The contribution of syndemic conditions to cardiovascular disease risk

Abstract

The syndemic conditions of low education, childhood maltreatment, depression, HIV, alcohol and cocaine use, and obesity have been established as independent risk factors for cardiovascular risk, but research examining the association between syndemic conditions and cardiovascular risk in high-risk populations is lacking. A total of N = 503 participants underwent an ultrasound of the carotid artery to assess for atherosclerotic plaque. Participants, HIV-infected (n=202) and HIV-uninfected (n=301) with and without a history of cocaine use, were a mean age of 36.13 years (SD =9.51); 50% were male, and 62% were African-American. Each syndemic condition was associated with 8% greater odds of atherosclerotic plaque (OR = 1.08), 9% greater odds of systolic blood pressure (OR = 1.09), and 10% greater odds of diastolic blood pressure (OR = 1.10). Multilevel research, interventions, and public policy initiatives are needed to activate stakeholders at each level to maximize their impact at a community level among populations with high rates of syndemic conditions.

Introduction

Cardiovascular disease (CVD) is the leading cause of mortality in the United States and worldwide, and the risk of developing CVD is exacerbated by atherosclerosis and hypertension (Galkina and Ley, 2009; National Center for Health, 2017). Atherosclerosis is the chronic inflammation and narrowing of the arterial lumen due to the accumulation of leukocytes and deposition of lipid plaques in the arteries, which leads to increases in blood pressure, and predict the development of CVD (Eckard et al., 2017; Galkina and Ley, 2009; Naqvi and Lee, 2014; Pant et al., 2014; van den Oord et al., 2013; Weiner et al., 2014). Atherosclerosis is a pathology of the vascular intima, the innermost layer of blood vessels, and is caused by hyperlipidemia and lipid oxidation (Baradaran, 2012; Hennekens and Gaziano, 1993; Rafieian-Kopaei, Setorki, Doudi, Baradaran, & Nasri, 2014). In patients with atherosclerosis, the intimal layer of blood vessels becomes thickened secondary to fat buildup. The disease can be divided into two distinct parts: 1) atherosis, which refers to fatty deposit buildup and involvement of macrophages and 2) sclerosis, the thin fibrous layer that surrounds the plaque (Rafieian-Kopaei, et al., 2014; Rahimi, 2012; Ross, 1999). Although the pathogenesis involves a variety of cytokines and inflammatory markers that are still being investigated, the overall methodology is known: atheromatous plaques are formed when cholesterol crystals deposit in the intima. As the plaque increases in size, less blood is able to flow through arteries. Over time, fibroblasts insert calcium into the blood vessels, resulting in sclerotic, hardened arteries. Eventually, the plaque’s irregular surface results in the buildup of a clot and thrombosis, resulting in an abrupt disruption of blood flow (Rafieian-Kopaei, et al., 2014; Tavafi, 2013). Ultrasonographic measurement of carotid atherosclerotic plaques is a noninvasive imaging procedure used to detect early carotid atherosclerosis that can be used to predict future thrombotic and cardiovascular events (van den Oord, et al., 2013).

