Association between a novel nutritional–inflammatory index and testosterone levels: a cross-sectional study

Abstract

Introduction

Testosterone deficiency (TD) has been closely associated with systemic inflammation and metabolic disturbances. The C-reactive protein–albumin–lymphocyte (CALLY) index is a composite biomarker integrating inflammatory, nutritional, and immune parameters. However, its association with TD remains unclear. This study aimed to investigate the relationship between the CALLY index and TD in adult men.

Methods

This multicenter cross-sectional study included 434 adult men recruited from urology and andrology outpatient clinics between December 2024 and February 2026. The CALLY index was calculated using serum albumin, lymphocyte count, and C-reactive protein levels. TD was operationally defined as a morning total testosterone (TT) concentration <300 ng/dL in the presence of relevant clinical symptoms. Calculated free testosterone (cFT) was estimated from TT, SHBG, and albumin using the Vermeulen equation, and cFT-based sensitivity analyses were performed. Multivariable logistic regression was performed after adjustment for age, body mass index, smoking status, alcohol use, regular exercise, hypertension, diabetes, total cholesterol, and triglycerides. Receiver operating characteristic (ROC) curve analysis was conducted to evaluate discriminative performance.

Results

Among the participants, 198 (45.6%) were classified as having TD. The prevalence of TD decreased progressively across increasing quartiles of the CALLY index (70.6% in Q1 vs. 23.2% in Q4; P < .001). After adjustment for demographic, lifestyle, metabolic factors, comorbidities, and SHBG, higher CALLY index values remained independently associated with lower odds of TD (adjusted OR = 0.779, 95% CI: 0.721–0.843, P < .001). Compared with Q1, the adjusted ORs for TD were 0.683 (95% CI: 0.531–0.873) for Q2, 0.497 (95% CI: 0.368–0.662) for Q3, and 0.360 (95% CI: 0.329–0.499) for Q4 (all P < .001). cFT increased significantly across CALLY quartiles (P < .001), and a cFT-based sensitivity analysis showed a consistent inverse association (adjusted OR = 0.851, 95% CI: 0.799–0.906, P < .001). The CALLY index demonstrated moderate discriminative ability for TD (AUC = 0.777), with an optimal cut-off value of 4.10 (sensitivity 64.6%, specificity 81.7%).

Discussion

These findings indicate that the CALLY index is associated with TD and provides moderate discrimination. It may serve as a readily accessible composite marker of the inflammatory and nutritional milieu associated with TD; however, external validation is required before any clinical application can be considered.

Introduction

Testosterone is a principal androgen predominantly synthesized and secreted by Leydig cells in the testes and plays a central role in the maintenance of male reproductive function and secondary sexual characteristics. Although small amounts of testosterone are also produced by the adrenal glands,¹ circulating testosterone levels in men are substantially higher than those in women and are essential for male fertility and sexual development.² Testosterone production peaks during puberty and early adulthood and subsequently declines progressively with aging. Longitudinal studies have shown that serum testosterone levels decrease at an estimated rate of 0.4%-2.6% per year after the age of 40, resulting in a 20%-50% reduction by the age of 80 compared with younger individuals.^3,4

Beyond its role in sexual development,⁵ testosterone contributes to the maintenance of musculoskeletal and metabolic health, and its deficiency has been associated with a range of adverse clinical outcomes.^6–8 Reduced testosterone levels have been associated with a broad spectrum of adverse health outcomes, such as sexual dysfunction, mood disorders, cognitive impairment, sarcopenia, osteoporosis, and metabolic abnormalities. Epidemiological evidence further indicates that low testosterone levels are linked to an increased risk of cardiovascular diseases, including hypertension, stroke, and coronary heart disease,^9–11 as well as metabolic disorders such as dyslipidemia, type 2 diabetes, and obesity.^12,13 With the ongoing global trend toward population aging, declining testosterone levels have become an increasingly prevalent clinical concern. Population-based studies estimate that approximately 7% of men over the age of 50 exhibit TD,¹³ with prevalence rising markedly with advancing age. In this context, the identification of reliable biomarkers associated with changes in testosterone levels is of considerable clinical importance, as it may facilitate early risk stratification and contribute to the prevention of testosterone-related physical and psychological morbidity.

