Prevalence and associated factors of metabolic syndrome among people live with HIV in a medical center of Northern Taiwan -- a cross sectional study

doi:10.21203/rs.3.rs-4817270/v1

Download PDF

Article

Prevalence and associated factors of metabolic syndrome among people live with HIV in a medical center of Northern Taiwan -- a cross sectional study

https://doi.org/10.21203/rs.3.rs-4817270/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

HIV acquisition has been found to be associated with increased risk of cardiometabolic syndrome in the past. In this study, we attempt to delineate the risk factors of metabolic syndrome among persons living with HIV in a single medical center in Taiwan using mostly integrase inhibitors-based regimen as HAART.

Methods

A cross-sectional study involving all people live with HIV (PLWH) aged ≥ 18 years who visited MacKay Memorial Hospital, Taipei, Taiwan between September 7, 2022 to January 31, 2023 was included. A modified National Cholesterol Education Program ATP III definition of metabolic syndrome was used to define metabolic syndrome and body mass index was used as surrogate for waist circumference.

Results

809 PLWH participated in the study, in which 81.3% of patient were on integrase inhibitor-based HAART regimen. The prevalence of metabolic syndrome was 10.0%, which is markedly lower than those reported in the literature. Age was a significant risk factor for metabolic syndrome. HIV-related factors (CD4, viral load, HAART regimen) did not affect the presence of metabolic syndrome.

Conclusions

Prevalence rate of metabolic syndrome appeared to have declined among PLWH compared to previous reports, possibly secondary to widespread use of integrase inhibitors. Age remains a strong risk factor for the development of metabolic syndrome among PLWH.

Biological sciences/Microbiology

Health sciences/Endocrinology

metabolic syndrome

HIV

Taiwan

HAART

integrase inhibitor

According to WHO, there are 39 million people living with HIV globally in 2022.¹ While once thought of as a deadly disease, HIV is now considered a manageable disease with the advent of highly-active antiretroviral therapy (HAART) in 1995. Now, a growing number of people are aging with HIV. As people age, chronic diseases such as type 2 diabetes mellitus (DM), hypertension (HTN), and other cardio-metabolic diseases may develop,² similar to the general population. People live with HIV are also uniquely susceptible to disease-related vulnerabilities, such as inflammation caused by carrying HIV infection,³ impaired immune system,⁴ or toxicities due to long-term antiviral medication use.⁴

Metabolic syndrome (MetS) is a disorder consisting of obesity, dyslipidemia, hypertension, and insulin resistance. HIV infection has been found to be associated with increased risk of cardiometabolic syndrome in the past.³ Recent literature found that age,^5–11 body mass index,^5,6 gender,^6,8,9 socioeconomic status (as evidenced by education,⁷ wealth,^8,9 occupation,⁷ residence⁹), physical activity,^7,8 use of protease inhibitors,^2,12–17 and CD4 level^10,11,17 were risk factors for metabolic syndrome among PLWH. Inflammation (as evidenced by hs-CRP level) was occasionally found to be related to metabolic syndrome,¹⁸ and smoking may play a role.^19,20 Most of these studies were done prior to the widespread use of single tablet integrase inhibitors (II) as first line therapy. Besides, there is a paucity of studies to study metabolic syndrome among people live with HIV in Taiwan. In this study, we attempt to delineate the associated factors of metabolic syndrome in a predominant integrase inhibitor era in a Taiwanese population.

Patient Data Collection

This is a cross-sectional study involving all people live with HIV (PLWH) aged ≥ 18 years who visited MacKay Memorial Hospital between September 7, 2022 to January 31, 2023. Data were collected in this time interval as most of our patients came back for follow-up every 3 months. For any patients who visited more than one time, only data from the first visit was included. Data collected included smoking, sex, type of HAART used, anti-hepatitis B surface antibody (Anti-HBs), anti-hepatitis B core antigen (Anti-HBc), anti-hepatitis B surface antigen (HBsAg), anti-hepatitis A antibody (Ant-HAV), anti-hepatitis C antibody (HCV), patient’s viral load (VL), rapid plasma reagin (RPR), age, CD4 count, fasting glucose, serum aspartate transaminase (AST), serum alanine transaminase (ALT), serum total cholesterol, triglycerides (TG), creatinine, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), glycosylated hemoglobin (HbA1c), presence of proteinuria on urinalysis, urine protein to creatinine ratio (UPCR), years since diagnosis, and whether they are on treatment for diabetes mellitus (DM), hyperlipidemia, or hypertension (HTN). Patients with no information on body weight, blood pressure, smoking status, either fasting glucose or HbA1c, or lipid profile were excluded from the study.

We also calculated Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) score,²¹ atherosclerotic cardiovascular disease (ASCVD) risk score,²² and Framingham risk score²³ in these patients.