Syndemic theory posits that multiple, co-occurring psychosocial comorbidities act synergistically to exacerbate the risk and consequences of disease (Tsai, Mendenhall, Trostle, & Kawachi, 2017). Low education (Winkleby, Jatulis, Frank, & Fortmann, 1992), childhood maltreatment (Batten, Aslan, Maciejewski, & Mazure, 2004), depression (Suls, 2018), HIV (De Socio, Pucci, Baldelli, & Schillaci, 2017), alcohol and cocaine use (Benzaquen, Cohen, & Eisenberg, 2001; Lu et al., 2014), and increased body weight are syndemic conditions that may interact to increase the risk of CVD. Illegal drug use and alcohol intake can accelerate CVD progression and other HIV comorbidities (Beires et al., 2018; Napuri et al., 2013; Pandhare et al., 2014; Weiner, et al., 2014). Molecular and epidemiological evidence suggests that increased cocaine use can result in increased HIV viral load and immunological decline due to gene expression changes, and alcohol use is detrimental to the cardiac health of PLWH (Adjemian, Volpe, & Adjemian, 2015; Goel, Sharma, & Garg, 2018; Kelly et al., 2016; Raghavan, Rimmelin, Fitch, & Zanni, 2017; Stein, Currier, & Hsue, 2014). Syndemic theory further proposes that the clustering of these conditions may increase disease burden by worsening disease progression and increasing the risk for related diseases. The clustering of these conditions in the presence of HIV infection may affect immune function, increasing vulnerability to diseases directly or indirectly related diseases. Overall, the goal of syndemic theory is to explain why many diseases can plaque the same individuals, the biological pathways these diseases manipulate, and how interactions with social circumstances and inequality can overall worsen the morbidity and mortality (Singer, Bulled, Ostrach, & Mendenhall, 2017). In order to adequately treat individuals with complex comorbid conditions, healthcare providers cannot only target the disease, they must also holistically manage the social environment (Singer, et al., 2017).

Among people living with HIV (PLWH), there are a number of syndemic risk factors that may increase disease burden, such as the interaction of HIV and CVD. People living with HIV have increased incidence of CVD due to biological factors and social health disparities and are at greater risk of atherosclerotic plaque deposition and mortality from CVD, even when controlling for other cardiometabolic risk factors. The accelerated progression of CVD among PLWH is likely associated with other syndemic conditions as well as with abnormalities in the inflammatory response and lipid processing facilitated by chronic infection (Beires, et al., 2018; Freiberg et al., 2013; Hanna et al., 2015; Kelly, et al., 2016). The etiology of hyperlipidemia in PLWH is unclear but is likely due to a combination of the infection itself in addition to the use of antiretrovirals (Carr et al., 1998; Echevarria, Hardin, & Smith, 1999; Geletko and ZuWallack, 2001) For this reason, it is difficult to apply the same general principles that the National Cholesterol Education Program (NCEP) uses when evaluating the general population. Furthermore, given the unclear etiology of hyperlipidemia, it is suspected that traditional lipid lowering agents may not be effective in PLWH (Geletko and ZuWallack, 2001). Illegal drug use and alcohol intake can also accelerate CVD progression and other HIV comorbidities (Beires, et al., 2018; Napuri, et al., 2013; Pandhare, et al., 2014; Weiner, et al., 2014). Molecular and epidemiological evidence suggests that increased cocaine use can result in increased HIV viral load and immunological decline due to gene expression changes, and alcohol use is detrimental to the cardiac health of PLWH (Adjemian, et al., 2015; Goel, et al., 2018; Kelly, et al., 2016; Raghavan, et al., 2017; Stein, et al., 2014).

Other well-documented social determinants of health that influence the development of CVD among HIV-uninfected individuals, such as mental illness and low education, also present significant burdens to PLWH (Davila et al., 2012; Galobardes, Smith, & Lynch, 2006; Gutierrez and Williams, 2014; Havranek et al., 2015; National Center for Health, 2012). In addition to an increase risk for developing CVD, PLWH are also less likely to receive evidence-based lipid controlling treatment in comparison with HIV-uninfected individuals (Policarpo, Valadas, Rodrigues, & Moreira, 2014).

Despite the fact that the syndemic conditions of low education, childhood maltreatment, depression, obesity, alcohol and cocaine use, as well as HIV infection, are established independent risk factors for cardiovascular risk, research examining the rates and the association between syndemic conditions and the risk for CVD in high-risk populations is lacking. The aim of this study was to evaluate the association between syndemic conditions and the risk for CVD, as measured by atherosclerosis and elevations in blood pressure among people living with and without HIV, with and without a history of cocaine use. It was anticipated that syndemic conditions would be associated with the presence of atherosclerotic plaque and elevations in blood pressure. It was hoped that results from this study could inform CVD prevention strategies for those at high risk for cardiovascular disease, particularly those experiencing high rates of syndemic conditions.

