Prevalence of metabolic syndrome in HIV-infected patients naive to antiretroviral therapy or receiving a first-line treatment

Abstract

Background: The combination antiretroviral therapy (cART) has dramatically improved the life expectancy of patients with HIV infection, but may lead to several long-term metabolic abnormalities. However, data about the frequency of metabolic syndrome (MS) in HIV-infected people vary considerably across different observational studies.

Methods: The prevalence of MS among HIV-infected patients was evaluated by a cross-sectional study conducted among subjects naive to cART or receiving the first antiretroviral regimen and referring to our Clinics from January 2015 to December 2015. The diagnosis of MS was made based on the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III), and International Diabetes Federation (IDF) criteria.

Results: The study recruited 586 patients: 98 naive to cART and 488 under the first antiretroviral treatment. The prevalence of MS, according to NCEP-ATP III criteria, was significantly higher among treated patients than among naive ones (20.9% vs. 7.1%; p = 0.014). The most frequently reported components of MS among treated patients were high triglycerides (44.3%), low high-density lipoprotein cholesterol (41.1%), and hypertension (19.7%). On multivariate analysis, long duration of HIV infection, low nadir of CD4 lymphocytes, high body mass index, current use of one protease inhibitor, and long duration of cART were significantly associated with a higher risk of MS, while current use of one integrase inhibitor was significantly associated with a lower risk of MS.

Conclusions: The non-negligible prevalence of MS among HIV-infected patients under cART requires a careful and periodic monitoring of its components, with particular attention to dyslipidemia and hypertension.

Keywords:

• Metabolic disorders
• Dyslipidemia
• Diabetes mellitus
• Antiretroviral therapy
• Cardiovascular risk

Introduction

Several age-related comorbidities and metabolic alterations have been reported with increasing frequency among HIV-infected persons as a consequence of aging, chronic inflammation, immune activation, and long-term exposure to the combination antiretroviral therapy (cART).^1,2

The metabolic abnormalities observed in HIV-positive patients include changes in body fat (mostly visceral fat accumulation with central obesity), dyslipidemia, insulin resistance, and type 2 diabetes mellitus. Moreover, patients with HIV infection show an increased risk of myocardial infarction and cardiovascular disease compared to the general population, caused by an overrepresentation of traditional risk factors, several metabolic disorders, and a chronic systemic inflammation.^3,4

Metabolic syndrome (MS) is a complex disorder, characterized by an aggregation of some physical and metabolic abnormalities, including central obesity, endothelial dysfunction, increased blood pressure, dyslipidemia, and hyperglycemia. ^5,6 The overall age-adjusted prevalence of MS in the United States increased from 29 to 34% over the last decade, while it seems to be lower in the European countries, where it ranges from 6 to 30%.⁷⁻⁹ Moreover, there is a remarkable controversy about the completeness and precision of the current MS definitions, but most authors agree that MS originates from the insulin resistance and usually moves forward diabetes mellitus, atherosclerosis, and cardiovascular diseases. Consequently, MS usually predicts an increased cardiovascular risk and it could explain in part the increased risk of myocardial infarction observed among HIV-infected patients.⁵⁻⁹

The real prevalence of MS among HIV-infected patients is still debated and varies in variable of observational studies, ranging from 7 to 52%. This large variability is probably due to several differences in the employed criteria, sampled population, and study design, most being cross-sectional with low patient number and different populations enrolled.¹⁰⁻¹³

The present cross-sectional study aims to evaluate the prevalence of MS in HIV-infected individuals naive to cART or receiving their first antiretroviral regimen at our tertiary care center.

Patients and methods

A cross-sectional study was conducted among patients with HIV-1 infection naive to cART or receiving the first antiretroviral regimen and referring for routine clinical and laboratory follow-up from January 2015 to December 2015 to the Clinic of Infectious Diseases, Outpatient Unit, of the S.Orsola-Malpighi Hospital in Bologna, Italy.

