The Influence of Cardiac Rehabilitation on Inflammation and Metabolic Syndrome in Women with Coronary Heart Disease

Abstract

Background

Metabolic syndrome (MetS) and elevated inflammatory markers, both predictors of future cardiovascular events, are more prevalent in women with coronary heart disease (CHD). The influence of cardiac rehabilitation (CR) on MetS and inflammatory biomarkers is not well characterized for women.

Purpose

The purpose of this paper is to examine the effects of a 12-week behaviorally enhanced CR exclusively for women compared to traditional CR on components of the MetS and inflammatory markers in women with CHD.

Methods

The randomized clinical trial employed 2 treatment groups; both receiving a comprehensive 12-week CR program, with 1 group receiving a motivationally enhanced intervention exclusively for women. A subset of 91 women (mean age 61.6 years) from the parent study provided serum samples to examine the effects of CR on high-sensitivity C-reactive protein (hsCRP), interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), and intercellular adhesion molecule-1 (ICAM-1).

Results

Following CR, the total sample of women demonstrated significant reductions in hsCRP (P=.002), IL-6 (P<.001), TNF-α (P=.010), and ICAM-1 (P=.016). Women in the gender-tailored CR program, significantly improved all biomarker levels compared to baseline (P<.05 for all) while those in the traditional group improved only hsCRP (P=<.05) and IL-6 (P<.05) levels. The combined study group demonstrated improvements in several components of the MetS (triglycerides, waist circumference, and systolic blood pressure) but not others (high-density lipoprotein cholesterol, fasting glucose, and diastolic blood pressure).

Conclusion

Cardiac rehabilitation promotes greater improvements in inflammatory biomarkers than in components of MetS for women with CHD. Improvements in body composition or weight may not be a precondition for the benefits of exercise due to loss of abdominal fat. Examining components of MetS as continuous variables is recommended to prevent lost information inherent in dichotomization.

Coronary heart disease (CHD) remains the leading cause of disability and death in North American women.¹ Heart disease, a chronic multistage inflammatory disease, is influenced by environmental exposures, lifestyle factors, and genetic determinants that are reflected in traditional risk factors, inflammatory biomarkers, and metabolic status. Metabolic syndrome (MetS), a constellation of CHD risk factors, includes atherogenic dyslipidemia (elevated triglycerides and decreased high-density lipoprotein cholesterol (HDL-C)), hypertension, abdominal obesity, hyperglycemia, insulin resistance, and a proinflammatory state. ^2–6 While controversy continues regarding which operational definition is the most appropriate and clinically useful, ⁷ it is generally agreed that MetS confers increased risk for CHD⁸ and poor long-term prognosis.^9–12

Obesity, a central feature of MetS, is associated with a proinflammatory state⁷ mediated by dysregulated adipose tissue releasing numerous inflammatory cytokines.¹³ Inflammation, a principal driver of MetS, is more prevalent in women,^14–18 is a stronger predictor of CHD than in men ^{8, 19} and women with established CHD are at higher risk for adverse cardiovascular outcomes.²⁰ The most intensively studied inflammatory biomarkers include high-sensitivity c-reactive protein (hsCRP),^{21, 22, 23} an acute phase reactant, interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α), both inflammatory cytokines, and intercellular adhesion molecule-1 (ICAM-1),^{21, 24} a marker of endothelial function. C-reactive protein, produced by hepatocytes, is regulated by IL-6 and TNF-α and is found in the endothelium of atherosclerotic plaques, smooth muscle cells, macrophages, and adipocytes.^25–28 C-reactive protein affects the endothelium by altering the bioavailability of nitric oxide, thus, up-regulating ICAM-1.²⁹