For this study, we used a modified National Cholesterol Education Program ATP III definition of metabolic syndrome to define metabolic syndrome. As waist circumference were not routinely collected during office visits, we used body mass index (BMI) as a surrogate for waist circumference, similar to another study in the past.¹² Metabolic syndrome was defined as the presence of three or more of the following five manifestations: high serum triglycerides, as defined by a triglyceride level of greater or equal to 150 mg/dL; low HDL-C, as defined by a serum HDL-C < 40 mg/dL in man or < 50 mg/dL in woman; hypertension, as evidenced by a clinical diagnosis or in-office blood pressure of a systolic blood pressure ≥ 140 mmHg, or diastolic pressure of ≥ 90 mmHg; obesity, as defined by BMI ≥ 30; and diabetes mellitus, as evidenced by a clinical diagnosis, or laboratory values showing fasting glucose ≥ 126 mg/dL, or HbA1c > 6.5%. Individuals would also be diagnosed with hypertriglyceridemia, HDL-C hypocholesterolemia, hypertension, or diabetes mellitus if they were receiving specific treatments for these conditions. This study complies with the Declaration of Helsinki and was approved by the MacKay Memorial Institutional Review Board, which waived the need to obtain informed consent from patients. The protocol number from the MacKay Memorial Hospital Institutional Review Board was 23MMHIS280e

Analysis

Continuous variable of age was grouped into five groups (age 18–30, 31–40, 41–50, 51–60, and 61+). CD4 count was grouped into five groups (CD4 1–50, 51–100, 101–200, 201–400, and more than 400 cell/µL). Serum HIV viral load was grouped into ≤ 200 and > 200 copies/mL. Smoking variable was a categorical variable with either current smoker, never smoked, or past smoker. Syphilis was categorized as a binary variable of serum RPR positivity or negativity. HAART use was divided into three categories based on the medication's third agent. The three categories were: 1) regimens that used integrase strand inhibitors (II) (including ABC/3TC/DTG, 3TC/DTG, DTG/RPV, BIC/FTC/TAF, EVG/COBI/FTC/TAF), 2) regimens that included protease inhibitors (PI) (DRV/COBI), and 3) regimens that included non-nucleoside reverse transcriptase inhibitors (NNRTI) (including EFV/TDF/FTC, TDF/3TC/DOR, RPV/TAF/FTC).

Cross tabulation, Chi-squared tests and univariate logistic regression was used to investigate the relationship between categorical variables and metabolic syndrome. Student’s t-test was used for continuous variables. For the factors that were significant in univariate analysis that were considered a possible risk factor, Chi-squared tests were performed to see whether the possible risk factor was associated with any of the independent variables. Those variables that showed evidence of association with both outcome (metabolic syndrome) and the exposure factor were considered to be potential confounders if they were not on the causal pathway. All variables were assessed as potential confounders or effect modifiers by using the Test for Homogeneity. Finally, a final multivariate logistic regression was performed with variables which confounded or modified the association between possible associated factors and metabolic syndrome. All tests were 2-tailed, and p-values of less than 0.05 was considered significant. The statistical analyses were performed either using SPSS 21.0 (SPSS Inc., Chicago, IL, USA) or STATA 16 (StataCorp LLC., College Station, TX, USA).

A total of 1130 PLWH were followed in our hospital. After excluding 321 people who lacked data in either height, body weight, smoking, blood pressure, fasting glucose, HbA1c or lipid profile, a total of 809 were included into the study (Fig. 1). We compared gender and age between subjects that were included or excluded, and no significant difference was noted. (Table 1) In the study group, 97.7% were men, while those excluded had 96.0% men (p = 0.162). Mean age of study group was 39.44 while in those excluded, the mean age was 39.5 (p = 0.940).

Table 1

Characteristics of study group and excluded group
	Total	In study	Excluded
	n = 1130	n = 809	n = 321	p value
Male, n (%)	1098 (97.2)	790 (97.7)	308 (96.0)	0.162
Mean age (Year)	39.46	39.44	39.5	0.940

In our study, 81 people had been diagnosed with metabolic syndrome, resulting in a prevalence rate of 10.0%. The most common HAART regimen was II-based (81.3%), which included BIC/FTC/TAF, 3TC/DTG, EVG/COBI/FTC/TAF, DTG/RPV, and ABC/3TC/DTG. 97.9% of our patients had a viral load below 200 copies/ml, and 75.1% had a CD4 count above 400 cells/µl. There were eleven treatment naïve subjects and two of them was diagnosed with metabolic syndrome. (Table 2)

