Cyclones raise acute coronary syndrome risk and lengthen the time to the catheter laboratory

Design

time-stratified case-crossover study with distributed lag non-linear models

Population

2,563,780 people with acute coronary syndrome in mainland China, 2015-2022

Primary outcome

acute coronary syndrome incidence after tropical cyclone exposure

Effect

14% higher risk over lag 0-3 days (95% CI 2% to 27%)

A time-stratified case-crossover analysis of a nationwide Chinese registry covering 2,563,780 people with acute coronary syndrome between 2015 and 2022 matched presentations against modelled tropical cyclone wind fields, with exposure defined as daily maximum sustained winds of 17.5 m/s or more. Over the 0 to 3 days after exposure, acute coronary syndrome risk rose by 14% (95% CI 2% to 27%).

The second finding is the operational one. On cyclone days patients took longer to present - a median 5.8 hours to self-referral against 5.3 - and once in hospital, admission-to-catheterisation lengthened from 0.9 to 1.0 hours. The hazard arrives at the same time as the system's ability to treat it degrades. Associations were stronger in men, in those with less education, and in those carrying more risk factors.

For Indian practice this is directly transferable: the east coast from Odisha to Tamil Nadu takes cyclone landfalls every season, and the same combination of a presentation surge and a transport system that stops working is what a district catheter laboratory faces. The actionable part is anticipatory, not clinical - stock, staffing and a fibrinolysis fallback decided before the warning, not during it.

• Build a cyclone or flood contingency into the primary PCI pathway before the season, including when to fall back to fibrinolysis.
• Expect presentation delay to lengthen, so widen the window in which you consider a late presenter for reperfusion.
• Prioritise counselling of high-risk patients about not deferring chest pain during a storm warning.
• Check that antiplatelet and statin supply at home covers a week of disrupted access.
• Treat the effect modifiers - male, lower education, more risk factors - as the group to target with pre-season advice.

Why it matters

It reframes extreme weather from a public health abstraction into a predictable stress on the reperfusion pathway, which can be planned for.

Don't overread it

This is an observational case-crossover design: it shows association between cyclone exposure and presentations, not that the storm caused the infarct.

The statistics, in plain English

The 14% increase has a confidence interval of 2% to 27%, which excludes no effect but only just - the true increase could be as small as 2%. The delay differences (5.8 vs 5.3 hours, 1.0 vs 0.9 hours) are precise because the registry is enormous, but small in absolute terms; their importance is that they point the same way as the incidence signal rather than their individual size.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.