Skip to main content
Download PDF

Development and validation of a predictive model for metabolic syndrome in a large cohort of people living with HIV

Virology Journal volume 21, Article number: 321 (2024) Cite this article

Abstract

Background

The global prevalence of metabolic syndrome (MetS) in people living with HIV (PLWH) is on the rise in the post era of antiretroviral therapy (ART). Nevertheless, there are no validated predictive models available for assessing the risk of MetS in this specific population.

Methods

This study included PLWH who participated in annual follow-ups at Southern Medical University Nanfang Hospital from September 2022 to November 2023. Participants enrolled in this study were divided into the training set and validation set based on the follow-up duration. We employed both multivariate logistic regression and lasso regression to develop three distinct prediction models. Subsequently, the optimal model was determined through comprehensive analyses, including receiver operating characteristic (ROC) curve analysis, calibration curve, and decision curve analysis (DCA). Ultimately, we generated a nomogram for the optimal model and analyzed the correlation between the model score and the components of MetS.

Results

A total of 1017 participants were included in this study, with 814 in the training set and 203 in the validation set. The ultimate prediction model of MetS risk in PLWH incorporated five factors: age, CD8 + T cell counts, controlled attenuation parameter (CAP), gamma-glutamyl transferase (γ-GT) and lactate dehydrogenase (LDH). The area under the ROC curve (AUC) of the model in the training set and validation set was 0.849 and 0.834, respectively. Furthermore, we revealed a significant correlation between the model score and the MetS components. Additionally, the model score revealed significant group differences in MetS and related metabolic disorders.

Conclusions

This study established a potential model for predicting MetS in PLWH.

Introduction

The post antiretroviral therapy (ART) era has witnessed a remarkable increase in the life expectancy of people living with HIV (PLWH) [1, 2]. However, this population is now confronted with age-related and antiretroviral treatment-related complications such as metabolic disorders, renal impairment, osteoporosis, neurocognitive impairment, and cardiovascular disease (CVD) [3,4,5]. Among these, the adverse prognostic implications associated with metabolic syndrome (MetS) are gradually emerging as a significant threat to the well-being of PLWH [6, 7].

MetS constitutes a complex interplay of components, encompassing abdominal obesity, impaired fasting glucose (IFG), elevated blood pressure, hypertriglyceridemia (HTG), and low levels of high-density lipoprotein cholesterol (HDL-C) [8]. Its primary clinical consequence is the occurrence of cardiovascular events, and it is intrinsically linked to an elevated risk of atherosclerosis and mortality [9, 10]. Immune activation and inflammation persist even with successful viral suppression from ART among PLWH, potentially leading to metabolic disturbances that contribute to MetS [11]. Additionally, the metabolic side effects of ART may exacerbate issues such as insulin resistance, dyslipidemia, and central obesity [12]. Therefore, the combination of sustained low-level immune activation, the metabolic toxicity associated with ART, and intricate interactions involving traditional risk factors may collectively elevate the risk of MetS among PLWH [13, 14].

The identification and management of MetS have become a critical issue for PLWH receiving combination antiretroviral therapy [4]. Nevertheless, despite these challenges, there is still a lack of effective and viable models for predicting MetS in this population. Therefore, the primary objective of this study was to establish an effective MetS prediction model in PLWH through regression analysis of a large cohort. We aim for this model to provide a theoretical basis for clinical decision-making and the implementation of health interventions, ultimately contributing to the reduction of poor prognosis in PLWH.

Methods

Study design

This cross-sectional study was conducted from September 2022 to November 2023 at Nanfang Hospital, affiliated with Southern Medical University. We included PLWH who participated in annual follow-ups during this period. Exclusion criteria encompassed individuals with (1) any history of cancer, (2) systemic infections, (3) pregnancy status, and (4) an absence of essential clinical data. Subsequently, the participant cohort was divided into the training cohort and the validation cohort based on follow-up dates, with the first 80% assigned to the training cohort for model construction and the remaining 20% to the validation cohort for model evaluation.

