Short Communication: Risk of Elevated Total Cholesterol/High-Density Lipoprotein Cholesterol Ratio After Antiretroviral Therapy in HIV/Hepatitis C Virus Patients

Abstract

Dyslipidemia from highly active antiretroviral therapy (HAART) use has been reported to be less severe among persons with HIV and hepatitis C (HCV) compared to those with HIV monoinfection. However, the effect on lipoprotein ratios is less clear. The total cholesterol/high-density lipoprotein ratio (TC/HDL-C ratio) is a robust measure of cardiovascular disease (CVD) risk but has not been examined in the context of HIV/HCV-coinfected patients. We compared the TC/HDL-C ratio before HAART initiation and after at least 6 months on HAART between patients monoinfected with HIV and coinfected with HIV and HCV. Pre- and post-HAART TC, HDL-C, and non-HDL-C were also assessed. Although TC, HDL-C, and non-HDL-C significantly increased after HAART initiation in both HIV and HIV/HCV patients, the TC/HDL-C ratio did not. In addition, although the pre- and post-HAART TC, HDL-C, non-HDL-C, and TC/HDL-C ratio were significantly different between HIV and HIV/HCV patients, the magnitude in the change from pre- to post-HAART was not significantly different between infection groups. These results persisted after controlling for age, sex, race, current pharmacotherapy for lipoproteins, body mass index, and current CD4 cell count. The magnitude of change in the TC/HDL-C ratio after HAART initiation is not significantly different between HIV and HIV/HCV patients, suggesting subsequent CVD risk in HIV/HCV patients may be greater than currently appreciated.

M any highly active antiretroviral therapies (HAART) lead to increased lipoprotein levels in HIV-infected patients.^1,2 However, lipoprotein increases after HAART have been reported to be attenuated among persons coinfected with HIV and hepatitis C (HCV) compared to those with HIV alone,^3

–6 resulting in HCV coinfection having a “protective effect” against hyperlipidemia.^7,8 Nevertheless, cardiovascular disease (CVD) events among HIV/HCV persons have been noted to be higher than those with HIV, suggesting a better predictor of CVD is needed.^9

–12 Previously, lipoprotein comparisons of total cholesterol (TC), or high-density lipoprotein cholesterol (HDL-C) from pre- to post-HAART between HIV and HIV/HCV groups were primarily assessed separately, a reportedly less accurate depiction of CVD risk than lipoprotein ratios. Because the TC/HDL-C ratio is a significant predictor of future CVD,¹³ and has minimal variability in nonfasting states,¹⁴ the objective of this cohort study was to investigate the changes of the TC/HDL-C ratio before and after HAART. Our study was designed (1) to determine if the TC/HDL-C ratio significantly changed 6 months after HAART initiation in either the HIV or the HIV/HCV group, and (2) to determine if the change in the TC/HDL-C ratio from pre- to post-HAART initiation significantly differed between patients in these groups. Changes in TC and HDL and non-HDL-C were also assessed.

This research study was a retrospective cohort study of HIV and HIV/HCV patients receiving medical care at one of three sites (University of Rochester Medical Center's Infectious Disease Clinic, AIDS Community Health Center, and Unity Health System's Infectious Disease Clinic) from Monroe County in Rochester, NY. Approximately 85% of all patients with known HIV in Monroe County receive their care from one of these three clinics. Patients were identified through an internal database and were assessed for eligibility. To be eligible, patients had to be over the age of 18, and had to be either HAART naive, or not on HAART therapy for 60 days prior to a baseline lipoprotein measurement. HIV/HCV patients had to be chronically infected with HCV. For patients with multiple lipoprotein profile measurements, the measurement that was closest, but not after the initiation of HAART, was used for the baseline measurement. Patients were followed after starting HAART, and a post-HAART lipoprotein assessment was obtained after at least 6 months on HAART. HIV patients were randomly sampled from Monroe County's primary HIV service care provider (University of Rochester's Infectious Diseases); however, due to smaller clinic sample sizes, all HIV/HCV patients receiving care from the three study sites were sampled. The study was approved by the University of Rochester Research Subjects Review Board.

