Review of saturation decompression procedures used in commercial diving

SOURCES

The documentation related to the development of commercial diving in general, and saturation diving in particular, is dispersed, mainly located in company internal and restricted documents. The authors have been involved in commercial diving operations during their career. Non-referenced information in this review should be considered as sharing of personal experience. In particular, two authors worked for Comex and Stolt Comex Seaway and provided information on their historical diving manuals.

UNITS

The documents reviewed are operational procedures where depth is used instead of ambient pressure and where expressions such as ‘shallower’, ‘deeper’, ‘ascending’ or ‘surface’ are common terms. With editorial consent we deliberately kept this technical jargon for consistency. For the same reason:

• Gas partial pressures are expressed in mbar (1 mbar = 0.1 kPa). Gas fractions are expressed in percentages.

• Pressures are expressed in msw (1 msw = 10.0381 kPa according to EN 13319).

• Imperial units have been converted using 1 foot of seawater (fsw) = 0.30643 msw as specified in the US Navy diving manual. When procedures used both imperial and metric systems, only the provided metric values were considered.

DIVING COMPANIES

The IMCA website lists 50 companies that hold a certificate for ‘unrestricted diving’ which covers saturation diving. This number must be reduced to around 30 considering that some companies have multiple registrations. We selected 11 leading international companies for which we had connections through our professional activities. The conditions for participation were defined in a memorandum of understanding, signed by the authors and each diving company, stating that:

• The procedures could be used scientifically with the name of the company being unidentifiable.

• The company could review the final paper and retain the right to withdraw from the publication.

The eleven companies are:

• Boskalis, Papendrecht, The Netherlands

• DOF Subsea, Perth, WA

• Helix Well Ops, Aberdeen, UK

• K Subsea, Singapore

• McDermott, Houston, Texas, USA

• Mermaid Subsea Services, Bangkok, Thailand

• Rever Offshore, Aberdeen, UK

• Shelf Subsea, Perth, WA

• Subsea7, Aberdeen, UK

• TechnipFMC, Aberdeen, UK

They are later de-identified as company A, company B, etc in an order unrelated to the above list. For each of these diving manuals received we associated a saturation procedure called procedure A, procedure B, etc. Note that we compared procedures in the range of 200 msw to surface, independently of the deepest storage depth specified in the manual. We excluded from the study any specific procedures used for deeper diving. We considered procedures in use in 2020 and disregarded any subsequent versions.

It should be noted that changes have occurred in the industry since 2022. Rever Offshore was taken over by the company Boskalis Subsea Services, but their procedures were reviewed by the former Rever Offshore diving manager. Fugro have ceased manned diving operations and no longer maintain their diving manuals. Simon Binsted, the former Fugro Diving Manager received authorisation from Fugro management to review this paper on their behalf.

HELIOX SATURATION DECOMPRESSION PROCEDURES

We focused on the final decompression of saturation dives for which we identified several operational characteristics:

• Boskalis, Papendrecht, The Netherlands

• The minimal hold period at storage depth that is generally required after the divers have returned from their last excursion dive, before final decompression.

• An initial ‘pull-up’. This corresponds to a rapid pressure drop equivalent to an upward excursion that was historically used to initiate decompression.

• The decompression protocol. The decompression generally takes place as a continuous pressure reduction (‘continuous bleed’), or alternatively through incremental steps of typically 0.2–0.3 msw.

• The daily decompression period. Decompression can be continuous (24 h/24 h) or include interruptions for divers’ comfort.

• These interruptions are called ‘rest stops’ by the US Navy. The rest stop can be set at a fixed time (at night for instance) or after a given daily decompression time. In that case, the time of the rest stop depends on the final decompression start time (‘sliding rest stop’).

• The chamber oxygen. The decompression starts with a constant chamber PO₂. However, the chamber oxygen fraction increases as the pressure decreases and must be limited to less than 23% because of the fire risk. A common chamber PO₂ of 500 mbar will exceed 20% at 15 msw. From 15 msw to surface, the decompression proceeds with a constant oxygen fraction.

• The decompression rates, which vary depending on depth ranges. The term ‘decompression profile’ characterises the distribution of decompression rates over depth.

• The daily decompression rate which is the pressure reduction achieved in 24 h, including rest hold periods.

