What Should we do About Lung Squeeze?
So far, competition organisers have responded to the tragic death of Nicholas Mevoli in 2013 by looking for ways to exclude divers who seem to be at risk of serious injury. At the moment this seems like the only sure way of reducing the likelihood of another serious accident. One of the most common approaches has been to test for small injuries sustained during competition using SaO2 monitoring, symptom reports and examination by a medic. It now looks as though previous injuries make later, larger injuries more likely and divers who fall below a given threshold, or who report symptoms of lung squeeze are now often excluded for part, or all of the rest of the competition.
One of the problems of this approach is that it can be seen to encourage dishonesty amongst the athletes. In the absence of good objective tests of existing injury, reckless divers will feel they can profit by concealing their symptoms and trying to appear well when they are not. What we need is an objective test that can tell us whether or not someone is fit to dive and for a long time the feeling has been that measuring O2 Saturations after the dive might be the best candidate for such a test.
We now have the first real scientific evidence that post-dive SaO2, a completely objective measure, usually corresponds to the presence or absence of lung injury. How useful is this information? How should it inform the way we design our processes for competition safety? The answer, of course, isn’t simple.
What Does a Good Test for Lung-Squeeze Look Like?
When looking at the usefulness of any kind of test for a medical condition, 2 factors are considered to be important: these are Sensitivity and Specificity.
Sensitivity is a measure of how likely the test is to be positive if you actually have the condition. A good example of a sensitive test is the pregnancy testing kits that you pee on. If you’re actually more than 5 weeks pregnant, it’s almost certain that the test will be positive. This is useful because a negative test is a good indication that you’re not pregnant.
Specificity, on the other hand is a measure of how likely it is that a positive result on the test corresponds to the presence of the disease (or injury) in question. Unlike the pregnancy test, a positive result on a test for, say, high levels of Ferritin (a measure of low iron stores) might mean that you have low iron levels but it can also mean that you have some kind of inflammation going on, because Ferritin also rises in response to inflammation. Ferritin is still quite sensitive for low iron stores, though, because if it’s normal, it’s quite likely that your iron stores are too.
So how good is SaO2 as a marker of lung injury?
Research commissioned by AIDA from Professor of Environmental Physiology Erika Schagatay has recently demonstrated that having low O2 Saturations 10min after surfacing from a dive is a good indicator of lung injury. In her study which (by her own admission) contains a relatively small number of cases, 8 out of 9 divers with symptoms of lung squeeze had an SaO2 less than 97% 10mins after surfacing. This equates to a Sensitivity of almost 89%. In simple terms, this means that if we exclude all divers with a saturation lower than 97% (a ‘positive test’), 10-15min after surfacing, we should catch around 89% of lung squeezes. Not bad, you might think!
If we look at the Specificity of this test, however, things aren’t so good: Lots of divers who triggered a “positive” test at the 97% saturation level (6 out of 14), had no symptoms at all. This equates to a specificity of only 57%. In simple terms, if we used this measure alone to exclude divers, according to this research, 43% of the people we excluded would very likely not have a lung squeeze at all!
At the 95% saturation cutoff things look a bit different. The Specificity is much better: Here, 5 out of 6 divers (83%) who triggered a positive test actually had symptoms of a squeeze. On the other hand, sensitivity was worse: Only 5 of the 9 divers who reported symptoms was caught by the test, that’s a sensitivity of just 56%, meaning that many true squeezes would be missed. The numbers of divers in the study are small but already we can see that it’s not easy to find a system that works both ways.
There are a few qualifiers to these conclusions too: First, the quoted numbers equate symptoms of lung squeeze to actual squeeze. It’s not clear that
a) all divers with actual squeeze will have symptoms or
b) that all symptoms of lung squeeze reported actually equate to a real squeeze (reporting symptoms also has a sensitivity and specificity, of course).
Ultrasound testing is perhaps a better way to identify actual squeeze and the results from preliminary testing of this still show a good correlation with SaO2.
Another possible qualifier is that this study too relies on honesty on the part of the participants. It’s possible that, since the divers were all involved in competition, there was a bias towards down-playing or concealing symptoms of squeeze. Finally, the question of “how long should I rest” which is important to answer if we’re going to make meaningful decisions about excluding divers from competition, is outside the scope of this study. Far more work is needed to iron out some of these issues, and fortunately a lot of it is already in the pipeline.
How Should we Act to Reduce the Risk of Serious Accidents in Freediving Following This Research?
Reading the analysis above, you might be left with the feeling that, from a practical perspective, all Erika’s hard work has been a bit of a waste of time. Although she’s conclusively demonstrated that low Post-Dive SaO2 is associated with lung-squeeze, we’re faced with a choice between, on the one hand excluding a lot of divers who have no injury at all, and on the other, missing a lot of squeezes which could cause serious accidents in later dives. Is this really the case?
No. In fact, what this piece of information really tells us is that using a single test as a means of excluding divers from competition is not currently a good option and as long as we understand that, this research, far from being a waste of time, is extremely valuable.
Testing for post-dive O2 saturations is an excellent ‘first screening’ test to filter out people who “almost certainly” (i.e. with 89% certainty) don’t have a lung-squeeze. Once we’ve identified the population of people who might have a squeeze, we need to rely on other measures of fitness to dive in order to make exclusion decisions (if, indeed, that’s the way we want to go). I’ve written before about the need for qualified professional people to fill the roles of competition medic, platform medic and to make exclusion decisions, and highlighted the fact that a completely different skill set is required for each job. Once at-risk divers are identified, the only sure way to protect them is to open a dialogue between the individual and a trained medical professional who is able to offer them constructive advice. To me this research serves to highlight that need more than ever.
Post-Dive SaO2 is a great field test. It’s cheap and easy to administer and, when the right cutoff point is used, has good sensitivity. It’s specificity, however, is poor at the level of sensitivity that’s required for meaningful screening. Once again we’re confronted with the fact that the real backstop for competition safety is a culture of openness and honesty amongst divers themselves and an unwillingness to reach the plate at any cost.