{"http:\/\/dx.doi.org\/10.14288\/1.0045255":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Medicine, Faculty of","type":"literal","lang":"en"},{"value":"Population and Public Health (SPPH), School of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/alternative":[{"value":"BCOHTA 92:3","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/contributor":[{"value":"University of British Columbia. Centre for Health Services and Policy Research","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"Sheps, Samuel Barry","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2014-10-06T21:45:00Z","type":"literal","lang":"en"},{"value":"1992-10","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"This Technology Assessment Report from the B.C. Office of Technology Assessment (BCOHTA) was requested by the Vancouver General Hospital (VGH) in response to suggestions by Clinical Staff that Hyperbaric Oxygen (HBO) use for chronic osteomylelitis (COM) and osteoradionecrosis (ORN) should be assessed. Thus, the VGH felt it would be useful to obtain an overview of the available scientific evidence on the efficacy and effectiveness of HBO for these two conditions.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/50610?expand=metadata","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"Centre for Health Servicesand Policy ResearchHYPERBARIC OXYGEN FOR OSTEO~YELITISAND OSTEORADIONECROSISBCOHTA92:3 OCTOBER 1992B.C. Office of Health Technology AssessmentDiscussion Paper SeriesIII ' iSf:!,,:..-.:J,.:...J . THE UNIVERSITY OF BRITISH COLUMBIA\" ,HYPERBARIC OXYGEN FOR OSTEOMYELITISAND OSTEORADIONECROSIS8.8.8hepsBCOHTA92:3B.C. Office of Health Technology AssessmentCentre for Health Services & Policy ResearchS 184 - Koerner Pavilion2211 Wesbrook MallVancouver, B.C., CanadaV6T 2B5 - October 1992TABLE OF CONTENTSFOREWORDEXECUTIVE SUMMARY i1.0 INTRODUCTION \u00b711.1 General considerations 11.2 Local availability of HBO and recent data on its use 32.0 l\\1ETIIODS 'C 43.0 REVIEW OF TIlE LITERATURE: CHRONIC OSTEOMYELITIS 53.1 Reviews : 53.2 Case series 73.3 Trial-s : 93.4 Costs : 113.5 Summary , 114.0 REVIEW OF TIlE LITERATURE: OSTEORADIONECROSIS 124.1 Reviews 124.2 Case series 134.3 Trials ; 134.4 Summary 145.0 CONCLUSION AND COST CONSIDERATIONS 155.1 Costs 155.2 Conclusion 16REFERENCES 18FOREWORDThe British Columbia Office of Health Technology Assessment (BCOHTA) was established onDecember 1, 1990 by a grant to the University of British Columbia from the Province to promoteand encourage the use of assessment research in policy and planning activities at thegovernment level and in policy, acquisition and utilization decisions at the clinical, operations andgovernment levels. It is important to understand that the role of the Office is to appraise thescientific evidence only, and not to be involved in actual policy development.Assessments are performed in response to requests from the public sector such as hospitals,physicians, professional associations, regional districts, government; private sector groups suchas manufacturers; and individuals from the general public.One or more of the following criteria are used to determine the-priority of an assessment and thelevel of analysis: 1) number of users and potential change in quality of life; 2) acquisition andoperating costs to the health care system; 3) potential to influence provider and consumerbehaviour as a result of a review and 4) availability of accurate information and appropriateresearch skills. The collection of information for assessment is done in a systematic way to getall information from traditional sources such as libraries as well as fugitive information from othersources. A comprehensive search of the literature is done through the U.S. National Library ofMedicine databases such as Medlars (Medline-journals, HEALTH-planning and administration,CANCERlIT, etc), BMEDSS (Biomedical engineering), NORDSER (Nordic countries) and otherinformation networks.This literature is reviewed by BCOHTA staff, medical residents or consultants . The evaluationof a health technology is based upon the analysis of the information for quality of the evidenceand strength of the findings. Logical and defensible conclusions about the technology areformulated. A document is then prepared in a predetermined format.Scientific papers and health technology reviews _are read by at least two members of theBCOHTA Steering Committee. The documents are then sent out to one of a team of experts fromavariety of academic or clinical disciplines for external review. Comments and suggestions areconsidered by the Steering Committee and incorporated into the paper. Distribution of the papersis by request from the Office or through inclusion on the BCOHTA mailing list; summary versionsappear in the quarterly