Data collection

We systematically gathered demographic and clinical information from the study participants. These data encompassed age, gender, body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), the duration of ART, ART regimen, CD4 + T cell counts, CD8 + T cell counts, CD4/CD8 ratio, acquired immunodeficiency syndrome (AIDS) stage, white blood cell count (WBC), lymphocyte count (LYM), neutrophil count (NEU), monocyte count (MONO), eosinophil count (EOS), red blood cell count (RBC), hemoglobin (HGB), platelet count (PLT), alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin (ALB), globulin (G), total bilirubin (TBIL), direct bilirubin (DBIL), alkaline phosphatase (ALP), gamma-glutamyl transferase (γ-GT), uric acid (UA), creatinine (CR), blood urea nitrogen (BUN), estimated glomerular filtration rate (eGFR), triglycerides (TG), total cholesterol (CHOL), HDL-C, low-density lipoprotein cholesterol (LDL-C), very low-density lipoprotein cholesterol (VLDL-C), fasting blood glucose (FPG), glycosylated hemoglobin A1c (HbA1c), lactate dehydrogenase (LDH), hydroxybutyrate dehydrogenase (HBDH), creatine kinase (CK), creatine kinase myocardial band (CKMB), C-reactive protein (CRP), erythrocyte sedimentation rate (ESR). Concurrently, we conducted hepatic assessments on the participants utilizing transient elastography. The controlled attenuation parameter (CAP) was to evaluate liver steatosis and liver stiffness measurements (LSM) to assess liver fibrosis. The above data were sourced from medical records or databases.

Diagnosis of metabolic disorders

MetS was defined according to the International Diabetes Federation (IDF) criteria [8], which require the presence of any three of the five following components: (1) TG levels ≥ 1.70 mmol/L or being on treatment for dyslipidemia. (2) Low HDL-C levels were defined as < 1.0 mmol/L for men and < 1.3 mmol/L for women, or if the participant was receiving treatment for dyslipidemia. (3) Elevated blood pressure was defined as ≥ 130 mmHg and/or DBP ≥ 85 mmHg or if the participant was using antihypertensive medication. (4) FBG levels ≥ 5.6 mmol/L, a diagnosis of type 2 diabetes, or treatment with oral hypoglycemic agents. (5) Obesity was defined as waist circumference ≥ 90 cm for men or ≥ 80 cm for women, or BMI > 27.5 kg/m², based on the World Health Organization guidelines for the Asian population [15, 16].

In addition, metabolic disorders associated with MetS were defined as follows: HTG was defined as TG ≥ 1.7mmol/L [17]. Low level of high-density lipoprotein, also known as hypoalphalipoproteinemia (HA), was defined as HDL-C ≤ 1.03 mmol/L in men and ≤ 1.29 mmol/L in women [18]. Hypertension was defined as ≥ 140 mmHg, and/or DBP ≥ 90 mmHg or if the participant was using antihypertensive medication [19]. Elevated fasting glucose includes diagnostic criteria for IFG and diabetes, defined as FPG ≥ 5.6mmol/L [20, 21].

Construction and evaluation of the prediction model

In our study, we performed univariate and multivariate binary logistic regression analyses in the training set to identify factors associated with MetS in PLWH. Variables directly related to the definition of MetS were excluded, while all other variables were included in the univariate analysis. Variables demonstrating a p-value < 0.05 in the univariate analysis were subsequently incorporated into the multivariate model. The independent predictors identified through this process served as the foundation for Model 1. Subsequently, we applied Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis, using the same set of variables as in the logistic regression model. While cross-validation initially identified an optimal λ that selected a larger number of variables, we opted for a λ value that yielded the same number of variables as Model 1 to maintain model simplicity and interpretability. This selection allowed us to construct Model 2 with a comparable level of complexity. Drawing from the insights of these two models, we developed a simplified prediction model, denoted as Model 3.

To evaluate and compare the predictive performance of these models, we generated receiver operating characteristic (ROC) curves and calculated the area under the ROC curve (AUC). Furthermore, we employed calibration curves to assess the concordance between actual risk and predicted risk, and decision curve analysis (DCA) curves to ascertain the clinical net benefit. With the refinement of our optimal model, we created a nomogram, which serves as a visual tool to enhance the representation of its clinical utility.

Statistical analysis

In our analysis, variables with a normal distribution were presented as mean ± standard deviation, with comparisons between two groups conducted using Student’s t-test. For non-normally distributed variables, the median (interquartile range) was used, and comparisons were made with the Mann-Whitney U test. Categorical data were expressed as percentages and subjected to comparison through the Chi-square test or Fisher’s exact test. The relationship between model scores and clinical variables was explored using Pearson correlation analysis. It is noteworthy that all data analysis and graphical representations were performed using R version 4.2.1.