ANALYSIS

We first studied the operational features of the decompression such as initial pull-up, decompression hold, daily rate of ascent, etc. We have compared decompression procedures from these companies as well as reference procedures such as:

• The procedures published in the regulations of Norway, Brazil and France.

• The procedures of the US Navy Diving Manual.

• The procedures from historical diving companies, Comex and Stolt Comex Seaway, for which two authors worked.

We then attempted to discuss the safety performances by using four endpoints:

• The decompression sickness (DCS) incidence recorded during operations.

• The venous gas emboli (VGE) grade measured during or after the decompression.

• The oxygen exposure and its level of pulmonary toxicity.

DIVER POPULATION

To characterise the diving population, at least in the frame of the North Sea operations, one of the participating companies provided the age distribution of 131 divers who rotated onboard one of their vessels in 1979.

It is the authors’ view that saturation divers have significant experience; They traditionally start as air divers at 30 years old, move to saturation diving 10 years later and stay in the career. At the time of the study, the mean age of the saturation divers was 47 (range 30–61) (Figure 2).

HISTORICAL REFERENCE PROCEDURES

INTERNATIONAL REFERENCE PROCEDURES

Three countries, France, Norway and Brazil, have regulated diving to the level where ascent rates, rest stops, chamber PO₂, etc. are specified for saturation. These regulations cannot be used directly for operations but build a strict frame for editing saturation procedures. We mention hereafter the French, Norwegian and Brazilian procedures in this context.

REVIEW OF COMPANY PROCEDURES

These procedures are summarised in Tables 1, 2, 3, 4, along with the ones from the references already mentioned.

SOURCES OF C URRENT SATURATION PROCEDURES

The US Navy procedures have played a major role in the development of commercial diving because they were readily and freely available at the time it all started. The diving managers liked to claim they were complying with the US Navy because it was an unchallenged reference. Companies have since gained experience and introduced their own modifications such as increasing the PO₂ during the decompression and restricting excursion distances during the bell dives. Yet procedures C, F, G, H, I and K (six out of 11 companies) still reference the 2016 Rev 7 edition of the US Navy Diving Manual as the source of their saturation procedures.

The Comex manual largely inspired the Brazilian NORMAM-15 procedures. It also influenced procedures A, B, D and E that include a constant rate of decompression from bottom to 15 msw (four out of 11 companies).

The source of company J procedures are unknown.

These reference procedures have been empirically modified by the companies in consultation with their respective diving experts (medical advisors, consultants, etc.). The authors have participated in several of these diving manual revisions. However, the rationale behind these changes has often been lost. Only companies A, B, C, D and K (five out of 11 companies) have formally compiled documents called ‘Justifications’ or ‘Provenance’ that trace and explain the evolution of their diving procedures.

STORAGE DEPTH PO₂

The procedures all specify a chamber PO₂ at storage depth set around 400 mbar (Table 1).

The PO₂ at storage depth is raised to avoid complications in the event of hypoxia due to improper mixing of oxygen and inert gases. However, two other reasons are identified.

The first reason is historical. In the early 70’s, Comex mobilised a saturation system on the deck of the Choctaw I barge with an external regeneration system. The barge was experiencing bad weather, the system at 120 msw and the chamber PO₂ at 400 mbar, when a large wave swept the regeneration plant away. The 2-inch hoses burst and the pressure rapidly dropped inside the chambers. Divers started closing all the valves in panic, including the pressurisation valve, and the surface team had to use the bell pressurisation valve to re-establish the pressure. By that time, the chamber had dropped half of the initial pressure. Since then, the industry policy has been to set the storage PO₂ around 400 mbar so that the atmosphere could remain breathable if the pressure were to accidentally drop to half of its initial value.

The second reason is related to excursion dives. In case of an ascending excursion, the storage depth becomes the deepest depth. The storage depth PO₂ therefore influences the permitted excursion distance. If storage PO₂ and dive mix PO₂ are too different, ascending and descending excursions become asymmetrical. With modern procedures that use a sliding excursion window, a higher storage PO₂ provides a higher flexibility.