All HIV-infected subjects who were 18-year old or above, fulfilled inclusion and exclusion criteria and signed a written informed consent were enrolled into the study to evaluate the prevalence of MS. Exclusion criteria were duration of current antiretroviral therapy <12 months, no regular clinical or laboratory data of follow-up for at least 12 months, active opportunistic diseases or severe infectious diseases, acute or chronic inflammatory diseases (diagnosed by a Rheumatologist), alcohol or drug abuse, hypothyroidism, Cushing’s syndrome, acute or chronic kidney diseases, acute hepatitis, liver cirrhosis, pregnancy, an underlying treatment with corticosteroids, nonsteroidal anti-inflammatory drugs (NSAIDs), or immune-modulatory agents. Patients receiving the first antiretroviral therapy were excluded from the study if they presented diagnosis of MS or diabetes mellitus or arterial hypertension before the start of the antiretroviral therapy. Patients with duration of current cART lower than 12 months were excluded because the real contribute of the antiretroviral regimen (beyond the viral suppression) to the MS occurrence could not be appraised in subjects with a more recent initiation of cART. A periodical check about concomitant medications was performed by a specific questionnaire presented to the patients at each control visit.

Current alcohol use and intravenous drug dependence were defined as a daily alcohol consumption >30 g and ≥1 intravenous drug use within the previous six months, respectively. Liver cirrhosis was excluded by liver biopsy or elastometry. The HBV and HCV coinfections were diagnosed by the persistent positivity of HBs antigen or HCV serum antibodies positivity plus HCV-RNA positivity, respectively.

The following demographic, clinical, and laboratory data were obtained from the patients’ medical records: sex, age, race, physical examination (thoracic auscultation and abdominal palpation), waist circumference, body mass index (BMI), arterial pressure, clinical manifestations, current and past medications, current smoking status; spot urinalysis, and serum levels of triglycerides, total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, glucose, complete liver and kidney function tests, CD4+ and CD8+ T lymphocyte count, and HIV viral load. Serum glucose and lipid levels were measured after a 12-h overnight fast.

All medical and laboratory tests, such as all the antiretroviral drugs were provided free of charge for all patients, in conformity with the Italian National Health System organization, so patient income/education attainment and access to health care were not potential confounders.

All the plasma samples were analyzed for the HIV-RNA level using the automated COBAS AmpliPrep Instrument for specimen processing and the COBAS TaqMan Analyzer for amplification and detection (Roche CobasAmpliPrep/Cobas TaqMan HIV-1 tests version 2.0; Roche Diagnostics, Mannheim, Germany). Besides gag primers and a FAM-labeled gag probe, additional LTR primers and a FAM-labeled LTR probe were included in the assay. The two targets, gag and LTR, are amplified with the same efficiency and the limit of quantification (LOQ), as defined by the manufacturer, was reduced to 20 copies/mL.

Diagnosis of type 2 diabetes mellitus and disorders of glucose metabolism were defined using the American Diabetes Association criteria.¹⁴ Diabetes mellitus was diagnosed by a fasting glucose ≥126 mg/dL or a non-fasting glucose > 200 mg/dL in the absence of symptoms of diabetes. Diabetes was also diagnosed if patients were taking insulin or oral anti-diabetic drugs.

Waist circumference was measured to the nearest 0.1 cm horizontally at the narrowest point between the lower end of the rib cage and iliac crest. BMI was classified based on the World Health Organization (WHO) definition for adults as underweight (<18.5 kg/m²), normal (18.5–24.9 kg/m²), overweight (25–29.9 kg/m²), and obese (≥30 kg/m²).¹⁵ Lipodystrophy was determined by physical examination which recorded fat loss and/or fat accumulation (in the face, neck, dorso-cervical region, arms, breasts, abdomen, buttocks, and legs), or mixed form. Physical examination was performed by two physicians and diagnosis of lipodystrophy was made only if both evaluations were in agreement. Blood pressure was measured by a physician in the outpatient clinic with a mercury sphygmomanometer, with the study subject seated for ≥10 min before measurement. The average of three consecutive measurements was repeated on at least two different visit days 3–4 months apart. Patients were classified as having hypertension if the blood pressure was ≥ 140/90 mm Hg or if they were taking blood pressure lowering drugs.