Despite heated debates surrounding the clinical utility of MetS, the cornerstone for improving the components of MetS and inflammation includes weight reduction and exercise training.^30–33 Evidence supports the anti-inflammatory effects of exercise on biomarkers such as hsCRP in various settings and populations.^{34, 35} Cardiac rehabilitation (CR) programs consisting of prescriptive exercise, health education, and psychological counseling have compelling benefits for improving both physiological and psychosocial outcomes for individuals with CHD.^36–39 The effects of lifestyle modifications on inflammatory biomarkers have been examined ^40–43 with fewer exploring the effects of CR on MetS and inflammation in CHD participants.^{31, 44–48} Scientific inquiry of the impact of CR on MetS and inflammatory biomarkers in women is particularly meager.^{30, 44} The aim of this paper is to examine the effects of a 12-week behaviorally enhanced CR exclusively for women compared to traditional CR on components of MetS and inflammatory biomarkers in women with CHD.

Methods

Participants

Data analyzed here are from a sub-study of a larger randomized clinical trial that was already underway. Receipt of a supplemental funding award afforded us the opportunity to analyze blood samples from all women subsequently enrolled. Thus, the inflammatory biomarkers were available for analysis from 91 women (Figure 1). Recruitment information for the parent trial⁴⁹ and the methodological design ^50,51 are described elsewhere. The experimental design employed 2 treatment groups; both receiving a comprehensive, 12-week CR program, with 1 group receiving a motivational behavioral enhancement exclusively for women with CHD. The institutional review boards of the university and the study hospital approved the study.

Figure 1.

Flow of Participants in Study

Women in the sub-study were recruited from 2006 to 2008. Inclusion criteria were 1) diagnosed with an acute myocardial infarction (AMI), angina, or having undergone coronary artery bypass graft (CABG) surgery or percutaneous coronary intervention (PCI) within the last year; 2) able to read, write, and speak English; and 3) willing and able to participate. The exclusion criteria were 1) health insurance coverage for less than 36 ECG-monitored exercise sessions; 2) cognitive impairment; 3) inability to ambulate; or 4) insertion of an automatic internal cardiac defibrillator (AICD) in the last year.

Measures

Clinical and Physiological Characteristics

Physiological assessments including risk factors, anthropomorphic measures, medication use, fasting lipids and glucose tests, and a symptom-limited exercise tolerance tests (ETT) were conducted at baseline and after CR completion. Using the modified Bruce protocol for the ETT, ^52–54 peak exercise capacity was expressed in units of metabolic equivalents (METS) where a MET equals 3.5 ml oxygen consumption per kg per minute. Blood pressure (BP) was measured with one calibrated automated oscillometric BP monitor (Datascope, Mahwah, NJ) according to established guidelines.⁵⁵ Body mass index (BMI) was calculated as weight (kg)/height (m²). Body fat composition was determined using the mean of three measures of skin folds taken at three sites (suprailium, triceps, and thigh). The percent of body fat was calculated from standardized tables.⁵⁶ Lipid profiles and serum glucose were measured after a 12-hour fast using the Cholestech LDX System.

The National Cholesterol Education Program Adult Treatment Panel III (ATP III) criteria were used to define MetS and comprises at least 3 of the following criteria: waist girth >88 cm, triglycerides >150 mg/dL; HDL-C <50 mg/dL; blood pressure (BP) >130 mm Hg systolic and/or >85 mm Hg diastolic; and fasting glucose >110 mg/dL.⁵⁷ Following the lead of those who recommend avoiding lost information inherent in dichotomization,^{58, 59} the individual MetS components were analyzed as continuous variables.

