and osteoporosis, have been observed in patients with NFA [3,9–14]. It has been remarked that such complications are related to an increased frequency in AI patients of hypercortisolemia, including its subclinical [14,15]. Moreover, it has been reported that some glucose metabolism disturbances (IR, and impaired glucose tolerance) are significantly higher in patients with NFA, regardless of their body mass index (BMI) [10,16,17]. The prevalence of disturbed glucose tolerance among patients with NFA has been reported to be remarkably higher than in the general population [16,17]. The World Health Organization (WHO), the National Cholesterol Education Program-Adult Treatment Panel III (NCEP-ATP III) and the American Association of Clinical Endocrinologists/American College of Endocrinology (AACE/ACE) classification have revealed that NFA are independent predictors of MetS [10]. Cardiovascular diseases are prevalent in the general population, and they are the leading cause of death in developed countries. Studies of the cardiovascular risk in patients with AI in comparison with the background population could be of a special interest [14,15].

The European Society of Cardiology (ESC) has developed Systematic Coronary Risk Evaluation (SCORE2) model designed to predict the 10-year risk of initial cardiovascular events in individuals aged 40–69 years without prior CVD or diabetes, recalibrated to European population target across 4 risk regions [18]. For populations aged over 70, ESC has provided a SCORE2 for Older Persons (SCORE2-OP) utilizing similar methodologies. SCORE2 incorporates risk estimates for fatal and non-fatal CVD events [19]. Depending on the established model of diagnosis, it is expected that between 5% and 28% of NFA will develop mild autonomous cortisol secretion (MACS) [20]. MACS is considered to be related to an increased cardiometabolic risk [5,20,21]. Patients with NFA have a higher risk of developing atherosclerosis compared to patients with healthy adrenal glands and low-intermediate CVD risk profiles [22]. It has been suggested that the evaluation of the coronary-artery-calcium (CAC) score may serve as a valuable predictor of atherosclerosis risk in elderly patients [22]. Furthermore, increased carotid intima-media thickness [14], higher D-dimer levels, and elevated uric acid levels have been observed in patients with NFA [12]. Patients with NFA and those with secreting adenomas have been presented with a higher prevalence of cardiometabolic risks compared to individuals with normal adrenal glands [23,24]. Given this discrepancy, the precise diagnosis of adenomas is crucial to reaching treatment goals and implementing measures to address the prevalence of cardiometabolic risks in these patients, using simple methods such as the SCORE2 risk prediction algorithms.

Therefore, this study aimed to evaluate cardiovascular risk in 106 patients with non-functioning adrenal adenomas using the SCORE2 and SCORE2-OP algorithms, and to evaluate metabolic deregulation, and inflammation in patients with NFA.

Material and Methods

PARTICIPANTS:

Retrospective analysis evaluated 408 NFA patients (283 females, 69.36%) with incidentally found adrenal adenoma/s referred to the Department of Endocrinology, Piekary Medical Centre, St. Luke’s Local Hospital in Piekary Slaskie, Poland.

Inclusion criteria were: 1) age: 40 or older 2) no signs or symptoms of cortisol hypersecretion; 3) normal early morning basal serum ACTH and cortisol levels, preservation of ACTH and cortisol circadian rhythm, and normal 24-hour urinary free cortisol (UFC) concentration.

Exclusion criteria were: 1) history of previous malignancy and/or inflammation, 2) medication known to affect any of the measured parameters, 3) presence of secreting adenoma or adrenals insufficiency, and 4) renal and liver failure define as: eGFR <60 ml/min/1.73 m²/bilirubin >2 mg/dL; INR >1.5 and albumin <3.5 g/dL.

After exclusion criteria, 106 patients were eligible for final analysis.

DIAGNOSIS OF NFA:

Morning cortisol level after dexamethasone suppression test below <1.8 ng/mL. When the post-dexamethasone serum cortisol concentration was over 1.8 ug/dL, the absence of cortisol excess in other tests confirmed:

Figure 1 shows CT a non-functional right adrenal adenoma.