newsletter.Arrnlnee Kazanjian, Dr. Soc.Chair, BCOHTA SteeringCommitteeCopies may be obtained at no charge from:B.C. Office of Health Technology AssessmentS184 Koerner Pavilion2211 Wesbrook MallVancouver, B.C.V6T 2B5; (604) 822-7049Kathryn D. Friesen, M.Sc.o Program Manager, BCOHTAEXECUTIVE SUMMARYThis Technology Assessment Report from the B.C. Office of Technology Assessment (BCOHTA)was requested by the Vancouver General Hospital (VGH) in response to suggestions by ClinicalStaff that Hyperbaric Oxygen (HBO) use for chronic osteomylelitis (CO~) and osteoradionecrosis(ORN) should be assessed. Thus, the VGH felt it would be useful to obtain an overview of theavailable scientific evidence on the efficacy and effectiveness of HBO for these two conditions.The literature was systematically searched, as per the B.C.O.H.T.A. routine database searchstrategy, for the use of HBO for the two indicated conditions for the years 19&8-1991. Relevantadditional papers from prior years were obtained from the reference lists of the papers identifiedand were reviewed in depth. This process yielded approximately 40 papers on the two topics ofinterest.With two exceptions, all the papers identified were either review articles, case series, case reports,or animal studies. Animal studies and case reports are not included in this assessment of theliterature. Two trials, one non-randomized in COM and one randomized in ORN, were found.Most reviews comment on the lack of prospective trials, although some go to great lengths tojustify the absence of trials by pointing out how difficult they would be to carry out.The paucity of well designed clinical trials makes assessment of HBO for COM and ORN difficult.. -Not only is the literature limited essentially to animal studies, case reports and case series but these,although relatively large in number, tend to be by the same authors (often reporting the same data,which in tum are cited in reviews and textbooks edited by the same authors), and thus provide littlemore than (often repeated) anecdotal evidence.The two controlled trials were: 1) Esterhai et. al. with 14 patients in each group .(non-random1yallocated) who had uncomplicated COM which found no difference between the HBO group andthe control group in length of stay, rapidity of wound repair, initial clinical outcome or recurrencerate. These negative results must be viewed with caution since the power of the study to detectmeaningful differences is low; and 2) Marx et. al., on the prevention of ORN, compared HBO withprophylactic penicillin in patients requiring tooth extraction following radiation of the mandible: 37patients were randomly assigned to each group - there was a significant increase in socket woundhealing at six months in the HBO group.Data from case series is difficult to interpret because of poor descriptions of methodology (e.g.patient selection) and mixed results with success rates ranging from 50% to 100% both with andwithout HBO.The Annual Report from the Division Head for HBO at the VGH was also reviewed. It providesutilization data for the year 1990-91, allowing an assessment of the relative importance of COMand ORN in terms of the use of the HBO chamber in Be. These two conditions make up only 20%of the case load for the year in question. The VGH Annual Report makes no comment on unmetneed. Moreover, cost data were provided and these are discussed in the report.In summary, for COM there is neither evidence that HBO clearly makes a difference in the outcomenor support to expand the use of HBO, although, as noted before, the literature is so poor that adefinitive assessment is impossible. There is certainly no clear evidence that expanded use of HBOfor COM is warranted, or indeed required. With regard to ORN, although the data are equally poor,the complex nature of the condition, the results reported in the case series, and- the singlerandomized trial suggest -that there may be stronger indications for HBO use than is the case withCOM, although the trial was limited to the evaluation of prophylactic HBO in the context of toothextraction for the prevention of ORN.One conclusion is clear: all experts agree and the data support the fact that successful managementof COM relies first on adequate surgical and antibiotic treatment. HBO is regarded by virtually all-authors as an ac!.iunct to this primary management. For ORN, differential success with and withoutHBO appears to be greater and although primary management is surgery and appropriateantibiotics, the complexity of the tissue injury in ORN (particularly of the mandible the site mostoften studied) suggests thatHhO may be efficacious and effective. However, as noted above, L.1elack of evidence based on well designed trials makes this, at best, a tentative, conclusion.The