Clinical characteristics and survival of hospitalized anthrax patients with pleural, pericardial, or peritoneal fluid collections.

Abstract

The prevalence of pleural, pericardial, or peritoneal effusions among anthrax patients is unknown, as is the impact of drainage on mortality. We identified hospitalized anthrax patients with effusions published 1920-2018 worldwide (n=1,108), excluding cases with insufficient clinical data. Two manuscript authors independently abstracted fluid collection and morbidity data. We evaluated how effusion drainage impacted mortality using logistic regressions. Effusions were present in 99 (13%) of 744 eligible anthrax patients; 65 (66%) of these died. Pleural effusions developed in 72% of patients with inhalation anthrax (N=61), most often bilaterally. Similarly, 55% of ingestion anthrax patients (N=55) developed peritoneal effusions. Pericardial effusions were rare (1.7%). Most pleural Gram-stains or cultures obtained pre-antimicrobials were positive (83%). Of the 44 patients with clinically significant pleural effusions, only 59% received drainage. Drainage of these effusions was associated with survival (OR 38.3, 95% CI: 4.3-339.0), even when controlling for antimicrobials and/or antiserum. Most patients with fatal outcomes following drainage had either bilateral effusions with single-sided drainage or effusion recurrence. Although drainage of clinically significant peritoneal effusions was not associated with survival, laparotomy ±resection for peritonitis was associated (OR 53.6, 95% CI: 9.4-inf). Pleural drainage appears associated with survival. Most inhalation anthrax patients develop bilateral collections, and may require chest-tube insertions. Surgical source control is associated with survival for anthrax-associated peritonitis. Following a wide-area release of Bacillus anthracis, public health authorities should anticipate many patients may require source control (e.g., chest tube drainage or surgery) and plan accordingly.