Growing evidence suggests that reduced testosterone levels are closely linked to systemic inflammation, nutritional status, and immune dysregulation. Chronic low-grade inflammation can impair hypothalamic–pituitary–gonadal axis signaling and suppress Leydig cell steroidogenesis, thereby contributing to declines in circulating testosterone.¹⁴ Elevated inflammatory markers, particularly C-reactive protein (CRP), have consistently been associated with lower testosterone concentrations in aging and metabolically compromised men.^15,16 In addition, serum albumin, a marker of nutritional reserve and systemic inflammatory burden, has been positively correlated with testosterone levels in several population-based studies.¹⁷ Hypoalbuminemia often reflects a catabolic and pro-inflammatory milieu that may be unfavorable for androgen production.¹⁸ Lymphocyte count, a conventional indicator of systemic immune competence and immune homeostasis, further provides insight into the balance of inflammatory and stress-related responses.^19,20 Alterations in lymphocyte levels have been linked to immune aging and chronic physiological stress,²¹ both of which may adversely influence endocrine regulation. Collectively, these findings indicate that inflammation, nutritional status, and immune function are biologically interconnected with testosterone regulation.

The C-reactive protein–albumin–lymphocyte (CALLY) index is a composite biomarker calculated as serum albumin multiplied by lymphocyte count and divided by CRP. By integrating markers of nutritional status, immune competence, and systemic inflammation, it provides a multidimensional assessment of overall physiological condition.²² Originally developed in oncology research, the CALLY index has been associated with clinical outcomes in colorectal cancer and hepatocellular carcinoma and has shown potential utility in cardiovascular and other chronic inflammatory conditions.^22–25 Compared with isolated inflammatory markers, this composite index may better reflect systemic homeostasis. However, despite the established links between inflammation, nutritional status, and testosterone regulation, the relationship between the CALLY index and circulating testosterone levels remains unclear.

Therefore, the present multicenter cross-sectional study aimed to investigate the association between the CALLY index and testosterone deficiency (TD) in adult men and to evaluate its discriminative performance relative to its individual components. Given the established links among systemic inflammation, nutritional status, immune regulation, and androgen homeostasis, we further examined whether this association persisted when calculated free testosterone (cFT) was considered. Specifically, we asked: Is a higher CALLY index independently associated with a lower likelihood of TD, does this association remain evident when androgen status is assessed using cFT, and does the CALLY index provide greater discriminative ability than albumin, lymphocyte count, or CRP alone?

Methods

Study design and participants

This multicenter, cross-sectional study was conducted at the three hospitals. The study aimed to evaluate the association between a composite nutritional–inflammatory index and TD in adult men. Participants were consecutively recruited from the urology and andrology outpatient clinics of the three hospitals between December 2024 and February 2026. All eligible individuals provided written informed consent prior to enrollment. Each participant underwent a standardized clinical assessment, including collection of demographic information, medical history, physical examination, and laboratory measurements of serum testosterone and hematological parameters. The study protocol was approved by the institutional ethics committees of the participating centers.

Adult male patients attending the participating outpatient clinics who provided written informed consent were eligible for inclusion. Individuals were excluded if they were receiving or had received within the previous 3-6 months testosterone replacement therapy, anti-androgen therapy, or gonadotropin-releasing hormone analogs; were undergoing radiotherapy or using medications known to influence androgen levels, including neuroleptics, antidepressants, or systemic corticosteroids; had primary hypogonadism, pituitary disorders, thyroid disease, chronic liver disease, severe hepatic insufficiency, chronic renal failure, or other severe systemic conditions that could affect hormone metabolism or albumin synthesis; had active malignancy; or presented with acute infection or significant inflammatory conditions (eg, markedly elevated CRP, fever, or acute hospitalization). Participants with missing key laboratory data required for calculation of the composite index were also excluded. A total of 434 participants met the eligibility criteria and were included in the final analysis (Figure 1).

Figure 1.

Flow diagram illustrating the recruitment and selection process of study participants.