Data collected from electronic and paper medical records included demographic characteristics, metabolic and lipoprotein profiles, anthropometric measurements, blood pressure, history of current or past smoking or drug use, and indicators of HIV severity (CD4 cell counts and viral load) and HCV severity (HCV viral load, fibrosis score). Because the fasting status of the patients was not always known, diabetes was conservatively defined as a single measurement of blood sugar ≥200 mg/dl, or self-report of current or former diagnosis of diabetes or pharmacotherapy. Hypertension was defined as a single measurement of systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, self-report of current or former diagnosis of hypertension, or pharmacotherapy for hypertension. Non-HDL-C was defined as TC-HDL-C.

The HIV and HIV/HCV groups were compared at baseline for differences in demographic and health characteristics with t-tests and chi-squares for continuous and categorical variables, respectively. The change in metabolic measures (TC, HDL, non-HDL-C, and TC/HDL-C ratio) from baseline to follow-up was assessed with paired t-tests for within group comparisons, and t-tests for between group comparisons. Covariates controlled for in the multivariable linear regression analyses included age, sex, race, current use of lipoprotein lowering medications, post-HAART CD4 cell count, change in body mass index (BMI) from pre- to post-HAART, and duration of current HAART use. Analyses were performed with SAS 9.2 (SAS, Cary, NC).

There was no evidence of heterogeneity across clinics, thus we did not control for clinic. The characteristics of the study population (n=194) at baseline are presented (Table 1). The HIV/HCV participants were on average 5 years older than the HIV group (45.9 vs. 40.6 years, respectively, p=0.0001) and were more likely to be smokers (58% vs. 36%, p=0.004) and current or former drug users (89% vs. 29%, p<0.0001). The groups were comparable in CD4 cell count, HIV viral load, sex, marital status, medication use for hypertension or hyperlipidemia, previous HIV therapy use including duration, and time since last HIV medication. Among the HIV/HCV patients, approximately half of the HIV/HCV patients previously had a biopsy performed (data not shown). Of these patients, two-thirds had no signs of fibrosis (n=9) or low fibrosis (fibrosis score of 1 or 2, n=10). Five patients had a moderate fibrosis score of 3 or 4, and the remaining five patients had cirrhosis. Excluding patients with cirrhosis did not affect our results (data not shown). Five HIV/HCV patients had ever received HCV antiviral therapy; three discontinued therapy more than 5 years prior to study follow-up, one discontinued therapy 6 months prior to study follow-up, and one was currently on treatment. Excluding these five participants did not affect our results (data not shown).

Table 1.

Sociodemographic Characteristics of the Study Population at Baseline

Subject characteristics ^a	HIV only (n=132)	HIV/HCV (n=62)	p
Age, mean (SD)^b	40.6 (10.8)	45.9 (7.8)	0.0001
Male	91 (69%)	39 (63%)	0.40
Body mass index, mean (SD)	27.1 (5.4)	24.8 (5.0)	0.005
Ethnicity
White	65 (49%)	14 (23%)	0.001
Black	53 (40%)	31 (50%)
Hispanic	12 (9%)	17 (27%)
Other	2 (1%)	0 (0%)
Marital status
Single	82 (62%)	35 (56%)
Married	29 (22%)	12 (19%)	0.62
Divorced/widowed	21 (16%)	13 (21%)
Current smoker	48 (36%)	36 (58%)	0.004
Current or former drug use	38 (29%)	55 (89%)	<0.0001
HIV risk behavior
Heterosexual contact	64 (48%)	26 (42%)
MSM	59 (45%)	1 (1%)
IDU	3 (2%)	26 (42%)	<0.0001
MSM and IDU	1 (1%)	4 (6%)
Other	5 (4%)	1 (1%)
Hypertensive^c	21 (16%)	9 (14%)	0.80
Hyperglycemic^d	1 (1%)	5 (8%)	0.01
On lipoprotein medications	6 (4%)	2 (3%)	0.67
On blood pressure medications	18 (14%)	7 (11%)	0.65
On glucose medications	8 (6%)	6 (10%)	0.36
Family history of CHD^e	68 (51%)	25 (40%)	0.38
Family history of diabetes	50 (38%)	22 (35%)	0.17
Duration of HIV, mean years (SD)	3.79 (5.26)	7.4 (6.1)	<0.0001
CD4 count, mean cells/μl, (SD)	250 (216)	215 (199)	0.28
HIV viral load, mean log (SD)	5.13 (5.3)	5.11 (5.3)	0.85
Previous HAART use	48 (37%)	24 (39%)	0.75
Duration of previous HAART, mean years (SD)	3.20 (2.07)	2.78 (1.90)	0.40
Time since last on HAART, mean years (SD)	0.7 (1.0)	1.5 (1.8)	0.04