Biomarkers

Blood samples (5 mL) were obtained in heparinized tubes after a 12-hour fast. Samples were centrifuged at 3500 rpm for 25 minutes at 0° C and serum samples were then stored in 1.5 ml microcentrifuge tubes and immediately placed at −80°C until analysis. The Luminex 200 IS system, a laser-based fluorescent analytic system that is used in conjunction with LINCOplex Bio-assays (Millipore Corporation, St. Charles, Missouri), was used to analyze hsCRP, IL-6, TNF-α, and ICAM-1. All assays were analyzed according to manufacturers’ protocols. All samples were run in duplicates and concentrations were analyzed using a 5-parameter logistic curve-fitting method. Measures of cytokines that were less than the minimum detectable concentration (<3.2 pg/mL) were set equal to half the detection level.⁶⁰

Serum hsCRP was measured using a LINCOplex Human Cardiovascular Disease kit. The minimum detectable concentration (sensitivity) is 6 pg/mL. The cutpoints of low risk (<1.0 mg/L), average risk (1.0 to 3.0) mg/L and high risk (>3.0 mg/L) correspond to tertiles of hsCRP in the adult population.⁶¹ Elevated hsCRP levels (>3.0 mg/L) add prognostic information to traditional risk factors of future CHD risk.^{62, 63} The Human Cytokine LINCOplex two-plex kit was used to measure TNF-α and IL-6 in serum samples. Sensitivities for TNF-α and IL-6 were 0.22 pg/mL and 0.79 pg/mL, respectively. ICAM-1 was measured using the Human Sepsis/Apoptosis LINCOplex kit with a minimum detectable concentration of 30.0 pg/mL. An overnight incubation of 18 hours at 4°C on a plate shaker set at 600 rpm was performed to provide better sensitivity for the analyses of TNF-α, IL-6, and ICAM-1. Manufacturer-provided quality controls for calculating intra- and interassay coefficients of variation (CVs) were used. The intra-assay CVs for hsCRP, TNF-α, IL-6, ICAM-1 were 8.0%, 8.0%, 13.6%, and 5.0, respectively. The interassay CVs for hsCRP, TNF-α, IL-6, ICAM-1 were 17.5%, 10.9%, 12.7%, and 8.7%, respectively.

Interventions

Traditional CR Intervention

The traditional CR program, nationally certified by the American Association of Cardiovascular and Pulmonary Rehabilitation (AACVPR), was delivered by nurses and exercise physiologists (EP) using a case management model. The exercise protocol consisted of combined aerobic exercise and upper body resistance training 3 days per week for 12 weeks. Exercise consisted of a 5 minute warm-up and 35–45 minutes of aerobic exercise (treadmill walking, cycling, or rowing) with exercise heart rates maintained at 60–85% of maximal heart rate calculated from their baseline ETT. Resistance training included wall-pulleys and hand weights followed by 5 minutes of cool-down exercises. Both the aerobic and resistance exercise training protocols were progressed weekly in duration and intensity according to participant’s functional capacity, response to exercise, and rate of perceived exertion. Eight hours of education classes (1 hour each week), focusing on CHD risk factor modification, were provided by the CR personnel.

Gender-Tailored Intervention

The gender-tailored exercise protocol was identical to that of the traditional CR program except that participants exercised exclusively with women in their cohort and the time of the intervention was restricted to one time slot when the traditional CR facility was closed to avoid crossover contamination. The psychosocial intervention, guided by the Transtheoretical Model (TTM) of behavior change ⁶⁴ and delivered with a motivational interviewing (MI) counseling style⁶⁵ was administered by female research nurses and EPs. Women were assessed for their motivational readiness to change regarding physical activity, stress reduction, and dietary modification. The TTM assessment provided an individualized 5-page report providing feedback on change processes relevant to the participant’s stage of change for each behavior that were subsequently reinforced by their case manager. Participants received 1-hour individualized MI sessions at weeks 1 and 6 with a clinical psychologist or a clinical nurse specialist (CNS) formally trained in MI. The CNS and psychologist facilitated weekly psychoeducational classes, totaling 10 hours, focused on gender-based practice guidelines, relaxation exercises, and social support. Details of the intervention are described elsewhere.⁵⁰ Both groups received a 30 minute consultation with a dietician.