Method

The study was conducted from December 2014 to June 2018 in Miami, Florida, a Southeastern urban center in the US with high rates of syndemic conditions. Candidates were aged 18 to 50. As part of the study screening process, candidates who self-reported having been diagnosed with medical conditions associated with increased CVD risk (diabetes mellitus, hypertension, co-infection with hepatitis C, or hypercholesterolemia receiving lipid lowering agents), or history of vascular diseases (myocardial infarction, transient ischemic attack, bypass surgery, or angioplasty) were not eligible. Participants with a history of cocaine use who had not used cocaine in the past 3 months were excluded from the study.

Study Design

The study was designed to explore the occurrence of CVD among those who were HIV-infected and using cocaine, in comparison with HIV-uninfected cocaine users, HIV-infected non-cocaine users, and healthy controls not using cocaine or being HIV-infected. All participants (n=503) were recruited via community outreach, using flyers, snowball sampling, and referrals by physicians or clinic coordinators. To enhance recruitment of HIV-infected participants, flyers were distributed to infectious disease physicians and HIV providers throughout the community. To ensure recruitment of individuals who were using cocaine users, flyers were also distributed to substance use and mental health treatment professionals, clinics, alcoholic and narcotic anonymous meetings, and facilities throughout the community. HIV-uninfected non-drug users were recruited from the local community by flyers and word of mouth.

Participants

Measures

Demographic Characteristics.

Participants completed a questionnaire assessing demographic characteristics. Demographic questions included age, race and ethnicity, gender, sexual identity, and education.

Cardiovascular Risk Measures

Count of Atherosclerotic Plaques in the Carotid Artery.

An ultrasonography of the carotid artery was performed using a high-resolution B-mode carotid ultrasound machine by a proficient sonographer who was blinded to the participants’ risk factors for CVD, HIV status, and reported drug use. To ensure accuracy, the ultrasonography device was calibrated prior to study onset, and only one machine was used to examine all participants. In addition, the sonographer did not have any other health information about the participants and was not involved in the physical examination of the participants; the only information provided was the participant’s study identification number. Four hundred carotid ultrasonography images were gathered at 1 cm segment of the distal common carotid artery near and far wall, 1 cm of bifurcation and 1 cm proximal internal carotid artery near and far wall. These images were then analyzed using a software measured off-line with an automated edge tracking system M’Ath (Intelligence in Medical Technologies, Inc., Paris, France). M’Ath is used to automatically search in the vicinity of reference for the wall boundaries using an intensity gradient detection algorithm. Then, the presence of carotid atherosclerotic plaques in the vicinity of reference was assessed and used as the outcome for the current study. Left and right carotid bifurcations and internal and common carotid arteries were evaluated for the presence of plaque.

Blood Pressure.

Participants underwent a physical assessment performed by registered nurses in which three resting systolic and diastolic blood pressure measurements were obtained; the three measurements were averaged for data analyses. Average systolic and diastolic blood pressure values were dichotomized according to the latest guidelines from the American Heart Association, which indicate that systolic values above 120 and diastolic values greater than 80 are considered elevated (American Heart Association, 2018).

Syndemic Factors

Low Education.

Education was assessed on the demographic questionnaire. Education was dichotomized as low versus high education, with low being less than high school, and high being high school education or more.

Childhood Maltreatment.

Childhood maltreatment was be assessed using the Childhood Trauma Questionnaire (CTQ) (Bernstein, Ahluvalia, Pogge, & Handelsman, 1997), a 28-item Likert scale (1= never true to 5= very often) assessing emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect, and denial about abuse and neglect in childhood. Internal reliability for this scale in this sample was excellent (α = 0.91).

Depressive Symptoms.