Table 2

Epidemiological and clinical characteristic of PLWH and odds ratio of metabolic syndrome
Characteristic	All	MetS	No MetS	p value	odds ratio (confidence interval)
Characteristic	n = 809	n = 88	n = 721	p value	odds ratio (confidence interval)
Sex, n (%)
male	790	80 (98.8)	710 (97.5)	0.711	1
female	19	1 (1.2)	18 (2.5)	0.711	0.49 (0.07–3.74)
Smoking, n (%)
non-smoker	538	58 (71.6)	480 (66.8)	0.737	1
smoker	250	22 (27.2)	228 (31.7)		0.80 (0.48–1.34)
quit	12	1 (1.2)	11 (1.5)		0.75 (0.01–5.93)
Age, n (%)
18–30	162	7 (8.6)	155 (21.3)	< 0.001	1
31–40	328	28 (34.6)	300 (41.2)		2.07 (0.88–4.84)
41–50	188	22 (27.2)	166 (22.8)		2.94 (1.22–7.06)
51–60	105	19 (23.5)	86 (11.8)		4.89 (1.98–12.10)
61+	26	5 (6.2)	21 (2.9)		5.27 (1.53–18.12)
HAART group, n (%)
II	658	66 (81.5)	592 (81.3)	0.240	1
PI	7	2 (2.5)	5 (0.7)		3.59 (0.68–18.86)
NNRTI	144	13 (16.0)	131 (18.0)		0.89 (0.48–1.66)
VL group (copies/ml), n (%)
< 200	777	78 (97.5)	699 (97.9)	0.686	1
>=200	17	2 (2.5)	15 (2.1)	0.686	1.20 (0.27–5.32)
CD4 level (cells/µl), n (%)
1–50	4	0 (0.0)	4 (0.6)	0.620	1
51–100	7	1 (1.3)	6 (0.8)		N/A
101–200	26	2 (2.5)	24 (3.4)		N/A
201–400	151	11 (13.8)	140 (19.6)		N/A
400+	607	66 (82.5)	541 (75.7)		N/A
Treatment naïve	11	2 (18.2)	9 (81.8)	0.34	0.54 (0.12–2.56)
RPR positive, n (%)	281	21 (26.3)	260 (36.5)	0.84	0.62 (0.37–1.04)
Anti-HAV positive, n (%)	476	48 (62.3)	428 (61.8)	> 0.999	1.03 (0.63–1.67)
Anti-HBs positive, n (%)	460	44 (56.4)	416 (59.6)	0.628	0.88 (0.55–1.41)
Anti-HBc positive, n (%)	195	26 (49.1)	169 (34.8)	0.05	1.81 (1.02–3.19)
HbsAg positive, n (%)	59	9 (11.8)	50 (7.2)	0.168	1.74 (0.82–3.70)
Anti-HCV positive, n (%)	118	14 (17.7)	104 (14.9)	0.509	1.23 (0.67–2.27)
HTN, n (%)	223	68 (84.0)	155 (21.3)	< 0.001	19.34 (10.41–35.92)
DM, n (%)	60	31 (38.3)	29 (4.0)	< 0.001	14.94 (8.35–26.74)
Hypertriglyceridemia, n (%)	292	66 (81.5)	164 (22.5)	< 0.001	15.13 (8.41–27.22)
HDL-C hypocholesterolemia, n (%)	236	69 (85.2)	189 (26.0)	< 0.001	16.40 (8.69–30.99)
Obesity, n (%)	60	34 (42.0)	16 (2.2)	< 0.001	32.19 (16.58–62.50)
Age, mean (SD)		43.4 (10.7)	39 (10.3)	< 0.001	1.04 (1.02–1.06)
Duration of treatment (years), mean (SD)		9.1 (5.8)	7.9 (5.4)	0.075	1.04 (1.00-1.08)
Biochemical data, mean (SD)
Fasting Glucose (mg/dL)		125.1 (59.1)	95.1 (15.1)	< 0.001	1.04 (1.02–1.05)
AST (IU/L)		29.8 (16.9)	25.2 (26.9)	0.139	1.00 (0.99–1.01)
ALT (IU/L)		43.0 (32.5)	28.2 (37.8)	< 0.001	1.01 (1.00-1.02)
Total cholesterol(mg/dL)		189.9 (37.5)	25.2 (34.8)	0.992	1.00 (0.99–1.01)
Triglyceride (mg/dL)		262.7 (238.6)	121.6 (111.4)	< 0.001	1.01 (1.00-1.01)
Creatinine (mg/dL)		1.1 (0.3)	1.1 (0.6)	0.93	0.98 (0.65–1.49)
LDL-C (mg/dL)		112.6 (33.5)	117.8 (29.8)	0.154	0.99 (0.99-1.00)
HDL-C (mg/dL)		37.2 (5.7)	49.1 (10.9)	< 0.001	0.86 (0.83–0.89)
HbA1c (%)		6.4 (1.7)	5.4 (0.5)	< 0.001	3.07 (2.23–4.22)
Proteinuria (UA)		0.5 (0.8)	0.1 (0.4)	< 0.001	2.92 (2.00-4.28)
Protein/Cr Ratio		10.1 (78.3)	0.2 (1.1)	0.318	1.04 (0.86–1.27)
hs-CRP (mg/dL)		0.3 (0.4)	0.2 (0.5)	0.115	1.35 (0.91–1.99)
HOMA-IR index		5.7 (6.8)	2 (3.1)	< 0.001	1.19 (1.12–1.27)
ASCVD Risk Score		8.7 (9.7)	3.2 (4.9)	< 0.001	1.11 (1.07–1.15)
Framingham Risk Score		6.6 (7.0)	3.5 (4.7)	< 0.001	1.09 (1.05–1.14)
HAART, highly active antiretroviral therapy; II, intergrase inhibitor based regimen; PI, protease inhibitors based regimen; NNRTI, non-nucleoside reverse transcriptase inhibitor; VL, viral load; HTN, hypertension; DM, diabetes mellitus; HDL-C, high-density lipoprotein cholesterol; MetS, metabolic syndrome

