Association between inflammation, body mass index, and long-term outcomes in patients after percutaneous coronary intervention: A large cohort study

To the Editor: Despite optimal secondary prevention treatments, a large proportion of patients with coronary artery disease (CAD) after percutaneous coronary intervention (PCI) are still at a high risk of recurrent cardiovascular events. Inflammation is a well-known component of residual cardiovascular risk and contributes to progression of atherosclerosis, leading to destabilization and rupture of atheroma plaques.^[1] As a downstream protein in the activated inflammatory pathway that mediates the progression of atherosclerosis, high-sensitivity C-reactive protein (hsCRP) is widely used as a biomarker of inflammatory status and predictor of adverse outcomes in the settings of primary and secondary prevention of cardiovascular disease. Obesity is regarded as a growing global burden and also a remarkable risk factor for cardiovascular disease. Several studies have previously proposed and tested the hypothesis that body mass index (BMI) could modify the association between hsCRP and cardiovascular risk.^[2,3] However, the relationship is not clear in CAD patients. Therefore, we aimed to examine the interrelationship of BMI, hsCRP, and long-term outcomes after PCI based on a large, prospective cohort.

Between January and December 2013, a total of 10,724 patients undergoing PCI were enrolled at Fuwai Hospital, Chinese Academy of Medical Sciences (Beijing, China). Patients without hsCRP or follow-up data and those with high inflammatory risk (hsCRP >10 mg/L), acute myocardial infarction, or BMI <18.5 kg/m² were excluded, leaving 7396 patients for analysis. The study was approved by the Institutional Review Board of Fuwai Hospital (Approval No. 2021–1501) and conducted in accordance with the Declaration of Helsinki. All participants provided written informed consent.

The primary endpoint was all-cause mortality. The secondary endpoints were cardiac mortality and major adverse cardiac and cerebrovascular events (MACCE), a composite of all-cause mortality, myocardial infarction, revascularization, and stroke. All mortality events were considered cardiac mortality unless there was an unequivocal non-cardiac cause. Participants were interviewed at a clinic visit or by telephone at 1 month, 6 months, 1 year, 2 years, and 5 years after discharge.

BMI was calculated as weight (kilogram) divided by the square of height (meter squared). High hsCRP was defined as >3 mg/L. Height and weight data and blood samples were collected within 24 h after admission for PCI. The hsCRP level was measured by an immunoturbidimetry (Array 360, Beckman Coulter, Brea, CA, USA) at the biochemistry center of Fuwai Hospital.

Categorical variables are reported as percentages and continuous variables as the mean ± standard deviation or the median (interquartile range). Categorical data were examined for significance using the chi-squared test and continuous data using one-way analysis of variance or the Kruskal–Wallis H test. Event rates are presented as Kaplan–Meier curves and were compared using the log-rank test. Multivariable Cox proportional hazards regression models were performed to evaluate the relationship between hsCRP and clinical outcomes, as well as interactions between hsCRP and BMI. All multivariable models were adjusted for age, sex, left ventricular ejection fraction, prior PCI/coronary artery bypass grafting, serum creatinine, diabetes, hypertension, smoking, triglyceride and low-density lipoprotein cholesterol (LDL-C) levels, and statin use. A collinearity analysis was performed to ensure that variables in the models had a low-degree of collinearity. All covariates were associated with a variance inflation factor of <2 [Supplementary Table 1,https://links.lww.com/CM9/B563]. Proportional hazard assumption was assessed by Schoenfeld residuals. The 7396 eligible patients were divided into three groups according to the Chinese criteria for BMI: normal weight (18.5 kg/m² to <24.0 kg/m², n = 1870), overweight (24.0 kg/m² to <28.0 kg/m², n = 3762), and obese (≥28.0 kg/m², n = 1764). Obese patients tended to be younger and male and to have higher rates of prior PCI. Traditional cardiovascular risk factors including CAD family history, diabetes, hypertension, hyperlipidemia, and smoking were also more common in the obese group. They were also more likely to have higher levels of hsCRP, creatinine, LDL-C and triglyceride levels, and greater number of medications prescribed at discharge [Supplementary Table 2,https://links.lww.com/CM9/B563].