INITIAL ‘PULL-UP’

The US Navy Diving Manual paragraph 13.23 allows starting a saturation decompression with an ascending excursion (initial pull-up), based on the concept of the diver’s deepest depth, which directs the selection of saturation excursion distance. The excursion amplitude can be significant, as for instance, a 30 msw excursion from 120 msw to 90 msw. It is, however, specified that this initial pull-up remains within the discretion of the person in charge.

Even if the initial pull-up is limited, the problem is to measure the influence of this sudden pressure change on potential bubbles remaining from the last excursion dive and the impact of these bubbles on the following decompression. Flook used a bubble growth algorithm developed by Van Liew and Burkard to estimate this impact.[ 13] Flook modeled the bubble population after excursions and during saturation decompression and concluded that neither the excursion nor the decompression alone was likely to cause DCS. However, she pointed out that if a decompression was to follow an excursion with a too short interval, the residual bubble population from the excursion could interfere with the final decompression process and carried a risk.[ 14]

All the procedures reviewed have removed the possibility of starting a decompression with an ascending excursion.

DECOMPRESSION HOLD

Companies now specify a minimal time interval, called decompression hold, after an excursion dive, before starting a final decompression (Table 1).

This decompression hold is only required after a descending excursion dive for procedures C and G. All the other procedures request a pre-decompression hold regardless of the type of excursion dive (nine out of 11 companies).

The hold duration varies from 2 h (procedure H), to 6 h (procedures E, F, J) and 8 h (procedures A, B, C, D, G, I and K, i.e., seven out of 11 companies). In practice, a few hours are needed to raise PO₂ inside the chamber before decompression and this hold has a minimal impact on the operations.

Note that Comex authorised a 1 msw ascent performed in 10 minutes at the start of the decompression but this was only intended to create a small pressure drop to seal the door of adjacent chambers. A similar procedure is proposed by company K.

CHAMBER BLEED VERSUS STAGED DECOMPRESSION

Saturation decompressions are slow, e.g., a 90 min·msw^-1 rate of decompression corresponds to 11 millimetres depth change every minute. They require continuous attention from the chamber operators. Operationally, two methods are available for decompressing the chamber:

• Continuous decompression (‘continuous bleed’) typically controlled by computers onboard modern diving support vessels.

• Staged decompression with repeated small decrements of depth.

Several procedures have been identified for staged decompression:

• Comex 1979 used an optional 1 msw step decrement with 10 min ascent time to the next stop when deeper than 50 msw.

• Procedure D proposes an optional 0.33 msw decrement as in US Navy Rev 7 procedures.

• Procedure F proposes an optional 5 msw step with 5 min ascent time to the next stop.

• Procedure K proposes an optional 0.2 msw step resulting in an ascent rate equivalent to continuous decompression using the last minute of the stop time to travel to the next stop depth.

A problem of staged decompression is what Comex divers used to call ‘passage de bulles’ during the step changes near the surface (literally translated as feeling bubbles passing by, or, alternatively, feeling ‘niggles’). Because the effect of Boyle’s law becomes more important close to the surface, it can be speculated these small but sudden pressure variations increase bubble volume, resulting in a greater likelihood to produce symptoms in whatever tissue in which they are present (such as periarticular connective tissue). The staged decompression option was removed in the later versions of the Comex manuals and the symptoms disappeared. The other companies only propose continuous chamber bleed (eight out of 11 companies).

REST STOPS VERSUS CONTINUOUS DECOMPRESSION

Rest stops were first defined in the US Navy Diving Manual as a daily interruption in the decompression process (or ‘night stops’ when stops take place during sleeping time). The justification presented at the time was to avoid divers sleeping in a cramped position that could reduce perfusion during decompression. Another story told was that in the early times, the US Navy doctors were annoyed by awakening every night and decided to stop decompression to get some sleep (personal communication with Dr Spaur).

We identified several rest stop patterns in our review (Table 2):

• Rest-stops set at fixed times identical to the US Navy pattern (00:00 to 06:00 and 14:00 to16:00) or the NORSOK pattern (00:00 to 06:00) as in procedures F, G, H and I (four out of 11 companies)

• A rest stop set after a given decompression duration that slides around the clock depending on the start decompression time (procedures A, B, D, E and K, i.e., five out of 11 companies)

• Continuous decompression without any rest stop (procedures C and J)

The stop durations vary from 4 h (procedure E) to 5 h (procedures A, B, D) and the classic US Navy 8 h split over two stops (procedures F, G, H, I, K, i.e., five out of 11 companies).

