Quantification of lobar gas exchange: a proof-of-concept study in pigs

Abstract

Editor—The prediction of postoperative lung function helps to stratify lung cancer patients' risk of mortality and potential suitability for resection.1Lim E. Baldwin D. Beckles M. et al.Guidelines on the radical management of patients with lung cancer.Thorax. 2010; 65 (iii1–27)Crossref Scopus (371) Google Scholar Carbon monoxide transfer (TLCO) and forced expiratory volume in 1 s (FEV1) tests are commonly used to determine diffusion capacity and degree of airway obstruction respectively, and to predict post-resection dyspnoea, assuming they would decrease in proportion to the number of lung segments resected. These tests provide global indicators of lung function, but cannot distinguish between functional contributions from different lobes. Gas exchange occurs heterogeneously, differing between lung regions in health and disease,2Wagner P.D. The physiological basis of pulmonary gas exchange: implications for clinical interpretation of arterial blood gases.Eur Respir J. 2015; 45: 227-243Crossref PubMed Scopus (61) Google Scholar and this heterogeneity can be exaggerated by lung cancer, limiting the usefulness of whole-lung function tests. Estimates of postoperative FEV1 based upon CT measurements of resected lung volume3Papageorgiou C.V. Antoniou D. Kaltsakas G. Koulouris N.G. Role of quantitative CT in predicting postoperative FEV1 and chronic dyspnea in patients undergoing lung resection.Multidiscip Respir Med. 2010; 5: 188Crossref PubMed Scopus (9) Google Scholar or quantitative ventilation/perfusion scintigraphy1Lim E. Baldwin D. Beckles M. et al.Guidelines on the radical management of patients with lung cancer.Thorax. 2010; 65 (iii1–27)Crossref Scopus (371) Google Scholar may be more informative. However, these methods are limited by poor correlation between postoperative FEV1 and functional capacity, where maximal oxygen uptake is more sensitive.4Wang J.S. Abboud R.T. Wang L.-M. Effect of lung resection on exercise capacity and on carbon monoxide diffusing capacity during exercise.Chest. 2006; 129: 863-872Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar Quantifying lobar contribution to overall pulmonary gas exchange could provide more accurate predictions of postoperative lung function. For example, if a diseased lobe contributes minimally to gas exchange, its resection will be unlikely to have a major functional impact. The aim of this study was to measure oxygen uptake at the lobar level in a proof-of-concept experimental study. A fibreoptic sensor was used to measure tidal variation in lobar partial pressure of oxygen (Po2), and whole-lung CT to quantify lobar tidal volume; these measurements combined were used to calculate lobar oxygen uptake. This study, conducted in the Hedenstierna Laboratoriet at Uppsala University, Sweden, received ethical approval (AREC ref.C98/16) and conformed with the National Institutes of Health (NIH) and Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines.5Kilkenny C. Browne W.J. Cuthill I.C. Emerson M. Altman D.G. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research.PLoS Biol. 2010; 8e1000412Crossref PubMed Scopus (4264) Google Scholar Two male domestic pigs (29 kg) were studied in dorsal recumbency under general anaesthesia and mechanical ventilation via tracheostomy; details of the anaesthesia protocol are presented elsewhere.6Lachmann B. Robertson B. Vogel J. In vivo lung lavage as an experimental model of the respiratory distress syndrome.Acta Anaesthesiol Scand. 1980; 24: 231-236Crossref PubMed Scopus (504) Google Scholar Mechanical ventilation was delivered by a Servo-I ventilator (Maquet, Rastatt, Germany) in pressure-control mode with tidal volume of 10 ml kg−1, ventilatory frequency of 12 breaths min−1, inspiratory/expiratory (I/E) ratio of 1:2, and inspiratory rise time of 0 s so each breath consisted of 1.67 s of inspiration and 3.33 s of expiration. A saline lavage surfactant-depletion lung-injury model was induced in one pig to study heterogeneous lungs.6Lachmann B. Robertson B. Vogel J. In vivo lung lavage as an experimental model of the respiratory distress syndrome.Acta Anaesthesiol Scand. 1980; 24: 231-236Crossref PubMed Scopus (504) Google Scholar Cardiopulmonary variables (IntelliVue M8004A, Philips Healthcare, Amsterdam, the Netherlands; Capnomac Ultima, Datex-Ohmeda, Madison, WI, USA) and lobar Po2 (OxyLite Pro, Oxford Optronix, Abingdon, UK) were continuously monitored, with analogue signals digitised using PowerLab (ADInstruments, Dunedin, New Zealand) and recorded with LabChart version 8.2.1 (ADInstruments) at a sampling rate of 10 Hz throughout. Data were processed using R version 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria; www.r-project.org). Po2 was recorded with fibreoptic sensors with a response time of <150 ms in air.7Chen R. Formenti F. Obeid A. Hahn C.E. Farmery A.D. A fibre-optic oxygen sensor for monitoring human breathing.Physiol Meas. 2013; 34: N71-N81Crossref PubMed Scopus (13) Google Scholar The sensor working principle is based on luminescence quenching by oxygen of a fluorophore embedded in a polymer material with technical details as described elsewhere.7Chen R. Formenti F. Obeid A. Hahn C.E. Farmery A.D. A fibre-optic oxygen sensor for monitoring human breathing.Physiol Meas. 2013; 34: N71-N81Crossref PubMed Scopus (13) Google Scholar, 8Formenti F. Bommakanti N. Chen R. et al.Respiratory oscillations in alveolar oxygen tension measured in arterial blood.Sci Rep. 2017; 7: 7499Crossref PubMed Scopus (15) Google Scholar, 9Formenti F. Chen R. McPeak H. Matejovic M. Farmery A.D. Hahn C.E. A fibre optic oxygen sensor that detects rapid PO2 changes under simulated conditions of cyclical atelectasis in vitro.Respir Physiol Neurobiol. 