N (%) unless otherwise indicated. Not all % add up to 100 due to rounding or missing values.

SD, standard deviation.

Hypertension: systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg or self-report of current or former diagnosis of hypertension or pharmacotherapy for hypertension.

Hyperglycemic: random glucose ≥200 mg/dl or self-report of current or former diagnosis of diabetes or pharmacotherapy for diabetes.

Family history available for 80 and 32 HIV and HIV/HCV participants, respectively, for CHD and 60 and 31 HIV and HIV/HCV participants, respectively, for diabetes.

HCV, hepatitis C virus; MSM, men who have sex with men; IDU, intravenous drug use; CHD, coronary heart disease; HAART, highly active antiretroviral therapy.

At baseline, the TC/HDL-C ratio was significantly higher in the HIV patients (4.39) than in the HIV/HCV patients (3.68, p<0.0001). The mean TC was significantly higher in the HIV group compared to the HIV/HCV group (175.3 mg/dl vs. 150.7 mg/dl, respectively, p=0.0003) (Table 2), as was non-HDL-C (132.9 mg/dl vs. 106.9 mg/dl, p<0.0001), but HDL cholesterol was not significantly different (42.4 mg/dl vs. 43.9 mg/dl, p=0.55).

Table 2.

Comparison of Lipoprotein Levels in HIV and HIV/Hepatitis C Virus Patients at Baseline and After at Least 6 Months of Highly Active Antiretroviral Therapy

	HIV				HIV/HCV				HIV vs. HIV/HCV
	Baseline	Follow-up	Delta	p	Baseline	Follow-up	Delta	p	p of comparison at baseline	p of comparison at follow-up	p of delta comparison
Total cholesterol	175.3 (46.3)	202.2 (47.9)	26.9 (45.5)	<0.0001	150.7 (41.2)	165.8 (45.0)	15.1 (40.3)	0.004	0.0003	<0.0001	0.08
HDL cholesterol	42.4 (14.9)	47.2 (14.7)	4.9 (12.5)	<0.0001	43.9 (17.2)	48.5 (18.5)	4.7 (15.4)	0.02	0.55	0.61	0.93
TC/HDL-C ratio	4.39 (1.2)	4.53 (1.2)	0.1 (1.1)	0.13	3.68 (0.9)	3.74 (1.4)	0.06 (1.1)	0.68	<0.0001	0.0003	0.61
Non-HDL-C	132.9 (39.6)	155.0 (44.6)	22.0 (40.0)	<0.0001	106.9 (32.5)	117.3 (42.8)	10.4 (33.3)	0.02	<0.0001	<0.0001	0.05

Data presented are unadjusted mean values (standard deviation). Unless otherwise noted, p-values are for the differences in baseline to follow-up.

HCV, hepatitis C virus; HDL, high-density lipoprotein; TC/HDL-C, total cholesterol/high-density lipoprotein cholesterol.

All patients had been placed on HAART for a minimum of 6 months, but average duration was approximately 12 months for both HIV (383 days) and HIV/HCV patients (391 days, p=0.81). At follow-up, the HIV and HIV/HCV groups were comparable in terms of their statins or lipoprotein lowering medication use, as well as their HAART regimen, with similar proportions on protease inhibitors and nonnucleoside reverse transcriptase inhibitors-based regimens (p>0.10, data not shown).

