ICYMI: Delirium Risk After Major Surgery, and What to Do About It

An unusual kind of meta-analysis gives us some answers we can use

ICYMI: Delirium Risk After Major Surgery, and What to Do About It

What’s the Claim?

A meta-analysis that used individual patient data (very unusual!) from 21 studies of patients undergoing noncardiac surgery identified a number of factors associated with an increased risk of postoperative delirium. Those most useful to surgeons performing spine surgery were:

  • American Society of Anesthesiologists (ASA) status 4 (odds ratio [OR], 2.4 [95% CI 1.4 to 4.1])
  • Older age (OR for 65 to 85 years, 2.7 [2.2 to 3.3]; OR for > 85 years, 6.2 [4.7 to 8.4])
  • Low body mass index (OR for body mass index < 18.5, 2.3 [1.6 to 3.1])
  • A history of delirium (OR, 3.9 [2.7 to 5.7]) or preoperative cognitive impairment(OR, 4.0 [2.9 to 5.4])

How’s It Stack Up?

This study was appealing to us for its size (almost 8400 patients) and scope, but most importantly for its use of individual patient data. This allows a level of analytic detail that is seldom possible in meta-analyses. The tradeoff is, as you’d imagine, not every study’s corresponding author provides his or her dataset when asked; in this case, 21 studies provided data (out of 175 that were eligible). This is a shortcoming, but there’s no particular reason to believe it would push findings in one direction over another. One would expect that the major impact of the non-participation of the other studies would be a decrease in statistical power, resulting perhaps in some patient factors being important but not identified as such. But with over 8000 patients’ data included, this seems unlikely to be a serious shortcoming.

You might reasonably wonder whether a study that we originally covered in our Joint Replacement newsletter, and one that includes studies on any noncardiac operation, is applicable to your spine surgery practice. These findings are relevant to you. The included studies were a mixed bag, to be sure, but they were comparable in magnitude to major spinal surgery even if they didn't mostly consist of spine patients. Twelve of the 21 included studies were either THA, TKA, hip fracture, or “any elective surgery” including spinal surgery. Others included major abdominal and vascular procedures performed for different indications. The reality is that most of the same risk factors consistently come up as relevant in general surgery and orthopaedic surgery when dealing with older hospitalized patients having general anesthesia.

What’s Our Take?

Although surgeons may minimize postoperative delirium because it’s usually transient, and patients often forget they had it, delirium after surgery is one of the most upsetting postoperative complications for family members. No one ever forgets when mom or dad forgets where they are or who you are.

It’s also been associated in some studies with serious complications after major orthopaedic surgery; I speculate that this is the result of confounding factors like other comorbid conditions and frailty. Regardless, family members will just remember the hospital stay as a stormy and frightening experience, or worse.

  • Because postoperative delirium is both important and disturbing to patients’ families, I try to speak to family members of all my patients who are over the age of 80 so that they know it may occur (it happens in 10% to 20% of patients in that age group). 

What I learned from this study is that I’m not talking to enough family members. I’ll probably start warning my “younger old patients” (those over the age of 65, and not just my octogenarians) — an age cutoff that has been suggested before in several spine surgery-specific studies on the topic — and I’ll try to ask my older patients if they’ve had delirium following other operations they’ve had, since a high percentage of patients undergoing spinal surgery have had things like prior joint replacements or other major procedures before. 

Last words to CORRelations' advisor in Spine, Charlie Reitman, MD:

Delirium among surgical inpatients is a topic that does not get enough attention. While it's not preventable per se, it's possible to reduce the risk that it will occur. It increases length of stay, may result in an unnecessary stop in acute rehab (rather than home discharge), and certainly decreases family satisfaction. It also adds a tremendous workload to our already-overworked nursing staffs, which is something to think about as recruiting nurses is very difficult in many parts of the country. As value-based care becomes more widespread, it will also affect reimbursement, which will really get everyone’s attention. We should attend to it before that happens.


Sadeghirad B, Dodsworth BT, Schmutz Gelsomino N, et al. Perioperative Factors Associated With Postoperative Delirium in Patients Undergoing Noncardiac Surgery: An Individual Patient Data Meta-Analysis. JAMA Netw Open. 2023;6:e2337239.