Metabolic syndrome was defined using the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATPIII),⁶ and International Diabetes Federation (IDF) criteria.¹⁶

The NCEP ATPIII diagnosis requires three or more of the following criteria: (a) waist circumference ≥88 cm in women and ≥102 cm in men; (b) systolic blood pressure ≥130 mm Hg and/or diastolic blood pressure ≥85 mm Hg or specific drug treatment for hypertension; (c) triglycerides ≥150 mg/dL or specific drug treatment for high triglycerides; (c) high-density lipoprotein (HDL) cholesterol <50 mg/dL in women and <40 mg/dL in men or specific drug treatment for dyslipidemia; and (d) glucose ≥100 mg/dL or specific drug treatment for hyperglycemia. ⁶

In the IDF definition, central obesity is a mandatory criterion for establishing the diagnosis, and it is defined for Europeans by a waist circumference ≥ 80 cm in women and ≥94 cm in men. MS is diagnosed if there is central obesity plus at least two of the following features: (a) systolic blood pressure ≥130 mm Hg and/or diastolic blood pressure ≥85 mm Hg or specific drug treatment for hypertension; (b) triglycerides ≥150 mg/dL or specific drug treatment for high triglycerides; (c) HDL cholesterol <50 mg/dL in women and <40 mg/dL in men or specific drug treatment for dyslipidemia; (d) glucose ≥100 mg/dL or previously diagnosed type 2 diabetes mellitus.¹⁶

Continuous variables are expressed by mean values ± standard deviation (SD) for all levels of comparison. Categorical variables are presented in frequency (proportion) and test of association between proportions was done using χ² test. Statistical evaluation was carried out by Student’s t-test, Mantel–Haenszel chi-square test, or Fisher exact test (where appropriate), with significance levels placed at p < 0.05. A univariate analysis of collected data was performed to determine the significance of dependent and independent variables. The t-test was employed to compare the subjects who presented MS with those who did not present MS. The χ² or Fisher exact test was employed to assess the predictive value of age >55 years, duration of HIV infection >5 years, AIDS diagnosis, current and nadir value of CD4 T lymphocytes <200 cells/mm³, current and zenit value of HIV RNA >100,000 copies/mL, chronic hepatitis C, lipodystrophy, BMI >25 kg/m², smoking status, duration of cART >3 years, current use of tenofovir, abacavir, one non-nucleoside reverse transcriptase inhibitors (NNRTI), one ritonavir-boosted protease inhibitor (PI), or one integrase strand transfer inhibitor (INSTI). Significant variables with p ≤ 0.05 were included in a multivariate stepwise logistic regression analysis to predict the occurrence of MS (GraphPad Statistical Software, San Diego, CA, USA).

The adherence to the current therapy was carefully checked on the outpatient visits by self-reported questionnaires. The study was approved by the Ethic Committee of the S.Orsola-Malpighi Hospital and all participants signed an informed consent after receiving information about the purpose of the study.

Results

A total of 586 patients were enrolled in the study: 98 patients were naive to cART and 488 were receiving their first antiretroviral treatment. The sociodemographic, clinical, and laboratory characteristics of the enrolled patients stratified by the antiretroviral therapy status are summarized in Table 1.

Table 1 Sociodemographic, clinical, and laboratory characteristics of the 586 enrolled patients stratified by the antiretroviral therapy status