Results

Characteristics of participants enrolled

According to the flowchart in Fig. 1, a total of 1017 participants were ultimately included in the cohort, with 814 in the training set and 203 in the validation set. The demographic and clinical characteristics of our cohort were presented in Table 1. Notably, no statistically significant differences in clinical data were observed between the validation and training sets. Subsequently, when we grouped participants according to their MetS diagnosis as listed in Supplementary Table 1, several differences were observed between the Non-MetS and MetS groups, including, but not limited to, age (P < 0.001), BMI (P < 0.001), and CD8 + T cell counts (P < 0.001).

Fig. 1
figure 1

Study flow diagram. Abbreviations ROC, receiver operating characteristic; DCA, decision curve analysis; CAP, controlled attenuation parameter; γ-GT, gamma-glutamyl transferase; LDH, lactate dehydrogenase

Full size image
Table 1 Characteristics of participants or the training and validation cohorts
Full size table

Regression analysis and model construction

We conducted both univariate and multivariate logistic regression analyses on the training set (Supplementary Table 2). The outcomes of these analyses revealed that age (OR = 1.062, P < 0.001), CAP (OR = 1.017, P < 0.001), and γ-GT (OR = 1.007, P = 0.002) were independent risk factors for MetS in PLWH, while CD4/CD8 ratio (OR = 0.431, P = 0.024) and CKMB (OR = 0.985, P = 0.021) were identified as protective factors. Based on these independent factors, we constructed Model 1 for MetS prediction.

Additionally, we employed LASSO regression analysis on the training set (Supplementary Fig. 1). Since Model 1 comprised five variables, we selected the LASSO Model including five variables to build Model 2, which included age (OR = 1.065, P < 0.001), CD8 + T counts (OR = 1.001, P < 0.001), CAP (OR = 1.020, P < 0.001), γ-GT (OR = 1.008, P < 0.001), and LDH (OR = 1.008, P = 0.007). To streamline the model, common variables from both Model 1 and Model 2 were extracted to form Model 3, which included age (OR = 1.059, P < 0.001), CAP (OR = 1.021, P < 0.001) and γ-GT (OR = 1.008, P < 0.001). Ultimately, we proposed three distinct models for predicting MetS in PLWH and provided their model score calculation formulas separately (Table 2).

Table 2 The models for predicting MetS in PLWH
Full size table

Model evaluation and nomogram of optimal model

To assess and compare the predictive performance of the three models in both the training and validation sets, we initiated our analysis with the construction of ROC curves (Fig. 2A-B) and calculated the corresponding AUC values. In the training set, Model 1 achieved an AUC of 0.843, Model 2 achieved 0.849, and Model 3 achieved 0.831. In the validation set, the AUC values were 0.829 for Model 1, 0.834 for Model 2, and 0.824 for Model 3. These results indicated robust predictive capabilities for all three models, with Model 2 displaying the highest AUC in both sets. Additionally, we generated calibration curves (Fig. 2C-D), which demonstrated a close alignment between predicted probabilities and actual probabilities, affirming the well-calibrated nature of the models. Furthermore, the DCA curves (Fig. 2E-F) provided evidence of substantial clinical benefit associated with the predictive models. Importantly, Model 2 consistently outperformed the other models, exhibiting superior predictive performance in both the training and validation sets.

Fig. 2
figure 2

Performance evaluation of the models. (A) ROC curve for predictive models in the training cohort. (B) ROC curve for predictive models in the validation cohort. (C) Calibration plot of models in the training cohort. (D) Calibration plot of models in the validation cohort. (E) DCA curve of models in the training cohort. (F) DCA curve of models in the validation cohort

Full size image

For the sake of enhanced clinical applicability and more intuitive representation, we formulated a nomogram for the optimal model (Fig. 3). The nomogram provided a practical tool for healthcare professionals, facilitating the estimation of the risk of developing MetS in PLWH.

Fig. 3
figure 3

Nomogram of the optimal model. This nomogram illustrates the predictive model for MetS in PLWH. To use the nomogram, find each variable’s value (e.g., age, CD8 + T cell count, CAP, γ-GT, and LDH) on its respective axis and draw a line upward to get the corresponding points. Sum all points and locate the total score on the “Sum of all points” axis. This total score maps to a predicted risk of MetS on the bottom scale, with higher scores indicating a greater risk. Abbreviations: CAP, controlled attenuation parameter; γ-GT, gamma-glutamyl transferase; LDH, lactate dehydrogenase; MetS, metabolic syndrome; PWLH, people living with HIV

Full size image

Correlation between model score with components of MetS among PLWH

According to the formulation of the optimal model, we calculated the model score for the validation set. Subsequently, we carried out correlation analysis between the model score and each component of MetS (Fig. 4), which encompassed BMI (r = 0.661, P < 0.001), TG (r = 0.512, P < 0.001), HDL-C (r = -0.163, P = 0.020), SBP (r = 0.344, P < 0.001), DBP (r = 0.166, P = 0.018) and FPG (r = 0.237, P < 0.001). The results revealed a significant and noteworthy correlation between the model score and each component of MetS, which underscored the utility of the model in effectively predicting and assessing the risk factors associated with MetS in PLWH.