From our experience, when divers are asked their comments on rest stops, they typically provide the following answers:

• Stops prevent the ‘popping’ of my ears when I sleep in the last part of the decompression.

• Fixed stops permit synchronising back to normal day rhythm (for divers on night shift).

• I do not care, I sleep a lot anyway during decompression, at any time.

Operationally, rest stops set at a fixed time raise the problem of calculating the end decompression time because the number of rest stops depends on the start decompression time. Rest stops may also happen a few metres from the surface, causing technical problems (toilet no longer in operation, risk of sudden surfacing) and frustration. One way around this is to forbid rest stops shallower than 3 msw (procedures H and K) or to carry them out at 3 msw (procedure A). Another way is to remove any rest stop in the last 24 hours of the decompression and adapt the ascent rate accordingly, which corresponds to a slow and conservative end of decompression (procedures B, D, E and I).

Theoretically, rest stops reduce the daily time available for decompression. For a given daily decompression rate, rest stops require a faster rate during the active decompression phase. The question remains whether the recovery during the rest stop exactly balances the increase of the decompression rate during the ascent phase. No theoretical work could be found on the subject. In 1984, an attempt was made during the DeepX II experiment at NUTEC to compare the performances of the two methods of decompression. Three divers were decompressed continuously while another group of three divers were decompressed with rest stops, both groups with the same 24 h decompression rate. No difference could be documented.[ 15] A similar conclusion was derived from the Comex database (Imbert JP, presentation at the NPD conference, 1988).

The presence of rest stops therefore remains more a matter of company culture than a strategy for improving the decompression safety.

DECOMPRESSION RATES AND DURATIONS

Decompression rates are linked to the chamber PO₂ and govern the decompression duration. They have a critical operational and commercial importance (Table 3 and 4). Decompression rate patterns determine two characteristics of the decompression: duration and profile.

We calculated the decompression durations and the instructions specified in the procedures. We compared these decompression durations for several typical storage depths. Note that the final decompression duration may vary depending on the starting time (we used 06:00 in the program) and the conversion factor used for fsw and msw. We added the NORSOK procedures for comparison.

Table 5 and Figure 3 display the difference between the slowest and the fastest decompression durations. This difference reaches 25.7 h at 150 msw. However, Table 5 also indicates that for procedures A, B, C, D, E, F and H, this difference is less than 5 h over the 60–150 msw range. This means that seven out of 11 companies have very similar decompression durations.

We then plotted the daily decompression rate versus depth to compare the decompression profiles between procedures. Figure 4 shows that procedures have a similar decompression rate to 60 msw while there are greater differences in the shallower depths.

Deeper than 60 msw, all the procedures reviewed are characterised by a constant decompression rate. This has been a characteristic of the Comex saturation decompression since 1984 (and derived procedures like the NORMAM-15).[ 16] It is a consequence of the Comex method of calculation that used a safe ascent criterion based on Hennessy’s critical volume assumption.[ 17] The same result can be obtained using Vann’s model that predicts a linear relation between the rate of decompression and the PO₂.[ 18] This is also a consequence of the concept of extended oxygen windows.[ 19] According to these algorithms, the rate of decompression is a linear function of the PO₂ and should remain constant as long as the chamber PO₂ remains constant, regardless of depth.

The shallower part from 60 to 15 msw is also conducted with a constant PO₂ but reveals two practices, one with a constant decompression rate (as per the Comex algorithm), and the other with varying decompression rates (as per the US Navy tradition). The US Navy has never published the way their saturation decompressions were computed. It is likely that in the early 70’s, they used trial and error and involved a combination of various models. This profile consisting of deep constant decompression rates and shallow varying decompression rates is typical of procedures that use the US Navy as parent procedures. It is also found in the NORSOK procedures that adopted this profile as a best practice at the time they were written.

INTERMEDIATE DECOMPRESSION

Projects often require intervention at various working depths. When depth variations exceed the possibilities of bell excursion dives, the storage depths can be adjusted by intermediate compressions and/or decompressions.

