Laser Photoacoustic Detection of Pulmonary and Tracheobronchial Suture Leaks

Vollständiger Abstract

Worum geht es in dieser Arbeit?

ABSTRACT Background Prolonged postoperative air leak (PAL) after thoracic surgery remains a major cause of morbidity, extended hospital stays, and bronchopleural fistula. Existing intraoperative aerostasis assessment methods, including water submersion testing, iodine indicator instillation, and ventilator flowmetry, are subjective, anatomically restricted, or provide limited information about the precise location and magnitude of individual leaks. This experimental study evaluates a quantitative intraoperative aerostasis monitoring approach employing a laser photoacoustic leak detector with sulfur hexafluoride SF 6 as a biologically inert trace gas marker, configured in a rabbit thoracotomy model. Methods Validation proceeded in two sequential phases. Bench‐top experiments used a latex‐balloon respiratory analogue ventilated under positive end‐expiratory pressure conditions (PEEP 490–1500 Pa), with controlled microdefects introduced using a 26 G needle. Acute in vivo studies were conducted in a New Zealand White rabbit (3.2 kg) anesthetized with isoflurane and undergoing left‐sided lateral thoracotomy with controlled visceral pleural and bronchial stump defects. SF 6 was introduced into the ventilator circuit at 0.02 vol.% (1:5000 v/v). The laser photoacoustic leak detector probe (3 mm inner diameter) was hand‐scanned across the operative field while photoacoustic signals were recorded continuously at 10 Hz. A mass‐balance model tailored to rabbit ventilatory parameters was derived and experimentally validated to relate measured SF 6 concentration to defect efflux rate. Results All 12 controlled defects (10 visceral pleural and 2 bronchial stump) were detected. During systematic scanning of the exposed lung surface, peak signal‐to‐background ratios ranged from 155:1 to 540:1 at 3–10 mm probe stand‐off distance, with a median localization error of 2 mm (0–4 mm range) relative to the known defect sites. Calculated SF 6 efflux rates agreed with theoretical predictions within 14%. Continuous ambient monitoring showed operative‐field SF 6 concentrations of 0.012–0.058 ppm, more than 13,000‐fold below the occupational exposure limit. Introduction of 0.02 vol.% SF 6 into the ventilator circuit produced no detectable changes in end‐tidal carbon dioxide or ventilatory mechanics compared with baseline. Conclusions In this rabbit thoracotomy model, laser photoacoustic SF 6 leak detection is feasible, safe, and quantitatively reliable for intraoperative aerostasis assessment. The method offers high sensitivity and millimeter‐scale spatial resolution when the operative field is systematically scanned, thereby complementing conventional aerostasis tests that rely on visual detection of air bubbles. These findings should be interpreted as proof‐of‐concept in an experimental setting. Further work is required to refine the human–machine interface, to assess performance in more complex operative fields and minimally invasive procedures, and to compare this approach with existing acoustic and gas‐analysis techniques. The validated mass‐balance framework provides a basis for scaling efflux‐rate estimation to human ventilatory parameters and for future clinical studies linking intraoperative leak characteristics with PAL.