Fig. 4
figure 4

Correlation between model score with MetS components among PLWH in the validation cohort. The Scatter plots of model score and BMI (A), TG level (B), HDL-C level (C), SBP level (D), DBP level (E), and FPG level (F). Abbreviations: BMI, body mass index; TG, total triglycerides; HDL-C, high-density lipoprotein cholesterol; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose

Full size image

The relationship between model score with metabolic disorders

Our analysis extended to the exploration of the relationship between model score and MetS as well as associated metabolic disorders (Fig. 5). The results indicated significant differences in the model score when comparing the training set and the validation set across groups with or without MetS (P < 0.001), obesity (P < 0.001), HTG (P < 0.001), HA (P < 0.05), IFG (P < 0.001), and hypertension (P < 0.05). These findings collectively underscored the efficacy of the model in effectively predicting and discerning metabolic-related conditions in PLWH.

Fig. 5
figure 5

The model score between metabolic disorders among PWLH. (A) The model score in PLWH between Non-MetS group and MetS group. (B) The model score in PLWH between Non-Obesity group and Obesity group. (C) The model score in PLWH between Non-HTG group and HTG group. (D) The model score in PLWH between Non-HA group and HA group. (E) The model score in PLWH between Non-IFG group and IFG group. (F) The model score in PLWH between Non-Hypertension group and Hypertension group. Abbreviations: MetS, metabolic syndrome; HTG, hypertriglyceridemia; HA, hypoalphalipoproteinemia; IFG, impaired fasting glucose

Full size image

Discussion

In this study, we divided the cohort of 1017 participants into training and validation cohorts. Employing both multivariate logistic regression and Lasso regression, we systematically constructed predictive models. These models were rigorously evaluated for sensitivity, specificity, and calibration to determine the best model for further development into a nomogram scoring system. Furthermore, we delved into the correlation analysis between model score and the components of MetS. Our investigation also extended to the relationship between model score and related metabolic disorders. The final risk score model incorporated five robust risk predictors: age, CD8 + T counts, CAP, γ-GT, and LDH. Notably, our model differed from those predicting MetS in the general population by incorporating the CD8 + T cell counts, which accounted for the complex interplay between inflammation and immunosuppression and their potential role in metabolic disease in PLWH.

Researchers suggest that age-related diseases are primarily driven by chronic inflammation and immune system activation, which are commonly observed in older adults [22]. These factors play a significant role in the development of cardiovascular diseases and MetS [23,24,25]. The increased life expectancy has exposed PLWH to the effects of aging itself, which was defined as “inflammatory AIDS” towards the later stage of HIV infection [26]. In parallel, our findings emphasized the potential of age as a predictor of MetS in PLWH, aligning with the conclusions of other researchers.

It has been reported that the CD4/CD8 ratio has emerged as a potential indicator for predicting MetS in individuals with HIV/AIDS [27,28,29]. Additionally, some researchers have suggested that elevated CD8 + T cell counts may constitute a moderate risk factor for stroke in PLWH [27, 30]. In addition, elevated CD8 + T cell counts have been closely associated with immunosenescence in PLWH, suggesting that CD8 + T cells may play a pivotal role in linking immunosenescence with metabolic disorders [31]. Notably, in our study, we observed a significant difference in CD8 + T cell counts between the non-MetS and MetS groups in PLWH, and CD8 + T cell counts were also identified as independent predictors in our predictive model. This finding supports the potential of CD8 + T cells as a risk marker for MetS in PLWH and provides a direction for future research to explore the role of CD8 + T cells in predicting and managing MetS in this population.

γ-GT is widely used as a diagnostic marker for liver diseases [32], but increasing evidence suggests that it also plays an important role in cardiometabolic diseases such as obesity, hypertension, and diabetes [33]. Elevated γ-GT levels have been identified as predictive of the occurrence of MetS, CVD events and mortality [34,35,36]. Independent studies by Nguyen and Fourie have concurred that γ-GT is also linked to CVD risk in PLWH populations [37, 38]. In this study, γ-GT emerged not only as an independent risk factor for MetS in PLWH but also as one of the key elements in the predictive model, further highlighting its significance in predicting MetS occurrence in this population. Nevertheless, we acknowledge that elevated γ-GT may be a result of metabolic syndrome rather than a cause.