Clinical assessment and laboratory measurements

Demographic and clinical information were obtained through structured interviews and review of electronic medical records. Age and body mass index (BMI) were recorded during routine clinical examinations. Comorbid conditions, including hypertension and diabetes, were identified based on self-reported physician diagnosis and verified through medical records or corresponding medication use. Hypertension was defined as a documented history of hypertension or current use of antihypertensive agents, and diabetes was defined by a prior medical diagnosis or the use of glucose-lowering medications.

After an overnight fast of at least 12 hours, venous blood samples were collected between 7:00 and 10:00 AM to minimize diurnal variation in hormone levels. Samples were centrifuged and analyzed in the central laboratories of the participating hospitals following standardized procedures.

Complete blood counts, including lymphocyte count, were determined using an automated hematology analyzer (Sysmex XN-900, Sysmex Corporation, Kobe, Japan). Serum albumin, total cholesterol (TC), and triglycerides (TG) were measured using an automated biochemical analyzer with standard enzymatic methods. Serum total testosterone (TT), CRP, sex hormone–binding globulin (SHBG), and estradiol (E2) were quantified using chemiluminescent immunoassay. The CALLY index was calculated to reflect both inflammatory and nutritional status using the following formula: CALLY index = Albumin (g/L) ^* Lymphocyte count (10⁹/L)/[CRP (mg/L)^*10]. Calculated free testosterone was estimated from measured TT, SHBG, and albumin concentrations using the Vermeulen mass–action equation.²⁶

Clinical and sexual symptoms potentially related to TD were assessed during the routine clinical evaluation through structured patient interviews and review of the medical record. Relevant symptoms included decreased libido, erectile dysfunction, fatigue, and reduced muscle mass. A validated symptom–specific questionnaire was not systematically administered.

For the purposes of this cross-sectional analysis, participants were classified as having testosterone deficiency (TD) using an operational study definition based on the AUA-recommended biochemical threshold, namely a morning TT concentration < 300 ng/dL in combination with relevant clinical symptoms.²⁷ Total testosterone was measured once between 7:00 and 10:00 AM after an overnight fast. Because a second confirmatory morning TT measurement was not available, this study definition should not be considered equivalent to a definitive guideline-based clinical diagnosis of TD. Clinical assessments were performed by urologists or andrologists at the participating centers. For sensitivity analyses, low cFT was defined as <220 pmol/L (approximately 64 pg/mL), consistent with the threshold proposed in the European Male Ageing Study.²⁸

Statistical analysis

All statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). A two-sided P value <.05 was considered statistically significant. Continuous variables were assessed for normality using the Shapiro–Wilk test. Normally distributed variables are presented as mean ± standard deviation (SD), whereas non-normally distributed variables are expressed as median with interquartile range (IQR). Categorical variables are presented as numbers and percentages. Comparisons across quartiles of the CALLY index were performed using one-way analysis of variance or the Kruskal–Wallis test for continuous variables, as appropriate, and the chi-square test for categorical variables.

The CALLY index was analyzed both as a continuous variable and as quartiles (Q1-Q4). Multivariable logistic regression analysis was conducted to evaluate the association between the index and TD. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated after adjustment for several covariates. Covariates were selected a priori based on their established clinical relationships with testosterone levels and systemic inflammatory–metabolic status. The multivariable model included age, BMI, smoking status, alcohol use, regular exercise, hypertension, diabetes, TC, and TG to account for potential demographic, lifestyle, and cardiometabolic confounding. SHBG was additionally included because of its direct influence on circulating testosterone concentrations. To assess the discriminative performance of the index and its individual components (albumin, lymphocyte count, and CRP) for identifying TD, receiver operating characteristic (ROC) curve analysis was performed. Pairwise comparisons of the AUCs between the CALLY index and its individual components were performed using DeLong’s test for correlated ROC curves. A two-sided P value <.05 was considered statistically significant. A supplementary ROC analysis was also conducted for the CALLY index using low cFT as the outcome. The area under the curve (AUC) with 95% CIs was calculated.