Comparing the characteristics between the group with and without metabolic syndrome, there were no significant differences in the distribution of sex (p = 0.711), smoking (p = 0.737), HAART regimen (p = 0.240), VL (p = 0.686), CD4 level (p = 0.620), treatment experience (p = 0.34), RPR (p = 0.84), Anti-HAV (p > 0.999), Anti-HBs (p = 0.628), HBsAg (p = 0.168), Anti-HCV (p = 0.509), or duration of treatment (p = 0.075). MetS group had higher prevalence of Anti-HBc positivity (p = 0.05, MetS: 49.1%, No MetS: 34.8%), HTN (p < 0.001, MetS: 84.0%, No MetS: 21.3%), DM (p < 0.001, MetS: 38.3, No MetS: 4.0%), hypertriglyceridemia (p < 0.001, MetS: 81.5, No MetS: 22.5%), HDL-C hypocholesterolemia (p < 0.001, MetS: 85.2%, No MetS: 26.0%), and obesity (p < 0.001, MetS: 42.0%, No MetS: 2.2%). Those in MetS group were also older (p < 0.001, MetS: 43.4, No MetS: 39). In biochemical data, level of AST (p = 0.139), total cholesterol (p = 0.992), creatinine (p = 0.930), LDL-C (p = 0.154), UPCR (p = 0.318), and hs-CRP (p = 0.115) were not significantly different. (Table 2) Those in MetS group had higher fasting glucose (mg/dL) (p < 0.001, MetS: 125.1, No MetS: 95.1), ALT (IU/L) (p < 0.001, MetS: 43.0, No MetS: 28.2), HbA1c (%) (p < 0.001, MetS: 6.4, No MetS: 5.4), proteinuria (p < 0.001, MetS: 0.5, No MetS: 0.1) but lower HDL-C (mg/dL) (MetS: 37.2, No MetS: 49.1). Those in MetS group also had higher scores in HOMA-IR (p < 0.001, MetS: 5.7, No MetS: 2.0), ASCVD risk (p < 0.001, MetS: 8.7, No MetS: 3.2), and Framingham risk (p < 0.001, MetS: 6.6, No MetS: 3.5). (Table 2)

Initially, we divided the HAART regimen into three groups according to third agents and no statistical significance was noted (p = 0.240). (Table 2) To further explore this relationship, we regrouped HAART regimen into six smaller groups, as detailed in Table 3, and none of them showed statistical significance compared with BIC/FTC/TAF (p = 0.403, 0.195, 0.286, 0.219, 0.541, respectively).

Table 3

Odds ratio of detailed HAART regimen
	MetS	No MetS	p value	odds ratio (confidence interval)
BIC/FTC/TAF	35 (11.67%)	265 (88.33%)	reference
TDF/FTC/EFV TDF/3TC/DOR TAF/FTC/RPV	13 (9.03%)	131 (90.97%)	0.403	0.751 (0.384–1.469)
Combivir, Prezcobix	2 (28.57%)	5 (71.43%)	0.195	3.029 (0.566–16.206)
ABC/3TC/DTG	16 (8.60%)	170 (91.40%)	0.286	0.713 (0.383–1.327)
EVG/COBI/FTC/TAF	1 (3.57%)	27 (96.43%)	0.219	0.280 (0.037–2.128)
DTG/RPV 3TC/DTG	14 (9.72%)	130 (90.28%)	0.541	0.815 (0.424–1.569)