Elevated LDH indicates a pathological condition of acute tissue or cell injury [39, 40]. Previous studies have shown an association between elevated LDH and MetS, with LDH levels being closely related to frailty and all-cause mortality [39, 41]. Some researchers have also proposed that LDH tests can be utilized to monitor HIV disease progression and treatment response [42]. Although elevated LDH levels may be a consequence rather than a cause of metabolic syndrome, LDH emerged as a predictor of MetS in the PLWH population in this study, suggesting its potential as a biomarker for MetS in this group. However, further longitudinal studies are needed to clarify the causal relationship between LDH and MetS in PLWH.

In recent years, Fibroscan has emerged as a widely adopted diagnostic tool, primarily employed for the prediction of fatty liver and liver fibrosis in diverse populations. Notably, previous studies have reported the use of a combination of CAP and LSM for the prediction of nonalcoholic steatohepatitis or liver fibrosis, particularly in PLWH or those coinfected with HCV [43,44,45]. However, the role of CAP and LSM in predicting MetS has been relatively underexplored. Our findings suggest that CAP may serve as a valuable predictive marker for MetS in PLWH, which could have important implications for the management and prognosis of this population. Nonetheless, we acknowledge that elevated CAP might also reflect the presence of fatty liver as a consequence of MetS, rather than serving as a causal factor. Further longitudinal studies are needed to clarify the directionality of this association.

Our study possesses several strengths. Firstly, it proposes a model that may assist in identifying MetS in PLWH, offering clinicians a possible tool to help assess MetS risk and inform early intervention strategies. Secondly, this study uniquely integrates a multivariate prediction model that considers patient-host characteristics specific to PLWH, revealing potential associations between new variables and MetS in this population. Nonetheless, we acknowledge certain limitations in our study. Firstly, as this is an exploratory analysis, we did not adjust for multiple comparisons across the variables examined, which may increase the likelihood of false positives. Additionally, its retrospective nature may be subject to inherent limitations associated with this study design, as some observed associations may represent consequences rather than contributing factors to metabolic syndrome. Future research should focus on conducting longitudinal studies to assess the model’s predictive effectiveness over time. Finally, our model lacks external validation using data from independent sources, underscoring the need for further research to confirm its robustness across diverse cohorts.

Conclusions

In conclusion, our study provides a potential model for assessing the risk of MetS in PLWH, further research is needed to validate its predictive capacity and explore the causal relationships underlying these associations.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Abbreviations

AIDS:

Acquired immunodeficiency syndrome

ALB:

Albumin

ALP:

alkaline phosphatase

ALT:

Alanine aminotransferase

ART:

Antiretroviral therapy

AST:

Aspartic aminotransferase

AUC:

Area under the ROC curve

B/F/T:

Bictegravir group

BMI:

Body mass index

BUN:

Blood urea nitrogen

CAP:

Controlled attenuation parameter

CHOL:

Total cholesterol

CK:

Creatine kinase

CKMB:

Creatine kinase myocardial band

CR:

Creatinine

CRP:

C-reactive protein

DBIL:

Direct bilirubin

DBP:

Diastolic blood pressure

DCA:

Decision curve analysis

DTG:

Dolutegravir group

EOS:

Eosinophils count

EFV:

Efavirenz group

eGFR:

Estimated glomerular filtration rate

ESR:

Erythrocyte sedimentation rate

FPG:

Fasting plasma glucose

G:

Globulin

HA:

Hypoalphalipoproteinemia

HBDH:

Hydroxybutyrate dehydrogenase

HDL-C:

High-density lipoprotein cholesterol

HGB:

Hemoglobin

HTG:

Hypertriglyceridemia

IFG:

Impaired fasting glucose

LDH:

Lactate dehydrogenase

LDL-C:

Low-density lipoprotein cholesterol

LPV/r:

Lopinavir/ritonavir group

LSM:

Liver stiffness measurement

LYM:

Lymphocyte count

MetS:

Metabolic syndrome

MONO:

Monocyte count

PLT:

Platelets count

ROC:

Receiver operating characteristic

RBC:

Red blood cell count

SBP:

Systolic blood pressure

TBIL:

Total bilirubin

TG:

Total triglycerides

UA:

Uric acid

VLDL-C:

Very low-density lipoprotein cholesterol

WBC:

White blood cell count

References

  1. Mills EJ, Bakanda C, Birungi J, Chan K, Ford N, Cooper CL, et al. Life Expectancy of persons receiving combination antiretroviral therapy in low-income countries: a cohort analysis from Uganda. Ann Intern Med. 2011;155:209–16.