	All patients	cART-naive patients	On cART patients	p value*
	(n = 586)	(n = 98)	(n = 488)
Males [No. (%)]	479 (81.7)	82 (83.7)	397 (81.3)	0.667
Mean age [years (SD)]	45.2 (17.4)	44.1 (15.9)	46.7 (18.5)	0.529
white race [No. (%)]	564 (96.2)	92 (93.9)	472 (96.7)	0.409
Current smokers [No. (%)]	349 (59.6)	55 (56.1)	294 (60.2)	0.094
Mean duration of HIV infection [years (SD)]	8.3 (2.4)	3.8 (1.4)	9.2 (2.9)	0.021
Mean nadir CD4 T lymphocyte count [cells/mm^3 (SD)]	372 (159)	547 (208)	235 (109)	<0.001
Mean current CD4 T lymphocyte count [cells/mm^3 (SD)]	437 (231)	589 (251)	648 (294)	<0.001
Patients with plasma HIV RNA <50 copies/mL [No. (%)]	480 (81.9)	9 (9.2)	471 (96.5)	0.66
Mean current plasma HIV RNA in patients with detectable viremia [log10 copies/mL (SD)]	3.66 (0.38)	4.34 (0.56)	3.42 (0.21)	0.032
Mean zenit plasma HIV RNA [log10 copies/mL (SD)]	4.97 (0.98)	4.89 (0.82)	5.04 (0.95)	0.064
HIV stage [No. (%)]
Asymptomatic	383 (65.3)	98 (100)	285 (58.4)	<0.001
Symptomatic non-AIDS	147 (25.1)	0	147 (30.1)	<0.001
AIDS	56 (9.6)	0	56 (11.5)	<0.001
Patients treated with [No. (%)]:
2 NRTIs + 1 NNRTI	219 (37.4)	0	219 (44.9)	n.a.
2 NRTIs + 1 PI	167 (28.5)	0	167 (34.2)	n.a.
2 NRTIs + 1 INSTI	102 (17.4)	0	102 (20.9)	n.a.
Mean duration of antiretroviral therapy in patients on cART [years (SD)]	4.3 (1.9)	0	4.3 (1.9)	n.a.
Patients with lipodystrophy syndrome [No. (%)]:	90 (15.3)	6 (6.1)	84 (17.2)	0.721
Fat loss	4 (0.7)	1 (1)	3 (0.6)	0.897
Fat accumulation	86 (14.7)	5 (5)	81 (16.6)	0.721
Patients with chronic HCV infection [No. (%)]	45 (7.7)	8 (8.2)	37 (7.6)	0.365
Patients with chronic HBV infection [No. (%)]	13 (2.2)	1 (1)	12 (2.5)	0.188
Mean total cholesterol [mg/dL (SD)]	221 (51)	188 (43)	227 (58)	0.018
Mean HDL cholesterol [mg/dL (SD)]	43 (9)	48 (8)	41 (11)	0.044
Mean LDL cholesterol [mg/dL (SD)]	131 (35)	112 (27)	138 (39)	0.019
Mean triglycerides [mg/dL (SD)]	244 (87)	207 (81)	256 (93)	0.006
Mean glucose [mg/dL (SD)]	94 (36)	92 (38)	95 (32)	0.068
Mean systolic blood pressure [mmHg (SD)]	126 (25)	122 (29)	127 (24)	0.074
Mean diastolic blood pressure [mmHg (SD)]	76 (16)	74 (14)	78 (18)	0.118
Mean waist circumference [cm (SD)]	87.1 (16.3)	84.2 (15.8)	88.7 (17.6)	0.039
Mean BMI [kg/m^2 (SD)]	24.1 (6.4)	23.8 (5.9)	24.4 (5.6)	0.067

Values are presented as mean + standard deviation (SD) and frequency (%).

* p value defines the level of significance in the comparison between cART naive and on cART patients.

cART, combination antiretroviral therapy; NRTIs, nucleoside reverse transcriptase inhibitors; NNRTIs, non-nucleoside reverse transcriptase inhibitors; PIs, protease inhibitors; n.a., not applicable; HCV, hepatitis C virus; HBV, hepatitis B virus; HDL, high-density lipoprotein; LDL, low-density lipoprotein; BMI, body mass index.

Patients on cART showed a significantly longer mean duration of HIV infection, a lower mean nadir value of CD4 T lymphocytes, a higher mean current value of CD4 T lymphocytes, a lower mean HIV RNA, and a higher percentage of symptomatic patients than those naive to cART. Current antiretroviral regimen in treated subjects included two nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) plus one non-nucleoside reverse transcriptase inhibitor (NNRTI) in 219 cases (44.9%), one ritonavir-boosted protease inhibitor (PI) in 167 (34.2%), and one integrase strand transfer inhibitor (INSTI) in 102 (20.9%). The NRTI backbone was represented by tenofovir/emtricitabine in 372 cases (76.2%) and abacavir/lamivudine in 104 (21.3%). Current NNRTI was rilpivirine in 129 cases (26.4%) and efavirenz in 64 (13.1%). Current PI was darunavir/ritonavir in 112 cases (22.9%) and atazanavir/ritonavir in 52 (10.6%). Current INSTI was raltegravir in 65 cases (13.3%), elvitegravir/cobicistat in 24 (4.9%) and dolutegravir in 13 (2.7%). Among treated subjects, plasma HIV RNA was <50 copies/mL in 471 cases (96.5%).

With regard to the laboratory features, patients on cART presented a significantly higher mean concentration of total cholesterol, LDL cholesterol, and triglycerides and a significantly lower mean concentration of HDL cholesterol than naive ones. Mean waist circumference was significantly higher in patients on cART, while mean values of blood pressure and BMI were comparable in both groups. The prevalence of lipodystrophy was higher among patients on cART, but difference was not statistically significant (Table 1).

