Differential long term survival following decortication for empyema and non-infectious etiologies.

Author(s):
Christopher Towe; Sudershan Srinivasan; Vanessa Ho; Stephanie G. Worrell; Malavika Kesavan; Katherine Wu; Yaron Perry; Philip Linden

Background:

Long-term outcomes for patients undergoing pulmonary decortication and the association with underlying medical conditions is largely unknown.

Hypothesis:

We hypothesize that mortality after decortication is associated with whether the indication for decortication was infection and that 30-day outcomes underestimate the mortality burden of decortication.

Methods:

A prospective database of consecutive pulmonary decortications from 1/1/2010 to 10/1/2017 was reviewed and categorized as empyema or non-infectious disease (NID). NIDs are were performed for complex pleural effusions, hemothorax, or malignancy. Follow-up data was recorded for all patients. Kaplan Meier Survival estimates and Cox proportional hazard estimates were performed.

Results:

312 decortication patients were identified (185 (59%) empyema and 127 (41%) NID). Preoperative variables differed between NID and empyema in several ways, including age (64 v 58, p=0.001), hypertension (61% v 49%, p=0.04), congestive heart failure (17% vs 9%, p=0.03), coronary artery disease (28% v 12%, <0.001), and diabetes (24% vs 14%, p=0.01). Post-operative variables were similar between NID and empyema, including length of stay (14 v 14.4 days, p=0.82) and rate of complications (32% v 35%, p=0.60).

Median follow-up was 33.5 months (IQR 7.2-82.6), and was shorter among NID patients (30.5 vs 37.2 months, p=0.03). Decortication for empyema had a 30-day survival of 95%, 91% at 1-year, and 88% at 3-years, see Figure 1. Decortication for NID was associated with worse survival at all time points (p<0.01); 88% at 30-days, 76% at 1-year, 69% at 2-years, and 66% at 3-years.

In a Cox proportional hazard analysis adjusting for decortication etiology, age, hypertension, CHF, CAD, and diabetes, NID (HR 2.34, p=0.005) and age (HR 1.02, p=0.007) were associated with mortality.

Conclusions:

Despite considerable perioperative mortality after decortication, decortication for empyema has less risk of long-term mortality than for NID, likely due to the association of malignancy or other comorbid medical conditions.  Poor perioperative outcomes after decortication suggest ongoing need to improve post-operative care.