Data Analysis

Data were analyzed using SPSS version 17 for Windows (SPSS, Inc, Chicago, Ill). Descriptive statistics included means, standard deviations, Pearson correlations, and percentages. Categorical variables were compared using the χ² test, and continuous variables were compared using the t-test. ANOVA was used to compare baseline to post-test changes between the 2 treatment groups. All tests were 2-tailed and evaluated for statistical significance using an α of .05 criterion. The inflammatory biomarkers and triglyceride levels were logarithmically transformed for statistical analyses and then back-transformed to their natural units for presentation in tables.

Results

Baseline Characteristics

With a mean age of 61.6±10 years (range, 42–82 years) most of the women were Caucasian (82%), married (51.6%), retired (49.5%), with ≥ high school education (93%). Women in this study qualified for CR because they had undergone either a PCI (48%) or CABG surgery (35.5%) or were diagnosed with an AMI (4.4%) or stable angina (12%). Medical record documentation revealed a previous AMI in 18%, a previous CABG in 8% and a previous PCI in 21% of participants. At baseline, women were taking evidence-based medications including β-blockers (78%), lipid lowering agents (92%), aspirin (88%), clopidogrel (67%), and angiotensin-converting enzyme inhibitors (32%). About 26% of women were taking nonsteroidal anti-inflammatory agents and 6.6% were on estrogen replacement. There were no group differences in medication use and medication consumption remained stable over the study period. At baseline, women randomized to the traditional CR program were not different than those randomized to the gender-tailored program regarding lipid profiles, anthropomorphic indices, weight, or exercise capacity (Table 1). Groups also demonstrated similar distributions of the components of MetS and inflammatory biomarkers at baseline (Table 2). However, those in the traditional compared to the gender-tailored group had higher baseline triglyceride levels (P<.05).

Table 1.

Physiological Characteristics Before and After Cardiac Rehabilitation by Intervention Group

Variable	Total Sample			Traditional (n=39)			Gender-Tailored (n=48)			F Group × Time
	Baseline	Post-CR	F Time	Baseline	Post-CR	F Time	Baseline	Post-CR	F Time
Total cholesterol mg/dL	162±38	152±35	7.40*	167±39	157±41	3.18	158±37	148±29	4.34*	<1
LDL-cholesterol mg/dL	92±34	84±29	5.33*	95.2±33	89±34	1.57	89±34	79.5±24	4.24*	<1
Total/HDL cholesterol ratio	4±2	3.6±1	9.31*	4.4±2	4±2	3.74	3.7±1.5	3.3±1	5.82*	<1
Body mass index, kg/m2	32±7	31±7	13.00*	32±7	31±7	8.40*	32±7	31±7	4.71*	<1
Body weight, pounds	183±43	180±42	10.53*	182±42	179±42	5.47*	184±44	181.5±43	5.08*	<1
Percent body fat	38.88±3	38.27±4	6.39*	38.75±3	38.41±4	<1	38.98±3	38.16±4	7.21*	1.14
Resting heart rate, bpm	72±9	70±9	3.62	72±8	71±10	<1	73±10	69±8	5.78*	1.68
Peak exercise capacity, METS	6±2	8±3	59.03*	6±2	8.1±3	36.74*	6±2.5	7.4±2	22.58*	1.74
Peak treadmill time, minutes	8.7±3	11±3	68.65*	9±3	11±3.5	31.23*	8.5±3	11±3	38.14*	<1

Abbreviations: bpm, beats per minute; CR, cardiac rehabilitation; HDL, high density lipoprotein; LDL, low density lipoprotein; METS, metabolic equivalent units

Results are presented as mean±SD.

^*p <.05

Table 2.