In the early time of North Sea installation, some clients insisted in decompressing divers during bad weather, based on the idea that they would be safer closer to the surface if the situation was to deteriorate. Divers could be subjected to a significant series of intermediate decompressions just because of bad weather. Following three intermediate decompressions in a row imposed on Comex divers by the November weather conditions in the Shetlands, the practice was eventually banned.

In 1986, Comex introduced a limitation to intermediate decompression based on the understanding that decompression was stressful and limitations should be based on a ‘maximal acceptable dose’. The maximum dose was defined as a decompression distance of 200 msw, which corresponded to the deepest available storage depth in their manual. The principle was that this distance could be split into a series of intermediate decompressions. Intermediate decompressions could be cumulated as long as their total distance would not exceed 200 msw. The system could be ‘pushed’ beyond reasonable limits if multiple small intermediate compressions/decompressions were used (saw tooth-shaped profile).

Around the same time, another company, Rockwater, introduced a limitation based on the number of intermediate decompressions followed by a compression to a new storage depth, known as the ‘W’ profile (Figure 5). The W-profile was restricted to one intermediate decompression and one intermediate compression before final decompression. The system could also be ‘pushed’ by using an intermediate decompression/compression of high amplitude but many companies adopted it because of its simplicity.

Finally, in Norway, NPD referred to a publication from the Hades database to justify the notion that the dive planning should be based on minimum change of storage depths and excursion exposures. The NORSOK standards thus included the more restrictive rule of the ‘V’ profile where divers can work at intermediate storage depths during decompression but cannot be recompressed to any deeper storage depth.[ 20] This position was later judged as a misinterpretation by JE Jacobsen, one of the main authors of the Hades paper, during a presentation at a DMAC meeting in 2017.[ 21]

The review shows that current practice for limiting intermediate decompression is a mixture of the V, W profiles and Comex cumulative decompression distance.

EXERCISE DURING DECOMPRESSION

Exercise can be associated with muscle stretch, impacts on joints and vibrations. Preconditioning studies on divers have shown that exercise reduces VGE levels after the dive.[ 22 , 23] It has been postulated that vibrations and vasodilatation affect a pre-existing population of bubble precursors and therefore could reduce the source of bubble formation.[ 24 , 25] On the other hand, Madden et al. showed that exercise after diving increases the incidence of bubbles appearing in the arterial circulation, possibly because of the increase in the pulmonary artery pressure.[ 26] It is therefore assumed that a light and measured exercise could be beneficial during decompression.

Such potential benefit was subjectively evaluated via a questionnaire survey performed in 2017 onboard a North Sea diving support vessel by one of the authors but not published in the associated paper.[ 27] Answers evoked a matter of lifestyle. Old divers are reluctant to exercise because, during their long careers, they have been consistently told that exercising is harmful. Younger and fitter divers exercise during decompression and claim that they feel and sleep better during this long boring period.

Our review has shown that eight out of 11 companies encourage divers to lightly exercise during the decompression. This takes the form of a small paragraph named good or healthy practice during decompression stating: “Move around regularly. Do not maintain a cramped position that restricts blood circulation” (procedures G,E and J). The more detailed form (procedure A for instance) specifies that the exercise must remain moderate such as steppers, bungees and static bicycles. In the absence of definitive scientific evidence, light exercise during saturation decompression remains a matter of diver’s personal choice.

PUBLISHED DECOMPRESSION SICKNESS INCIDENCE

The evaluation of the DCS risk in modern saturation operations is difficult. Companies do not share information. We did not request this information from the 11 companies participating in the study.

Among the historical procedures, Comex documented their safety performances with an exposure database and published their saturation diving track records. The average DCS incidence was 1.02% (59 DCS cases in 5,744 exposures) for dives performed using procedures described in the 1979 Comex manual.[ 28] It decreased to 0.54% (12 DCS in 2,200 exposures) for dives performed with Comex procedures later implemented in the French 1992 diving regulations (unreferenced report). All symptoms were exclusively articular pain, reported during the last metres of the ascent and never at surface.

Seaway also developed a database (Hades) for the monitoring of their diving operations.[ 20] In 1978, the results published for their saturation decompressions indicated an overall DCS incidence of 0.83% (22 DCS cases in 2,662 exposures,). As in the Comex database, all cases were articular pain occurring in the last part of the decompression.