In the group of with metabolic syndromes, the most prevalent component among the five diagnostic criteria of metabolic syndrome was HDL-C hypocholesterolemia, which was found in 69 (85.2%) people of the group. The next common condition was hypertension (84.0%), followed by hypertriglyceridemia (81.5%). Less than half of this group had obesity or DM (42.0% and 38.3% respectively). (Table 2) Among the components of metabolic syndrome, obesity had the strongest association with metabolic syndrome (OR: 32.19, CI: 16.58–62.50), followed by hypertension (OR: 19.33, CI: 10.41–35.92), HDL-C hypocholesterolemia (OR: 16.40, CI 8.69–30.99) and hypertriglyceridemia (OR: 15.13, CI 8.41–27.22). Although DM was the most weakly associated with metabolic syndrome, having DM was still linked with an over tenfold risk of developing metabolic syndrome (OR = 14.94; CI: 8.35–26.74).

Among factors that showed significant association with metabolic syndrome, some were related to the diagnostic criteria of metabolic syndrome and thus association was expected. Factors which were unrelated to the diagnostic criteria and yet showed evidence of strong association with metabolic syndrome in univariate analysis included age, Anti-HBc, ALT, and proteinuria. We hypothesized that high ALT level was caused by fatty liver, which is a complication of metabolic syndrome, and that proteinuria is a complication of DM nephropathy. Therefore, we excluded these variables when during further analyses.

After performing tests of homogeneity, it was found that fitting the other independent variables (sex, smoking, HAART group, viral load group, CD4 level, RPR, Anti-HAV, Anti-HBs, HBsAg, Anti-HCV, duration of treatment) into the multivariate logistic model either did not change the odds ratio significantly or failed due to issues of collinearity. Therefore, only the variables Anti-HBc and age group were fitted into the final model (Table 4). In the final multivariate logistic regression, age was still a significant risk factor in metabolic syndrome but Anti-HBc showed no more association with metabolic syndrome (p value = 0.796, OR: 1.10, CI: 0.54–2.21) (Table 4). There was a clear trend of having higher odds of having metabolic syndrome as individuals age, with significantly higher risk of metabolic syndrome after age 50 (Table 4). The age groups 51–60 years and > 61 years showed significant association with metabolic syndrome (p = 0.028, p = 0.015 respectively). Those older than 61 years old were nearly twice as likely to have metabolic syndrome than those 51–60 years old (61+: OR = 6.49, CI: 1.44–29.30; 51–60: OR = 3.60, CI: 1.15–11.22).

Table 4

Odds ratio of age and Anti-HBc after using multivariate logistic regression
	p value	Odds ratio
Age
18–30	0.105	1
31–40	0.280	1.69 (0.65–4.40)
41–50	0.107	2.42 (0.83–7.08)
51–60	0.028	3.60 (1.15–11.22)
61+	0.015	6.49 (1.44–29.30)
Anti-HBc	0.796	1.10 (0.54–2.21)

The global prevalence of metabolic syndrome in PLWH is approximately 16.7%.¹⁹ Many studies have showed that the prevalence rates of metabolic syndrome in PLWH varied from 19.2–40.1%.^{5,8,11,12,18,24–27} These variations could be attributed to differences in regional diets, genetic predispositions, socioeconomic factors, and local healthcare practices.

In our study, the prevalence of metabolic syndrome in PLWH was 10.0%, which was lower than the global average and other documented prevalences. In similar studies in Asia, the prevalence rate was 23.6% in Singapore¹² and 40.1% in India.⁵ A previous study in Taiwan in 2012 reported a prevalence of metabolic syndrome in PLWH of 26.2%.²⁷ In the study in Singapore by Ang et al., researchers analyzed 2231 treatment-experienced PLWH. Most of them were men (93.9%). All of them had been exposed to NRTI as the first line of treatment, 93.9% to NNRTI, 28.6% to protease inhibitors and 12.8% to integrase strand transfer inhibitors. ¹² In the study by Mally et et al., 182 PLWH were included, both treatment-experienced and treatment-naïve. Most of them were using NRTI and NNRTI.⁵ In the study by Wu et al., 877 PLWH were included. 81.7% of them were using NRTI as HAART but no one used II for treatment.²⁷ We suspected that the marked discrepancy between our number and those documented previously could be, at least in part, attributed to the evolution in the choice of HAART regimens. In the past, we often used protease inhibitors to suppress HIV. PIs had been reported to influence lipid and glucose metabolism, which may lead to metabolic syndrome.^15,16,28 On the other hand, use of integrase inhibitors was associated with a lower risk of metabolic syndrome.² As the Guidelines for diagnosis and treatment of HIV/AIDS, published by Taiwan AIDS society, suggested the usage of II-based regimen as first-line HAART since 2016,²⁹ most of our patients were using II-based regimen (81.3%) instead of PI-based regimen (0.9%). The fact that our prevalence rate was lower than previously reported could be the result of a fall in the prevalence of metabolic syndrome after the adoption of II-based regimen, despite the weight gain implications of these medications.^30,31 This finding is comparable with previous finding from Taramasso et al.³²