Results

Baseline characteristics

A total of 434 participants were included in the final analysis, among whom 198 (45.6%) were classified as having TD. Baseline characteristics stratified by quartiles of the CALLY index are shown in Table 1.

Table 1.

Demographic and clinical characteristics of participants according to quartiles of the CALLY index.

Variables	Total	Q1 (≤ 3.394)	Q2 (3.394-5.751)	Q3(5.751-8.784)	Q4(>8.784)	P
	n = 434	n = 109	n = 109	n = 108	n = 108
Age (years)	61.44 ± 5.77	62.31 ± 4.93	62.18 ± 5.71	60.82 ± 6.48	60.43 ± 5.69	.030
BMI (kg/m2)	28.05 ± 2.75	27.96 ± 2.64	28.47 ± 2.52	27.81 ± 3.03	27.95 ± 2.78	.303
Personal history
Smoking, n (%)	124 (28.57)	35 (32.11)	37 (33.94)	31 (28.70)	21 (19.44)	.085
Alcohol user, n (%)	150 (34.56)	39 (35.78)	45 (41.28)	37 (34.26)	29 (26.85)	.165
Regular exercise, n (%)	163 (37.56)	36 (33.03)	42 (38.53)	42 (38.89)	43 (39.81)	.726
Hematologic parameters
TG (mmol/L)*	1.83 (1.39-2.27)	1.89 (1.44-2.35)	1.87(1.32-2.43)	1.79 (1.44-2.15)	1.84 (1.43-2.25)	.354
TC (mmol/L)	5.09 ± 0.62	5.30 ± 0.63	5.11 ± 0.62	5.04 ± 0.56	4.90 ± 0.61	<.001
SHBG (nmol/l)	35.11 ± 9.74	31.27 ± 9.22	35.61 ± 10.40	36.78 ± 8.53	36.80 ± 9.76	<.001
E2 (pg/ml)	27.97 ± 7.73	31.38 ± 6.65	27.33 ± 7.99	26.91 ± 7.91	26.25 ± 7.31	<.001
TT (ng/dL)*	393.50 (273.59-513.42)	263.50 (245.13-281.88)	393.10 (269.73-516.48)	465.45 (341.33-589.58)	464.60 (377.48-551.73)	<.001
cFT (pg/mL)*	70.87 (53.58-94.03)	57.51 (48.69-65.21)	68.72 (54.16-84.36)	82.46 (64.53-96.87)	87.31 (64.53-103.07)	<.001
Albumin (g/L)*	42.80 (38.20-47.40)	38.90 (34.93-42.88)	41.70 (37.45-45.95)	43.95 (39.40-48.50)	45.15 (40.61-49.69)	<.001
Lymphocyte (109/L)*	2.37 (1.88-2.86)	1.77 (1.31-2.23)	2.33 (1.97-2.70)	2.53 (2.15-2.92)	2.71 (2.36-3.07)	<.001
CRP (mg/L)*	1.75 (1.17-2.65)	3.35 (2.63-4.31)	2.08 (1.72-2.64)	1.53 (1.17-1.77)	1.00 (0.80-1.18)	<.001
CALLY index*	5.73 (3.04-8.43)	2.35 (1.75-2.95)	4.45 (3.92-4.98)	7.22 (6.53-7.92)	11.33 (8.91-13.75)	<.001
Hypertension, n (%)	219 (50.46)	66 (60.55)	57 (52.29)	50 (46.30)	46 (42.59)	.046
Diabetes, n (%)	128 (29.49)	32 (29.36)	31 (28.44)	31 (28.70)	34 (31.48)	.961
TD, n (%)	198 (45.62)	77 (70.64)	61 (55.96)	35 (32.41)	25 (23.15)	<.001

Values are percentages or mean ± SD unless noted otherwise.

^*Values are presented as median with interquartile range. Regular exercise: Twice a week, at least 30 minutes each time.

Abbreviations: CALLY, C-reactive protein-albumin-lymphocyte; Q, quartile; BMI, Body Mass Index; TG, Triglyceride; TC, Total Cholesterol; SHBG, Sex hormone-binding globulin; E2, estradiol; CRP, C-reactive protein; TT, total testosterone; cFT, calculate free testosterone; TD, testosterone deficiency. ^a independent sample t tests; ^b Kruskal–Wallis test was used; ^c chi-square test.