The only modern source is PSA in Norway that has been collecting and publishing saturation safety records since 1990.[ 29] The cases are all recorded in the Norwegian sector and therefore associated with use of the NORSOK procedures. The site accessed on 21/09/2023 indicates one DCS case recorded since 2000. The corresponding incidence cannot to be evaluated because the exposures are expressed as men x hours in saturation and not saturation dives.

Finally, the diving manager of one of the participating companies indicated to the authors that DCS has become a rare event and stated “we have not had a bend in the last 10 years”. This is in line with the authors’ experience with other companies. Nowadays, across the industry, DCS is not observed in saturation diving.

MEASURED VGE GRADES

Venous gas emboli are commonly observed after asymptomatic dives using acoustic doppler or ultrasound imaging. Although the presence of VGE is by itself not predictive of DCS for a given diver, a statistical link is observed in surface-oriented diving between the amount of observed VGE in a group of divers and the incidence of DCS.[ 30 , 31]

Significant VGE grades were detected during deep saturation decompression in the experimental dives and operational dives conducted in Norway.[ 32 , 33] These high levels of VGE contrast with recent measurements performed during operations in the North Sea. In 2017, we monitored 49 saturation divers after decompression from 120 and 140 msw using echocardiography and detected no bubbles.[ 30 , 34] In 2022, we monitored 15 divers using ultrasound subclavian Doppler detection during and after saturation decompression from 40–50 msw storage depths, and found no bubbles.

Our experience with bubble monitoring is that high grades of VGE are no longer a concern in the saturation decompression we monitored, at least to 140 msw.

PULMONARY OXYGEN TOXICITY

Oxygen plays a major role during decompression. It increases the inert gas gradient between tissue and blood as well as the oxygen window and permits accelerating the decompression.[ 19 However, high levels of oxygen generate reactive oxygen species that interfere with normal cell functions. While central nervous system toxicity is not a concern in saturation diving, pulmonary toxicity can be a limiting factor.

The PO₂ should not be too low. An experimental saturation dive to 240 msw was conducted in Norway with slow decompression rates and reduced PO₂ during decompression (500 mbar during the day, 300 mbar during the 8 h night stop). No change in pulmonary function was observed. However, one case of DCS was recorded among the eight divers.[ 35] This suggests the difficulty of safely decompressing from saturation with a low oxygen level. Prof. Lambertsen had a definite position on the subject that he summarised as “better cope with the effects of oxygen than with the ones of DCS”.[ 36]

On the other hand, a high PO₂ cannot be sustained for too long. Early saturation decompressions used a 600 mbar chamber PO₂ (Comex 1979) but when diving moved to deeper depths in Norway and in Brazil, longer decompressions raised the problem of pulmonary toxicity. The PO₂ was reduced to 500 mbar and the use of a 600 mbar PO₂ was restricted to less than 155 msw (Comex 1986 manual). Ultimately, a 500 mbar chamber PO₂ became the company standard for all depths.

Our survey has shown that while one company uses a chamber PO₂ of 530 mbar (procedure E), most companies use 500 mbar as an optimal value (procedures A, B, F, G, H, I, J and K, i.e., eight out of 11 companies) or 490 mbar (procedures C and D).

The divers’ tolerance of pulmonary oxygen toxicity is difficult to measure and predict.[ 37] The industry approach was based on ‘units of pulmonary toxicity dose’ (UPTD) because of their simplicity.[ 38] However, it has been shown that the UPTD dose is not an appropriate tool for measuring oxygen toxicity, in surface oriented diving.[ 39]

Arieli developed a dose index accounting for recovery and validated it against a sample of saturation exposures.[ 40] However, given the paucity of exposure data at the lower end of the hyperoxic spectrum, the model remains to be validated for operational use in saturation diving. It is therefore currently not possible to reliably estimate a pulmonary toxic threshold dose for the company procedures.

It must be noted that the higher PO₂ to which the divers are exposed is the one used during the bell dives, for six continuous hours, daily. Therefore, most of the hyperoxic exposure takes place during the bottom phase.

It appears that the PO₂ used in current saturation procedures is the result of successive empirical adjustments and a better understanding of both the excursion and decompression design. Hence, the oxygen toxicity dose calculations could help optimising the diver’s hyperbaric oxygen exposure.

