NEW YORK (Reuters Health) – Treating septic shock with steroids produces hyperglycemia, but patients are no better off with intensive insulin therapy than with conventional glucose control, according to results of the multicenter COIITSS trial.

Furthermore, the investigators report in the January 27th Journal of the American Medical Association, adding oral fludrocortisone to hydrocortisone treatment does not improve survival either.

International guidelines suggest that physicians consider prolonged low-dose corticosteroids for adults with septic shock who respond poorly to fluids and vasopressors, Dr. Djillali Annane, at Hopital Raymond Poincare, Garches, France, and coauthors note.

The randomized, 2 x 2 factorial, open-label COIITSS trial was designed to compare various insulin regimens for steroid-induced hyperglycemia in these patients.

In 11 French hospitals, patients with severe sepsis were assigned as follows:
— Continuous IV insulin and hydrocortisone (n = 126),
— Continuous IV insulin, hydrocortisone and fludrocortisones (n = 129),
— Conventional insulin therapy and hydrocortisone (n = 138), or
— Conventional insulin therapy, hydrocortisone and fludrocortisone (n = 116).

All of the patients had multiple organ dysfunction, were on vasopressors, and were receiving hydrocortisone 50 mg IV every 6 hours.

For intensive therapy, IV insulin was strictly titrated, with target glucose levels between 80 and 110 mg/dL. In the conventional-therapy groups, the target glucose level was <150 mg/dL, as advised in the 2004 Surviving Sepsis campaign. 9-alpha-Fludrocortisone 50 mcg daily was administered via nasogastric tube. The median daily dose of insulin was 71 UI in the intensive therapy groups vs 46 UI in the control groups (p < 0.001). As a result, blood glucose levels in the ICU were markedly lower in the experimental group, and time spent with glucose levels in the range of 80 to 110 mg/dL was greater (p < 0.00001 for both). More patients receiving intensive therapy experienced serious hypoglycemic episodes (16.4% vs 7.8%, p = 0.003). Nevertheless, in-hospital mortality did not differ significantly between the two groups (45.9% vs 42.9%). In addition, secondary outcomes during 180 days of follow-up – number of vasopressor and mechanical ventilation-free days, length of stay, serious adverse events, and presence of posttraumatic stress disorders – did not differ significantly between groups. Fludrocortisone made little difference in hospital mortality, which was 42.9% in the fludrocortisone group and 45.8% in the control group. The only secondary outcome to differ between groups was a higher rate of urinary tract superinfection with fludrocortisone (p = 0.02). In an editorial, Dr. Greet Van den Berghe from Catholic University of Leuven, Belgium contends that the COIITSS trial was underpowered to detect a significant difference between the two insulin strategies, and that there was substantial overlap between blood glucose levels. “Clinicians caring for patients with septic shock treated with hydrocortisone will still be left with uncertainty as to whether insulin should be given and to what level the blood glucose should be lowered,” she writes. The “lack of adequately robust evidence” on how best to care for patients with septic shock is “a glaring deficiency,” Dr. Van den Berghe says. In light of the “huge global burden” of septic shock, she calls not only for large, adequately powered trials but also for cooperation between national and international trials groups and their funding sources. Reference:
JAMA 2010;303:341-348,365-366.