Estimating the Effect of Lung Collapse and Pulmonary Shunt on Gas Exchange During Breath-Hold Diving. The Scholander and Kooyman Legacy
Authors: A. Fahlman, S. K. Hooker, A. Olszowka, B. L. Bostrom, D. R. Jones
DOI / Source: https://doi.org/10.1016/j.resp.2008.09.013
Date: 26 September 2008
Reading level: Advanced
Why This Matters for Freedivers
This paper explains (using a detailed model) why “lung collapse depth” is not a single fixed number—it depends on lung volume, depth, and how gas exchange gradually shuts down as the lungs compress. That matters for freedivers because it shapes how nitrogen and oxygen might move during deeper or repetitive profiles, and why simple all-or-nothing explanations can be misleading.
Synopsis
For decades, diving physiology relied on a neat idea: as you descend, your lungs compress, the alveoli eventually collapse, and gas exchange stops—reducing nitrogen uptake and helping protect against decompression sickness (DCS). The problem is that real measurements in diving animals don’t fit a simple “open until a magic depth, then closed” model. This paper tackles that gap by building a mathematical model that links lung compression to a gradual loss of gas exchange through increasing pulmonary shunt (blood bypassing the lungs).
The authors model how oxygen (O₂), carbon dioxide (CO₂), and nitrogen (N₂) move between lung, blood, and tissues while the lungs are being squeezed by pressure. A key insight is that compression can briefly spike arterial gas pressures early in descent (because the same gas is packed into a smaller volume), but as the dive continues and gas is absorbed, the effective gas-exchange surface and function drop—eventually approaching a point where exchange becomes minimal.
One of the most practical takeaways is how strongly “diving lung volume” (how full the lungs are at the start) changes everything. In the model, starting with a larger or smaller lung volume shifts the depth and timing of lung collapse/shunt and causes big differences in predicted nitrogen levels in blood and tissues. That means an animal (or diver, in principle) can influence gas exchange with behavior—by changing pre-dive lung volume. The paper also shows that different assumptions about collapse depth (instant collapse vs graded shunt vs no collapse) can produce dramatically different nitrogen predictions, which is crucial because those predictions are often used to estimate DCS risk.
Overall, the message is: lung compression, circulation, and gas exchange interact in complex ways. Models that assume a simple “on/off switch” for lung collapse can be very wrong, and interpreting results without considering graded shunt and lung-volume effects can lead to false confidence.
Abstract
We developed a mathematical model to investigate the effect of lung compression and collapse (pulmonary shunt) on the uptake and removal of O₂, CO₂ and N₂ in blood and tissue of breath-hold diving mammals. We investigated the consequences of pressure (diving depth) and respiratory volume on pulmonary shunt and gas exchange as pressure compressed the alveoli. The model showed good agreement with previous studies of measured arterial O₂ tensions (PaO₂) from freely diving Weddell seals and measured arterial and venous N₂ tensions from captive elephant seals compressed in a hyperbaric chamber. Pulmonary compression resulted in a rapid spike in PaO₂ and arterial CO₂ tension, followed by cyclical variation with a periodicity determined by total cardiac output. The model showed that changes in diving lung volume are an efficient behavioural means to adjust the extent of gas exchange with depth. Differing models of lung compression and collapse depth caused major differences in blood and tissue N₂ estimates. Our integrated modelling approach contradicted predictions from simple models, and emphasised the complex nature of physiological interactions between circulation, lung compression and gas exchange. Overall, our work suggests the need for caution in interpretation of previous model results based on assumed collapse depths and all-or-nothing lung collapse models.