Age differed significantly across quartiles, with participants in the highest quartile being younger than those in the lowest quartile (P < .05). No significant differences were observed in BMI, smoking status, alcohol use, regular exercise, or diabetes prevalence across quartiles. In contrast, hypertension prevalence decreased progressively from Q1 to Q4 (P < .05). As expected, albumin and lymphocyte count increased, whereas CRP levels decreased across increasing quartiles of the index (all P < .001), confirming the internal consistency of the composite measure. Median TT levels increased significantly across quartiles (P < .001). Similarly, median cFT increased from 57.51 pg/mL in Q1 to 68.72 pg/mL in Q2, 82.46 pg/mL in Q3, and 87.31 pg/mL in Q4 (P < .001) (Table 1). The prevalence of TD declined markedly from 70.6% in Q1 to 23.2% in Q4 (P < .001). Figure 2 illustrates the distribution of TT across CALLY index quartiles. Total testosterone levels were significantly higher in Q2-Q4 than in Q1 (P < .001), whereas no significant difference was observed between Q3 and Q4.

Figure 2.

Distribution of total testosterone across quartiles of the CALLY index. Violin plots showing the distribution of serum total testosterone (TT) levels across quartiles (Q1-Q4) of the CALLY index. Solid lines indicate the median, and dotted lines indicate the interquartile range. Statistical comparisons between groups were performed as described in the Methods. P < .05, ^**P < .001; ns, not significant. TT, total testosterone; CALLY, C-reactive protein–albumin–lymphocyte index.

Association between the CALLY index and TD

Multivariable logistic regression analysis demonstrated a significant inverse association between the composite nutritional–inflammatory index and TD (Table 2). When analyzed as a continuous variable, higher CALLY index values remained independently associated with lower odds of TD after further adjustment for SHBG in addition to age, BMI, smoking status, alcohol use, regular exercise, hypertension, diabetes, TC, and TG (adjusted OR = 0.779, 95% CI: 0.721–0.843, P < .001).

Table 2.

Multivariate logistic regression analysis for TD (TD vs. without TD).

Variable	Adjusted model
	OR	95% CI	P
CALLY (continuous)	0.779	0.721-0.843	<.001
CALLY (Quartiles)
Q1	Reference	Reference	Reference
Q2	0.683	0.531-0.873	<.001
Q3	0.497	0.368-0.662	<.001
Q4	0.360	0.329-0.499	<.001

Statistical Analysis: Adjusted model: adjusted for age, BMI, smoking status, alcohol user, regular exercise, hypertension, diabetes, TC, TG, and SHBG.

Abbreviations: TD, testosterone deficiency; CALLY, C-reactive protein-albumin-lymphocyte; BMI, Body Mass Index; TC, Total Cholesterol; TG, Triglyceride; Q, quartiles.

In quartile-based analyses, compared with participants in the lowest quartile (Q1), those in Q2, Q3, and Q4 exhibited progressively lower odds of TD. In the fully adjusted model including SHBG, the adjusted ORs were 0.683 (95% CI: 0.531–0.873) for Q2, 0.497 (95% CI: 0.368–0.662) for Q3, and 0.360 (95% CI: 0.329–0.499) for Q4 (all P < .001).

Sensitivity analysis using calculated free testosterone

To further assess whether the observed association was attributable to differences in testosterone-binding proteins, we performed sensitivity analyses using cFT. Higher CALLY index values remained significantly associated with lower odds of low cFT after multivariable adjustment, including SHBG (adjusted OR = 0.851, 95% CI: 0.799–0.906, P < .001; Table S1). In the supplementary ROC analysis, the CALLY index showed an AUC of 0.715 (95% CI: 0.662–0.763, P < .001) for identifying low cFT. The optimal CALLY cut-off was 4.36, with a specificity of 78.4% and a sensitivity of 58.9% (Table S2).

