Subclinical myocardial injury and cardiovascular mortality: Racial differences in prevalence and risk (from the third National Health and Nutrition Examination survey)

Abstract Background Subclinical myocardial injury (SCMI) determined from the Electrocardiographic Cardiac Infarction/Injury Score (CIIS) is associated with increased risk of cardiovascular disease and mortality. We hypothesized that SCMI prevalence and association with mortality would differ by race, categorized as non‐Hispanic White (White), non‐Hispanic Black (Black), and Mexican American. Methods Our analysis included 5,852 participants (age 58.5 ± 13.2 years; 54% women, 52% Whites, 23% Blacks, and 25% Mexican American participants) from the National Health and Nutrition Examination Survey (NHANES III, 1988–94) who were free of cardiovascular disease at the time of enrollment. SCMI was defined as the presence of CIIS ≥ 10 score points on the 12‐lead ECG. Prevalence of SCMI and its association with cardiovascular mortality were examined in each race/ethnic group in models adjusted for sociodemographics and common cardiovascular risk factors. Results SCMI prevalence was 23.4% in Whites, 21.8% in Blacks, and 18.0% in Mexican Americans. Compared to Whites, Blacks were as likely to have SCMI (odds ratio [OR] 0.95, 95% confidence interval [CI] 0.80–1.13), while Mexican Americans were less likely (OR 0.74, 95% CI 0.62–0.88). SCMI was not associated with increased risk of cardiovascular mortality in either Whites (hazard ratio [HR] 1.18, 95% CI 0.95–1.48) or Blacks (HR 1.19, 95% CI 0.79–1.80). In contrast, SCMI in Mexican Americans was associated with increased risk of cardiovascular mortality (HR 1.74, 95% CI 1.13–2.67, p < .05). Conclusion Mexican Americans had a lower prevalence of SCMI, but increased risk of cardiovascular mortality. Screening for SCMI may identify individuals at increased risk and improve targeted prevention efforts.


| INTRODUC TI ON
The Cardiac Infarction Injury Score (CIIS) was developed in 1981 as an ECG classification to improve diagnostic accuracy of myocardial injury compared to conventional criteria (e.g., the Minnesota Code).

CIIS uses a combination of features measured on a continuous scale
to evaluate 12 items across multiple leads of a standard electrocardiogram (e.g., T-wave amplitude and Q-wave duration). (Rautaharju et al., 1981) Increased CIIS scores have been associated with identifying individuals at high risk for cardiovascular disease (CVD) mortality. (Dekker et al., 1994;Domburg et al., 1998;Richardson et al., 2005;Siscovick et al., 1996) While the prognostic ability of CIIS among a diverse US population without established CVD has been previously studied, there has been limited assessment of racial differences in the significance of subclinical myocardial injury (SCMI). (ONeal et al., 2014) Therefore, the purpose of this study was to evaluate SCMI prevalence and association with CVD mortality among different races using data from the Third National Health and Nutrition Examination Survey (NHANES III).

| ME THODS
NHANES is a periodic survey that samples a cohort representative of the US population and was designed to estimate disease prevalence and risk factors. The survey methods have been well described previously. (NHANES III, 1988-94) Briefly, baseline data for NHANES III were collected from an in-home interview and a subsequent visit to a mobile examination center between 1988 and 1994.
The interview portion includes demographic, medication information, socioeconomic, dietary, and health-related questions, and the examination component consists of medical, dental, and physiologic measurements, as well as laboratory tests. All participants gave written informed consent at the time of study enrollment. Participant characteristics, electrocardiography methodology, and ascertainment of mortality in NHANES III have been previously published. (Yang et al., 2012) Blood samples were obtained at mobile centers, and basic laboratory values were recorded for each participant (total cholesterol, high-density lipoprotein cholesterol, and plasma glucose). The present analysis included 5,852 participants free of baseline CVD and who had available baseline demographic, laboratory, medication, and mortality data. Participants with baseline CVD were excluded, determined by a self-reported history of heart attack and/or stroke, electrocardiographic evidence of myocardial infarction, or major ST-T depression by Minnesota Electrocardiogram Classification (Prineas et al., 2010).
Age, sex, race/ethnicity, smoking history, and household income were self-reported. Medication history, including the use of antihypertensive, antidiabetic, and lipid-lowering medications, was also self-reported. Smoking was defined as ever or never smoker. Blood pressure measurements were obtained, and the average reading from 3 in-home measurements and 3 mobile center measurements was used. Body mass index was computed as the weight in kilograms divided by the square of the height in meters. Obesity was defined as a body mass index ≥ 30 kg/m 2 . Diabetes was defined as a fasting plasma glucose level ≥ 126 mg/dl, glycosylated hemoglobin A1c value ≥ 6.5, or a history of antidiabetic medication use. (NHANES III, 1988-94).
Standard 12-lead electrocardiograms were recorded by trained technicians using a Marquette MAC 12 system (Marquette Medical Systems, Milwaukee, Wisconsin) during each participants' visit to a mobile examination center. Computerized automated analysis of the electrocardiographic data was performed with visual inspection of outlier values by a trained technician in a central electrocardiographic core laboratory. The calculation of CIIS and methodology has been previously described. (Rautaharju et al., 1981) Briefly, the score is defined by a set of 10 discrete binary (e.g., a single threshold) and ternary (e.g., high and low threshold) features in combination with 2 features measured in continuum (linear numerical values), and provides a simple scoring scheme suitable for both visual and computer classification of the conventional 12-lead electrocardiogram.
Features utilized to calculate CIIS scores include Q-wave duration and amplitude, R-wave amplitude, and T-wave amplitude, among others, in various precordial and limb leads. The CIIS values were multiplied by a factor of 10 in NHANES III to avoid using decimal points. We divided the reported CIIS values by 10 to remain consistent with prior studies. (ONeal et al., 2014) Subclinical myocardial injury was defined as CIIS values ≥ 10, representing the limit for abnormal CIIS. (Rautaharju et al., 1981).
Mortality data for NHANES III participants were available through December 31, 2006, and methods for mortality ascer- Participants who were unable to be matched with a death record were categorized as alive throughout follow-up.
Continuous variables were reported as mean ± standard deviation and categorical variables by their distributions. Statistical significance for continuous variables was tested using the t test procedure and the Rao-Scott chi-square method for categorical variables. Unadjusted CVD mortality rates (per 1,000 person-years) were calculated. Kaplan-Meier estimates were used to compute unadjusted survival estimates for CVD mortality, and the differences in estimates between presence and absence of SCMI were compared using the log-rank procedure. (Gray & Tsiatis, 1989) Cox proportional hazards regression was used to generate hazard ratios (HR) and 95% confidence intervals (95% CI) for the association between SCMI and CVD mortality. Multivariable-adjusted models were constructed | 3 of 7 BROUGHTON eT al. with incremental adjustments as follows: model 1 unadjusted; model 2 adjusted for age, sex, and total annual income; and model 3 adjusted for model 2, plus smoking, hypertension, diabetes, hyperlipidemia, and obesity. Statistical significance was defined as p ≤ .05.