SCORE2 AND SCORE2-OP ALGORITHM:

Systematic Coronary Risk Evaluation Model 2 (SCORE2) is an algorithm derived to predict an individual’s 10-year risk of fatal and non-fatal CVD in individuals aged 40–69 years in Europe. For healthy population aged >69 years, the SCORE-Older Persons (SCORE2-OP) algorithm estimates 5-year and 10-year fatal and non-fatal CVD across Europe [18,19]. For Poland, the algorithm is specifically tailored for a high-risk region. To estimate cardiovascular (CV) risk using SCORE2 and SCORE-2OP algorithms, the following predictors were selected: sex, age, current smoking status, history of diabetes mellitus, systolic blood pressure (SBP), and non-HDL cholesterol levels. Based on these parameters and information, the algorithms classify patients into 3 categories of CV event risk: 1) low-to-moderate risk, 2) high risk, and 3) very high risk. In our study, 16 patients were in the low-to-moderate risk group, 28 patients were in the high-risk group, and 62 patients were in the very high-risk group. The CVD risk categories based on SCORE2 and SCORE2-OP in study population are presented in Figure 2.

CV risk based on the SCORE2, and SCORE2-OP algorithms was estimated using data obtained from patient medical history assessments and laboratory test results. To calculate the risk, we collected information on sex, age, smoking status, history of diabetes mellitus, SBP, and non-HDL cholesterol levels, which were calculated using laboratory results for total cholesterol (TC) and low-density lipoprotein cholesterol (LDL). Patients aged <50 with an estimated CV event risk of <2.5% were classified as low-to-moderate risk, 2.5% to <7.5% as high risk, and ≥7.5% as very high risk. For patients aged 50–69 years, risks <5% were low-to-moderate risk, 5% to <10% were high risk, and ≥10% were very high risk [18]. The SCORE2-OP algorithm, used for patients ≥70 years old, classified risks <7.5% as low-to-moderate, <7.5% to <15% as high risk, and ≥15% as very high risk [19].

LABORATORY MEASUREMENTS:

The study consisted of 2 parts: a structured medical interview and a physical examination. The medical assessment covered age, sex, medical history (smoking status), and measurements of anthropometric and biochemical parameters. All biochemical tests were performed after fasting, in the morning. Fasting blood glucose [mmol/L], total cholesterol (TC) [mmol/L], HDL cholesterol [mmol/L], LDL cholesterol [mmol/L], non-HDL cholesterol [mmol/L], and triglycerides [mmol/L] were measured. Blood pressure [mmHg] was measured in a seated position, after waking up. Based in medical history and the latest test results, patients were categorized into those with diabetes and those without. Patients with diabetes were excluded according to the guidelines of the Polish Diabetes Society (PTD), which define diabetes as: fasting plasma glucose ≥7.0 mmol/l, random plasma glucose ≥11.1 mmol/l with symptoms of hyperglycemia, 2-h post-load glucose ≥11.1 mmol/l during an oral glucose tolerance test (OGTT), and HbA1c ≥6.5% [25].

ANTHROPOMETRIC INDICES:

Height and weight were measured by standard methods in the morning, and body mass index (BMI) was calculated by the following formula:

WHR was calculated as:

ATHEROGENIC INDICES:

Atherogenic indices was computed as:

Triglyceride-glucose body mass index (TyG BMI index) was defined as:

Lipoprotein Combine Index (LCI) was computed using the following formula:

Castelli’s risk index-I was computed as:

Castelli’s risk index-II was computed as:

Atherogenic coefficient (AC) was calculated as:

Atherogenic index of plasma (AIP) was computed as:

Triglyceride to HDL cholesterol was calculated as:

The metabolic score for insulin resistance (METS-IR) was computed as:

STATISTICAL ANALYSIS:

Data distribution was assessed using the Shapiro-Wilk test. Analyzed variables are shown as a median with interquartile range when the deviations from Gaussian distribution were significant, and as a mean and standard deviation (SD) for normally distributed data. Differences in each variable at the 3 groups of patients (low-to-moderate CVD risk, high CVD risk, very high CVD risk) were determined using the Kruskall-Wallis test with the post hoc multiple comparison test performed for non-normally distributed data. Differences in evaluated parameters were visualized with boxplots showing interquartile range of variables and significance levels of post hoc comparisons. Correlations between the examined variables were evaluated using Pearson’s or Spearman’s coefficients, for normally and non-normally distributed data, respectively. The chi-square test was used to determine associations between categorical variables. A receiver operating characteristic (ROC) analysis was performed to evaluate laboratory parameters as potential biomarkers for high and very high risks of CVD. Youden Index method was used to determine optimal cut-off value. Significance was defined at P values ≤0.05. The statistical analysis was conducted using STATISTICA 13 software (StatSoft, Tulsa, OK, USA) and R Studio (Integrated Development for R. RStudio, PBC, Boston, MA, USA).

