Randomised controlled trials (RCTs) that compared glucocorticoid treatment with a placebo or no treatment control in people with sepsis or septic shock were eligible for inclusion. Studies had to report either shock reversal or survival outcomes. The criteria used for determining sepsis or septic shock in the primary studies had to be consistent with the American College of Chest Physicians/Society of Critical Care Medicine consensus definitions.
Nine included studies were published prior to 1989 and 12 after 1997. In studies published before 1989, median corticosteroid dose (expressed as hydrocortisone equivalent) was 23,975mg (range 1,050 to 42,000mg) and median duration of treatment was one day (range four hours to six days); none used a tapered steroid dose. After 1997, median dose was 1,209mg (range 330 to 1,880mg) and median duration of treatment was six days; more than half the studies used a tapering steroid regimen. Where specified, steroids used were hydrocortisone, betamethasone, dexamethasone, methylprednisolone and prednisolone. Methylprednisolone was most commonly investigated in the studies prior to 1989 and hydrocortisone after 1997. Cointerventions included antibiotics, fluids, vasopressors and standard care.
The authors stated neither how the papers were selected for the review nor how many reviewers performed the selection.