issue of the cost of managing COM and ORN and the components of these costs attributable toHBO is complex. On the one hand, HBO use for several conditions (e.g. decompression sickness,gas gangrene, etc.) is unquestioned so that operating costs, although not insignificant (on the orderof $300,000), are justified by more than simply the treatment of COM and ORN. The specificoverall management costs for COM and ORN are high (approximately $100,000 - $250,(00), butestimates suggest that the subset of costs associated with HBO account for only 5% of the total costper case . On the other hand, the HBO physician costs specifically for COM and ORN are nottrivial (ranging between $5,000 to $14,000 for the average course of HBO treatment). Thus, whilethe impact of expanding the use of HBO for COM and ORN has definite cost implications, theseiiare relatively small compared with the basic and substantial total management cost associated withthese conditions (including multiple hospitalization, multiple surgery, antibiotics, etc.)Conclusions regarding the value of HBO, in terms of costs versus benefits, are currently notpossible since these relationships for COM and ORN are complex and no formal economicevaluations were found in the literature.iii1.0 INTRODUCTIONThe first use of hyperbaric oxygen therapy (HBO) occurred in 1860 in North America when achamber for increasing oxygen pressure was .built in Oshawa Ontario, most likely for thetreatment of decompression sickness (1). The first International Conference on the use of HBOwas held a century later, in 1963, with eight subsequent conferences up to 1987 (including one inVancouver in 1973). Over the last thirty years interest in, and knowledge about this form oftherapeutic intervention has increased geometrically. Indeed, the HANDBOOK OFHYPERBARIC OXYGEN TIIERAPY published in 1988 cites about 1550 references (2). Arecent textbook, \"PROBLEM WOUNDS, TIIE ROLE OF OXYGEN\", edited by Davis andHunt, also provides a comprehensive overview of the use of HBO (3). This paper reviews theevidence for the use of HBO in chronic osteomyelitis (COM), also known as refractoryosteomyelitis, and osteoradionecrosis (ORN), a common\"condition seen after radiation therapy,especially of the mandible . The pathophysiology of these conditions is similar in that bothinvolve significant bone infection associated with severe tissue hypoxia, hence the use of HBO.intheir management.1.1 General considerationsHBO has been proposed in the management of infection generally (except viral infections) and- .COM and ORN specifically because oxygen (02) has antibiotic properties (2,3). These are oftwo types: 1) leukocytes utilize hydrogen peroxide, superoxide, and other reduced 02components as part of their mechanism of bactericidal action; and 2) 02 itself inhibits both gram+ and gram - bacterial growth. The latter effect is of less importance since it is variabledepending on the organism, the amount of pressure used, and other conditions. For example,aerobic bacteria show a biphasic response to 02 with growth enhancement at 0.6-1.3 ATA, butgrowth inhibition above 1.3 ATA. Moreover, the killing action is time dependent. In addition,02 is also an indiscriminate killer of cells, hence its' toxicity for host cells.The proposed mechanisms of HBO action are based on the fact that both COM and ORN areanoxic infectionswith tissue p02 usually less than 30rnmHg (2,3). Thus, indirect action throughleukocyte function is of primary importance; direct action on bacteria is said not to be a factorsince HBO used prophylactically to prevent infection does not seem to work. (2,3) . In addition,optimal oxygenation of the infected bone enhances osteogenesis, collagen formation andneovascularization, which are critical in filling dead space and enhancing the entry of leukocytesand antibiotics into affected bone (2,3).With regard to ORN, radiation therapy not only produces the direct effects of disrupting mitoticactivity (the effect more pronounced in proportion to the mitotic activity of the target cells andsurrounding tissue), but there is also a late effect: obliterative endarteritis which in tum producesprofound hypoxia. REO appears to work by two mechanisms. First by counteracting the anoxiceffects of obliterative endarteritis (with HBO, although it produces vasoconstriction, p02 levelsactually increase and initial vasoconstriction is followed by vasodilatation); and second,prolonged treatment with HB0 produces neovascularization and collagen formation, the samebeneficial effect seen in COM. Although COM and ORN share several pathogenic mechanisms,ORN of the mandible unlike COM has a continuous focus of bacteria (the teeth and gums). Thisdifference is the basis for evidence to suggest that HBO is useful prophylactically in ORN of