In previous research, the duration of an HIV diagnosis, the severity of the infection, and the length of HAART usage have been identified as risk factors for metabolic syndrome.^2,17,18,27 Associations have been drawn between an increased risk of metabolic syndrome and virological failure (viral load > 1000 copies/ml)¹⁸, low CD4 counts¹¹, and long-term HAART use.²⁷ Contrary to these findings, our study did not find a strong correlation between CD4 levels or viral load and metabolic syndrome. Additionally, we observed no correlation between the duration of HIV diagnosis (as a proxy for HAART duration) and metabolic syndrome. Using protease inhibitors as part of HAART has been reported as a risk factor of metabolic syndrome in previous study.^2,4,15 In our study, the use of protease inhibitors was not significantly associated with metabolic syndrome (OR: 3.59, CI: 0.68–18.86), although it should be noted that our confidence intervals were rather wide. The wide confidence interval may be attributed to a small subgroup sample size, which is likely why in our study the use of protease inhibitors did not show significant association with metabolic syndrome. Thus, although we did not find statistically significant association between protease inhibitors and metabolic syndrome, our study is still in line with previous research.

In our study, age remains one of the factors associated with metabolic syndrome.^33–35 It had a strong positive correlation with HTN (p < 0.001), DM (p < 0.001), hypertriglyceridemia (p < 0.001), and HDL-C hypocholesterolemia (0.003). (Table 5) Consequently, the likelihood of developing metabolic syndrome increases as individuals age.

Table 5

Prevalence of HTN, DM, hypertriglyceridemia, HDL-C hypocholesterolemia, obesity and metabolic syndrome in different age group
Age	18–30 (n, %)	31–40 (n, %)	41–50 (n, %)	51–60 (n, %)	60+ (n, %)	p value
HTN	33 (20.2)	75 (22.7)	58 (30.4)	41 (38.7)	17 (63.0)	< 0.001
DM	6 (3.7)	13 (3.9)	15 (7.9)	24 (22.6)	3 (11.1)	< 0.001
Hypertriglyceridemia	30 (18.4)	110 (33.2)	91 (47.6)	51 (48.1)	12 (44.4)	< 0.001
HDL-C hypocholesterolemia	40 (24.5)	80 (24.2)	70 (36.6)	41 (38.7)	10 (37.0)	0.003
Obesity	6 (3.7)	21 (6.4)	16 (8.5)	5 (4.8)	2 (7.7)	0.383
MetS	8 (4.9)	28 (8.5)	26 (13.8)	21 (20.0)	5 (19.2)	< 0.001
HTN, hypertension; DM, diabetes mellitus; HDL-C, high-density lipoprotein cholesterol; MetS, metabolic syndrome
PLWH, people live with HIV; MetS, metabolic syndrome

Our study had a few limitations. First, our study is a single-institution study. The results may not accurately represent the prevalence of metabolic syndrome among the PLWH in Taiwan. Second, as this is a cross sectional analysis, causality could not be inferred. Third, we cannot fully evaluate the implication of prior HAART use, as our patients often relocate or change institutions of care. However, prior to the adoption of IIs as first-line treatment, most of our patients were on an NNRTI-based regimen as first-line or on PIs if they had virological resistance. It can be seen from the study that treatment duration was not associated with metabolic syndrome, hence the use of prior types of HAART regimen likely did not affect the outcome of the study. Finally, there could be residual confounding that are not accounted for within the scope of our study.

In conclusion, our findings indicate a lower prevalence rate compared to global studies, which may be related to broad adoption of II-based HAART. PI-based HAART appeared to be a risk factor, but it did not reach statistical significance in this study. Age was strongly associated with metabolic syndrome. It was also associated with other factors (HTN, DM, hypertriglyceridemia, HDL-C hypocholesterolemia) in our study. The insights from our research can guide both clinicians and patients to focus on managing traditional risk factors, including hypertension, diabetes mellitus, dyslipidemia, and obesity, to prevent metabolic syndrome and its associated complications.

Author contributions statement

Chien An Tu and Alice Ying Jung Wu conceived and designed the research. All authors provided the data and Chien An Tu, Alice Ying Jung Wu, and Fang Ju Sun analyzed it. Alice Ying Jung Wu took the lead in writing the manuscript. All authors provided critical feedback and helped shpae the research, analysis and manuscript.

