Are deep stops better after all?
A new study by Alessandro Marroni and colleagues has stirred up the debate on deep decompression stops: in two trimix dive profiles, divers produced fewer venous gas bubbles (VGE) following the DEEP profile than following the SHALLOW profile. Three cases of skin decompression sickness occurred only after the SHALLOW profile. The authors see this as an apparent contradiction to the well-known NEDU study by Doolette and colleagues and as an indication that deep stops are, after all, superior.
How was the study designed?
The authors investigated the effect of different decompression profiles on venous gas bubble formation in 18 experienced divers. Each participant completed, in random order, two dives to 40 metres with a bottom time of 40 minutes. Trimix 21/35 was used as the breathing gas, with EAN50 used from 21 metres onwards for both profiles. With a comparable total dive time of approximately 80 minutes, the profiles differed primarily in the distribution of the decompression stops: in the DEEP profile, these began at 21 metres, whilst in the SHALLOW profile they began only at 9 metres.
The authors initially used GF 20/85 to determine the common decompression time. According to the publication, the profiles derived from this and manually adjusted correspond approximately to GF 30/80 for DEEP and GF 100/75 for SHALLOW. This was chosen to simulate a technical dive with a long decompression stop at 6 m (although the aim there is to decompress at 6 m using 100% O₂ rather than EAN 50).
Dekostopps | DEEP | SHALLOW |
21 m | 4 min | – |
18 m | 2 min | – |
15 m | 2 min | – |
12 m | 2 min | – |
9 m | 4 min | 5 min |
6 m | 16 min | 22 min |
3 m | 8 min | 10 min* |
Gradient Factors | 30/80 | 100/75 |
Figures as per Table 1 of the publication. The methodology section, however, states 11 minutes for the final SHALLOW stop.
During the first 90 minutes after surfacing, venous gas bubble load was measured every 15 minutes using echocardiography. According to the DEEP profile, it was significantly lower – averaging 6.7 bubbles per heartbeat compared with 10.7 – than after the SHALLOW profile (p = 0.0006). Furthermore, three divers developed cutaneous decompression sickness with cutis marmorata following the SHALLOW profile, whilst no such cases occurred following the DEEP profile. However, due to the small sample size, this difference was not statistically significant.
The authors conclude that, under the conditions studied, the DEEP profile resulted in fewer venous gas bubbles. They see this as an apparent contradiction to the NEDU study by Doolette and colleagues, in which deeper stops were associated with a higher bubble load and more cases of decompression sickness.
But is the result of this study really surprising, and does it truly contradict the NEDU study? I don’t think so.
What the study really examined...
The DEEP vs SHALLOW study does not examine the benefits of deep stops, but rather compares, using EAN50, two different decompression strategies: a more conservative profile during the early ascent and a significantly more aggressive one.
In my view, it is hardly surprising that the more aggressive SHALLOW profile, with approximately GF 100/75, exhibits a higher VGE. Following the gas change, the supersaturation tolerance according to Bühlmann is practically fully utilised here as early as the ascent to the first stop. With the DEEP profile, on the other hand, the tissues are given time to release inert gas under the decompression gas whilst still at a greater depth, before the ambient pressure drops further. Against this background, the higher bubble load observed with the SHALLOW profile seems plausible and was to be expected.
Even the best medicine takes time to take effect. The situation is similar with a decompression gas: whilst the gas change immediately alters the partial pressures of the inhaled gases, the desaturation of the tissues takes time. It is therefore crucial to allow the decompression gas the opportunity to reduce the inert gas load before high supersaturations arise as a result of further ascent. This was not the case with the SHALLOW profile, but it was with the DEEP profile.
... and what Doolette tested.
Doolette’s NEDU study examined a completely different scenario: air dives to around 52 metres with a bottom time of 30 minutes, followed by decompression using air without any decompression gas. Nitrox was therefore not used.
Both profiles included 174 minutes of pure stop time. In the deep profile, decompression began at 70 feet, i.e. around 21 metres. At 21, 18 and 15 metres alone, the divers spent a cumulative total of 44 minutes. The shallow profile, by contrast, only began at around 12 metres. These are very different circumstances from four minutes at 21 metres and two minutes at 18 metres whilst using EAN50.
The NEDU results showed a higher incidence of decompression sickness following the deep profile: after a final assessment of the cases, there were 10 cases out of 198 dives, compared with 3 out of 192 following the shallow profile. The burden of bubbles was also higher.
The explanation: long, deep stops on air can delay the desaturation of slower-saturating tissues or even promote their further saturation.
No contradiction in the results
The Marroni study therefore does not contradict that of the NEDU. Rather, it is a case of comparing apples with pears. However, the authors do cite the different decompression strategy – in particular the longer stops at 21 and 18 metres on EAN50 – in their own study as the reason for the discrepancy.
Conclusion
The DEEP vs SHALLOW study does not contradict current accepted theory; on the contrary, it confirms it: an aggressive decompression profile at the limit of supersaturation tolerance leads to greater VGE, even when using a decompression gas.
For me, the results are therefore consistent with a sound basic principle: make sensible use of the supersaturation tolerance during the ascent without maximising it too early – and allow the appropriate decompression gas time to reduce the inert gas load. At the same time, one must not remain at depth for so long that slower-reacting tissues become unnecessarily oversaturated.
Marroni’s study is important because it confirms this strategy. However, it neither proves a specific optimal GF setting nor does it refute the disadvantages of extended deep stops using bottom gas, as demonstrated by Doolette.
Literature
Marroni A et al. Deep versus shallow decompression stops: illustration of different decompression strategies used in technical diving. International Maritime Health. 2026;77:157–167. DOI: 10.5603/imh.106791.
Doolette DJ, Gerth WA, Gault KA. Redistribution of decompression stop time from shallow to deep stops increases incidence of decompression sickness in air decompression dives. NEDU Technical Report 11-06, 2011.
further reading on dekoblog.ch:
Kommentare