METHODS

Subjects were drawn from the coronary artery surgery inflammation study (CASIS). Briefly, all patients undergoing elective first time CABG at the Middlesex Hospital, London, UK, between October 1999 and September 2000 were invited to participate. The study had hospital ethics committee approval, and all patients gave written informed consent.

Subjects undergoing additional surgical procedures (such as valvar surgery or aneurysmectomy) were excluded.

All patients were taking aspirin 75 mg/day, and ceased taking this 10 days before surgery. Other drugs were withheld on the morning of surgery. ACE inhibitor treatment was given as usual the night before surgery. No patient had clinical evidence of active inflammatory or immunomodulatory disease (such as acute coronary syndromes, malignancy, intercurrent infection, renal failure, or autoimmune disease) at recruitment, nor were any receiving anti-inflammatory agents. Similarly, before analysis all data were excluded from those subjects who experienced significant postoperative complications, even if these were only apparent after the six hour sample time point, as IL-6 rises within hours of an inflammatory stimulus. Study exclusion criteria thus included infections requiring antibiotic treatment, prolonged respiratory support, circulatory support, or renal failure requiring haemofiltration.

Surgical procedure

CABG was performed through a midline sternotomy by one of four consultant surgical staff, with subsequent hypothermic cardiopulmonary bypass employing right atrial and ascending aortic cannulation. Myocardial protection was maintained by intermittent cold cross clamp fibrillation, and heparin anticoagulation was reversed postoperatively by protamine sulfate.

A 4.5 ml citrated blood sample was drawn before surgery (with the patient supine and resting) and again six hours after cardiopulmonary bypass. Venous blood was immediately centrifuged (at 3500 g for 10 minutes), and the plasma was separated and stored at −20°C until analysis. IL-6 concentrations were measured by enzyme linked immunosorbent assay (R&D Systems, Abingdon, UK) by staff blind to all subject data. Interassay and intra-assay coefficients of variation were 5% and 3%, respectively, and assay sensitivity was < 0.70 pg/ml).

Statistical analysis

All data were analysed using SPSS for Windows version 9 (SPSS Inc, Chicago, Illinois, USA). We analysed raw data when they were normally distributed. However, the IL-6 values were skewed, so they were normalised by log transformation (the data shown represent geometric means and standard deviations). The effect of ACE inhibitor treatment on IL-6 was assessed by analysis of variance and Student's t test for unpaired data. One way analysis of covariance was performed using age, sex, smoking, diabetic status, left ventricular ejection fraction, statin treatment, duration of operation, and cardiopulmonary bypass and aortic cross clamp times as covariates. Following backwards stepwise linear regression modelling, only variables that had a significant effect on IL-6 concentrations were included in the final model. Pearson's correlation coefficient was used to test the relation between IL-6 concentrations and baseline clinical and surgical variables. The study was powered to detect a 10% difference in post-CABG IL-6 with 80% power at p < 0.05.

RESULTS

Of 269 elective CABG cases, 172 who fulfilled the inclusion criteria were randomly approached (depending on investigator availability), and of these 161 chose to participate. At enrolment, 41 (25%) were receiving ACE inhibitor treatment: 11 (27%) were on lisinopril, 10 (24%) on ramipril, six (15%) on enalapril, six (15%) on perindopril, four (10%) on captopril, and four (10%) were taking other agents. In all, 111 subjects (67%) were receiving statin treatment. Twenty one patients experienced confounding postoperative complications and were thus excluded, according to protocol. This left a final study group of 140 subjects, including 33 (24%) receiving ACE inhibitors.

Clinical and operative characteristics (table 1) were largely independent of ACE inhibitor treatment, although body mass index was slightly lower. As anticipated, a past medical history of diabetes or hypertension was more common in ACE inhibitor treated patients. Similarly, a higher proportion of subjects receiving ACE inhibitors had a left ventricular ejection fraction below 50% (18/33 v 29/107, p < 0.01).

The correlation between baseline and post-CABG IL-6 values and the major clinical and surgical variables is shown in table 2. While both baseline and post-CABG IL-6 concentrations were lower in non-diabetic patients, men, and non-smokers, none of these factors was associated with a significant effect on IL-6 concentrations; however, post-CABG IL-6 concentrations were lower in patients receiving statin treatment (p = 0.029).

