- Design
- Meta-analysis of 17 randomised trials, random effects
- Population
- 14,033 patients undergoing drug-eluting stent PCI; 10,155 in East Asian trials
- Primary outcome
- Cardiac death, with geographic interaction
- Effect
- Overall RR 0.71 (0.53-0.94); East Asian 0.56 (0.46-0.69) vs non-Asian 1.23 (0.85-1.76)
This meta-analysis pooled 17 randomised trials of intravascular ultrasound (IVUS)-guided versus angiography-guided drug-eluting stent PCI: 14,033 patients, 10,155 of them in 10 East Asian trials. Whether geography modified the effect on cardiac death was the prespecified primary question.
Overall, IVUS guidance reduced cardiac death (RR 0.71, 95% CI 0.53 to 0.94). The split was stark: RR 0.56 (0.46 to 0.69) in East Asian trials against 1.23 (0.85 to 1.76) in non-Asian trials, interaction p<0.001. Target vessel infarction, revascularisation and major adverse events followed the same pattern. Definite stent thrombosis fell consistently everywhere (RR 0.39, 0.19 to 0.79). All-cause death did not differ (RR 0.89).
The authors suggest the difference lies in how operators acted on the images, with predefined optimisation criteria more consistently applied in the Asian trials. The benefit of imaging seems to depend on what is done with it.
- Use IVUS with explicit optimisation targets, such as minimum stent area and expansion, rather than as a look-and-see.
- Expect imaging to reduce stent thrombosis regardless of setting.
- Prioritise imaging in complex lesions such as left main, long lesions and bifurcations.
- Audit your own lab: record how often IVUS findings led to post-dilatation or an extra stent.
Why it matters
Indian labs follow East Asian practice patterns closely, and this suggests the protocol around imaging matters as much as the catheter.
Don't overread it
Geography stands in for differences in protocol and patients; it does not show that IVUS works only in Asian patients.
The statistics, in plain English
A risk ratio of 0.56 means cardiac death was about 44% lower with IVUS in East Asian trials. The non-Asian estimate of 1.23 has a confidence interval running from 0.85 to 1.76, so it is compatible with no effect or some harm. An interaction p value below 0.001 says the two estimates are very unlikely to differ by chance, but differences between groups of trials can reflect design and populations as well as geography.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for cardiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free