2014; 191: 1-8Crossref PubMed Scopus (26) Google Scholar, 10Crockett D.C. Cronin J.N. Bommakanti N. et al.Tidal changes in PaO2 and their relationship to cyclical lung recruitment/derecruitment in a porcine lung injury model.Br J Anaesth. 2019; 122: 277-285Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar The fine-bore sensor was inserted into a bronchoscope until the sensor tip was visible inside the main bronchus entering a lobe, with the bronchoscope itself remaining within the large airways. Data for ∼12 breaths were collected in each lobe and averaged to produce a single breath per lobe. Po2 tidal variation was calculated as the peak-to-trough difference in the averaged breath. Tidal variation in lobar Po2 (kPa) was then converted to tidal variation in oxygen concentration (DeltaPo2, %), dividing by 101.3 kPa. This experiment could not be completed in the CT scanner, so lobar Po2 measurements were completed just before CT imaging. Whole-lung volume CT scans (Somatom Definition Flash; Siemens Healthcare, Forchheim, Germany) were recorded during end-inspiratory and end-expiratory breath-holding manoeuvres (under the same ventilation conditions as during the lobar Po2 experiment) to measure lobar tidal volume. Lobar volumes were calculated via segmentation at 3-mm intervals using 3D Slicer version 4.1 (https://www.slicer.org); Figure 1a–c illustrates the segmentation process. Gas volumes were then calculated asLobar gas volume (ml)=Lobar volume (cm3)×−[Mean voxel density (HU)]1000,and lobar oxygen uptake was calculated by multiplying tidal gas volume (difference between end-inspiratory and end-expiratory lobar gas volumes) by DeltaPo2 (%). Further methodological details are presented in Supplementary Table S1, which shows the baseline characteristics of the animals studied. Cardiorespiratory parameters were within the normal or expected range, with a lower P/F ratio in the lung injury model (15.1 kPa) than in the control (61.5 kPa). In order to achieve normoxaemia, inspired oxygen concentration was higher in the lung injury model (0.7 vs 0.4 [control]). Figure 1d illustrates the lobar Po2 tidal variation, greater in the saline lavage lung injury model than in the control pig. Figure 1 inset table presents lobar Po2, end-inspiratory, end-expiratory and tidal volumes, and the associated lobar oxygen uptake. Po2 tidal variation ranged from 4.7 to 20.3 kPa in different lobes. Lobar tidal volume ranged from 12.9 to 75.1 ml, with lobar oxygen uptake ranging from 0.6 to 11.9 ml breath−1. Figure 1e shows oxygen uptake in both animals, visualising the difference between lobes, and overall pulmonary oxygen uptake between animals. These results demonstrate the feasibility of a novel technique to calculate lobar oxygen uptake in a mechanically ventilated control and saline lavage lung injury pig model. Study limitations include the small sample size and use of a porcine model that do not fully replicate all features of the human respiratory system. Data were collected at rest, assuming proportional contributions during exercise, when functional limitation likely occurs. Nevertheless, the lung injury model used resulted in heterogeneous lung injury and hypoxaemia, and was sufficient to determine differences in lobar and overall oxygen uptake compared with the control. The distribution of lobar contribution to gas exchange can vary significantly, even in patients with similar overall lung function, but tests currently used to predict lung function after lung cancer resection cannot determine gas exchange at the lobar level. After refinement, the technique proposed here could provide valuable insight into the proportion of oxygen uptake associated with a given lobe and support the decision on patients' suitability for lobar resection. Although this technique was tested under conditions of general anaesthesia and mandatory ventilation, its working principle would remain valid during spontaneous breathing, allowing use of the technique in the outpatient setting. The technique is clinically feasible given the widespread use of both bronchoscopy and CT imaging in lung cancer diagnosis. We are grateful to Rongsheng Chen and Oxford Optronix for making the fibreoptic oxygen sensors available, to Dr Douglas Crockett, to Dr João Batista Borges and to the staff in the Hedenstierna Laboratoriet and in the Radiology Department, Uppsala University Hospital, including Agneta Roneus, Kerstin Ahlgren, Mariette Anderson, Liselotte Pihl, Maria Swälas, Monica Segelsjö, Göran Hedenstierna, Anders Larsson, and Miklos Lipcsey for their support with data collection. The authors declare that they have no conflicts of interest.