Discriminative performance of the CALLY index

The discriminative ability of the CALLY index and its individual components for identifying TD was summarized in Table 3 and illustrated in (Figure 3). Among the individual markers, CRP demonstrated the highest area under the curve (AUC = 0.705, 95% CI: 0.655–0.754), followed by albumin (AUC = 0.690, 95% CI: 0.640–0.739) and lymphocyte count (AUC = 0.648, 95% CI: 0.594–0.702) (all P < .001). The CALLY index demonstrated the highest discriminative performance, with an AUC of 0.777 (95% CI: 0.732–0.822). Pairwise comparison using DeLong’s test showed that the AUC of the CALLY index was significantly higher than those of albumin (AUC = 0.690, P = .005), lymphocyte count (AUC = 0.648, P < .001), and CRP (AUC = 0.705, P < .001). These findings indicate that the composite CALLY index provided significantly better discrimination for TD than any of its individual components. The optimal cut-off value was 4.10, corresponding to a sensitivity of 64.6% and a specificity of 81.7%.

Table 3.

Diagnostic performance of albumin, lymphocyte, CRP, and CALLY for identifying TD.

Indicator	Optimal cut-off	Specificity	Sensitivity	AUC (95%CI)	P value	P vs. CALLY (DeLong test)
Albumin, g/L	41.45	0.758	0.513	0.690 (0.640-0.739)	<.001	.005
Lymphocyte, 109/L	2.39	0.677	0.631	0.648 (0.594-0.702)	<.001	<.001
CRP, mg/L	2.40	0.797	0.550	0.705 (0.655-0.754)	<.001	<.001
CALLY index	4.10	0.817	0.646	0.777 (0.732-0.822)	<.001	Reference

Abbreviations: CRP, C-reactive protein; TD, testosterone deficiency; CALLY, C-reactive protein-albumin-lymphocyte; AUC, Area Under the Curve; CI, Confidence Interval.

Figure 3.

Receiver operating characteristic (ROC) curves for identifying testosterone deficiency. Receiver operating characteristic (ROC) curves comparing the discriminative performance of the CALLY index (D) and its individual components (albumin (A), lymphocyte count (B), and CRP (C)) for identifying testosterone deficiency.

Discussion

In this multicenter cross-sectional study, we observed a significant inverse association between the CALLY index and TD in adult men. Participants with lower index values exhibited a higher prevalence of TD, and the association remained robust after adjustment for demographic characteristics, lifestyle factors, metabolic parameters, comorbidities, and SHBG. A clear dose–response pattern was observed across index quartiles, with progressively lower odds of TD in higher quartiles. Importantly, cFT also increased significantly across CALLY quartiles, and cFT-based sensitivity analyses yielded a consistent inverse association. In addition, the composite index demonstrated moderate discriminative performance for identifying TD. These findings suggest that an integrated assessment of nutritional and inflammatory status may be more informative than isolated biomarkers when examining the systemic correlates of low androgen status. The present findings should be interpreted in the context of an already well-established biological relationship between inflammation, metabolic health, nutritional status, and testosterone regulation. Accordingly, the contribution of this study is not the identification of a novel mechanistic pathway, but rather the evaluation of the CALLY index as an integrated marker combining inflammatory, nutritional, and immune information in men with TD. Our results extend the potential application of the CALLY index to the field of male endocrine and reproductive health.

Accumulating evidence has demonstrated a close association between systemic inflammation and TD. Elevated inflammatory markers, including CRP, interleukin-6, and tumor necrosis factor-α, have been correlated with lower circulating testosterone levels and a higher prevalence of TD in aging and metabolically compromised populations.^29–32 Among these markers, CRP has been most consistently associated with hypogonadal status. Beyond inflammatory mediators, nutritional status has also been linked to testosterone regulation. Serum albumin, commonly regarded as an indicator of nutritional reserve and chronic inflammatory burden,³³ has been positively associated with circulating testosterone levels in several population–based studies.^17,34 Lower albumin concentrations often accompany systemic catabolic and pro-inflammatory states, which have been observed more frequently in men with reduced testosterone levels. In recent years, composite inflammatory indices such as the neutrophil-to-lymphocyte ratio and the systemic immune–inflammation index have been explored in relation to androgen deficiency,^35,36 reflecting growing recognition of immune–inflammatory imbalance in male reproductive health. However, these indices primarily capture inflammatory or immune activation and do not account for nutritional status. In contrast, the composite nutritional–inflammatory index evaluated in the present study integrates markers of inflammation, immune competence, and nutritional reserve. By incorporating CRP, lymphocyte count, and albumin simultaneously, this index may provide a more comprehensive representation of systemic physiological status than isolated inflammatory parameters when examining the correlates of TD.

