Relative blood pressure falls, not fixed thresholds, track with postoperative pneumonia

Design

Retrospective multicentre cohort, competing-risk models

Population

151,036 adults having non-cardiac surgery under general anaesthesia at three Chinese hospitals

Primary outcome

Postoperative pneumonia within 30 days

Effect

20% fall from baseline: any hypotension sub-HR 1.108 (95% CI 1.068 to 1.149); per episode 1.079 (1.050 to 1.109)

In a retrospective cohort of 151,036 adults having non-cardiac surgery under general anaesthesia at three Chinese hospitals, researchers assessed intraoperative hypotension with 12 metrics across nine thresholds, from absolute mean arterial pressure below 50 to below 70 mmHg and relative falls of 20 to 50% from baseline. The outcome was pneumonia within 30 days, which occurred in 3,486 patients (2.3%).

Relative falls were more consistently linked to pneumonia than absolute thresholds. At a 20% fall from baseline, the number of episodes (sub-distribution hazard ratio 1.079, 95% CI 1.050 to 1.109) and the presence of hypotension (1.108, 1.068 to 1.149) were both associated with pneumonia. Absolute thresholds were not consistently associated after adjustment. Depth and recurrence metrics were more consistent than duration. Risk rose quickly at low exposure and then plateaued.

This is an association in observational data. Patients who dropped their pressure may have been sicker or had bigger operations in ways the models did not capture, and the study cannot show that keeping pressure near baseline prevents pneumonia.

• Pneumonia occurred in 2.3% within 30 days.
• A 20% fall from baseline was associated with a hazard ratio of about 1.1.
• Absolute thresholds were less consistently associated than relative ones.
• Risk rose quickly at low exposure then levelled off.
• Know the patient's baseline pressure before induction.

Why it matters

It questions whether one fixed number such as 65 mmHg is the right target for everyone.

Don't overread it

A retrospective cohort cannot show that avoiding relative hypotension would prevent pneumonia.

The statistics, in plain English

A sub-distribution hazard ratio of 1.11 means about an 11% higher rate of pneumonia in the presence of a 20% fall, a modest effect. In 151,036 patients, very small associations reach significance, so size matters more than the p value.

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.