The negative effect of initial high-dose methylprednisolone and tapering regimen for acute respiratory distress syndrome: a retrospective propensity matched cohort study

Makoto Takaki, Kazuya Ichikado, Kodai Kawamura, Yasuhiro Gushima, Moritaka Suga

The efficacy of corticosteroid use in acute respiratory distress syndrome (ARDS) remains controversial. Generally, short-term high-dose corticosteroid therapy is considered to be ineffective in ARDS. On the other hand, low-dose, long-term use of corticosteroids has been reported to be effective since they provide continued inhibition of the systemic inflammatory response syndrome (SIRS) that accompanies ARDS. Thus far, no reports have been published on the efficacy of initiating treatment with a high-dose corticosteroid regimen with tapering. We conducted a retrospective observational study involving 186 patients treated at a teaching hospital (68% had sepsis, pneumonia, or aspiration pneumonia). ARDS was diagnosed according to the Berlin definition. Patients were divided into a high-dose (n = 21) or low-dose corticosteroid group (n = 165) to compare the effectiveness of a down-titration regimen. The primary medical team chose which treatment a patient would receive. We were careful to conduct a differential diagnosis of interstitial pneumonia (e.g., acute eosinophilic pneumonia) since corticosteroid treatment has been proven effective in that patient population. The primary outcome was the 60-day mortality rate. The secondary outcome was the number of ventilator-free days (VFD). Those started on a high-dose regimen had a significantly higher 60-day mortality rate (P = 0.031) with significantly fewer VFD (P = 0.021). Propensity scores were used to adjust patient backgrounds in a variable analysis that also showed the high-dose regimen was a factor in decreasing VFD (OR, 95.63; 95% CI, 1.74-5271.07; P = 0.026) and worsening the 60-day mortality rate (OR, 2.54; 95% CI, 0.92-7.02; P = 0.072). A tapering regimen after high-dose corticosteroids is likely to increase ventilator dependency and might aggravate the prognosis of patients with ARDS diagnosed according to the Berlin definition.