The observed association between the CALLY index and TD may reflect the integrated influence of systemic inflammation, nutritional status, and immune regulation on androgen homeostasis. Because this index simultaneously incorporates CRP, lymphocyte count, and albumin levels, it may serve as a surrogate marker of the broader inflammatory–metabolic milieu that accompanies hypogonadal states. Chronic low-grade inflammation represents a plausible biological pathway underlying this relationship. Pro-inflammatory cytokines, including interleukin-1 (IL-1), interleukin-6 (IL-6), interleukin-17 (IL-17), and tumor necrosis factor-α (TNF-α), have been reported to disrupt hypothalamic–pituitary–gonadal axis signaling.³⁷ These mediators may attenuate gonadotropin release and impair Leydig cell steroidogenesis, potentially contributing to reduced testosterone synthesis. Adipose tissue may further amplify this inflammatory influence. In men with increased visceral adiposity, inflamed adipose tissue acts as both a source of cytokine production and a site of enhanced aromatase activity, facilitating the conversion of testosterone to estradiol.^38,39 Such shifts in androgen–estrogen balance within a pro-inflammatory microenvironment may be associated with lower circulating testosterone concentrations. The relationship between inflammation and testosterone appears to be bidirectional. Experimental evidence suggests that androgens exert anti-inflammatory effects, whereas inflammatory activation may suppress androgen production.^40,41 Epidemiological studies have consistently demonstrated inverse associations between testosterone and CRP levels, supporting the notion that heightened inflammatory burden is linked to hypogonadal status.⁴² Given that the CALLY index captures inflammatory load, immune competence, and nutritional reserve simultaneously, its association with TD may reflect this complex systemic interplay rather than a single isolated pathway.

The CALLY index demonstrated moderate discriminative ability for identifying TD, with an AUC of 0.777, exceeding that of its individual components. The optimal cut-off value of 4.10 yielded a sensitivity of 81.7% and a specificity of 64.6%, indicating acceptable classification performance. Because the index is derived from routinely available laboratory parameters, it can be easily calculated in standard clinical settings without additional cost or specialized testing. As such, it may offer complementary information when evaluating men with suspected TD, particularly in the context of systemic inflammatory or metabolic disturbances. Nevertheless, given the cross-sectional design, the index should be regarded as an associated marker rather than a determinant of TD. Prospective studies are warranted to further clarify its potential role in clinical assessment and risk stratification.