The prevalence of MS among the study population was 18.6% for the NCEP ATPIII criteria and 16% for the IDF criteria. The prevalence of MS was significantly higher among patients under cART than among cART-naive subjects for both the NCEP ATPIII definition (20.9% vs. 7.1%; p = 0.014) and the IDF definition (18.2% vs. 5.1%; p = 0.005) (Table 2).

Table 2 Prevalence of metabolic syndrome diagnosed by the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATPIII) or by the International Diabetes Federation (IDF) criteria and prevalence of its clustering components among the 586 enrolled patients stratified by the antiretroviral therapy status

	All patients	cART-naive patients	On cART patients	p value*
	(n = 586)	(n = 98)	(n = 488)
Metabolic syndrome diagnosed by the
NCEP ATPIII criteria [No. (%)]	109 (18.6)	7 (7.1)	102 (20.9)	0.014
IDF criteria [No. (%)]	94 (16)	5 (5.1)	89 (18.2)	0.005
Diagnostic criteria:
0 [No. (%)]	125 (21.3)	36 (36.7)	89 (18.2)	0.022
1 [No. (%)	205 (35)	39 (39.8)	166 (34)	0.041
2 [No. (%)]	147 (25.1)	16 (16.3)	131 (26.8)	0.009
>2 [No. (%)]	109 (18.6)	7 (7.1)	102 (20.9)	0.014
Increased waist circumference [No. (%)]	100 (17.1)	7 (7.1)	93 (19.1)	<0.001
Increased blood pressure [No. (%)]	112 (19.1)	16 (16.3)	96 (19.7)	0.088
High triglycerides [No. (%)]	234 (39.9)	18 (18.4)	216 (44.3)	<0.001
Low HDL cholesterol [No. (%)]	215 (36.7)	14 (14.3)	201 (41.1)	<0.001
High glucose [No. (%)]	57 (9.7)	8 (8.1)	49 (10)	0.062

Values are presented as frequency (%).

* p value defines the level of significance in the comparison between cART naive and on cART patients.

cART, combination antiretroviral therapy; NCEP ATPIII, National Cholesterol Education Program Adult Treatment Panel III; IDF, International Diabetes Federation; HDL, high-density lipoprotein.

The prevalence of the clustering components of the MS was also significantly higher among treated patients compared to cART-naive counterparts. Patients with two diagnostic criteria of MS were 26.8% vs. 16.3% (p = 0.009) and those with more than two diagnostic criteria were 20.9% vs. 7.1% (p = 0.014). Particularly, among treated subjects there was a significantly higher percentage of patients with increased waist circumference, high triglycerides, and low HDL cholesterol compared to the cART-naive ones.

Table 3 shows predictors of MS assessed by univariate and multivariate analyses. Factors significantly associated with the occurrence of MS by the univariate analysis were age >55 years, duration of HIV infection >5 years, nadir value of CD4 T lymphocytes <200 cells/mm³, BMI >25 kg/m², duration of cART >3 years, and current use of one ritonavir-boosted PI, while current use of one INSTI had a protective effect. In the multivariate logistic regression analysis, only the following predictors were significantly associated with the occurrence of MS: duration of HIV infection >5 years [odds ratio (OR) 2.95, p = 0.029], nadir value of CD4 T lymphocytes <200 cells/mm³ (OR 2.27, p = 0.037), BMI > 25 kg/m² (OR 3.61, p = 0.008), duration of cART >3 years (OR 3.44, p < 0.001), and current use of one ritonavir-boosted PI (OR 1.85, p = 0.041). On the other hand, current use of one INSTI was significantly associated with a reduced risk of MS (OR 0.55, p = 0.037).