Author Contribution

C.T. and A.W. conceived and designed the research. All authors provided the data. C.T., A.W., and F.S. analyzed the data. C.T. and A.W. wrote and edited the manuscript. All authors provided critical feedback and helped shape the research, analysis and manuscript. All authors reviewed the final manuscript.

Data Availability

Dataset is available upon request from the corresponding author.

WHO. HIV statistics, globally and by WHO region, 2023, <https://www.who.int/teams/global-hiv-hepatitis-and-stis-programmes/hiv/strategic-information/hiv-data-and-statistics> (2023).
Calza, L. et al. Prevalence of metabolic syndrome in HIV-infected patients naive to antiretroviral therapy or receiving a first-line treatment. HIV Clin Trials 18, 110-117 (2017). https://doi.org/10.1080/15284336.2017.1311502
Maloberti, A. et al. Metabolic syndrome in human immunodeficiency virus-positive subjects: prevalence, phenotype, and related alterations in arterial structure and function. Metab Syndr Relat Disord 11, 403-411 (2013). https://doi.org/10.1089/met.2013.0008
Alvarez, C. et al. Metabolic syndrome in HIV-infected patients receiving antiretroviral therapy in Latin America. Braz J Infect Dis 14, 256-263 (2010).
Mallya, S. D., Reddy, T. S., Kamath, A., Pandey, A. K. & Saravu, K. Determinants of Metabolic Syndrome and 5-Year Cardiovascular Risk Estimates among HIV-Positive Individuals from an Indian Tertiary Care Hospital. AIDS Res Treat 2020, 5019025 (2020). https://doi.org/10.1155/2020/5019025
Mondy, K. et al. Metabolic syndrome in HIV-infected patients from an urban, midwestern US outpatient population. Clin Infect Dis 44, 726-734 (2007). https://doi.org/10.1086/511679
Bune, G. T., Yalew, A. W. & Kumie, A. Predictors of metabolic syndrome among people living with HIV in Gedeo-Zone, Southern-Ethiopia: A case-control study. (2020).
Chihota, B. V. et al. Metabolic syndrome among treatment-naive people living with and without HIV in Zambia and Zimbabwe: a cross-sectional analysis. J Int AIDS Soc 25, e26047 (2022). https://doi.org/10.1002/jia2.26047
Gebrie, A. The burden of metabolic syndrome in patients living with HIV/AIDS receiving care at referral hospitals of Northwest Ethiopia: A hospital-based cross-sectional study, 2019. 1, 1551-1556 (2020).
Shi, R. et al. Association of HIV infection with metabolic syndrome among normal or underweight young adults: evidence from the CHART cohort. 1, 450-456 (2021).
Ortiz, D. W. et al. Metabolic syndrome in people with HIV from Guatemala: analysis of components and risk factors. Int J STD AIDS 33, 987-994 (2022). https://doi.org/10.1177/09564624221119321
Ang, L. W., Ng, O. T., Boudville, I. C., Leo, Y. S. & Wong, C. S. An observational study of the prevalence of metabolic syndrome in treatment-experienced people living with HIV in Singapore. PLoS One 16, e0252320 (2021). https://doi.org/10.1371/journal.pone.0252320
Wu, P. Y. et al. Metabolic syndrome among HIV-infected Taiwanese patients in the era of highly active antiretroviral therapy: prevalence and associated factors. J Antimicrob Chemother 67, 1001-1009 (2012). https://doi.org/10.1093/jac/dkr558
Alencastro, P. R. et al. Metabolic syndrome and population attributable risk among HIV/AIDS patients: Comparison between NCEP-ATPIII, IDF and AHA/NHLBI definitions. (2012).
Duro, M. et al. Metabolic syndrome in human immunodeficiency virus-infected patients. Int J STD AIDS 29, 1089-1097 (2018). https://doi.org/10.1177/0956462418775188
Guira, O. et al. Features of Metabolic Syndrome and Its Associated Factors during Highly Active Antiretroviral Therapy in Ouagadougou (Burkina Faso). J Int Assoc Provid AIDS Care 15, 159-163 (2016). https://doi.org/10.1177/2325957415601503
Krishnan, S. et al. Metabolic syndrome before and after initiation of antiretroviral therapy in treatment-naive HIV-infected individuals. 1, 381-389 (2012).
Hamooya, B. M. et al. Metabolic syndrome in Zambian adults with human immunodeficiency virus on antiretroviral therapy: Prevalence and associated factors. Medicine (Baltimore) 100, e25236 (2021). https://doi.org/10.1097/MD.0000000000025236
Nguyen, K. A., Peer, N., Mills, E. J. & Kengne, A. P. A Meta-Analysis of the Metabolic Syndrome Prevalence in the Global HIV-Infected Population. PLoS One 11, e0150970 (2016). https://doi.org/10.1371/journal.pone.0150970
