Esketamine does not treat postoperative anxiety after cancer surgery

Design

Randomised, placebo-controlled clinical trial

Population

180 patients with preoperative anxiety undergoing major abdominal tumour surgery; 173 analysed

Primary outcome

Response rate at postoperative day 3, defined as a 50% or greater reduction in Hamilton Anxiety Rating Scale score

Effect

39.5% vs 42.5% (risk ratio 0.93, 95% CI 0.65-1.33, P=0.81); anxiety score difference 0 (-2 to 1) at day 3 and -1 (-3 to 1) at day 30; depression score difference -2 (-3 to -1, P=0.001) at day 30

Esketamine has accumulated perioperative uses faster than evidence for them, and reducing postoperative anxiety in already-anxious patients is a plausible one. This trial randomised 180 patients with preoperative anxiety scheduled for major abdominal tumour surgery to esketamine or placebo, with 173 analysed.

It did not work. The primary outcome — a 50% or greater fall in Hamilton Anxiety Rating Scale score by postoperative day 3 — occurred in 39.5% on esketamine and 42.5% on placebo (risk ratio 0.93, 95% CI 0.65-1.33, P=0.81). Anxiety scores were equivalent at day 3 (adjusted mean difference 0, 95% CI -2 to 1) and at day 30 (-1, -3 to 1). Depression scores were lower in the esketamine group at day 30 (adjusted mean difference -2, -3 to -1, P=0.001), a secondary outcome among several.

Two things follow. First, if anxiety is the indication, this is a negative trial and esketamine should not be given for it. Second, the day-30 depression signal is the kind of secondary finding that gets quoted as if it were the result, and it should not be: it is one of multiple secondary endpoints in a trial whose primary endpoint failed, and its effect size — two points on a symptom scale — is of uncertain clinical meaning. Note also the placebo arm: two in five patients met the response threshold without esketamine.

• Do not administer esketamine with the aim of reducing postoperative anxiety
• Treat the day-30 depression finding as hypothesis-generating, not as an indication
• Note that 42.5% of placebo patients met the anxiety response threshold
• Identify preoperative anxiety and address it with the measures that have evidence — information, expectation-setting, and psychological support
• Where esketamine is used for analgesia, that is a separate indication with separate evidence

Why it matters

It closes a question that was being answered informally in favour of the drug, and names the placebo response that made it look plausible.

Don't overread it

The day-30 depression difference is an uncorrected secondary outcome in a negative trial, not evidence that esketamine treats depression after cancer surgery.

The statistics, in plain English

A risk ratio of 0.93 with an interval from 0.65 to 1.33 sits squarely across 1.0: the trial found no difference, and its size means a modest benefit could still have been missed. The large placebo response, over 40%, is the reason an uncontrolled series would have made this drug look effective. The significant depression result at day 30 comes from a set of secondary outcomes with no correction for multiple comparisons, in a trial whose primary endpoint was negative — which is the standard recipe for a finding that does not replicate.

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.