Depressive symptoms were assessed using the Center for Epidemiological Studies-Depression Scale (CES-D) (Radloff, 1977). Participants provided the frequency of depressive symptomatology in the past week, with a possible range of 0 to 60, with greater scores indicating greater severity (α = 0.87).

HIV Status.

HIV testing was performed using the rapid test OraQuickADVANCE® Rapid HIV-1/2 Antibody Test among those who self-reported being HIV-uninfected. HIV-infected participants provided documentation to confirm their HIV status. The person conducting HIV testing and assessment was a trained HIV counselor and provided pre-and post-test counseling to participants.

Alcohol and Cocaine Use.

The Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders, 4^th Edition (DSM-IV), non-patient version (SCID-IV-NP) (First, Spitzer, Gibbon, & Williams, 1997), a well-validated, semistructured interview, was administered to assess cocaine and alcohol use. The SCID-IV-NP, which is based off the DSM IV, has been found to reliable in assessing substance use (Segal, Hersen, & Van Hasselt, 1994). Cocaine use was here defined as any cocaine use in the last 3 months.

Obesity.

Participants underwent a physical assessment performed by registered nurses in which their weight and height was measured. Body mass index (BMI) was calculated using weight and height obtained during the assessment.

Statistical Analyses

The analytic plan consisted of several steps. First, univariate analyses (means, standard deviations, and proportions) were used to describe demographic characteristics, smoking, and measures of depression and substance abuse. The, three logistic regression models were used to examine the association between number of syndemics and: 1. Presence of atherosclerotic plaque, 2. Elevated systolic blood pressure, and 3. Elevated diastolic blood pressure. The number of syndemics were included as a sum score, as in previous research (Friedman et al., 2015; Glynn et al., 2019; Harkness et al., 2019; Harkness et al., 2018; Sullivan, Messer, & Quinlivan, 2015). Statistically significant coefficients were transformed and reported as odds ratios with 95% confidence intervals. Statistical analyses were conducted using SPSS version 24 on a Macintosh operating system.

Results

Sample Characteristics

Table 1 summarizes the sociodemographic characteristics of the sample, as well as differences between groups with and without HIV. Overall, the sample was 36 years of age (range 18 to 50), two-thirds (61.8%) were African American, non-Hispanic (66%), half (49.3%) were male, and most (79.3%) identified as straight/homosexual.

Table 1.

Sociodemographic Characteristics N = 503

	M or n	(SD or %)	No HIV n (40%) M (SD)	HIV n (40%) M (SD)	p
Sociodemographics
Age	36.13	(9.51), range: 18–50	35.07 (8.94)	37.73 (10.13)	.003
Race/ethnicity					.002
Caucasian	56	(11.1%)	40 (13.2%)	16 (8.0%)
African-American	311	(61.8%)	166 (54.8%)	145 (72.5%)
Haitian/Haitian American/Haitian Bahamian	13	(2.6%)	7 (2.3%)	6 (3.0%)
Hispanic/Latino Caucasian	96	(19.1%)	69 (22.8%)	27 (13.5%)
Hispanic/Latino Black	12	(2.4%)	8 (2.6%)	4 (2.0%)
Other	15	(3.0%)	13 (4.3%)	2 (1.0%)
Gender
Male	249	(49.5%)	176 (58.1%)	73 (36.5%)	< .001
Female	248	(49.3%)	125 (41.3%)	123 (61.5%)
Transgender	6	(1.2%)	2 (0.7%)	4 (2.0%)
Sexual identity
Straight/heterosexual	398	(79.3%)	261 (86.4%)	137 (68.5%)	< .001
Homosexual	59	(11.8%)	20 (6.6%)	39 (19.5%)
Lesbian	5	(1.0%)	4 (1.3%)	1 (0.5%)
Bisexual	40	(8.0%)	17 (5.6%)	23 (11.5%)

As noted in Table 2, one-quarter of participants had atherosclerotic plaque (25.2%), 42.4% had elevated systolic blood pressure, and 28.5% had elevated diastolic blood pressure. In terms of syndemic factors, 30.4% had low education, 36.6% reported childhood emotional abuse, 53.3% childhood physical abuse, 32.4% childhood sexual abuse, 72.4% childhood emotional neglect, 39.6% childhood physical neglect, and 58.1% had clinically significant depressive symptoms. Almost half of the sample was HIV-infected (40.2%). Two-thirds (59.8%) of participants reported alcohol use, and 39.8% cocaine use. Almost three-fourths (70.2%) were overweight or obese.