Non-Hispanic Whites
Data were analyzed using SAS, version 9.3 (SAS Institute Inc).

| D ISCUSS I ON
In a large, diverse sample of US adults, the estimated prevalence of SCMI determined by CIIS values (CIIS ≥ 10) was highest among non-Hispanic Whites, followed by non-Hispanic Blacks, and then Mexican Americans. When compared to Whites, Blacks were just as likely to have SCMI, while Mexican Americans were less likely. After adjusting for potential confounding factors, the presence of SCMI in adults without baseline CVD was associated with an increased risk of CVD mortality in Mexican Americans, but not Whites or Blacks.
Prior studies have evaluated the prognostic significance of SCMI as determined by CIIS. (Dekker et al., 1994;Domburg et al., 1998;ONeal et al., 2014;Richardson et al., 2005;Siscovick et al., 1996) In a large cohort of apparently healthy men and women in the Netherlands, elevated CIIS values were associated with significantly higher rates of coronary heart disease and CVD mortality. (Dekker et al., 1994) In a Washington state population-based case-control study of patients free of clinical CVD, increasing CIIS values were linked with a higher risk of primary cardiac arrest among hypertensive patients. (Siscovick et al., 1996) High CIIS values (≥20) in a large series of postmyocardial infarction patients had an increased relative risk of total and CVD mortality at 1-and 3-year follow-ups.  an increased amount of CVD-related deaths among non-White individuals when compared to White individuals; however, the specific delineation of each non-White race was not identified. (ONeal et al., 2014) Additionally, prevalence and association between SCMI and race has not been investigated among a similar cohort.
Surveillance data have suggested that there are racial and socioeconomic disparities in the prevalence, morbidity, and mortality associated with CVD and major risk factors. For example, non-Hispanic Blacks have higher rates of CVD mortality and prevalence of hypertension, compared to non-Hispanic Whites and Mexican as there were no significant differences in baseline characteristics between race groups, and common confounders, such as demographics and traditional risk factors, were accounted for. Potential racial differences in access to care, proficiency of the English language, and health literacy were not explored among the study population, but these attributes are very likely to be playing a role in our results. It is also possible that there is a genetic component to the increased risk of CVD mortality in the presence of SCMI that may expose those of Mexican American background more than the other races in this study. Other previously established CVD risk factors, such as socioeconomic status (i.e., level of education, household income categorized by taxation rates, marital status, and others) were not included in this study. The inclusion of these additional socioeconomic factors might have also contributed to explaining our results, as there has been establishment of racial disparities within these categories. (Bell et al., 2018).
The electrocardiogram is a relatively inexpensive and noninvasive method of screening patients for underlying CVD that is regularly obtained. CIIS has been proven to accurately detect SCMI and has utility in prognostication of mortality among individuals without baseline CVD. (Dekker et al., 1994;Domburg et al., 1998;ONeal et al., 2014;Rautaharju et al., 1981;Richardson et al., 2005;Siscovick et al., 1996)  that could have changed with time (e.g., improvement of systolic blood pressure or cholesterol values with lifestyle modification).
Additionally, the participants' history was self-reported, subject to recall bias, and the classification of race does not account for those of bi-racial background. Lastly, CVD mortality was determined by

International Classification of Diseases codes and matching participant
identifiers (social security numbers, sex, date of birth), which may have been inadvertently misclassified.
In conclusion, we identified that among a diverse population of US adults without baseline CVD, Mexican Americans had a lower prevalence of SCMI and an increased risk of CVD mortality when compared to other race/ethnicities. These results implicate that addressing SCMI in Mexican Americans may enhance prevention and address cardiovascular disparities.

| E THI C S
The data utilized in this study from NHANES III have been approved by the NCHS Research Ethics Review Board, and documented consent was obtained from participants.