Metabolic Syndrome Components and Inflammatory Biomarkers Before and After Cardiac Rehabilitation by Intervention Group

Variable	Total Sample			Traditional (n=39)			Gender-Tailored (n=48)			F Group × time
	Baseline	Post-CR	F Time	Baseline	Post-CR	F Time	Baseline	Post-CR	F Time
Metabolic Syndrome
HDL-cholesterol mg/dL	43.4±13	45±13	2.66	41±12	42±11	<1	45±13	48±14	2.6	<1
Triglycerides, mg/dL	132±65	115±55	12.97*	154±76†	133±66	9.61*	113±47	101±39	3.76	1.01
Fasting glucose, mg/dL	101±25	103±22	<1	103±30	108±26	1.79	100±20	99±17	<1	1.09
Waist circumference, cm	100±15	97±15	39.59*	100±15	97±15	10.26*	100±15	97±14	34.16*	2.35
Systolic BP, mm Hg	118±12	115±12	4.54*	115±11	115±11	<1	121±12	115±12	7.91*	2.68
Diastolic BP, mm Hg	74±9	72±8	3.25	73±9	73±7	<1	75±9	71.5±8	8.35*	4.26*
Inflammatory Biomarkers
HsCRP, mg/L	5.8±8	3.6±5	9.67*	6±9	3.5±4	4.44*	6±8	4±5.5	5.33*	<1
Interleukin-6, pg/mL	18±47	8±15	13.87*	17±45	7±7	5.39*	18±48	8±19	8.93*	<1
TNF-α, pg/mL	10±22	6±5	6.23*	12±31	6±6	<1	8.5±9	5±4	6.99*	1.09
ICAM-1, pg/mL	211.5±117	177±81	5.04*	179±81	181±78	<1	228.5±134	173±83	14.11*	7.75*

Abbreviations: BP, blood pressure; CR, cardiac rehabilitation; HDL, high-density lipoprotein; HsCRP, high-sensitivity C-reactive protein; ICAM-1, intercellular adhesion molecule-1; TNF-α, tumor necrosis factor alpha

Results are presented as mean±SD.

^*P<.05;

^†P<.05, baseline comparison between groups

Effects of Cardiac Rehabilitation on Physiological Characteristics

We evaluated the effects of moderate-intensity exercise on the physiological characteristics for the entire sample of 87 women who completed CR; there were no significant between group differences (Table 1). Participants had a mean weight loss of 3±8.5 pounds, reduced their BMI by 0.6±1.5 kg/m², and reduced their body fat by 2±9% (P<.05 for all). The total sample achieved a 25% increased peak exercise capacity (2±2 metabolic equivalents) and an increased treadmill time by 2±2 minutes (P<.05 for both). The improved exercise capacity and weight loss was accompanied by a mean reduction in total cholesterol (TC) of 10.2±34 mg/dL, a mean reduction in low-density lipoprotein cholesterol (LDL-C) of 8.4±33 mg/dL, and a mean reduction in the TC/HDL-C ratio of 0.4±1.2 (P<.05 for all). Compared to baseline, participants of the gender-tailored intervention demonstrated significant improvements in all physiological parameters. The traditional CR intervention, in contrast, led to no improvements in TC, HDL-C, TC/HDL-ratio, or percent body fat.

Influence of Cardiac Rehabilitation on Components of the Metabolic Syndrome

The effects of 12 weeks of moderate-intensity exercise on components of the MetS and the inflammatory biomarkers were examined next (Table 2). For the combined study group, CR led to mean reductions in triglycerides of 17±44 mg/dL (P<.05), waist circumference of 3±4 cm (P<.05), and systolic BP of 3±14 mm Hg (P<.05). Improvements were not found for HDL-C, glucose, or diastolic BP. Compared to baseline, the gender-tailored participants had significantly reduced systolic and diastolic BP (P<.05) at follow-up with no detectable changes in the traditional group participants. Both interventions yielded similarly reductions in waist circumference (P<.05). The only between-group difference found was that that the gender-tailored group had greater reductions in diastolic BP over time compared to the traditional group participants (P<.05).