Table 2.

Cardiovascular Risk Measures and Syndemic Factors

Cardiovascular Risk Measures
Atherosclerotic plaque	122	(25.2%)
Elevated systolic	213	(42.4%)
vElevated diastolic	143	(28.5%)
Syndemic factors
1. Low education	153	(30.4%)
Childhood Maltreatment (1–6)
2. Emotional abuse	293	(36.6%)
3. Physical abuse	268	(53.3%)
4. Sexual abuse	163	(32.4%)
5. Emotional neglect	364	(72.4%)
6.Physical neglect	199	(39.6%)
7. Depression	292	(58.1%)
8. HIV	202	(40.2%)
9. Alcohol use	301	(59.8%)
10. Cocaine use	200	(39.8%)
11. Overweight or obese	353	(70.2%)

Syndemic conditions predicting cardiovascular disease risk

In examining number of syndemic conditions on atherosclerotic plaque, each syndemic condition endorsed resulted 8% greater odds of atherosclerotic plaque (OR = 1.08, 95% CI 1.01, 1.18, p = .030). In examining number of syndemic conditions on elevated systolic blood pressure, each syndemic condition endorsed resulted 9% greater odds of elevated systolic blood pressure (OR = 1.09, 95% CI 1.02, 1.16, p = .016). In examining number of syndemic conditions on elevated diastolic blood pressure, each syndemic condition endorsed resulted in 10% greater odds of elevated diastolic blood pressure (OR = 1.10, 95% CI 1.02, 1.19, p = .010). These models are summarized in Table 3.

Table 3.

Logistic regression models predicting # of syndemics on cardiovascular disease risk

Model 1: Atherosclerotic plaque	OR	95% CI	p
N = 485, X2(1) = 4.78, p = .029
# of syndemics	1.08	1.01, 1.18	.030
Model 2: Systolic blood pressure	OR	95% CI	p
N = 502, X2(3) = 5.84, p = 0.016
# of syndemics	1.09	1.02, 1.16	.016
Model 3: Diastolic blood pressure	OR	95% CI	p
N = 502, X2(3) = 6.69, p = 0.010
# of syndemics	1.10	1.02, 1.19	.010

Discussion

This study examined the association between syndemic factors (low education, childhood maltreatment, depressive symptoms, alcohol use, cocaine use, HIV infection, and body weight) and CVD risk among socioeconomically and racially diverse young and middle age men and women living in a city in the South of the US. A high prevalence of CVD risk measured by carotid atherosclerotic plaque (25.2%), and elevated systolic (42.4%) and diastolic blood pressure (28.5%), was identified. There was also high prevalence of syndemic conditions. Each syndemic condition was associated with increased odds of CVD risk, resulting in high risk for detection of atherosclerosis and elevations in blood pressure when multiple syndemic conditions were present.

Traditional risk assessment for CVD utilizes the Framingham Heart Study’s risk assessment tool, widely used in the general population (Pencina, D’Agostino, Larson, Massaro, & Vasan, 2009). Risk factors identified in the Framingham Heart Study have been adopted in national guidelines to establish CVD risk (Goff et al., 2014). The current study highlights that when syndemic conditions, such as drug use or HIV infection, are not included, Framingham-based risk estimates may less accurately predict CVD risk, especially among younger people with multiple syndemic conditions.