After a minimum of 6 months of HAART, mean TC, HDL-C, and non-HDL-C significantly changed after HAART initiation within both HIV-monoinfected and HIV/HCV-coinfected patients (all p<0.05). However, the TC/HDL-C ratio did not significantly increase in either group, and the magnitude of the changes in TC, HDL-C, non-HDL-C, and the TC/HDL-C ratio following HAART was not significantly different between infection groups (p=0.08, 0.93, 0.05, and 0.61, respectively). A sensitivity analysis using median values, rather than mean values, did not affect our main findings (data not shown).

After adjusting for age, sex, race, current use of lipoprotein lowering medications, current CD4 cell count, BMI, and duration of current HAART use, the mean change in the TC/HDL-C ratio was 0.04 greater among patients with HIV/HCV compared to HIV patients (p=0.86), the mean change in TC was 5.64 mg/dl lower (p=0.49), and the mean change in HDL was 0.22 mg/dl higher (p=0.93) (data not shown).

Consistent with the literature, TC, HDL-C, and non-HDL-C significantly increased after HAART exposure in our study.^15,16 However, our study improves upon the existing literature with the inclusion of the well-validated TC/HDL-C ratio. We found that the TC/HDL-C ratio did not change significantly after HAART use in either HIV or HIV/HCV persons. In addition, although the post-HAART TC, non-HDL-C and HDL-C were significantly different between HIV and HIV/HCV groups, the magnitude of the change from pre- to post-HAART was not significantly different. These results suggest that while post-HAART lipid values may have different absolute values between infection groups, the magnitude of the changes post-HAART is similar, possibly reflecting a delayed, but not significantly different CVD risk in those with HIV/HCV compared to HIV monoinfection. Increased CVD events and poorer surrogate markers in HIV/HCV patients compared to HIV patients have already been noted,^9

–12,17 and further research should be conducted to compare other traditional risk measures (such as the Framingham risk score) between HIV and HIV/HCV patients.

This study has a number of limitations. Sample size limited our ability to analyze specific HIV treatment regimens, and follow-up for this study may have been inadequate to observe large changes in lipoprotein profiles over time. Although all subjects were on HAART for an average of 12 months, the longitudinal effects of HAART as well as the effects of HAART interruption in these populations should be assessed with longer periods of follow-up time.¹⁸ In addition, the power to detect significant differences in our multivariable analyses may have been limited by our relatively small sample size. Severity of liver disease was not assessed for all persons with HIV/HCV and the association between cirrhosis and CVD remains unclear. Results may not be generalizable on a national level, however, the ethnicities of our participants are similar to that of the nationally infected populations.¹⁹ Lastly, because the fasting status of our subjects was oftentimes unknown, we focused our study on the lipids that have been shown to be stable between fasting and nonfasting states.

This study aimed to improve our understanding of the impact of coinfection with HIV and HCV on lipoproteins. With the advent of HAART, HIV-infected patients are living longer, with CVD accounting for 25% of the deaths due to non-AIDS-related causes.²⁰ However, the risk of CVD among people with HIV/HCV is still largely unknown.^9
–11 As approximately 2–4 million Americans are coinfected with HIV and HCV, chronic diseases such as CVD in this population represent a significant public health burden. Lower lipoprotein profiles in persons with HIV/HCV may incorrectly influence clinicians and patients into believing there is little risk of CVD, and additional studies on the risk of CVD among HIV/HCV persons are needed.

Footnotes

Acknowledgments

This study was supported by a grant from the National Heart Lung and Blood Institute (NHLBI) Preventive Cardiology Training Program #5T32HL007937-10. This publication was also made possible by Grant KL2 RR 024136 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and the NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NHLBI, NCRR, or NIH. Information on NCRR is available at http://www.ncrr.nih.gov/.

Information on Reengineering the Clinical Research Enterprise can be obtained from http://nihroadmap.nih.gov/clinicalresearch/overview-translational.asp. We gratefully acknowledge the support and assistance from the faculty and staff at the Infectious Diseases Division and Gastroenterology and Hepatology Division at the University of Rochester Medical Center, Unity Health, and AIDS Community Health Center.

Author Disclosure Statement

No competing financial interests exist.

References

Albuquerque

, de Faria

, Oliveira

, Magro

, Castilho

. High frequency of Fredrickson's phenotypes IV and IIB in Brazilians infected by human immunodeficiency virus. BMC Infect Dis, 2005; 5:47.