The Influence of Cardiac Rehabilitation on Inflammatory Biomarkers

After CR participation, the combined study group demonstrated significant improvements in hsCRP (P=.003), ICAM-1 (P=.027), IL-6 (P<.001) and TNF-α (P=.014) (Table 2). Mean baseline hsCRP levels decreased 40% from 5.8 mg/L to 3.6 mg/L after CR. Baseline TNF-α levels of 10 pg/mL were also reduced by 40% to 6 pg/mL after CR, and IL-6 levels were reduced by 55.5%, from a mean of 18 pg/mL to 8 pg/mL. ICAM-1 was more modestly reduced by a mean of 29±102 pg/mL (15%) after CR participation. The gender-tailored intervention produced significant reductions in all inflammatory biomarkers (P<.05 for all) while the traditional CR intervention led to lower hsCRP and IL-6 but not TNF-α or ICAM-1. The gender-tailored intervention led to greater reductions in ICAM-1 than the traditional CR intervention (P<.05).

To gain insight into differential effects of the 2 CR groups on the MetS components and inflammation we examined attendance by intervention group. The traditional CR participants attended a mean of 28±13 (78%) of the prescribed 36 exercise sessions while women in the gender-tailored group attended a mean of 33±9 (92%) (F=4.285, df (1, 89), P=.041). Compared to women in traditional CR, those in the gender-tailored intervention also attended a greater percentage (88.5±23 versus 62±31) of prescribed education sessions (F=20.794, df (1, 89), P<.001). All participants in the gender-tailored group completed the intervention while 4 in the traditional group did not; these noncompleters attended a mean of 6 sessions (range, 1–13).

Discussion

The major finding of this study is that 12 weeks of moderate-intensity exercise in the CR setting produces important improvements in markers of inflammation including hsCRP, IL-6, TNF-α, and ICAM-1 in women with CHD. Examination of the individual components of the MetS as continuous variables revealed that CR positively influenced abdominal obesity, triglyceride levels, and systolic BP. While both groups lost weight, only the gender-tailored group reduced their TC, LDL-C, and TC/HDL ratio. Both groups achieved significantly improved peak exercise capacity and treadmill time after completion of CR. Compared to traditional CR, the gender-tailored intervention was superior in reducing diastolic BP and ICAM-1 concentrations. Compared those in traditional CR, the gender-tailored intervention participants attended more prescribed exercise and education sessions. Explanations for greater attendance and reduced diastolic BP and ICAM-1 levels in the gender-tailored intervention remain speculative. The study was not designed to isolate the effect of any one of the intervention’s multifaceted components on the outcomes. It is, therefore, unknown which intervention component led to the statistically significant differences on these variables. That the intervention used motivational interviewing, stage-matching, gender-tailoring, and social support may have synergistically improved some of the outcomes.

At baseline, most women were taking evidence-based medications to treat components of the MetS. Not surprising then, baseline levels of BP, triglycerides and glucose fell at or below the ATP III cutpoint criteria for MetS leaving little room for improvement after CR. This potential floor effect for various MetS components presents a challenge for demonstrating improved outcomes in secondary prevention settings where patients may enter CR with relatively well-controlled BP and lipid profiles. Reduced waist circumference was observed after CR participation. Had we dichotomized waist circumference according to the ATP III criteria (< 88 cm versus >88 cm), this improvement would have been obscured because the mean value after CR remained above the somewhat arbitrary cutpoint of 88 cm.

Robust evidence links inflammation with central obesity and MetS,^{13, 19, 67} and with increased CHD risk.⁶⁸ Despite modest improvements in MetS components, we observed impressive reductions in the inflammatory protein and cytokine concentrations. The mechanism underlying these improvements remains elusive. While evidence exists that exercise was shown to independently lower CRP as much or more than statins,³⁵ others report a weak relationship between fitness and CRP in patients using medications that modify inflammation.⁶⁹ Independent of weight, regular muscle movement during exercise may modify inflammation locally while systemic inflammation is reduced through muscle-derived cytokines.³³ The anti-inflammatory effect of exercise in women with CHD may be even greater because of their lower baseline exercise capacity and higher inflammatory levels.⁴² Evidence points to an inverse relationship between inflammatory markers and cardiorespiratory fitness after adjustment for adiposity.⁷⁰ Even though the majority of our study participants were taking statins and aspirin therapy, they attained substantially improved inflammatory biomarkers; yet post-CR hsCRP remained in the high risk category (>3 mg/L).