Syndemic factors have been individually associated with the formation of carotid atherosclerosis and with the development of CVD among adults in the US. HIV infection alone or in combination with alcohol use increases the risk for CVD (Adjemian, et al., 2015; Goel, et al., 2018; Holloway and Boccara, 2017; Hsue and Waters, 2017; Kelly, et al., 2016; Raghavan, et al., 2017; Stein, et al., 2014; van Zoest, van den Born, & Reiss, 2017). Cocaine use has also been associated with both short-term and long-term effects on the risk of CVD, increasing the risk of cardiovascular events within minutes of use, and contributing to the accumulation of cardiovascular events with continued use (Bhargava and Arora, 2011; Sharma et al., 2016). In addition, syndemic factors such as low education, depression, and obesity, have also been identified as synergistically contributing to the risk for CVD (Batten, et al., 2004; Lu, et al., 2014; Suls, 2018; Winkleby, et al., 1992). Although individual and synergistic associations have been described, this study is unique in that it includes multiple syndemic factors that contribute to CVD risk and evaluates the additional risk conferred by each factor to the overall risk for CVD. The findings that each of the factors contribute to a significant increase in risk highlights the importance of developing a multilevel approach to effectively intervene and reduce CVD risk in this population (Paskett et al., 2016). Such multilevel interventions should consider environmental, social, cultural, and family factors increase the risk for CVD, and need the involvement of stakeholders at each of these levels to maximize their impact at a community level among high risk populations (Paskett, et al., 2016).

In interpreting study findings, some limitations must be noted. First, the cross-sectional design limits temporal and causal interpretations. Second, the measurement of carotid atherosclerotic plaques as a dichotomous variable in the analysis may limit the interpretability of the results and future studies should consider assessing plaque density and size (Alsulaimani et al., 2013; Tiozzo et al., 2014). Third, the sample was predominanly African American and low income, which limits the generalizability of results. Fourth, the tool used to assess cocaine and alcohol use was not used for systematic item-level data collection, although the SCID-IV-NP (First, et al., 1997) is a well-validated diagnostic interview. Nevertheless, a measure quantifying the specific levels of cocaine and alcohol use should be used in future studies to determine whether frequency or short- versus long-term use of alcohol and cocaine are associated with carotid atherosclerosis and elevated blood pressure. Fifth, the analysis did not account for the level of immunosuppression, which may have increased CVD risk (Buggey and Longenecker, 2017; McIntosh, Lobo, & Hurwitz, 2017). Sixth, cocaine use immediately before the study was not assessed, which may have influenced or altered ultrasound findings. Finally, there may be other syndemic factors that were not considered in this study. Lastly, an additional limitation of this study is the use of BMI as a risk factor for cardiovascular disease. While BMI is a simple, cost-effective measurement to approximate obesity (Prentice and Jebb, 2001), it does not always provide a clear portrayal of an individual’s health. Recent research has found that abdominal obesity (and not BMI) is an accurate predictor of atherosclerotic cardiovascular disease (ASCVD), and should therefore be evaluated in addition to BMI (Fan et al., 2016; Goff, et al., 2014; Melin, Thulesius, Hillman, Landin-Olsson, & Thunander, 2018; Wang et al., 2019).

This study measured the presence of carotid plaque by using carotid ultrasound. Although the use of this tool may not be feasible for clinical practice, its use in this study facilitated detection of plaques in an asyptomatic, undiagnosed sample of young and middle aged adults, and allowed the evaluation of the contributors to CVD in a population that may be considered low risk based on traditional Framingham-based risk scores (Beires, et al., 2018; Havranek, et al., 2015; Janssen, De Gucht, Dusseldorp, & Maes, 2013; Policarpo, et al., 2014).

Recommendations arising from this research include optimizing risk assessments by considerations of the inclusion of syndemic factors in addition to traditional CVD risk factors. In addition, results highlight the need to include a multifaceted approach when designing CVD prevention programs to decrease the burden of CVD in vulnerable populations experiencing multiple syndemic factors.