themandible (see below).HBO exposure is measured both in \"dives\", a term derived from its use in decompressionsickness, the pressure related to each dive, and the percent oxygen (% 02) used. Pressure ismeasured in atmospheres absolute (ATA). One ATA is equivalent to the air pressure at sea level(14.7 psi). . During a \"dive\" the pressure gauge is set to 0, and thus a chamber pressurized to14.7psi represents the pressure felt at 33 feet under sea level or 2 ATA; similarly a 3 ATA \"dive\"represents the pressure felt at 66 feet below sea level. HBO is said to be well tolerated at 1.5-2.0ATA (for up to 4D-90 minutes), but tolerance is dependent on the (% 02) used. Often intherapeutic situations 100% 02 is used. However, 02 is a toxic agent for multiple organs. Its'toxicity appears to be related to Mg++ and Ca++ effects since Mg++ and Ca++ blockers protectagainst HBO induced cellular injury. In addition, Vitamin E is often used prior to HBO,presumably to reduce the tissue damage by free oxygen radicals.Specific toxic reactions to HBO include, oxygen seizures, otic barotrauma, pulmonarybarotrauma, and myopia. Moreover, there is a long list of contraindications to its use, including:pneumothorax, upper respiratory tract infection, seizure disorders, emphysema, uncontrolled highfever, viral infections, malignant disease, pregnancy, optic neuritis, and others. It is useful to beaware of these since patients with COM or ORN may have any of these conditions.The indications for use of HBO include: decompression sickness (HBO has been the mainstay oftherapy for this condition for many years); bums; air embolism; carbon monoxide (CO),hydrogen sulfide (H2S) and methane (CH4) poisoning; as well as COM and ORN. However, asJacobson et. al. note in their 1965 historical perspective ofHBO (1),\"If this form of therapy is to achieve a worthwhile place in the medical armemantarium,it can only do so on a firm basis of accurate physiological data on the effect of both2pressure and oxygen obtained in experiments, as well controlled as clinical medicinewill permit.\"(emphasis ours)\"The purpose of this review is to examine the recent literature on HBO for COM and ORN todetermine if the evidence is sufficient to: 1) support its use generally; and 2) to assist decisionmaking regarding its ' use at VGH for these conditions.1.2 Local availability of HBO and recent data on its useIn order to put the discussion of the literature in perspective, it is of importance ~o note that theVancouver General Hospital has one of the two HBO chambers in the province (set up in 1966),and is thus a provincial resource .**Utilization of the VGH unit. over time and for 1990-91, has recently been summarized(4). Therehas been a steady increase in use since the chamber was installed: from less than 10 patients in1967 to between 25-40 patients from 1968-77. to less than 10 in 1978 (due to refitting thechamber in that year) , with a subsequent steady increase up to about 86 patients in 1990-91. Forthe year 1990-91 , mostpatients were treated for osteonecrosis unrelated to radiation therapy-21,decompression sickness-17, CO poisoning-IS. ORN-12, delayed wound healing-S , COM-5 andother conditions-8 (e.g. air embolism, gas gangrene. etc .). Thus , only 17\/86 (20%) of patientshad conditions of direct relevance to this report.In 1990-91, the chamber operated at 80% capacity, with 46 outpatients and 40 inpatients, 48 \u00b7 \u00adelective and 38 emergency patients, for a total of 1348 treatments (an average of 16 per patient).No data are provided regarding ranges or number of treatments by diagnosis. age, sex, etc .. .\u00adThere were 477 treatment dives (about 5.5 per patient, but a&ain no breakdown by relevantvariables), 1109 chamber hours, and an average of 2.8 patients per dive overall (3.2 on electivedives). Estimates of Unit operational costs were reported to be $300 .000 per year of which$30,000 were recovered from the Workers' Compensation Board (WCB) . These costs do notinclude physician fees. Thus the average operating cost per patient is about $3450, while theaverage cost per dive is $620, and the average cost per chamber hour (presumably chambertreatment hour) is $270 . These are, of course, average costs and thus do not reflect variationacross treatment groups which undoubtedly exists. Moreover, although on average about 10% ofthese costs are covered by the WCB, the proportion recovered may differ widely depending on'\" This point is emphasized because many authors feel that randomized controlled trials are not possibleto carry out for COM or ORN.