Lung Diffusing Capacity in Dutch Special Operations Forces Divers Exposed to Oxygen Rebreathers over 18 Years

Round 1

Reviewer 1 Report

This is a well-written and presented paper, depicitng interesting observational retrospective data.  Lung function data is shwon for 257 Navy divers who use specialised diving equipment that exposes them to high oxygen levels.   FEV1, VC,  CO diffusing capacity and alveolar volume were measured annually for 18 years. The cohort was split into self-reported smokers  and non-smokers. The data was analysed and presented appropriately.

While FEV1, VC, VA remain constant there is a slight decrease in CO diffusing capacity, which seems to be related to smoking and natural aging. 

The data presented supports the conclusions given.  The study supports the view that tobacco smoking is a health hazard and a habit notouriously difficult to give up. 

It is recommended that the title be amended by the authors to remove the phrase “no clinically significant reduction in” eg to  " Lung diffusing capacity in Dutch Special Operations Forces divers exposed to oxygen rebreathers over 18 years" .  This observation study provides lung function data only, so it is inappropriate to “headline”  what is a clinical opinion.

A further weakness of the study is not being able to explore if there is a link between exposure time and decline.

Minor points:

Table 1.  It is presumed that length here means height, please qualify if this different to standing height or relevance.

Figure 1 caption, remove plural: functions.

Omit reference 6 in text and reference list as this paper is not available to the public reader.

Author Response

Dear colleague,

Thank you for your time to review our manuscript and your kind words. We would like to respond to your brought up issues. 

Concerning the title: Thank you for this suggestion. Although we feel the clinical interpretation in this paper does follow from the presented data, specifically table 3, we can agree that a more neutral title is more approriate. We have adopted your suggestion.

Concerning the missing data on exposure: We are aware that this is a limitation of the study, as written in line 81-82. The Dutch Ministry of Defense does not allow public statements addressing this topic. However, though various public sources it can be derived that diving exposure can vary from a few hours per day to per week. We have not revised this in the paper due to the aforementioned restrictions, but does this give more insight in the reason why we could not include this?

Minor points: 
- You are correct. Although the paper has been checked by a native speaker with experience in revising scientific papers, this error has slipped through. Your revision has been adopted. 

- Thank you for your keen eye. This has been corrected.

- Thank you for addressing this omission: this thesis is only recently publicly available (due to an previously embargoed publication in Diving and Hyperbaric Medicine; which has become publicly available two years after publication). The URL has been added to the reference and can be accessed publicly.

Reviewer 2 Report

Exposure to oxygen rebreathers throughout a career induces no clinically significant reduction in diffusing capacity in Dutch Special Operations Forces divers.

den Ouden T, Wingelaar T, Endert E, van Ooij PJ.

This manuscript tests the hypothesis that tests of pulmonary function and diffusing capacity would not be affected by exposure to hyperoxia due to the use of oxygen rebreather systems in Special Operations Forces (SOF) of the Royal Netherlands Navy. The authors evaluated pulmonary function tests and diffusing capacity measurements (carbon monoxide) conducted during routine annual examinations over the course of the careers of divers in the SOF. In addition to years spent on the SOF diving group, the authors also evaluated the effects of age and smoking status. They found that there was a statistically significant effect of diving on TLco and Kco, but not the PFT results. These statistically significant results were not clinically significant, and were an order of magnitude smaller than the effects observed for smoking, which were also not clinically significant. Thus, the study finds that exposure to oxygen toxicity by these divers is not likely to result in appreciable lung injury.

The manuscript is well-written and easy to follow. The data analysis appears to be rigorous and the conclusions well-founded. I just have the following minor issues to point out:

1. The authors state they were unable to include any metrics of the number of dives or amount of time spent diving due to military restrictions. This is obviously lacking information that would greatly improve the quality of the study, but I assume that even including vague ballpark numbers for these endpoints is not permitted? For a non-military general audience like myself, I would not know whether the number of dives is in the range of a few hours per day, week, month, or year.
2. The authors do not explain clearly how they distinguish between changes in lung function attributed to normal aging vs diving years. I think maybe they were using z-scores based on the GLI-2012, but aren’t these just PFT results? What about the DLco metrics?
3. The x-y plots in Figure 1 are difficult to see. Could the font sizes of the axes labels, numbering, and R2 values be increased? Also, could the thickness of the regression lines be increased so they are easier to see?
4. It is established that both smoking and diving have an effect on lung function. Is it possible to determine whether diving had a greater effect on the smokers than it did in non-smokers?

Author Response

Dear colleague,

Firstly, we would like to thank you for your time and kind words. We would like to address your suggestions point by point:

1. Thank you for this comment. We are aware that this is a limitation of the study. The Dutch Ministry of Defense does not allow public statements addressing this topic. However, though various public sources it can be derived that diving exposure can vary from a few hours per day to per week. We have not revised this in the paper due to the aforementioned restrictions, but does this answer your question?

2. 

The effect of ageing and years of diving has been distinguished by analyzing years of diving as the main effect and dding age as a covariate (as described in line 94). With the data presented in table 2, any values with a p>0.05 are not significant contributing factors. For example, the value of -0.024 at IVC has a p-value of 0.002, therefore a statistically significant decline of 0.024 per year of diving. To enhance interpretation of the data, a few lines have been added between 115 and 124.

3. We have adopted your suggestion and revised the figures accordingly. Thank you for this suggestion.

4. This is quite difficult with our dataset, as the amount of datapoints in the smoking group is substantially smaller than the non-smoking group, both in terms of number of divers and years follow-up. The GEE that results from the smoking data alone is substantially less accurate (in our case: overfitting) than the total dataset. In short: although technically possible this would result in an inaccurate model.