EVOLUTION OF SATURATION PROCEDURES

In the 70’s, navies, universities and governments conducted research and provided diving procedures to the industry. The last large research led by the industry was the Norwegian deep diving program of the 80’s. Today, companies rely on themselves to improve their diving procedures.

The drive for such changes is no longer DCS occurrence but instead, the need for more flexible procedures. Managing operations of large and expensive diving vessels requires options and alternatives.

The ethical principles and the practical procedures for the development of decompression tables, were published by the Undersea Medical and Hyperbaric Society in the conclusions of a workshop on validation of decompression tables, in 1987.[ 41] These conclusions have been since considered as the reference for developing and improving decompression procedures. The principles developed in these conclusions are based on small step changes and careful evaluation. They include the following activities:

1. Evaluation of the latest scientific and medical information.

2. Definition of the models. Publication of the new procedures.

3. Monitoring operations to identify areas of improvement of the provisional procedures.

4. Validating the changes on selected worksites under controlled conditions before acceptance.

5. Review and analysis of data collected.

6. Approval of the new procedures or reiteration.

These principles are in line with the industry procedures of management of change that also stress the importance of validation and monitoring. They provide referenced standards permitting the companies to monitor and improve their diving procedures.

TRENDS

We have seen discrepancies between procedures. Table 5 shows a difference of 25.7 hours between the slowest and the fastest decompression time from 150 msw storage depth. However, as discussed during the DMAC 2014 meeting in Aberdeen, we do not have any information that would enable us to evaluate the consequence of these differences on the divers’ health.[ 42]

We have also observed a convergence within procedures (moving towards similar PO₂, similar daily decompression rates, similar pre-decompression holds, etc.). Table 5 also shows that for seven companies, there is less than 5 h difference in the decompression time from 60 to 150 msw storage depth.

Because no large-scale research project has been conducted since the Norwegian deep diving contracts, we identified three ways the company procedures have evolved through:

• Internal evolution based on empirical adjustments. This is facilitated by freelance personnel freely moving between companies and carrying along their knowledge and experience.

• Forced evolution after takeovers and mergers between companies.

• Guided evolution by regulations, industry standards and client’s requirements.

These evolutions have been made possible by the sharing of the company experiences within the industry associations such DMAC, IMCA and the International Association of Oil and Gas Producers (IOGP).

From this analysis, we foresee two possible paths for this evolution: standardisation and harmonisation:

• Standardisation assumes a stakeholder’s association, i.e., contractors or clients, that defines, endorses and publishes policies. As opposed to government authorities, such associations have the capacity to rapidly adapt and change their policies.

• Harmonisation results from free adhesion to a practice. It supposes a consensual objective and sharing of scientific evidence, experiences and policies. Harmonisation is likely to continue with the internationalisation of the offshore industry.

KNOWLEDGE GAP

We believe that saturation procedures will continue to evolve and that this evolution must be supported by scientific research. Our experience is that divers’ monitoring, which part of the Undersea and Hyperbaric Medical Society recommendations, brings valuable support to this evolution.

Commercial diving faces, at least two physiological challenges:

• The ageing of the population of divers and their capacity to cope with the various diving stresses.

• The oxygen partial pressures during saturation dive. We said that accurate models are required to evaluate pulmonary oxygen toxicity dose over a saturation. This dose must be managed considering that the bell excursion dives expose the divers to high PO2’s and this impacts the use of oxygen in decompression.

Companies seek flexibility to manage the modern and expensive diving support vessels that keep moving from one contract to the other. They need instructions on how to deal with these multi project campaigns that periodically change the working depths. They seek clear guidance for using all the possibilities of intermediate decompressions and ascending/descending/extended excursions.

Companies also need guidance in managing divers’ rotation onboard these vessels. It is known that saturation diving is associated with endothelial dysfunction and inflammatory stress, followed by a recovery.[ 34 , 43 , 44] Hence, the way the divers manage their careers, alternating saturations and rest periods, is important. The DMAC note 21 seems to provide adequate guidance in managing saturation duration and between dive intervals since we noted in the sample diver’s population that a 61-year-old diver can still obtain his saturation diving certificate. However, it is believed that more information is required to combine saturation diving and air diving, standard diving and deep diving, etc.