** The other chamber is at the Canadian Forces base at Esquimalt.3the patient's diagnosis, and thus treatment needs, in terms of numbers of dives and total hours fora course of HBO therapy. One issue not clear from these data is whether the chamber costs varyby the length or depth of the dive; i.e. by ATA into which the patient is placed. Physician feesdo. however, depend both on the length of the \"dive\" and whether the physician is in the chamberor monitoring the patient from outside the chamber (the fees are higher for inside chamber time).The relevance of these variables to costs are discussed below.The VGH data fail to provide any estimates of denominators. Thus rates of use by geographicregion, age, injury or illness type (and therefore costing the utilization of the HBO chamber forCOM and ORN) is difficult and can only be based on literature estimates of the number of divesand\/or hours needed to treat these specific conditions. In addition, total management costs forCOM or ORN also include hospitalization, surgery, antibiotics, and other direct patient care costs(not to mention indirect costs). Thus the figures cited in the VGH report are only a fraction ofoverall patient management costs.2.0 METHODSThe standard B.C.O.H.T.A. literature search routine was undertaken for the years 1988-91 usingthe \"National Library of Medicine database (MEDLINE), and the Health Planning andAdministration database (HE~TH), using the key words HBO, COM, ORN, bone infection,bone necrosis , infection, and necrosis. In addition, the relevant sections of the Handbook (2) andProblem Wounds (3) were reviewed to identify additional relevant citations published prior to1988. This process yielded a total of 61 papers which were sorted regarding type- of study:review articles; basic science and animal studies (these were excluded from our review); singlecase reports (also excluded -from this review); case series (mom than one patient); trials \u00ad(controlled or uncontrolled); and randomized controlled trials (RCTs). Twenty papers orabstracts met the selection criteria (although not all papers were available locally or throughinter-library loan), and thus 14 papers were available for in-depth review, and six abstracts wereassessed. The 41 papers and abstracts not discussed in this review, aside from being case reportsand animal studies, reported data on clinical conditions other than COM and ORN such as: softtissue infections; pulmonary infections; air embolism; decompression; hematuria; psychologicaleffects; nursing issues; and effects on physiological systems such as leukocytes in the immunesystem.Of the 20 recent papers or abstracts reviewed here: five were review articles; two were reviewarticles and case series; four were case series; and two were trials, one specifically on HBO in4COM, and one on the use of penicillin versus HBO in the management of ORN. Overall, 10papers and four abstracts on HBO in COM, and four papers and two abstracts on ORN wereassessed. \u2022 Thus, although not an exhaustive review of all possible citations on either COM orORN, the papers reviewed here are representative of the consensus among researchers in thefield. Notwithstanding the approach taken for this paper, several case series published prior to1988 (and thus not included in our search process) are briefly cited to enable the reader to get amore complete \"feel\" for the quality of this literature and the basis upon which clinicalimpressions of effectiveness have been formed.3.0 REVIEW OF THE LITERATURE: CHRONIC OSTEOMYELITISAlthough HBO has ~en promoted as an intervention in the management of this condition, themainstay of therapy is surgery (drainage, removal of foreign bodies , debridement and removal ofdead spaces), and appropriate antibiotics . Proponents of its use therefore view HBO is an adjunctto the primary therapy described above. Since surgical and antibiotic the.rapy is reported to beeffective in 70-80% of patients with osteomyelitis (2,3), HBO is seen as specifically useful inwhat is termed \"refractory or chronic\" osteomyelitis (COM). However, the proportion of thesepatients who might benefit from HBO is not clear. Since none of the clinicians currently activelyinvolved in research on HBO consider it to be other than an adjunct to standard therapy, the keyquestion for this review is whether HBO produces significant clinical improvement comparedwith standard therapy. While there are considerable animal and basic science data to suggest apositive therapeutic effect, studies in humans -are-less numerous and, with the exceptions notedabove, are generally restricted to case reports, case series, or reviews.3.1 ReviewsThe four main recent reviews all refer to numerous case reports, case series and animal studiesand all conclude that HBO is a useful adjunct to standard therapy (5-8). Being reviews, however,little is to be gained by describing them in detail since they provide no primary data. While theconsistency of opinion is clear, it should be noted