Hot Line - The power of meta-analyses – answering clinically important questions in CVD

In today’s Hot Line session, 4 important questions in CVD were answered using large individual patient-level data meta-analyses.

Firstly, Professor Colin Baigent (University of Oxford - Oxford, UK) presented a meta-analysis to resolve uncertainties around the possible effects of statins on muscle symptoms. Data were compiled from 23 Cholesterol Treatment Trialists’ Collaboration double-blind trials: 19 trials of statins versus placebo (123,940 patients) and 4 trials of intensive versus less intensive statin therapy (30,724 patients). In the statin versus placebo trials, during a median follow up of 4.3 years, muscle pain or weakness was reported by 27.1% of patients on statins and 26.6% on placebo (rate ratio [RR] 1.03; 95% CI 1.01 to 1.06). In the first year, there was a 7% relative increase in reports of muscle pain or weakness with a statin (RR 1.07; 95% CI 1.04 to 1.10), which corresponded to an absolute excess rate of 11 (95% CI 6 to 16) reports per 1,000 person years. In the remaining follow-up period, there was no evidence of any excess risk (RR 0.99; 95% CI 0.96 to 1.02). High-intensity regimens (e.g. atorvastatin 40 to 80 mg or rosuvastatin 20 to 40 mg) resulted in a larger relative increase in the rate of muscle pain or weakness than moderate-intensity regimens, with rate ratios of 1.08 (95% CI 1.04 to 1.13) and 1.02 (95% CI 1.00 to 1.05), respectively. Prof. Baigent points out, “Statins were not the cause of muscle pain in more than 93% of patients who reported symptoms. Statin therapy marginally increased the frequency, but not the severity, of muscle-related symptoms. The small excess risk of muscle symptoms occurred principally during the first year after commencing therapy.”

Doctor Alex Pitcher (University of Oxford - Oxford, UK) then presented a meta-analysis on behalf of the Marfan Treatment Trialists’ Collaboration, which aimed to assess the effects of angiotensin receptor blockers (ARBs) and beta-blockers on aortic root size in patients with Marfan syndrome. The analysis included individual data on 1,442 patients with no prior aortic surgery from 7 trials, where more than 80% of genotyped patients carried an FBN1 pathogenic variant. In 4 trials comparing an ARB with placebo or open control, ARBs approximately halved the annual rate of change in the aortic root Z score (+0.07 with ARBs versus +0.13 with control; absolute difference: –0.07; 95% CI –0.12 to –0.01; p=0.012). In 3 trials comparing an ARB with a beta-blocker, the annual change in the aortic root Z score was similar in the two groups (absolute difference ARB minus beta-blocker 0.03; 95% CI –0.05 to 0.10). When results from the two analyses were used to indirectly evaluate the effect of a beta-blocker compared with control, beta-blockers were estimated to approximately halve the annual change in the aortic root Z score (absolute difference: –0.09; 95% CI 0.18 to –0.0033; p=0.04). Dr. Pitcher remarks, “Our results suggest that ARBs and beta-blockers have similar, substantial and independent effects on reducing the aortic root Z score which, if maintained over a period of several years, would be expected to delay the need for elective aortic root surgery. The findings indicate that combination therapy, where tolerable, may benefit patients with Marfan syndrome.”

Attention then switched to the relative merits of aspirin versus P2Y12 inhibition in patients with coronary artery disease. Presented by Professor Marco Valgimigli (Cardiocentro Ticino Foundation - Lugano, Switzerland), the PANTHER meta-analysis of 7 trials included 12,178 patients assigned to P2Y12 inhibitor monotherapy (clopidogrel: 62.0%; ticagrelor: 38.0%) and 12,147 patients on aspirin monotherapy. With median treatment duration of 557 days, the risk of the primary efficacy outcome of CV death, myocardial infarction (MI) and stroke was lower with P2Y12 inhibitor monotherapy compared with aspirin monotherapy (5.5% versus 6.3%; hazard ratio [HR] 0.88; 95% CI 0.79 to 0.97; p=0.014). The number needed to treat to prevent one adverse outcome was 123 patients. The relative risk reduction in the primary composite endpoint was primarily driven by a 23% relative reduction in MI (2.3% versus 3.0%; HR 0.77; 95% CI 0.66 to 0.90; p<0.001). Of note, the risk of major bleeding was similar between P2Y12 inhibition and aspirin (1.2% versus 1.4%; HR 0.87; 95% CI 0.70 to 1.09; p=0.23). Prof. Valgimigli notes, “The findings from our analysis, which included all available randomised evidence, challenge the central role of aspirin and support a paradigm shift toward single P2Y12 inhibition for secondary prevention in the long-term antithrombotic management of patients with coronary artery disease.”

Doctor Giuseppe Gargiulo (Federico II University Hospital - Naples, Italy) ended the session by describing the first large individual patient-level data meta-analysis of high-quality trials to investigate the impact of a transradial approach (TRA) versus a transfemoral approach (TFA) on mortality and major bleeding. Investigators from the Radial Trialists’ Collaboration obtained individual patient data from 7 trials comparing TRA (n=10,775) versus TFA (n=10,825) among participants undergoing coronary angiography with percutaneous coronary intervention (PCI; 75.2%) or without PCI, of whom 95% presented with acute coronary syndromes. The incidence of all-cause death at 30 days was 1.6% in the TRA group and 2.1% in the TFA group (HR 0.77; 95% CI 0.63 to 0.95; p=0.012). Major bleeding was also significantly reduced with TRA versus TFA (1.5% versus 2.7%; odds ratio 0.55; 95% CI 0.45 to 0.67; p<0.001). In a multivariable model, TRA was independently associated with a significant 24% relative risk reduction of 30-day all-cause mortality and 51% reduction of major bleeding. Interestingly, mediation analysis showed that the benefit of TRA on mortality was only marginally driven by the prevention of major bleeding. Dr. Gargiulo concludes, “This analysis provides definitive evidence that TRA should be considered the gold standard for patients undergoing cardiac catheterisation with or without PCI, supporting the ‘radial-first’ approach.”