    Article  PubMed  Google Scholar 

  2. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. N Engl J Med. 2015;373:808–22.

    Article  Google Scholar 

  3. Althoff KN, Gebo KA, Moore RD, Boyd CM, Justice AC, Wong C, et al. Contributions of traditional and HIV-related risk factors on non-AIDS-defining cancer, myocardial infarction, and end-stage liver and renal diseases in adults with HIV in the USA and Canada: a collaboration of cohort studies. Lancet HIV. 2019;6:e93–104.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Shah ASV, Stelzle D, Lee KK, Beck EJ, Alam S, Clifford S, et al. Global Burden of Atherosclerotic Cardiovascular Disease in people living with HIV: systematic review and Meta-analysis. Circulation. 2018;138:1100–12.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Freiberg MS, Chang C-CH, Kuller LH, Skanderson M, Lowy E, Kraemer KL, et al. HIV infection and the risk of acute myocardial infarction. JAMA Intern Med. 2013;173:614–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Sudano I, Spieker LE, Noll G, Corti R, Weber R, Lüscher TF. Cardiovascular disease in HIV infection. Am Heart J. 2006;151:1147–55.

    Article  PubMed  Google Scholar 

  7. Naidu S, Ponnampalvanar S, Kamaruzzaman SB, Kamarulzaman A. Prevalence of metabolic syndrome among people living with HIV in developing countries: a systematic review. AIDS Patient Care STDs. 2017;31:1–13.

    Article  PubMed  Google Scholar 

  8. Alberti KGMM, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation. 2009;120:1640–5.

  9. Tune JD, Goodwill AG, Sassoon DJ, Mather KJ. Cardiovascular consequences of metabolic syndrome. Translational Res. 2017;183:57–70.

    Article  CAS  Google Scholar 

  10. Abate N. Obesity and cardiovascular disease: pathogenetic role of the metabolic syndrome and therapeutic implications. J Diabetes Complicat. 2000;14:154–74.

  11. Rönsholt FF, Ullum H, Katzenstein TL, Gerstoft J, Ostrowski SR. Persistent inflammation and endothelial activation in HIV-1 infected patients after 12 years of antiretroviral therapy. PLoS ONE. 2013;8:e65182.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Thet D, Siritientong T. Antiretroviral Therapy-Associated Metabolic Complications: Review of the Recent Studies. HIV/AIDS - Research and Palliative Care. 2020;12:507–24.

  13. Alencastro PR, Fuchs SC, Wolff FH, Ikeda ML, Brandão ABM, Barcellos NT. Independent predictors of metabolic syndrome in HIV-Infected patients. AIDS Patient Care STDs. 2011;25:627–34.

    Article  PubMed  Google Scholar 

  14. Jericó C, Knobel H, Montero M, Ordoñez-Llanos J, Guelar A, Gimeno JL, et al. Metabolic syndrome among HIV-infected patients: prevalence, characteristics, and related factors. Diabetes Care. 2005;28:132–7.

    Article  PubMed  Google Scholar 

  15. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157–63.

    Article  Google Scholar 

  16. Alberti KGMM, Zimmet P, Shaw J. Metabolic syndrome—a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med. 2006;23:469–80.

    Article  CAS  PubMed  Google Scholar 

  17. Berglund L, Brunzell JD, Goldberg AC, Goldberg IJ, Sacks F, Murad MH, et al. Evaluation and treatment of Hypertriglyceridemia: an endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2012;97:2969–89.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Singh V, Sharma R, Kumar A, Deedwania P. Low high-density lipoprotein cholesterol: current status and future strategies for management. Vasc Health Risk Manag. 2010;6:979–96.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Flack JM, Adekola B. Blood pressure and the new ACC/AHA hypertension guidelines. Trends Cardiovasc Med. 2020;30:160–4.

    Article  PubMed  Google Scholar 

  20. Petersen JL, McGuire DK. Impaired glucose tolerance and impaired fasting glucose — a review of diagnosis, clinical implications and management. Diabetes Vascular Disease Res. 2005;2:9–15.

    Article  Google Scholar 

  21. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2012;35:S64–71.

