Preoperative cannabis use had no effect on pain or outcomes after cardiac surgery | Research

The takeaway

In a prospective study of 73 cardiac surgery patients, those who used cannabis preoperatively had similar postoperative morphine requirements, pain scores, time to extubation, ICU length of stay, and complication rates as non-users.

Cannabis users facing surgery, anesthesiologists, and cardiac surgeons.

Cannabis users needed virtually identical morphine (60.98 vs 59.90 mg equivalents) after cardiac surgery

What the researchers found

Average morphine equivalents in the first 48 hours: cannabis users 60.98 vs non-users 59.90 (p=0.93). VAS pain at 24 hours: 5.52 vs 4.84 (p=0.414). VAS at 48 hours: 4.74 vs 3.90 (p=0.23). Time to extubation was nearly identical (718.41 vs 718.67 minutes, p=0.99). No differences in ICU length of stay, nausea/vomiting, reoperation, or in-hospital mortality.

Why it matters

There has been concern that cannabis users might require more opioids postoperatively due to cross-tolerance, or that cannabis might affect surgical outcomes. This study provides reassuring data that preoperative cannabis use did not worsen any measured outcome after cardiac surgery.

The numbers in context

73 patients (50 non-users, 23 cannabis users). 48-hour morphine equivalents: 60.98 vs 59.90 (p=0.93). 24h VAS: 5.52 vs 4.84 (p=0.414). 48h VAS: 4.74 vs 3.90 (p=0.23). Extubation: 718.41 vs 718.67 min (p=0.99). ICU stay: 2.91 vs 3.48 days (p=0.26).

How the study worked

Single-center prospective study of adults undergoing cardiac surgery via sternotomy. Patients were surveyed about cannabis use during preoperative consultation. 50 non-users and 23 cannabis users were compared. Primary outcomes: morphine equivalents in first 48 hours and VAS pain scores. Secondary outcomes: extubation time, nausea, ICU stay, reoperation, mortality.

What this study cannot tell us

Small sample, especially the cannabis group (n=23), limiting power to detect moderate effects. Single center. Cannabis use was self-reported and not verified biochemically. No information on dose, frequency, or recency of cannabis use. The study may be underpowered for rare outcomes like reoperation or mortality.

How to read the evidence

Small prospective single-center study providing reassuring null findings across multiple outcomes, but underpowered for definitive conclusions.

When this study was published

Published in 2026.

The bigger picture

As cannabis use becomes more prevalent, surgeons and anesthesiologists need data on how it affects perioperative outcomes. The null findings across every measured outcome are reassuring and suggest that cannabis use alone should not alter surgical planning for cardiac procedures, though the small sample size warrants confirmation.

Questions still open

Would heavier cannabis users show different patterns? Does the method of cannabis consumption (smoking vs edibles) matter for perioperative risk? Would larger studies reveal subgroup differences?

Common questions

Should I stop cannabis before heart surgery?

This study found no difference in pain, opioid needs, or outcomes between cannabis users and non-users after cardiac surgery. However, the small sample means the findings are preliminary, and you should discuss cannabis use with your surgical team.

Do cannabis users need more pain medication after surgery?

Not in this study. Cannabis users required virtually identical morphine equivalents (60.98 vs 59.90 mg) in the first 48 hours after cardiac surgery.

Read the original research

Effect of Preoperative Cannabis Use on Postoperative Pain and Outcomes Following Cardiothoracic Surgery.

Seminars in cardiothoracic and vascular anesthesia, 30(1), 21-27

View on PubMed ↗Publisher’s paper ↗

Citation

Shah, Sareena; Fletcher, Paul; Hamadah, Kareem; Gilmore, Drake; Staples, Bryant; Chadwick, Andrea; He, Jianghua; Kim, Jaromme; Flynn, Brigid. (2026). Effect of Preoperative Cannabis Use on Postoperative Pain and Outcomes Following Cardiothoracic Surgery.. Seminars in cardiothoracic and vascular anesthesia, 30(1), 21-27. https://doi.org/10.1177/10892532251374952