that three of the four reviews (5,7,8) are fromthe same institution (the University of Texas Medical Branch at Galveston) and, not surprisingly,these three reviews have at least two authors in common, thus the independence of opinion israther limited. The fourth review (6) is from the University of Chicago. All of the reviews note\u2022 In general, the literatureon this topic is highly repetitive and is, as can be seen in the reference list,limited to relatively few authors (who tend to cite and recite their own work or the work of colleagues).On occasion, different papers which are supposed to be original articles, use previously published data.5the lack of randomized controlled trials of HBO. Despite this, the preface to a recent textbook(3) makes the following statement:''There is no longer any question as to whether or not elevation of problem woundoxygen is efficacious - it is. Remaining questions are whether or not perfusion of agiven wound is adequate to achieve therapeutic P02 within safe limits of oxygenbreathing and then to determine the optimal dose.\"Despite the optimism of this statement, it would appear that even advocates of HBO therapy arestill uncertain about the effectiveness of this intervention, and given the literature available suchuncertainty is well founded .Davis and Heckman (5), review the animal and laboratory evidence for the biological plausibilityof HBO management of COM. They also review a number of case series, noting in the firstsection \"Clinical Series\" that: \"The one common feature to all (emphasis theirs) treatmentmod~ties for osteomyelitis is the absence of valid human prospective controlled clinical trials\".They cite several case series using therapeutic interventions without HBO reporting success ratesof 80% or more which provide useful background \"control\" data for the HBO papers (9-14).Thus Kelly (9) for example, in a study of closed irrigation and suction, compared 40 patients whohad this modality with 35 who did not. At 2-8 year follow up the closed irrigation and suctiongroup had an 80% success rate compared to 57% in the-control group. Davis and Heckman (5),although providing few details about this study, claim that it has many methodological flawswhich, unfortunately, they do not detail . Similarly, Papineau (10) reported on the use of openbone excisions and bone grafting in 39 patients and showed a 93% success rate. Other non-HBOcase series for COM (1l-l4) ;-using muscle flap transfer and antibiotics, shewed results on theorder of 90 to 100%. Case series of HBO cited by Davis and Heckman (5) will be discussedbelow.Grim et. al. (6) provide a very general overview of HBO therapy in many conditions but makereference to only three papers regarding COM: one is a case report, one another review and one isa case series discussed below.Mader et. al. (7) review the experimental literature and report on their own experimental work inrabbits. In addition, they cite three additional case series which are briefly considered below.They also discuss and criticize the only trial in the literature.6Finally, Calhoun et. al. (8) describe the same case series as the other reviews and cite but,curiously, do not comment on the only trial to date.3.2 Case seriesSlack et. al.' (15) treated five COM patients with HBO at 2 ATA all of whom showed clinicalimprovement. However, in the two review articles citing this paper, no details on subjectselection, number of treatments, or time to clinical improvement are provided.Perrins et. al. (16) treated 24 patients with COM and sinus tracts in whom previous procedures(sequestration, antibiotics, and marsupialization) failed to resolve the condition. However, withHBO and antibiotics, 17\/24 (71%) of these cases responded; four cases (16.6%) showed reducedsinus tract drainage; and three patients failed to respond at all (12.5%). Again, no details areprovided regarding patient selection (other than they all failed to respond to conventionaltherapy), intensity of HBO treatment, or other patient characteristics.Dupenbusch et. al. (17), in a larger case series, reported on 50 patients who had beenunresponsive to antibiotics and surgery. With HBO therapy. 