Definition - a persistent broncho-pleural airleak, lasting greater than 24 hours beyond placement of a chest tube
Incidence - very low. One series, out of 1700 adult patients with ARDS, 39 (2%) had Bronchopleural Fistulas develop
Significance - good question. Only 2 of the 39 patients with BPF were unventilatable. Survival was only 33% of patients who had BPF
4 year-old with Idiopathic Pulmonary Hemosiderosis, had death spell, pulmonary hemorrhage, rapidly deteriorated and arrested, required placement of multiple pleural tubes. Had persistent air-leak for several weeks, of fairly low volume. (Slow bubbles)
Management: HFOV for 4 weeks, chest tubes to suction.
Outcome: gradually weaned down on MAP, had good oxygenation, leaks subsided, patient was eventually converted to conventional ventilation and weaned very slowly to extubation. Patient was discharged from the hospital in good condition after 10 weeks.
Previously healthy 8 yo girl developed hip pain. The hip was aspirated and irrigated when pus was returned. Gram stain showed Gram positive cocci in clusters. Patient went on to develop sepsis and fulminant respiratory failure, and ARDS. Methicillin resistant Staph aureus was cultured from her blood and wound. On approximately day 5 of illness, developed pneumothorax. 5 chest tubes had been placed prior to transfer to FUHC. Airleak was present
Management: HFOV for 5 weeks, chest tubes to suction (some to water seal alone)
Outcome: patient expired following repeated episodes of pulmonary hemorrhage into the pleural space. Airleaks did not diminish over time. Required progressively higher MAP. Did not develop appreciable MSOF. Chest X-ray showed multiple pneumatoceles.
Previously "healthy" 95Kg 17 yo presents with Bi-lobar pneumonia at outside hospital. Develops respiratory failure requiring intubation. Weans to nearly extubatable settings then has episode requiring massive increase in support, developing ARDS. Transferred to FUHC, developed first pneumothorax 4 days later (tension PTX), completely evacuated without residual airleak, subsequently developed further pneumothoraces and BPF within the next several days.
Management: permissive hypercapnea with pressure control ventilation
Attempted HFOV x one week, without improvement
eventually on SIMV high rate, low volume, all tubes (11 total) to suction
Outcome: required PD secondary to renal failure, ran sats in low 80's on 60-70% FiO2, had a persistant airleak of approximately 50% of delivered tidal volume (~10l/min). Also required massive amounts of cardiovascular support. Over the 4 month treatment course his acute lung disease resolved, after three months all of his chest tubes were removed. He was gradually weaned from the ventilator and his tracheostomy was decannulated approximately 9 months after the initial illness.
Excellent overview of Bronchopleural fistula management. Key Points from this review
Presented some interesting information regarding incidence
Onset:
Prognosis correlated with leak size
A paper out of Vienna reviewing various high frequency ventilation techniques, including CHFV (conventional high frequency ventilation), SHFJV (Super High Frequency Jet Ventilation), VDR 4, and the Hayek Respirator (an external, non-invasive high frequency ventilator).
They cite indications for HFV as the following:
Looked at all ventilated patients in an adult ICU looking for risk of barotrauma, defined as a pneumothorax, pneumomediatinum. The only independently predictive risk factor for pneumothorax was the presence of ARDS. Their development of PTX was sooner than the previously mentioned study, but was not specifically looking for persistent air-leak.
Used the HFJV in 29 children with ARDS. 20/29 survived. (Young children, ranged from 0.3 to 4 years of age). Started when patients developed barotrauma - defined as PIE, PTX, pneumomediastinum, pneumoperitoneum, or BPF. They used the jet at lower rates (240-300) than used elsewhere. Survivors went on to the Jet earlier (3.7 v 9.6 days), and were on the jet shorter (4.4 v 7.3days). NO patients actually developed BPF (at least none that they reported) either before or after HFJV. Although they do discuss 'airleak' which may simply refer to any extrapulmonary air. Bronchoscopy showed no signs of necrotizing bronchitis. (interesting that they note multiple thromboemboli in one autopsy patient who had a large femoral venous catheter associated clot)
One of the papers refered to by the Pierson New Horizons review. They took a series of 9 patients with BPF and collected the gas excaping via the leak, then analyzed the gas along with the patients ventialtion and pulmonary status. About 25% of the Minute ventilation, on average, was expired through the chest tubes. CO2 was always present in the tube gas, often at higher concentrations than the ET-tube return. All chest tubes were maintained at -20cmH2O of suction. No attempts were made at changing the amount of suction, to see what affect this had on CO2 removal
A retrospective comparison of Pulmonary functions, radiographic anatomy and gas exchange in patients of differing duration of ARDS who all went on to ExtraCorporeal Support. All patients were at the same level of severity, as judged by the Murray Score. Only looked at presence of chest tube as a sign of barotrauma. The late ARDS patients showed somewhatbetter gas exchange (a better PaO2), and also required less PEEP. The late ARDS patients however had a much higher incidence of PTX (87% vs 30 and 46% for early and intermediate ARDS). This also correlated with a higher mortality (66% of patients with PTX, 42% of patients without PTX). There were not significant differences on CT, except the number of bullae corresponded to the length of ARDS (The difference was entirely within the dependent portion of the lung). Significant differences between survivors and non-survivors:
Non-survivors had higher pCO2, Higher PIP, larger DP, greater shunt fraction, more PTX, and more organ dysfunction when compared to survivors.
Briefly, VC - PIP - 60's (for 12 days) changed to PC with PIP of 42 then down to 38 and the airleak went away. This was a trauma patient - stabbing - hemopneumothorax.
Back to Weekly Conference list
Back to U of M: Peds PCCM home page
This page created on 10/15/97, last updated 10/2/98
The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota.