Autonomic dysfunction was near-universal, and predicted nothing

Design

prospective observational study with blinded assessment, single Indian tertiary centre

Population

69 adult women with breast cancer for surgery under general anaesthesia; 62 with complete autonomic testing

Primary outcome

post-induction hypotension (systolic fall over 30% or absolute below 90 mmHg)

Effect

autonomic dysfunction 79.0%, hypotension 32%; sympathetic OR 1.04 (95% CI 0.35-3.07), parasympathetic OR 1.17 (0.40-3.41)

Sixty-nine women with breast cancer scheduled for surgery under general anaesthesia at a Delhi tertiary centre had formal preoperative autonomic testing — heart rate response to deep breathing, Valsalva, handgrip, cold pressor and postural challenge — with dysfunction defined as two or more abnormal tests. Post-induction hypotension was defined as a fall of more than 30% in systolic pressure or an absolute value below 90 mmHg.

Autonomic dysfunction was present in 49 of 62 evaluable patients (79.0%): sympathetic in 61.3%, parasympathetic in 40.3%, both in 22.6%. Post-induction hypotension occurred in 22 of 69 (32%). The two did not connect. Odds ratios were 1.04 (95% CI 0.35-3.07) for sympathetic dysfunction, 1.17 (0.40-3.41) for parasympathetic and 1.11 (0.32-3.87) for combined — all sitting on 1.0. Older patients were more likely to have parasympathetic dysfunction (mean 52.3 against 42.8 years), while comorbidity was not significantly associated.

The value of a null result depends on what was being proposed, and here something specific was: that formal autonomic testing might identify the patient who will drop their pressure at induction. On this evidence it does not, and a battery of five bedside tests is a real preoperative cost to impose for no predictive return.

Two cautions. Sixty-nine patients gives wide intervals, and an odds ratio whose upper bound is above 3 has not excluded a meaningful association. And the finding is specific to this population — women with breast cancer, mostly without major comorbidity — not a general statement about autonomic testing before anaesthesia. What it does establish is that a 79% prevalence of dysfunction cannot be doing much work in explaining a 32% incidence of hypotension.

• Do not add formal autonomic function testing to the preoperative assessment to predict post-induction hypotension
• Expect post-induction hypotension in roughly a third of these patients regardless of autonomic status, and prepare for it
• Continue the established measures — cautious induction dosing, fluid status assessment, vasopressor drawn up before induction
• Treat a high prevalence of an abnormality as a reason to doubt its predictive value, not to confirm it
• Note that these are Indian data in women with breast cancer; the finding does not generalise to other surgical populations

Why it matters

An abnormality present in four-fifths of a population cannot explain an event occurring in a third of it.

Don't overread it

Sixty-nine patients with a very small comparison group cannot exclude a modest association — this is an absence of evidence in one specific population.

The statistics, in plain English

Every odds ratio here sits almost exactly on 1.0, which is a genuine null rather than a near-miss. But the intervals are wide — 0.35 to 3.07 for sympathetic dysfunction — so an association up to threefold has not been ruled out with 69 patients. With 79% of the cohort classified as having dysfunction, the comparison group is only 13 patients, and a comparison against a group that small cannot detect much. The study was observational, so no causal claim is made or possible in either direction.

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.