Table 3 Univariate and multivariate logistic regression analysis assessing predictors of metabolic syndrome

Variables	Univariate analysis		Multivariate analysis
	OR (95% CI)	p value	OR (95% CI)	p value
Age >55 years	1.29 (1.09 to 1.57)	0.033	1.23 (0.87 to 1.51)	0.058
Duration of HIV infection >5 years	2.81 (2.18 to 3.66)	0.004	2.95 (2.33 to 4.07)	0.029
AIDS diagnosis	0.98 (0.74 to 1.28)	0.072
Current CD4 T lymphocytes <200 cells/mm^3	1.09 (0.73 to 1.55)	0.086
Nadir CD4 T lymphocytes <200 cells/mm^3	2.35 (1.87 to 2.94)	0.011	2.27 (1.73 to 2.79)	0.037
Current HIV RNA >100,000 copies/mL	1.12 (0.81 to 1.38)	0.418
Zenit HIV RNA >100,000 copies/mL	1.02 (0.71 to 1.68)	0.275
Chronic HCV infection	1.14 (0.81 to 1.56)	0.311
BMI >25 kg/m^2	3.45 (2.88 to 4.26)	<0.001	3.61 (2.79 to 4.17)	0.008
Smoking	1.42 (0.91 to 1.89)	0.069
Lipodystrophy	1.37 (0.92 to 1.78)	0.288
Duration of cART >3 years	3.56 (2.74 to 4.11)	<0.001	3.44 (2.62 to 3.89)	<0.001
Current use of:
Tenofovir	1.13 (0.83 to 1.49)	0.408
Abacavir	0.94 (0.53 to 1.34)	0.322
One NNRTI	0.86 (0.52 to 1.27)	0.165
Rilpivirine	0.73 (0.43 to 0.92)	0.084
Efavirenz	0.94 (0.77 to 1.42)	0.288
One PI	1.76 (1.21 to 2.23)	0.045	1.85 (1.38 to 2.49)	0.041
Darunavir/ritonavir	1.82 (1.17 to 2.39)	0.004	1.89 (1.27 to 2.55)	0.014
Atazanavir/ritonavir	1.56 (1.04 to 1.89)	0.041	1.61 (1.13 to 1.97)	0.039
One INSTI	0.61 (0.37 to 0.92)	0.040	0.55 (0.29 to 0.86)	0.037
Raltegravir	0.56 (0.31 to 0.87)	0.026	0.51 (0.22 to 0.78)	0.017
Elvitegravir/cobicistat	0.78 (0.44 to 1.09)	0.064
Dolutegravir	0.67 (0.42 to 0.98)	0.046	0.77 (0.49 to 1.07)	0.061

Note: OR, odds ratio; 95% CI, 95% confidence interval; AIDS, acquired immune deficiency syndrome; BMI, body mass index; cART, combination antiretroviral therapy; HCV, hepatitis C virus; NNRTI, non nucleoside reverse transcriptase inhibitor; PI, ritonavir-boosted protease inhibitor; INSTI, integrase strand transfer inhibitor.

By the multivariate analysis, there was no difference in the risk of MS by type of NNRTI-based or PI-based regimen: both rilpivirine and efavirenz were not significantly associated with the occurrence of MS, while both darunavir/ritonavir (OR 1.89, p = 0.014) and atazanavir/ritonavir (OR 1.61, p = 0.039) were significantly associated with an increased risk of this syndrome. Relative to the INSTI-based regimen, only raltegravir showed a significantly lower risk of MS (OR 0.51, p = 0.017), while the risk decrease associated with the exposure to elvitegravir/cobicistat or dolutegravir was not significant (Table 3).

Discussion

The real prevalence of MS among HIV-positive persons is still controversial owing to the very wide-range estimates in observational studies, so it is difficult to establish if the prevalence of MS in patients with HIV infection is different or comparable to that reported in the general population.

Even though two cohort studies reported similar MS prevalence rates among HIV-infected patients compared to HIV-negative subjects in the United States,^12,17 an increasing MS prevalence in HIV-positive individuals was reported in recent years together with the aging of the HIV-infected people. An increase in the MS prevalence from 28 to 33% was registered in a cohort of 100 HIV-positive males between 1998 and 2010, while the prevalence of lipodystrophy declined from 69 to 53% over the same period.¹⁸

The Women’s Interagency HIV Study assessed 2393 female patients (1725 HIV-positive and 668 HIV-negative) between 2000 and 2004 and showed a significantly higher prevalence of MS among the HIV-infected than HIV-uninfected women (33% vs. 22%; p < 0.0001). Predictors of MS among HIV-positive women were older age, higher BMI, current smoking, higher HIV RNA, and current use of stavudine, while current use of nevirapine was protective.¹⁹