The median follow-up was 5.1 years (response rate, 91.5%). During follow-up, 276 (3.7%), 1620 (21.9%), and 155 (2.1%) patients experienced all-cause mortality, MACCE, and cardiac mortality, respectively. An hsCRP>3 mg/L was significantly associated with higher risks of all-cause mortality (hazard ratio [HR] 1.35; 95% confidence interval [CI] 1.01–1.79; P = 0.04), MACCE (HR: 1.20; 95% CI: 1.06–1.35; P = 0.004), and cardiac mortality (HR: 1.54; 95% CI: 1.08–2.21; P = 0.02) after multivariable adjustment.

Table 1 showed the outcomes of the hsCRP dichotomy (≤3 mg/L vs. >3 mg/L) in the three BMI groups. Both univariable and multivariable analyses showed that high hsCRP was associated with significantly higher risks of all-cause mortality (adjusted HR: 2.20; 95% CI: 1.34–3.60; P = 0.002), MACCE (adjusted HR: 1.37; 95% CI: 1.07–1.75; P = 0.01), and cardiac mortality (adjusted HR: 2.30; 95% CI: 1.22–4.34; P = 0.01) in the normal weight group. However, these associations weakened as BMI increased, with no significant associations in the obese group. There was a significant interaction for all-cause mortality between BMI and hsCRP dichotomy (P = 0.04).

Table 1 - Prognostic value of hsCRP dichotomy for endpoints across BMI categories.

Outcomes	BMI groups	hsCRP levels (mg/L)	Event/total (%)	Crude HR (95% CI)	P value	Adjusted HR (95% CI)*	P value
All-cause mortality
	Normal	≤3	60/1574 (3.8)	Reference	NA	Reference	NA
		>3	25/296 (8.4)	2.32 (1.45–3.70)	<0.001	2.20 (1.34–3.60)	0.002
	Overweight	≤3	95/3067 (3.1)	Reference	NA	Reference	NA
		>3	32/695 (4.6)	1.46 (0.97–2.21)	0.07	1.32 (0.86–2.01)	0.20
	Obese	≤3	49/1341 (3.7)	Reference	NA	Reference	NA
		>3	15/423 (3.5)	0.90 (0.50–1.62)	0.72	0.76 (0.40–1.43)	0.39
MACCE
	Normal	≤3	345/1574 (21.9)	Reference	NA	Reference	NA
		>3	88/296 (29.7)	1.42 (1.13–1.80)	0.003	1.37 (1.07–1.75)	0.01
	Overweight	≤3	635/3067 (20.7)	Reference	NA	Reference	NA
		>3	176/695 (25.3)	1.25 (1.06–1.48)	0.008	1.20 (1.01–1.42)	0.04
	Obese	≤3	280/1341 (20.9)	Reference	NA	Reference	NA
		>3	96/423 (22.7)	1.09 (0.87–1.38)	0.46	1.07 (0.84–1.37)	0.57
Cardiac mortality
	Normal	≤3	34/1574 (2.2)	Reference	NA	Reference	NA
		>3	15/296 (5.1)	2.41 (1.32–4.43)	0.004	2.30 (1.22–4.34)	0.01
	Overweight	≤3	45/3067 (1.5)	Reference	NA	Reference	NA
		>3	18/695 (2.6)	1.78 (1.03–3.07)	0.04	1.60 (0.92–2.78)	0.10
	Obese	≤3	31/1341 (2.3)	Reference	NA	Reference	NA
		>3	12/423 (2.8)	1.21 (0.62–2.36)	0.57	0.98 (0.48–2.00)	0.95

BMI: Body mass index; CI: Confidence interval; HR: Hazard ratio; hsCRP: High-sensitivity C-reactive protein; LDL-C: Low-density lipoprotein cholesterol: MACCE: Major adverse cardiac and cerebrovascular events; NA: Not available; PCI: Percutaneous coronary intervention. ^*Cox proportional hazards regression model was adjusted for age, sex, left ventricular ejection fraction, prior PCI/coronary artery bypass grafting, serum creatinine, diabetes, hypertension, smoking, triglyceride, LDL-C, and statin. P for interaction between BMI categories and hsCRP (≤ 3 mg/L or >3 mg/L): 0.04 for all-cause mortality; 0.40 for MACCE; and 0.17 for cardiac mortality.

