Increasing Steroid Dose Perioperatively May Not Be Necessary for Prevention of Hypotension in Patients With Recent or Current Steroid Use

EBM Focus - Volume 8, Issue 51

Read the complete EBM Focus and earn CME

Reference: Ann Surg 2014 Jan;259(1):32 (level 2 [mid-level] evidence)

Patients having major colorectal surgery while receiving steroid treatment are routinely given high-dose corticosteroids to prevent acute perioperative adrenal insufficiency. This practice has also become standard, even for patients not receiving steroids at the time of surgery but who had taken steroids in the previous year. However, high-dose steroids are associated with an increased risk of postoperative complications, such as wound infection or anastomotic dehiscence. Furthermore, observational studies in patients with inflammatory bowel disease (IBD) having colorectal surgery have previously shown no significant differences in the risk of hemodynamic instability for patients receiving high-dose steroids compared to low-dose steroids (Am J Surg 2012 Oct;204(4):481) or no steroids (Am Surg 2011 Oct;77(10):1295). Now, a prospective randomized noninferiority trial compares low-dose steroids to high-dose steroids in patients with IBD with current or recent steroid use who were scheduled to have major colorectal surgery.

A total of 121 patients with inflammatory bowel disease who were receiving steroids or were treated with steroids in the previous year were randomized to low-dose hydrocortisone (IV equivalent to presurgical oral dosing) vs. high dose hydrocortisone (100 mg IV 3 times daily). Hydrocortisone doses were gradually tapered for both groups. Among patients not taking steroids at the time of surgery, the median time from last steroid dose to surgery was 3 months for the low-dose steroid group and 4 months for the high-dose steroid group (not significant). Orthostatic, or postural, hypotension was defined as a decrease in systolic blood pressure of at least 20 mm Hg after sitting up from a supine position. Hemodynamic instability was defined as the presence of hypotension, tachycardia (heart rate > 100 beats/minute) or bradycardia (heart rate< 60 beats/minute).

The efficacy analysis excluded 29 patients (24% of those randomized) due to protocol violations or having been mistakenly randomized for a second time during a staged surgical procedure. Overall, 96% of those receiving low-dose steroids had an absence of orthostatic hypotension through postoperative day 1, compared with 95% of those receiving high-dose steroids (not significant, noninferiority criterion met). Similarly, there was an absence of orthostatic hypotension in 78% of those receiving low-dose steroids through postoperative day 7, compared with 79% of those receiving high-dose steroids (not significant). An absence of hemodynamic instability through postoperative day 7 was observed in 12% of patients in both groups. No patients in either group were treated with rescue high-dose steroids for adrenal insufficiency during the trial.

This randomized trial builds upon previous retrospective cohort studies performed by the same group, and similarly demonstrates no significant differences between low-dose and high-dose steroids for patients with IBD having major colorectal surgery with steroid use in the previous 12 months. These findings also add to those of a systematic review of 2 randomized trials and 7 cohort studies evaluating perioperative stress dose of corticosteroids in 315 patients taking daily corticosteroids. In this review, there were no differences in hemodynamic profiles in patients receiving stress dose compared to those receiving their regular steroid dose in the 2 randomized trials, and there were no episodes of unexplained hypotension or adrenal crisis in patients who did not receive stress dose in 5 cohorts studies (Arch Surg 2008 Dec;143(12):1222 full-text). Collectively, these data do not support increased perioperative dosing of steroids for the purposes of preventing adrenal insufficiency.

For more information, see the Adrenal insufficiency in adults topic in DynaMed.