Steroid Therapy and Outcome of Parapneumonic Pleural Effusions (STOPPE): A Pilot Randomized Clinical Trial.

Rationale: Pleural effusion commonly complicates community-acquired pneumonia and is associated with intense pleural inflammation. Whether antiinflammatory treatment with corticosteroids improves outcomes is unknown.

Objectives

To assess the effects of corticosteroids in an adult population with pneumonia-related pleural effusion.

Methods

The STOPPE (Steroid Therapy and Outcome of Parapneumonic Pleural Effusions) trial was a pilot, multicenter, double-blinded, placebo-controlled, randomized trial involving six Australian centers. Patients with community-acquired pneumonia and pleural effusion were randomized (2:1) to intravenous dexamethasone (4 mg twice daily for 48 h) or placebo and followed for 30 days. Given the diverse effects of corticosteroids, a comprehensive range of clinical, serological, and imaging outcomes were assessed in this pilot trial (ACTRN12618000947202). Measurements and Main

Results

Eighty patients were randomized (one withdrawn before treatment) and received dexamethasone (n = 51) or placebo (n = 28). This pilot trial found no preliminary evidence of benefits of dexamethasone in improving time to sustained (>12 h) normalization of vital signs (temperature, oxygen saturations, blood pressure, heart, and respiratory rates): median, 41.0 (95% confidence interval, 32.3-54.5) versus 27.8 (15.4-49.5) hours in the placebo arm (hazard ratio, 0.729 [95% confidence interval, 0.453-1.173]; P = 0.193). Similarly, no differences in C-reactive protein or leukocyte counts were observed, except for a higher leukocyte count in the dexamethasone group at Day 3. Pleural drainage procedures were performed in 49.0% of dexamethasone-treated and 42.9% of placebo-treated patients (P = 0.60). Radiographic pleural opacification decreased over time with no consistent intergroup differences. Mean duration of antibiotic therapy (22.4 [SD, 15.4] vs. 20.4 [SD, 13.8] d) and median hospitalization (6.0 [interquartile range, 5.0-10.0] vs. 5.5 [interquartile range, 5.0-8.0] d) were similar between the dexamethasone and placebo groups. Serious adverse events occurred in 25.5% of dexamethasone-treated and 21.4% of placebo-treated patients. Transient hyperglycemia more commonly affected the dexamethasone group (15.6% vs. 7.1%).

Conclusions

Systemic corticosteroids showed no preliminary benefits in adults with parapneumonic effusions. Clinical trial registered with www.anzctr.org.au (ACTRN12618000947202).