    Article  Google Scholar 

  22. Deeks SG. HIV infection, inflammation, immunosenescence, and aging. Annu Rev Med. 2011;62:141–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Pacheco AG, Tuboi SH, Faulhaber JC, Harrison LH, Schechter M. Increase in Non-AIDS related conditions as causes of death among HIV-Infected individuals in the HAART era in Brazil. PLoS ONE. 2008;3:e1531.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Ascaso JF, Romero P, Real JT, Lorente RI, Martínez-Valls J, Carmena R. Abdominal obesity, insulin resistance, and metabolic syndrome in a southern European population. Eur J Intern Med. 2003;14:101–6.

    Article  CAS  PubMed  Google Scholar 

  25. Paula AA, Falcão MC, Pacheco AG. Metabolic syndrome in HIV-infected individuals: underlying mechanisms and epidemiological aspects. AIDS Res Therapy. 2013;10:32.

    Article  Google Scholar 

  26. Nasi M, Pinti M, Mussini C, Cossarizza A. Persistent inflammation in HIV infection: established concepts, new perspectives. Immunol Lett. 2014;161:184–8.

    Article  CAS  PubMed  Google Scholar 

  27. Chammartin F, Darling K, Abela IA, Battegay M, Furrer H, Calmy A, et al. CD4:CD8 ratio and CD8 cell count and their prognostic relevance for coronary heart disease events and stroke in antiretroviral treated individuals: the Swiss HIV Cohort Study. JAIDS J Acquir Immune Defic Syndr. 2022;91:508.

    Article  CAS  PubMed  Google Scholar 

  28. Gojak R, Hadžiosmanović V, Baljić R, Zečević L, Ćorić J, Mijailović Ž. CD4/CD8 ratio as a predictor for the occurrence of metabolic syndrome in HIV / AIDS patients during 6 months of cART therapy. J Med Biochem. 2019;38:489–95.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Mazzitelli M, Fusco P, Brogna M, Vallone A, D’Argenio L, Beradelli G, et al. Weight of clinical and social determinants of metabolic syndrome in people living with HIV. Viruses-Basel. 2022;14:8.

    Article  Google Scholar 

  30. Quiros-Roldan E, Magro P, Raffetti E, Izzo I, Borghetti A, Lombardi F, et al. Biochemical and inflammatory modifications after switching to dual antiretroviral therapy in HIV-infected patients in Italy: a multicenter retrospective cohort study from 2007 to 2015. BMC Infect Dis. 2018;18:285.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Dock JN, Effros RB. Role of CD8 T cell replicative senescence in human aging and in HIV-mediated immunosenescence. Aging Disease. 2011;2:382.

    PubMed  PubMed Central  Google Scholar 

  32. Whitfield JB. Gamma Glutamyl transferase. Crit Rev Clin Lab Sci. 2001;38:263–355.

    Article  CAS  PubMed  Google Scholar 

  33. Ruttmann E, Brant LJ, Concin H, Diem G, Rapp K, Ulmer H, et al. Gamma-glutamyltransferase as a risk factor for cardiovascular disease mortality: an epidemiological investigation in a cohort of 163,944 Austrian adults. Circulation. 2005;112:2130–7.

    Article  CAS  PubMed  Google Scholar 

  34. Liu C-F, Zhou W-N, Fang N-Y. Gamma-glutamyltransferase levels and risk of metabolic syndrome: a meta-analysis of prospective cohort studies. Int J Clin Pract. 2012;66:692–8.

    Article  CAS  PubMed  Google Scholar 

  35. Lee MY, Koh SB, Koh JH, Nam SM, Shin JY, Shin YG, et al. Relationship between γ-glutamyltransferase and metabolic syndrome in a Korean population. Diabet Med. 2008;25:469–75.

    Article  CAS  PubMed  Google Scholar 

  36. Onat A, Can G, Örnek E, Çiçek G, Ayhan E, Doğan Y. Serum γ-Glutamyltransferase: independent predictor of risk of diabetes, hypertension, metabolic syndrome, and Coronary Disease. Obesity. 2012;20:842–8.

