Introduction Drainage of infection pleural fluid is essential in improving patient outcomes. The MIST-2 Trial (NEJM 2011) demonstrated intrapleural tissue plasminogen activator (t-PA) and DNase therapy improves drainage in patients with pleural infection from Day 1 to 7 compared to placebo. However, there is no evidence assessing whether this difference is sustained at 3 months. Methodology used in the MIST-2 trial demonstrated that a digital chest radiograph measurement strategy was an effective measurement of pleural fluid volume and correlated closely with computed tomography scanning in estimating effusion size.
To determine whether intrapleural treatment with t-PA, DNase, or both, showed a sustained significant difference in effusion size at 3 months.
To assess predictors of residual pleural shadowing 3 months post treatment.
Methods 132/193 (68%) patients in the MIST-2 trial had assessable chest radiographs at 3 months (remaining patients were lost to follow-up or died). Effusion size was measured as a percentage of the hemithorax occupied on digital radiograph image (JPEG format). Each image was analysed independently by a medical student and respiratory specialist. Where the difference in estimates exceeded 2%, the radiographs were reassessed together to generate consensus. Analysis was blind to treatment arm.
Results There was no significant difference between placebo and any treatment group in the 3 month radiology outcome. The 3 month effusion size was positively correlated with that at Day 7 (Pearsons 0.27, p=0.003). Neither Day 1 effusion size, nor the difference between Day 1 to 7 was significantly correlated with 3 month outcome. Linear regression analysis demonstrated significant correlation for Day 7 effusion size with 3 months size (p=0.004), but non-significance by treatment arm (p=0.071).
Conclusions In those that survive and have a chest radiograph at 3 months, there is no difference in the radiological outcome at 3 months between treatment groups on the basis of treatments given, including surgical intervention. However, the appearance at day 7 is significantly correlated with the 3 month chest radiograph. Further statistical modelling of this data is now required to assess the relative contributions of intrapleural interventions and surgery in those that survive to 3 months.