Sobieszczyk, M. E. et al. Prevalence and predictors of metabolic syndrome among HIV-infected and HIV-uninfected women in the Women's Interagency HIV Study. 1, 272-280 (2008).
Matthews, D. R. et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 28, 412-419 (1985). https://doi.org/10.1007/BF00280883
Goff, D. C., Jr. et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 129, S49-73 (2014). https://doi.org/10.1161/01.cir.0000437741.48606.98
Wilson, P. W. et al. Prediction of coronary heart disease using risk factor categories. Circulation 97, 1837-1847 (1998). https://doi.org/10.1161/01.cir.97.18.1837
Kiama, C. N. et al. Prevalence and factors associated with metabolic syndrome in an urban population of adults living with HIV in Nairobi, Kenya. Pan Afr Med J 29, 90 (2018). https://doi.org/10.11604/pamj.2018.29.90.13328
Sears, S. et al. Metabolic Syndrome Among People Living with HIV Receiving Medical Care in Southern United States: Prevalence and Risk Factors. AIDS Behav 23, 2916-2925 (2019). https://doi.org/10.1007/s10461-019-02487-8
Akl, L. D. et al. Metabolic syndrome in HIV-infected middle-aged women on antiretroviral therapy: prevalence and associated factors. Braz J Infect Dis 21, 263-269 (2017). https://doi.org/10.1016/j.bjid.2017.02.003
Wu, P. Y. et al. Metabolic syndrome among HIV-infected Taiwanese patients in the era of highly active antiretroviral therapy: prevalence and associated factors. 1, 1001-1009 (2012).
Aberg, J. A. et al. Metabolic effects of darunavir/ritonavir versus atazanavir/ritonavir in treatment-naive, HIV type 1-infected subjects over 48 weeks. AIDS Res Hum Retroviruses 28, 1184-1195 (2012). https://doi.org/10.1089/aid.2011.0327
Society., T. A. Guidelines for diagnosis and treatment of HIV/AIDS, 6th edition., <http://www.aids-care.org.tw/journal/treatment.php> (2020).
Calmy, A. et al. Dolutegravir-based and low-dose efavirenz-based regimen for the initial treatment of HIV-1 infection (NAMSAL): week 96 results from a two-group, multicentre, randomised, open label, phase 3 non-inferiority trial in Cameroon. Lancet HIV 7, e677-e687 (2020). https://doi.org/10.1016/S2352-3018(20)30238-1
Venter, W. D. F. et al. Dolutegravir plus Two Different Prodrugs of Tenofovir to Treat HIV. N Engl J Med 381, 803-815 (2019). https://doi.org/10.1056/NEJMoa1902824
Taramasso, L. et al. Metabolic syndrome and body weight in people living with HIV infection: analysis of differences observed in three different cohort studies over a decade. HIV Med 23, 70-79 (2022). https://doi.org/10.1111/hiv.13165
Aouam, A. et al. Metabolic syndrome among people with HIV in central Tunisia: Prevalence and associated factors. (2021).
Ayodele, O. E. et al. Prevalence and clinical correlates of metabolic syndrome in Nigerians living with human immunodeficiency virus/acquired immunodeficiency syndrome. Metab Syndr Relat Disord 10, 373-379 (2012). https://doi.org/10.1089/met.2012.0050
Jantarapakde, J. et al. Prevalence of metabolic syndrome among antiretroviral-naive and antiretroviral-experienced HIV-1 infected Thai adults. 1, 331-340 (2014).

No competing interests reported.

Download PDF

Editorial decision: Revision requested
17 Oct, 2024
Reviews received at journal
12 Oct, 2024
Reviews received at journal
11 Oct, 2024
Reviews received at journal
07 Oct, 2024
Reviewers agreed at journal
01 Oct, 2024
Reviewers agreed at journal
01 Oct, 2024
Reviews received at journal
25 Sep, 2024
Reviewers agreed at journal
11 Sep, 2024
Reviews received at journal
08 Sep, 2024
Reviewers agreed at journal
08 Sep, 2024
Reviewers agreed at journal
21 Aug, 2024
Reviewers invited by journal
21 Aug, 2024
Editor assigned by journal
21 Aug, 2024
Editor invited by journal
08 Aug, 2024
Submission checks completed at journal
07 Aug, 2024
First submitted to journal
28 Jul, 2024

You are reading this latest preprint version

Prevalence and associated factors of metabolic syndrome among people live with HIV in a medical center of Northern Taiwan -- a cross sectional study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Methods

Results

Discussion

Declarations

Author contributions statement

Author Contribution

Data Availability

References

Additional Declarations

Status:

Version 1