The cumulative incidence of all endpoints in patients with high hsCRP and normal weight was significantly higher than in other groups during 5.1 follow-up years [Supplementary Figure 1,https://links.lww.com/CM9/B563]. Compared with the reference group (normal weight, hsCRP ≤3 mg/L), individuals of normal weight with an hsCRP >3 mg/L were at significantly higher risks of all-cause mortality (HR: 2.06; 95% CI: 1.27–3.34; P = 0.003), MACCE (HR: 1.33; 95% CI: 1.05–1.70; P = 0.02), and cardiac mortality (HR: 2.24; 95% CI: 1.21–4.16; P = 0.01) [Supplementary Table 3, https://links.lww.com/CM9/B563].

In the sex-stratified analyses, a significant association between elevated hsCRP (>3 mg/L) and risk of all-cause mortality was only observed in the male group (HR: 1.45; 95% CI: 1.05–2.00; P = 0.03), but not in the female group (HR: 1.07; 95% CI: 0.58–1.97; P = 0.84). The interaction test between sex and hsCRP level (≤3 mg/L vs. >3 mg/L) for all-cause mortality was 0.42. Compared with the reference group (hsCRP ≤3 mg/L and normal weight), the risk of all-cause mortality was significantly higher in men with elevated hsCRP levels and normal BMI (HR: 1.99, 95% CI: 1.13–3.50; P = 0.02); however, the association was not significant among female groups (HR: 2.45, 95% CI: 0.97–6.19; P = 0.06) [Supplementary Table 4,https://links.lww.com/CM9/B563]. The interaction coefficient between BMI categories and hsCRP level (≤3 mg/L vs. >3 mg/L) for all-cause mortality risk was 0.16 in men and 0.09 in women.

In this large, real-world, prospective cohort of patients undergoing PCI, we found that BMI modified the effect of hsCRP on clinical outcomes during a 5-year follow-up. Patients with elevated hsCRP and normal weight were at highest risk of adverse outcomes. The results were similar after stratification by sex.

Numerous studies have demonstrated the prognostic value of hsCRP in secondary prevention, indicating that inflammation plays a key role in recurrence of cardiovascular events. The Canakinumab Anti-inflammatory Thrombosis Outcome Study, which confirmed the efficacy of anti-inflammatory therapy in CAD patients with elevated hsCRP levels, also provided strong evidence for the inflammatory hypothesis.^[4] Interestingly, obese individuals were found to have a higher hsCRP level, which may reflect excessive secretion of interleukin-6 by adipose tissues.^[5] Therefore, the predictive value of hsCRP in obese CAD patients may be overestimated given that a proportion of elevated hsCRP is induced by adiposity. Moreover, obesity could also be a protective factor. Accumulating epidemiological evidence shows a U-shaped association between BMI and mortality risk in patients with cardiovascular disease. The finding that obese or overweight patients with cardiovascular disease have a better prognosis is known as the "obesity paradox".^[6]

This study has several limitations. First, it had an observational cohort design and was conducted at a single center in China. Therefore, our findings may not be generalizable to other populations. Second, although multivariable models were used to adjust for common confounders, potential risk factors could not be fully adjusted for in view of the observational study design. Third, hsCRP and BMI were measured only at baseline, which may not represent the residual risk over a long period. Fourth, BMI is only a crude marker of body fat when compared with other measurements (e.g., body adiposity index or waist-to-hip ratio).

In conclusion, high hsCRP was associated with worse long-term outcomes after PCI, especially in patients with normal weight. An hsCRP-associated poor prognosis may be modified by BMI, which could help to identify individuals at higher residual inflammatory risk who may benefit from further anti-inflammatory treatment.

Funding

This research was supported by grants from the National Clinical Research Center for Cardiovascular Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences (No. NCRC2020013) and CAMS Innovation Fund for Medical Sciences (No. 2020-12M-C&T-B-049).

Conflicts of interest

None.