Our findings of reduced inflammatory biomarkers complement and extend the few studies that have examined the influence of CR on inflammation.^{31, 44–47} Our results support the premise that exercise training need not result in substantial weight loss to positively influence inflammation,^{30, 42, 45, 71} particularly in the CR setting where weight loss is typically minimal. Milani and colleagues ^{30, 44} found that CR for 235 patients (32% women) led to a 36% reduction in hsCRP independent of weight changes. Goldhammer et al. ⁴⁷ also reported that CR for 28 CHD patients (10 women) reduced hsCRP by 48% and IL-6 by 42%. Caulin-Glaser and associates ⁴⁵ assessed the effects of CR on hsCRP in 172 patients with CHD (22% women). Though not stratified by gender, they found reductions in hsCRP from 5.7 to 2.7 mg/L. Lavie and colleagues ⁴⁴ examined the effects of CR on hsCRP and MetS in 72 lean patients (BMI <25 kg/m²) and 73 obese patients (BMI > 30 kg/m²) with baseline hsCRP levels of 5.4 and 7.4 mg/L, respectively. After CR, the obese, but not the lean participants, attained a 43% reduction in hsCRP and a 1.5% reduction in BMI with little change in abdominal girth. This was accompanied by improvements in HDL-C (36.7 to 39.5) but no improvements in glucose, BP, or triglycerides.

Ades and colleagues⁴⁸ compared outcomes of 34 (8 women) participants randomized to a 5-month traditional CR program compared to 37 (5 women) randomized to a high-caloric-expenditure exercise protocol. While they found that the high-calorie-expenditure exercise resulted in double the weight loss and a 16% reduction in hsCRP, they found no group differences in the MetS components of BP, triglycerides, HDL-C or glucose. Our study expands on their work with our inclusion of only women who underwent moderate-intensity exercise. The generalizability of their results to women is tentative given the low exercise capacity in the majority of women enrolled in CR. ⁷²

Limitations of this study merit consideration. First, data must be interpreted with caution because individuals receiving statins may show falsely low hsCRP concentrations.⁷³ Second, the baseline biomarker levels demonstrated greater variability than those obtained after CR. The reduction in all the biomarker levels after CR likely represents a restricted range effect. We found no relationship between statin or aspirin therapy and inflammatory biomarkers (likely restricted range). Third, we did not exclude patients with undiagnosed inflammatory processes such as periodontitis, or subclinical infections. Finally, there are important genetic and environmental determinants of each of the measured biomarkers and metabolic profiles for which we have not accounted.

Conclusions

Inflammation plays a key role in metabolic risk and the progression of CHD. While we found few between-group differences, both the gender-tailored and traditional CR programs were effective in reducing inflammatory biomarkers in women. Beyond the benefits of evidence-based medications, CR led to reductions in some components of MetS (waist circumference and triglycerides) but not others. The sustainability of the observed reductions in inflammation and the consequences for disease progression require closer examination. Our study supports the benefits of moderate-intensity exercise for reducing inflammation and central adiposity and improving cardiorespiratory fitness in women with established CHD. Identifying the gene-gene and gene-environment interactions that influence inflammatory biomarkers, obesity, and MetS will increase our understanding of the biological explanations for disease progression and responses to treatment.

Summary and Implications

• Cardiac rehabilitation improved measures of inflammation in women with coronary heart disease.

• Cardiac rehabilitation typically leads to minimal weight loss but improvements in body composition may not be a precondition for the benefits of exercise perhaps due to reduced visceral fat.

• Cardiac rehabilitation improved some but not all components of the metabolic syndrome.

• Cardiac rehabilitation for women significantly reduced the mean waist circumference although it remained above the recommended limit of 88 cm.