Several limitations should be acknowledged. First, due to the cross-sectional design, causal relationships cannot be established. The relationship between testosterone and systemic inflammatory–metabolic status is likely bidirectional. Chronic low-grade inflammation may suppress hypothalamic–pituitary–gonadal axis activity and impair Leydig cell steroidogenesis, thereby contributing to reduced testosterone production. Conversely, low testosterone may itself promote visceral adiposity, insulin resistance, and adverse metabolic changes, all of which can amplify systemic inflammatory activity. In this context, a lower CALLY index may not only reflect an inflammatory and nutritional milieu associated with reduced testosterone but could also, at least in part, arise downstream of the metabolic and inflammatory consequences of hypogonadal status. Therefore, the biological direction underlying the observed association cannot be determined from the present cross-sectional data. Second, TD classification was based on a single morning TT measurement together with clinically reported symptoms. Current clinical guidelines recommend confirmation of low testosterone using at least two separate morning measurements because of substantial intra-individual and day-to-day variability in circulating testosterone. In addition, although relevant clinical and sexual symptoms were assessed during routine clinical evaluation, a validated symptom-specific questionnaire was not systematically administered. Consequently, some participants may have been misclassified with respect to TD status. This potential misclassification may have affected both the estimated prevalence of TD and the magnitude of its association with the CALLY index. Therefore, the present findings should be interpreted as associations based on an operational study definition rather than as evidence of a definitive guideline-confirmed diagnosis of TD. Future prospective studies incorporating repeated morning testosterone measurements and standardized validated symptom assessment are warranted. Third, albumin is an intrinsic component of the CALLY index and also contributes to circulating testosterone binding. This creates an inherent mathematical and biological coupling between the exposure and TT-related outcomes, such that part of the observed association may reflect differences in binding capacity rather than androgen production per se. SHBG also varied across CALLY quartiles and may influence TT concentrations. We therefore further adjusted for SHBG and performed sensitivity analyses using cFT; however, cFT itself is a calculated measure derived from TT, SHBG, and albumin and cannot fully remove this structural dependence. Accordingly, the CALLY index should not be interpreted as a direct marker of Leydig cell steroidogenic function. Future studies using repeated testosterone measurements and directly measured free testosterone, preferably by equilibrium dialysis, are warranted. Fourth, despite adjustment for multiple demographic and metabolic confounders, residual confounding cannot be excluded. Factors such as dietary patterns, detailed inflammatory conditions, and medication use beyond the recorded variables were not comprehensively evaluated. The possibility of overadjustment should also be considered. Several cardiometabolic covariates included in the multivariable model, particularly BMI, hypertension, diabetes, and lipid parameters, are closely interconnected with systemic inflammation and testosterone regulation. Although these variables were included as potential confounders because of their established associations with both metabolic health and androgen status, some may also represent intermediate factors within the inflammatory–metabolic pathway linking the CALLY index to testosterone. Because the cross-sectional design precludes determination of temporal or causal ordering, their precise roles as confounders or mediators cannot be established. Consequently, adjustment for these factors may have attenuated the overall association between the CALLY index and TD. The adjusted estimates should therefore be interpreted as conditional associations rather than as estimates of the total causal effect. Finally, participants were recruited from outpatient clinics in three hospitals within the same regional setting. Therefore, the findings may not be fully generalizable to broader populations or to individuals without healthcare-seeking behavior.

Conclusion

In this multicenter cross–sectional study, a higher CALLY index was independently associated with a lower likelihood of TD. The index showed moderate discriminative performance but should not be considered a standalone screening or diagnostic tool. Further prospective studies with external validation are needed to determine whether the CALLY index provides clinically meaningful complementary information in the assessment of men with low testosterone.

Author contributions

G.Z.: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Funding acquisition, Writing—original draft, Writing—review & editing. N.J.: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Funding acquisition, Writing—original draft, Writing—review & editing. Q.Z.: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Funding acquisition, Writing—original draft, Writing—review & editing. Y.M.: Data curation, Formal analysis, Investigation, Methodology, Writing—review &editing. C.C.: Data curation, Formal analysis, Investigation, Methodology, Writing—review &editing. H.L.: Data curation, Formal analysis, Investigation, Methodology, Writing—review & editing. N.S.: Data curation, Formal analysis, Investigation. X.F.: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing—original draft, Writing—review & editing. H.G.: Data curation, Formal analysis, Investigation.

Guoyang Zhang and Nuo Ji contributed equally to this work.

Funding

This work was supported by grants from the Changzhou Sci&Tech Program (CJ20259006, CJ20255003), the Changzhou Key Medical Discipline (KY20262960), the Youth Research Project Granted by Jiangyin Health Commission (Q202401), the Changzhou Municipal Health Commission Major Projects (ZD202432), the Top Talent of Changzhou “The 14th Five-Year Plan” High-Level Health Talents Training Project (2024CZBJ006).

Conflicts of interest

The authors declare that they have no competing interests.

Data availability

The data are currently not publicly available due to participant privacy, but, if necessary, they are available from the corresponding author upon reasonable request.

Consent for publication

Not applicable.

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. The study was approved by the Ethics Committee of Jiangyin People’s Hospital. All study participants submitted written informed consent.