Results

CHARACTERISTIC OF THE STUDY GROUP:

A total of 106 patients with NFA diagnosis were included in the study. The mean age of patients was significantly different between the analyzed groups of CVD risk based on SCORE2 and SCORE2-OP (P<0.0001). Among the groups of low and moderate CVD risk based on SCORE2 and SCORE2-OP the mean age was 47.94±5.37 years. In the high-risk group the mean age was 59.21±6.45 years, and in the very high-risk group the mean age was 66.22±7.48 years.

BASELINE CLINICAL PARAMETERS BETWEEN SCORE2 AND SCORE2-OP:

There was no statistically significant difference between the analyzed groups of CVD risk in most clinical parameters except for sodium (P=0.034), hemoglobin (0.016), urinary cortisol levels (0.0347) and systolic blood pressure (P=0.001). A comparison of baseline clinical parameters between the SCORE2 and SCORE2-OP groups is shown in Table 1 and Figures 2–6.

ANTHROPOMETRIC PARAMETERS AND INDICES BETWEEN SCORE2 AND SCORE2-OP GROUPS:

A comparison of baseline anthropometric parameters and indices between the SCORE2 and SCORE2-OP groups is shown in Table 2. Among anthropometric parameters and indices, only the WHR index was significantly different between the analyzed groups (P=0.016). Very high-risk patients had a higher WHR value (0.95; 0.89–0.98) compared to low- and moderate-risk patients (0.89; 0.84–0.91; P=0.0049). Also, high-risk patients had a higher WHR value (0.92; 0.87–0.97) compared to low and moderate risk patients (0.89; 0.84–0.91; P=0.0032) (Figure 7).

GLUCOSE METABOLISM PARAMETERS BETWEEN THE SCORE2 AND SCORE2-OP GROUPS:

Table 3 shows a comparison of glucose metabolism parameters between the SCORE2 and SCORE2-OP groups. Only fasting glucose [mmol/L] was significantly different between the analyzed groups (P=0.035). Patients with low and moderate CVD risk had significantly lower fasting glucose levels compared to patients with high CVD risk (5.35; 5.08–5.61 vs 5.78; 5.39–6.03; P=0.0093) as well as compared to patients with very high risk (5.35; 5.08–5.61 vs 5.83; 5.22–6.28; P=0.0230). There was no significant difference between patients with low and moderate CVD risk compared to those with very high CVD risk (P=0.85), as shown in Figure 8.

LIPID PARAMETERS BETWEEN THE SCORE2 AND SCORE2-OP GROUPS:

There was no significant difference in lipid parameters between the groups based on CVD risk according to SCORE2 and SCORE2-OP (Table 4).

ATHEROGENIC INDICES BETWEEN THE SCORE2 AND SCORE2-OP GROUPS:

There was no significant difference in atherogenic indices between the groups based on CVD risk according to SCORE2 and SCORE2-OP (Table 5).

The ROC analysis demonstrated poor predictive value of fasting glucose in identifying high and very high risk of CVD (Table 6, Figure 9).

Discussion

In this study, which included 106 patients, we aimed to assess CVD risk using SCORE2 and SCORE2-OP charts in individuals with NFA. We also analyzed metabolic disorders and various inflammatory and atherogenic indices (TyG index, TyG BMI, LCI, AIP, AC, Castelli risk-I, Castelli risk-II, TG/HDL ratio) across groups with varying CVD risk (low, high, very high). To our knowledge, this is the first study to analyze CVD risk groups according to the updated SCORE2 and SCORE2-OP charts, which is a notable strength of our work. Our analyses demonstrated statistical significance for variables such as WHR (P=0.016), very high-risk patients had a higher WHR value (0.95; 0.89–0.98) compared to low-to-moderate-risk patients (0.89; 0.84–0.91), P=0.0049, and high-risk patients had a higher WHR value (0.92; 0.87–0.97) compared to low and moderate risk patients (0.89; 0.84–0.91), (P=0.032), fasting glucose (P=0.035), urinary cortisol levels (P=0.0347), serum sodium concentration (P=0.034), hemoglobin levels (P=0.016), and systolic blood pressure (P=0.001).

We observed a significant correlation between WHR and CVD risk, with higher WHR associated with increased risk in the high-risk (P=0.032) and very high-risk group (P=0.0049) compared to the low-risk group (P=0.016). This finding aligns with the results of Cavalari et al [10]. Our study did not find a significant correlation between CVD risk groups based on BMI, WC, and hip circumference, consistent with findings in other studies [10,16,17]. The WHR, reflecting both central and peripheral fat distribution, may provide a more comprehensive assessment of CVD risk compared to individual measurements of WC or hip circumference.

