Use of Hyperbaric Oxygen in Patients with Necrotizing Soft Tissue Infections: A Scandinavian Multicenter, Prospective, Observational Cohort.

INTRODUCTION

Hyperbaric oxygen (HBO2) treatment is regularly used as adjuvant treatment in patients with necrotizing soft tissue infections (NSTI). Several observational studies have suggested an association between hyperbaric oxygen (HBO2) treatment and improved survival in patients with necrotizing soft tissue infections (NSTIs); however, the evidence remains inconclusive. Most patients with NSTI are severely ill, having sepsis or septic shock, and requiring mechanical ventilation and inotropic support in an intensive care unit (ICU). Although recent data suggest that the beneficial effect of HBO2 may be largest in the most severely ill, not all patients may receive it due to hemodynamic instability, pressure chamber capabilities, and availability, thereby potentially introducing selection bias.

METHODS

From January 2013 until June 2017, we conducted a multicenter, prospective, consecutive, observational study (The INFECT study, ClinicalTrials.gov number; NCT01790698) enrolling NSTI patients at five clinical centers delivering HBO2 to NSTI patients in an ICU setting in Denmark, Norway, and Sweden.

RESULTS

A total of 409 consecutive patients with necrotizing soft tissue infections were prospectively enrolled in the study, of whom 402 (98.3%) were admitted to the ICU. A total of 329 NSTI patients received HBO2, and 80 patients did not. Median time from arrival at a specialized hospital to first HBO2 was 4.3 h (IQR 2.5-7.7). Patients receiving HBO2 had less often chronic liver disease, penetrating trauma within 4 weeks before NSTI, lower extremity NSTI involvement, and acute kidney injury and lower severity scores (Simplified Acute Physiology Score; SAPS-2 and Sequential Organ Failure Assessment score; SOFA score) than patients not treated with HBO2. Patients receiving HBO2 were more frequently on mechanical ventilation. All-cause 30- and 90-day mortality for the HBO2-treated patients was 22/325 (7%) and 36/325 (11%) and for non-HBO2-treated patients 34/80 (43%) and 37/80 (46%). In exploratory analyses, HBO2 was associated with lower all-cause 30-day mortality in unadjusted and in that adjusted for sex, lactate, SAPS-2, and baseline norepinephrine infusion rate.

CONCLUSIONS

Patients receiving HBO2 in this cohort were less acutely ill than those not receiving HBO2 likely influencing physicians' decisions thereby introducing selection bias. In exploratory analyses, the use of HBO2 was associated with a reduced 30-day all-cause mortality. A randomized trial of HBO2 treatment seems warranted in patients with NSTI.

TRIAL REGISTRATION: ClinicalTrials.gov number NCT01790698.