Valerio

, Fontas

, Pradier

et al. Lopinavir/ritonavir combination and Total/HDL cholesterol ratio. J Infect, 2005; 50:229–235.

Diong

, Raboud

, Li

, Cooper

. the Ontario HIV Treatment Network Cohort Study Team. HIV/hepatitis C virus and HIV/hepatitis B virus coinfections protect against antiretroviral-related hyperlipidaemia. HIV Med, 2011; 12:403–411.

Polgreen

, Fultz

, Justice

et al. Association of hypocholesterolaemia with hepatitis C virus infection in HIV-infected people. HIV Med, 2004; 5:144–150.

Visnegarwala

, Chen

, Raghavan

, Tedaldi

. Prevalence of diabetes mellitus and dyslipidemia among antiretroviral naive patients co-infected with hepatitis C virus (HCV) and HIV-1 compared to patients without co-infection. J Infect, 2005; 50:331–337.

Patroni

, Torti

, Tomasoni

et al. Effect of highly active antiretroviral therapy (HAART) and hepatitis C co-infection on hyperlipidemia in HIV-infected patients: A retrospective longitudinal study. HIV Clin Trials, 2002; 3:451–461.

Lapadula

, Torti

, Paraninfo

et al. Influence of hepatitis C genotypes on lipid levels in HIV-positive patients during highly active antiretroviral therapy. Antivir Ther, 2006; 11:521–527.

Torti

, Patroni

, Tinelli

et al. Influence of hepatitis C virus coinfection on lipid abnormalities in HIV-positive patients after highly active antiretroviral therapy. J Acquir Immune Defic Syndr, 2002; 29:315–317.

Freiberg

, Chang

, Skanderson

et al. The risk of incident coronary heart disease among veterans with and without HIV and hepatitis C. Circ Cardiovasc Qual Outcomes, 2011; 4:425–432.

10.

Freiberg

, Cheng

, Kraemer

, Saitz

, Kuller

, Samet

. The association between hepatitis C infection and prevalent cardiovascular disease among HIV-infected individuals. AIDS, 2007; 21:193–197.

11.

Butt

, Xiaoqiang

, Budoff

, Leaf

, Kuller

, Justice

. Hepatitis C virus infection and the risk of coronary disease. Clin Infect Dis, 2009; 49:225–232.

12.

Weber

, Sabin

, Reiss

et al. HBV or HCV coinfections and risk of myocardial infarction in HIV-infected individuals: The D:A:D cohort study. Antivir Ther, 2010; 15:1077–1086.

13.

Manickam

, Rathod

, Panaich

et al.

Comparative prognostic utility of conventional and novel lipid parameters for cardiovascular disease risk prediction: Do novel lipid parameters offer an advantage?

J Clin Lipidol, 2011; 5,2:82–90.

14.

Mora

, Rifai

, Buring

, Ridker

. Fasting compared with nonfasting lipids and apolipoproteins for predicting incident cardiovascular events. Circulation, 2008; 118,110:993–1001.

15.

Asztalos

, Schaefer

, Horvath

et al. Protease inhibitor-based HAART, HDL, and CHD-risk in HIV-infected patients. Atherosclerosis, 2006; 184:72–77.

16.

Levy

, McCandless

, Harrigan

et al. Changes in lipids over twelve months after initiating protease inhibitor therapy among persons treated for HIV/AIDS. Lipids Health Dis, 2005; 4:4

17.

Boddi

, Abbate

, Chellini

et al. HCV infection facilitates asymptomatic carotid atherosclerosis: preliminary report of HCV RNA localization in human carotid plaques. Dig Liver Dis, 2007; 39,1:S55–60.

18.

Lampe

, Duprez

, Kuller

et al. Changes in lipids and lipoprotein particle concentrations after interruption of antiretroviral therapy. J Acquir Immune Defic Syndr, 2010; 54,3:275–284.

19.

Vital signs: HIV testing and diagnosis among adults—United States, 2001–2009. MMWR Morb Mortal Wkly Rep, 2010; 59:1550–1555.

20.

Sackoff

, Hanna

, Pfeiffer

, Torian

. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City. Ann Intern Med, 2006; 145:397–406.