The effect of ACE inhibitor treatment on IL-6 concentrations is shown in fig 1. A trend towards lower preoperative IL-6 concentrations among patients receiving ACE inhibitors (mean (SD), 3.7 (0.1) v 4.3 (0.1) pg/ml, p = 0.12) became less strong after multivariate analysis. Among the group as a whole, surgery was associated with a significant increase in IL-6 concentrations (mean rise, 177 (12) pg/ml, p < 0.0005 by paired t test). However, this response was significantly blunted among ACE inhibitor treated patients: median increases in IL-6 concentrations were 117 pg/ml in treated patients and 193 pg/ml in non-treated patients (Kruskal–Wallis, p = 0.02), with peak postoperative IL-6 concentrations lower among the 33 patients receiving ACE inhibitors than in the 107 untreated subjects (142 (19) v 196 (13) pg/ml, p = 0.02). The effect of ACE inhibitor treatment on post-CABG IL-6 concentrations remained significant after multivariate analysis (analysis of covariance, p = 0.018).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Effect of ACE inhibitor treatment on post-coronary artery bypass graft IL-6.

DISCUSSION

While the anti-inflammatory role of ACE inhibitors has been reported previously in vitro,^{35, 36} this study suggests for the first time that ACE inhibitor treatment is also associated with a reduction in IL-6 response to coronary artery bypass surgery. Although observational in nature and not derived from a randomised double blind control study, these data have important therapeutic implications in at least three fields.

Firstly, the pronounced inflammatory response to CABG is implicated in the pathogenesis of subsequent hyperdynamic circulatory instability, delayed myocardial recovery, and organ dysfunction.^{33, 37} More specifically, prolonged increases of circulating IL-6 are also associated with increased morbidity and mortality after cardiac operations.³⁸ It has already been shown that ACE inhibition may improve outcome in patients undergoing CABG,³⁹ although a mechanistic explanation for this observation remains elusive. Our data would support the suggestion that inherent anti-inflammatory properties of ACE inhibitor drugs may be responsible.³⁵ However, as the study was designed only to include subjects with an uncomplicated postoperative course, it is unsurprising that there was no significant difference in the length of hospital stay between those patients who did or did not receive an ACE inhibitor.

Secondly, there is evidence of cross signalling between the renin–angiotensin system and IL-6, so the former may contribute to the pathogenesis of atherosclerosis. Angiotensin II stimulates IL-6 production by smooth muscle cells, an effect that can be inhibited by captopril and ramiprilat.³⁶ Steady state mRNA for IL-6 can also be augmented after stimulation with angiotensin II, suggesting regulation of angiotensin induced IL-6 release at the pretranslational level.³⁶ Moreover, angiotensin II can also activate the pro-inflammatory transcription factor nuclear factor κB, which is necessary for transcription of most cytokine genes, including IL-6.⁴⁰ Thus it is likely that angiotensin can elicit an inflammatory response in human vascular smooth muscle cells by stimulating cytokine production and the activation of nuclear factor κB.³⁶ Our data would support an anti-inflammatory role for ACE inhibition in reducing vascular event rates in those at risk.³

Finally, the dramatic benefit of ACE inhibition in patients with heart failure remains to be explained, although our data once again support a potential anti-inflammatory role in mediating those effects.⁴¹

It remains possible, of course, that it is not the ACE inhibition on its own that is associated with a reduced IL-6 response, but the condition that led to such treatment in the first place. However, we think that unlikely.⁴² If anything, the inflammatory response to surgery might be expected to be greater among patients with pre-existing cardiac dysfunction or diabetes.⁴³ In any event, with the increasing indications for ACE inhibitor treatment (hypertension, diabetes,⁴⁴ coronary artery disease,³ and cardiac dysfunction^{1, 2}), there are ethical concerns about performing a randomised controlled trial of ACE inhibition in patients selected for coronary surgery.⁴⁴

Clearly further studies are required, both in vitro, to investigate the effect of ACE inhibitor on IL-6 production using preparations of cells implicated in IL-6 synthesis (such as macrophages, fibroblasts, and endothelial cells), and in vivo, to explore the effect of ACE inhibition on the IL-6 response to other types and grades of inflammatory stimulus—for example, following acute coronary syndromes or coronary intervention. Any such studies need to be both randomised and double blind in design. Nonetheless, these data are supportive of an in vivo anti-inflammatory effect of ACE inhibition in humans.