The biological role of pleural fluid PAI-1 and sonographic septations in pleural infection: Analysis of a prospectively collected clinical outcome study

Rationale

Sonographic septations are assumed to be important clinical predictors of outcome in pleural infection, but the evidence for this is sparse. The inflammatory and fibrinolysis-associated intrapleural pathway(s) leading to septation formation have not been studied in a large cohort of pleural fluid (PF) samples with confirmed pleural infection matched with ultrasound and clinical outcome data.

Objectives

To assess the presence and severity of septations against baseline PF PAI-1 (Plasminogen-Activator Inhibitor-1) and other inflammatory and fibrinolysis-associated proteins as well as to correlate these with clinically important outcomes.

Methods

We analyzed 214 pleural fluid samples from PILOT (Pleural Infection Longitudinal Outcome Study), a prospective observational pleural infection study, for inflammatory and fibrinolysis-associated proteins using the Luminex platform. Multivariate regression analyses were used to assess the association of pleural biological markers with septation presence and severity (on ultrasound) and clinical outcomes.

Measurements and Main Results

PF PAI-1 was the only protein independently associated with septation presence (P < 0.001) and septation severity (P = 0.003). PF PAI-1 concentrations were associated with increased length of stay (P = 0.048) and increased 12-month mortality (P = 0.003). Sonographic septations alone had no relation to clinical outcomes.

Conclusions

In a large and well-characterized cohort, this is the first study to associate pleural biological parameters with a validated sonographic septation outcome in pleural infection. PF PAI-1 is the first biomarker to demonstrate an independent association with mortality. Although PF PAI-1 plays an integral role in driving septation formation, septations themselves are not associated with clinically important outcomes. These novel findings now require prospective validation.