    Article  CAS  PubMed  Google Scholar 

  37. Fourie CMT, Botha-Le Roux S, Smith W, Schutte AE, Breet Y, Mels CMC, et al. Vascular function and cardiovascular risk in a HIV infected and HIV free cohort of African ancestry: baseline profile, rationale and methods of the longitudinal EndoAfrica-NWU study. BMC Infect Dis. 2020;20:473.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Nguyen KA, Peer N, Kengne AP. Associations of gamma-glutamyl transferase with cardio-metabolic diseases in people living with HIV infection in South Africa. PLoS ONE. 2021;16:e0246131.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Wu L-W, Kao T-W, Lin C-M, Yang H-F, Sun Y-S, Liaw F-Y, et al. Examining the association between serum lactic dehydrogenase and all-cause mortality in patients with metabolic syndrome: a retrospective observational study. BMJ Open. 2016;6:e011186.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Liao Y-T, Chen C-J, Li W-F, Hsu L-I, Tsai L-Y, Huang Y-L, et al. Elevated lactate dehydrogenase activity and increased cardiovascular mortality in the arsenic-endemic areas of southwestern Taiwan. Toxicol Appl Pharmacol. 2012;262:232–7.

    Article  CAS  PubMed  Google Scholar 

  41. Chen L-H, Wu L-W. Association between serum lactate dehydrogenase and frailty among individuals with metabolic syndrome. PLoS ONE. 2021;16:e0256315.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Ramana KV, Rao R, Kandi S, Singh PA, Kumar VBP. Elevated activities of serum lactate dehydrogenase in human immunodeficiency virus sero-positive patients in highly active antiretroviral therapy era. J Dr NTR Univ Health Sci. 2013;2:162.

    Article  Google Scholar 

  43. Benmassaoud A, Ghali P, Cox J, Wong P, Szabo J, Deschenes M, et al. Screening for nonalcoholic steatohepatitis by using cytokeratin 18 and transient elastography in HIV mono-infection. PLoS ONE. 2018;13:e0191985.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Michel M, Labenz C, Wahl A, Anders M, Armandi A, Huber Y, et al. Prevalence and risk factors of nonalcoholic steatohepatitis with significant fibrosis in people with HIV. Aids. 2022;36:1665–74.

    Article  CAS  PubMed  Google Scholar 

  45. Erman A, Sathya A, Nam A, Bielecki JM, Feld JJ, Thein H-H, et al. Estimating chronic hepatitis C prognosis using transient elastography-based liver stiffness: a systematic review and meta-analysis. J Viral Hepat. 2018;25:502–13.

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We express our gratitude to our colleagues in the Department of Infectious Diseases at Nanfang Hospital for their assistance and support.

Funding

This study was supported by the National Key Research and Development Program of China (No. 2022YFC2304800), the National Natural Science Foundation of China (No.82203300).

Author information

Author notes

  1. Suling Chen, Yuyuan Xu and Yuanhui Jiang contributed equally to this work.

Authors and Affiliations

  1. Department of Infectious Diseases, Nanfang Hospital, Southern Medical University, Guangzhou, 510515, China

    Suling Chen, Yuyuan Xu, Yuanhui Jiang, Hongjie Chen, Xuwen Xu, Huiqun Zhong, Jie Peng & Shaohang Cai

  2. Key Laboratory of Infectious Diseases Research in South China, Ministry of Education, Guangzhou, China

    Suling Chen, Yuyuan Xu, Yuanhui Jiang, Hongjie Chen, Xuwen Xu, Huiqun Zhong, Jie Peng & Shaohang Cai

  3. Department of Communicable and Endemic Disease Control and Prevention, Haizhu District Center for Disease Control and Prevention, Guangzhou, China

    Xiaoxuan Wu

  4. Second Department of Elderly Respiratory, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Guangdong Provincial Geriatrics Institute, Guangzhou, China

    Zhe Qian

Authors
  1. Suling Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Yuyuan Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Yuanhui Jiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Hongjie Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xiaoxuan Wu
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Zhe Qian
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Xuwen Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Huiqun Zhong
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Jie Peng
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Shaohang Cai
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

HQZ, JP and SHC contributed to the study conception and design. SLC, YYX and YHJ analyzed and interpreted the data, as well as drafted and finalized the manuscript. SLC, HJC, XXW, ZQ, and XWX participated in data collection and methodology. All the authors critically reviewed and approved the final version of the manuscript.

Corresponding authors

Correspondence to Huiqun Zhong, Jie Peng or Shaohang Cai.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the Institutional Ethics Committee of Nanfang Hospital (study identifier: NFEC-2021-448) and the study protocol was performed in accordance with the Helsinki Declaration of 1964 and its later amendments. Informed consent was obtained from all individuals.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chen, S., Xu, Y., Jiang, Y. et al. Development and validation of a predictive model for metabolic syndrome in a large cohort of people living with HIV. Virol J 21, 321 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12985-024-02592-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12985-024-02592-8

Keywords

Virology Journal

ISSN: 1743-422X