Previous research has indicated significant associations between NFA and glucose metabolism parameters, such as IR and MetS [26,27]. Emral et al [28] demonstrated increased risk of IR and MetS in patients with NFA. However, our study did not find statistically significant correlations between the prevalence of IR or impaired glucose tolerance, glucose, and insulin levels at 60 minutes and 120 minutes after the glucose tolerance test, fasting insulin levels, HbA1C, and HOMA-IR index. Notably, our study excluded patients with diagnosed diabetes, as diabetes is not considered in the SCORE2 and SCORE2-OP risk assessment charts. Fasting glucose levels were significantly higher in patients with high (P=0.0093) and very high (P=0.023) CVD risk groups compared to those with low risk (P=0.035). Although a ROC analysis for fasting glucose suggested a moderate discriminative ability for assessing CVD risk (Youden index value of 0.38), it highlights the need for further exploration of glucose metabolism in NFA patients. Araujo-Castro et al [27,29] suggested that patients with MACS have a higher risk of diabetes compared to those with NFA. In line with Krzyżewska et al’s [16] findings, we observed that the prevalence of impaired fasting glucose was significantly higher in NFA patients. Additionally, they identified diabetes in nearly 30% of their NFA patients [16].

No significant differences were observed between CVD risk groups concerning dyslipidemia and atherogenic indices. Cavalari et al [10] reported higher non-HDL levels among NFA patients compared to a control population. Our results did not include non-HDL in the analysis, as it is part of the SCORE2 and SCORE2-OP charts. The presence of lipid-lowering treatments among some patients may have influenced our results. Mintziori et al [30] investigated whether lipid parameters could predict autonomous cortisol secretion, finding no significant differences in the prevalence of dyslipidemia between NFA patients and those with cortisol-secreting adenomas. Lipid disorders among NFA patients may be associated with subclinical hypercortisolism [30,31]. There are reports suggesting improvements in lipid profile, glycemic profile, and blood pressure control after adrenalectomy [27,32,33], which may indicate that subtle cortisol hypersecretion occurs in NFA patients, potentially impacting CVD risk [31].

In our study, significant alterations in urinary cortisol levels were observed among CVD risk groups (P=0.035), particularly between the low- and high-risk groups (P=0.0079). This finding may suggest that assessing the steroid profile in urine among NFA patients could help identify those with glucocorticoid pathway hyperactivity [11].

Hypertension remains a major risk factor for CVD. The prevalence of hypertension in the adult population is estimated at around 31–32.5% and increases with age. In our study, the average age differed between CVD risk groups: low CVD risk (47.94 years), high CVD risk (59.21 years), and very high CVD risk (66.22 years). SCORE2 and SCORE2-OP risk assessment account for increased blood pressure with age. Our study demonstrated significant differences in systolic blood pressure (SBP) between the low and very high CV risk groups (P=0.0015) and between the high and very high CV risk groups (P=0.011). While many studies have reported higher SPB values in NFA patients [11,16,26,34], some studies did not find significant correlations [29]. This discrepancy may be attributed to the use of antihypertensive medications among some patients in our study. Despite this, our findings highlight a significant correlation between higher SPB and CVD risk, underscoring the need for effective blood pressure management in NFA patients.

Our study also identified significant differences in serum sodium concentration (P=0.02), aligning with finding from other study [26]. Electrolyte balance, particularly sodium is an intriguing area for further research on NFA.

Lastly, significant differences in hemoglobin levels were observed between the low and very high CV risk groups (P=0.0044). However, the role of hemoglobin in CVD in NFA patients requires further investigation.

The results of our study should be viewed in light of its limitations, including lack of a control group, and its relatively small sample size, resulting from strict inclusion criteria, which is common in studies involving NFA. Despite these limitations, our study makes a valuable contribution to research on NFA and CVD risk.

Conclusions

The use of SCORE2 and SCORE2-OP charts in 106 patients with NFA reveals that elevated fasting glucose levels are significantly associated with higher cardiovascular risk in this group, particularly in those classified in high- or very high-risk categories. Additionally, SBP and WHR were identified as cost-effective, simple measures for predicting of CVD among NFA patients. However, other metabolic markers and indices did not show any significant differences across CVD risk groups. Recognizing that NFA patients may exhibit elevated fasting glucose levels, higher SBP, and increased WHR highlights the need for increased vigilance and regular assessment. Monitoring these parameters may prompt clinicians to consider earlier pharmacological or surgical interventions.