The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) Study investigated changes in the MS prevalence rates in a cohort of 33,347 HIV-infected individuals under follow-up at 212 clinics in United States, Australia, and Europe at the end of six calendar periods, from 2000 to 2007. The prevalence of MS was evaluated using the NCEP definition modified to take account of the use of lipid-lowering and antihypertensive medications, measurement variability, and missing values. Using the primary definition, the MS prevalence increased from 19.4% in 2000/2001 to 41.6% in 2006/2007, while modification of the primary definition had a relatively little impact on the prevalence estimates. Among patients with MS, the majority of subjects met the elevated triglycerides, low HDL cholesterol, and hypertension criteria.¹¹

The MS prevalence varies very much with gender and ethnicity. Some observational studies report a significantly higher prevalence of MS in women than in men, and in white or Hispanic than in African-American Men.^19,20 Some recent reports show a very high rate of MS among African HIV-positive patients, ranging from 48 to 58%.^21,22

A systematic review and meta-analysis evaluated the global prevalence of MS in 65 studies across five continents comprising 55,094 HIV-infected patients aged 17–73 years. The overall prevalence of MS was 24.6% according to the ATPIII 2004–2005 criteria and 27.9% according to the ATPIII 2005 criteria. MS was significantly more frequent in women than in men, in cART users compared to cART-naive subjects, and in patients with longer duration of HIV infection.²³

The effect of cART initiation on the MS status was investigated using the ATPIII criteria in the AIDS Clinical Trials Group Longitudinal Linked Randomized Trials (ALLRT) cohort including 450 cART-naive patients. At the beginning of cART, MS was present in the 20% of patients, and after 96 weeks this prevalence declined to 12.6%. Therefore, 37% of patients with MS at cART initiation did not meet the MS criteria after 96 week of cART and this decrease was driven mostly by a significant elevation in the HDL cholesterol level. However, during the 96 weeks there was a significant increase in the proportion of patients with high triglycerides and with hyperglycemia, while the percentages of patients with large waist circumference and with high blood pressure did not change.²⁴ Immuno-virological parameters and antiretroviral regimen may also influence the risk of MS. After adjusting for demographics and BMI, the risk of MS in this cohort was decreased for CD4 + T-cell counts >50 cells/mm³, while the risk was increased for HIV RNA >400 copies/mL and use of a PI-based regimen.⁴

The beginning of cART might produce a significant improvement in metabolic profile of the naive patients, leading to a decreased MS prevalence, but specific antiretroviral agents can induce metabolic abnormalities associated with an increased risk of MS (such as dyslipidemia, insulin resistance, lipodystrophy with altered adipocytokine secretion, and endothelial dysfunction). Several observational retrospective and prospective studies have found an increased prevalence of MS in patients treated with thymidine analogs (mostly stavudine), older PIs (such as indinavir and lopinavir/ritonavir), and the NNRTI efavirenz.^25–29 On the other hand, patients exposed to the PI atazanavir or the NNRTI nevirapine were less likely to develop MS,^30,31 while there are no studies to date investigating the correlation between the exposure to INSTIs and the occurrence of MS.

The hypothesis that MS leads to an additional increase in the cardiovascular disease risk more than its single components among HIV-infected persons is debated still today.

Some studies have demonstrated a correlation between MS and surrogate markers of subclinical atherosclerosis, such as carotid intima-media thickness (IMT) and coronary artery calcium (CAC) score.^32,33

Authors of the D:A:D study investigated whether any particular combinations of the components of the MS are associated with an increased cardiovascular risk in 33,347 HIV-infected patients followed in a prospective observational analysis. Unadjusted, the presence of MS at study enrollment was associated with a 2.89 higher risk of cardiovascular disease, but after adjustment for each component, the MS was not a significant predictor of cardiovascular disease (OR 0.85, p = 0.32). Therefore, MS did not seem to increase the cardiovascular risk over that associated with its individual components.³⁴

Anyway, the MS includes a clustering of risk factors for cardiovascular diseases, so physicians should screen and periodically assess their patients for each physical and metabolic component of this syndrome in order to promptly reduce the global patient cardiovascular risk. Hypertension is becoming a prevalent feature of the MS among HIV-infected individuals because its prevalence is growing in an aging population. The prevalence of hypertension in HIV-infected people ranges from 4 to 54%, and the risk of hypertension usually increases with a longer duration of HIV infection and a longer exposure to cART.^35–37

