TOPIC: Procedures TYPE: Original Investigations PURPOSE: Postprocedural pneumothorax (iatrogenic) is a patient safety indicator resulting from invasive procedures such as insertion of subclavian central line, pacemaker/ implantable cardiac defibrillator (ICD) placement, transbronchial lung biopsy and transthoracic pulmonary biopsy. This study examines if any of these invasive procedures have a higher predisposition to iatrogenic pneumothorax in teaching hospitals as compared to non-teaching hospital. METHODS: Cohort of patients aged 18 or older who had percutaneous lung biopsy, trans-bronchial lung biopsy, subclavian central line placement, and pacemaker/ICD placement were identified from the National Inpatient Sample (NIS) database from the years 2016 to 2017 using ICD 10-procedure codes. Diagnosis of iatrogenic pneumothorax was identified using ICD 10 – diagnosis code. Incidence of post-procedural pneumothorax were estimated for overall hospital encounter and for each of the procedures. The incidence rate was compared between teaching hospital group (TH) and non-teaching hospital (NTH) using chi-square statistical test. Odds ratio (OR) comparing both group were estimated with logistic regression. RESULTS: Iatrogenic pneumothorax was identified in 2,951 of 13,575 (21.7%) transthoracic lung biopsy, with 21.6% in TH group vs 22.1% in NTH group ( OR 0.96, 95% CI 0.88 – 1.05, p = 0.41). Among patients who had trans-bronchial lung biopsy, post-procedural pneumothorax occurred in 890 out of 26,793 (3.3%) with 3.6% in TH vs 2.6% in NTH ( OR 1.37, 95% CI 1.17 – 1.59, p = <0.001). Across 568,677 subclavian central line insertions, incidence of post-procedural pneumothorax was also higher in TH group (0.43% in TH vs 0.34% in NTH, OR 1.26, 95% CI 1.15 – 1.39, p = <0.001). Pacemaker/ICD placement in 83,033 patients resulted in 1.17% incidence of iatrogenic pneumothorax in TH vs 1.44% in NTH (OR 0.81, 95% CI 0.71 – 0.92, p = 0.02). Overall, a total of 18,703 cases of iatrogenic pneumothorax out of the 14 million (unweighted) hospital admissions were identified between 2016 to 2017 with incidence of 144 events per 100,000 hospital encounters in the TH group and 103 events per 100,000 hospital encounters in the NTH. CONCLUSIONS: Our study showed a significantly higher incidence of post-procedural pneumothorax in teaching hospitals among patients undergoing trans-bronchial lung biopsy and subclavian central line insertion. And a lower incidence in patients undergoing pacemaker/ICD placement. Incidence of iatrogenic pneumothorax in patients undergoing transthoracic lung biopsy was not significantly different between the two groups. Overall, iatrogenic pneumothorax was more prevalent in teaching hospitals. Randomized controlled studies are needed to identify the etiology of our findings. CLINICAL IMPLICATIONS: Our study raises the concern that physicians-in-training due to lack of experience may be a major factor accounting for a higher incidence of iatrogenic pneumothorax in teaching hospital DISCLOSURES: no disclosure on file for Tewabe Belay; No relevant relationships by Sahai Donaldson, source=Web Response no disclosure on file for Daniel Larbi; No relevant relationships by Alem Mehari, source=Web Response No relevant relationships by Richard Ogunti, source=Web Response No relevant relationships by Lamiaa Rougui, source=Web Response No relevant relationships by Mahbubur Sumon, source=Web Response
Read full abstract