The common pathogenetic mechanism leading to the aggregation of abnormalities constituting the MS has been identified with a chronic state of inflammation. In fact, chronic inflammation may lead to insulin resistance, dyslipidemia, hypertension, and visceral adiposity. Several factors in patients with HIV infection (including residual viral load, microbial translocation, concurrent infections, and drug-related toxicity) induce a chronic inflammation with persistent immune activation which may favor the occurrence of MS in these subjects.^38,39

Results of our study were partially in agreement with the literature data. The overall prevalence of MS in our HIV-infected population (including subjects naive to cART or receiving the first cART) ranged from 16 to 19% and its value was comparable by both the NCEP ATP III and the IDF criteria. Even though the IDF definition includes a lower threshold for waist measurements, a slightly lower percentage of patients have MS by the IDF criteria compared to those having MS by the NCEP ATP III criteria. This occurred probably because a greater proportion of subjects had MS following dyslipidemia criteria and not central adiposity criteria.

On the other hand, the prevalence of MS was significantly higher among patients on cART than among naive ones, by employing both diagnostic criteria, and this fact is in disagreement with the literature data showing in general a decrease in the MS prevalence after the beginning of cART. The higher frequency of MS among treated subjects seems to be driven mostly by a higher prevalence of visceral adiposity and dyslipidemia in these individuals. In particular in our cohort, in contrast to other studies, patients on cART had a significantly lower mean value of HDL cholesterol than naive ones.

The analysis of predictors significantly associated with the MS occurrence demonstrated some risk factors reported also by the literature, such as a longer duration of HIV infection, a nadir of CD4 + T lymphocytes <200 cells/mm³, a higher BMI, a longer exposure to cART, and a current use of a ritonavir-boosted PI-based regimen. However, this is the first study, to the best of our knowledge, showing the protective effect of a INSTI-based regimen, because the current use of one INSTI was associated with a significantly lower risk of MS. Particularly, current use of raltegravir significantly reduced the risk of MS, while the correlation between exposure to elvitegravir/cobicistat or dolutegravir and the MS occurrence was not significant, probably because of a too limited number of patients treated with the two newest INSTIs.

The overall prevalence of MS found in the present study is higher than that (9.1%) reported in our previous study⁴⁰ investigating the prevalence of MS and diabetes mellitus in a cohort of 755 adult patients with HIV infection in the year 2009. The first reasons leading to this difference was certainly the MS definition employed in the previous study, or rather the NCEP ATPIII definition not considering the use of specific treatments for dyslipidemia, hypertension, and hyperglycemia. The second reason was probably the significantly younger mean age of the enrolled patients in the previous study compared to the current one.

The present study shows several evident limitations, and the first is the cross-sectional design and the lack of longitudinal follow-up data which could allow some inferences about causality. Second, the lack of data about nutrition/caloric update and exercise which could affect the MS status of the enrolled patients. Third, the limited number of subjects treated with some antiretroviral agents (such as etravirine, nevirapine, elvitegravir/cobicistat, and dolutegravir), leading to an insufficient statistical power to find a possible correlation with the MS.

Finally, further prospective studies are certainly needed in order to exactly define prevalence and predictive factors of MS among HIV-positive patients and to determine whether the presence of MS in HIV-positive people produces a multiplicative increase in cardiovascular risk above and beyond the additive risks of its individual components.

Conflicts of interest

Leonardo Calza has received research grants from ViiV and Gilead, and has received honoraria for consulting (as advisory board member) and speaking from Janssen, Merck and Bristol-Myers Squibb. Other Authors have no conflicts of interest to declare.

Notes on contributors

Leonardo Calza is a clinician and research scientist at the Clinics of Infectious Diseases, S.Orsola-Malpighi Hospital in Bologna, and associated professor in Infectious Diseases at the University of Bologna. His research interests include mainly HIV infection, antiretroviral therapy and HIV-associated comorbidities. Vincenzo Colangeli, Eleonora Magistrelli, Nicolò Rossi, Elena Rosselli del Turco, Linda Bussini and Marco Borderi are clinicians at the Clinics of Infectious Diseases, S.Orsola-Mapighi Hospital in Bologna. Pierluigi Viale is the Chief of the Clinics of Infectious Diseases, S.Orsola-Mapighi Hospital in Bologna, and ordinary professor in Infectious Diseases at the University of Bologna.