71% reported that their COM hadcompletely healed and the remaining 29% reported improvement in symptoms. Details regardingHBO treatment are not described nor is prior therapy described in detail.Davis and Hunt (18) reviewed 70 patients with chronic osteomyelitis in various sites (spine,pelvis, chest wall, skull, etc.) unresponsive to the usual therapy and found an average of 62%responded to HBO; the condition was arrested (a rather ambiguous term) and remained so at 5year follow-up. However, the proportion of patients whose disease was arrested ranged widelyfrom 37% with COM of the hip to 100% for the small number of patients with COM of thehumerous, hand, and frontal sinus. The most common site, the tibia (26nO patients-37%), had asuccess rate of 73%. The femur, the next most common site (15nO-21 %), was reported to have asuccess rate of only 40%.Further analysis presented by these authors (19), involving an additional 28 cases of COM ofvarious sites and at least two years follow-up (hence the figure of 98 cases cited in manyreferences), revealed a success rate of 71%.Morrey et. al. (20) reported a case series of 40 patients and long term follow-up. All the patientshad COM for at least 6 months, had at least one (often many) surgical procedure(s) prior to HBO,'and had received parenteral antibiotics. The average duration of COM was 30 months (range 67months to 23 years). All patients had contiguous focus, as opposed to hematogenous, disease.Treatment, in addition to HBO (which was used in all patients), included antibiotics and carefulsurgical debridement (all patients), autogenous bone grafting in 7 patients and soft tissue grafts in7 patients. Patients were followed for a mean period of 23 months (range 21-53 months) duringwhich 34 (85%) responded and remained clinically free of disease and 6 (15%) failed to respondor had a recurrence (mean time to recurrence was 4.3 months (range 0-12 months)). Thesepatients received an average of 42 HBO treatments (range 9-208) . The average number oftreatments in patients without recurrence was 39 (range 9-208), while for patients with recurrenceit was 78 (range 29-153). Recurrence rate was not correlated with the number of treatments. withthe type of organism, or with the duration of COM prior ~o. HBO treatment. Further analysispresented by these authors (20) with extended follow-up (average 8.4 years, range 7.5 - 10.5years) revealed 4 additional relapses giving an overall success rate of 75%.Eltorai (21) reported on the treatment of 44 patients, with HBO at 2 ATA for an average numberof 50 sessions, who had COM resulting from spinal cord injuries , and pressure sores around thepelvis . .Of these patients 30 (68%) were cured, although the length of time to resolution of theCOM and the correlation with the number of treatments was not described.Davis et. al (22) reported the results of HBO treatment in 38 patients followed for an average of34 months (range 24 to 59 months) and enrolled between 1 November 1979 to November 1 1982(reinforcing the relative rarity of this condition). All patients had non-hematogeneous COM andwere treated with careful surgical debridement and long term parenteral antibiotics based on. organism sensitivities. The mean duration of COM prior to enrollment was 8.9 years and manypatients had multiple surgical procedures. HBO was initiated on the first post-operative day at2.4 ATA per day with 100% humidified 02 (35 l\/min) for three 30 minute periods each followedby 10 minutes of air breathing. The mean number of daily treatments was 48 (range 8-103) andtreatment was stopped when the bone was fully covered by healthy vascular tissue. Threepatients required tympanostomy tubes to prevent otic barotrauma and two patients had transientoxygen induced myopia . Thirty-four of 38 patients (90%) became and remained clinically free ofCOM over an average of 34 months. All the HBO failures had significant infections: three withPseudomonas Aerogenosia and on~ with E. coli. These data are consistent with clinicalexperience that gram negative COM is more refractory to treatment with longer treatment periodsand more surgical procedures. It is useful to note that, despite these apparently positive results,the authors were cautious in interpreting their results:\"We wish to emphasize that the therapy with hyperbaric oxygen cannot be given salecredit for the results in our patients. The daily debridement and meticulous care of the8wound by the experienced physician and the use of culture specific antibiotics played alarge role in prolonging the disease-free interval in our patients ...Finally, in spite ofthese encouraging results and similarly encouraging results in a previous series ofpatients, we are careful not to regard our patients as cured of the chronic osteomyelitis.A much longer follow-up period than that used in the present report is essential before atrue rate of cure can be determined ...We believe that it was the combined treatmentprotocol. not anyone of its components, that was responsible for the reported success(emphasis mine).\"Hui-Chieh Lee et, all (23), in the largest case series of HBO patients published to date (N=1288), found over a 12 year period that for COM (not well defined) the cure rate was 59% withimprovement (undefined) in a further 37.3%. However, in their series, COM accounted for only83 of the 1288 patients (6.4%, not unlike the proportion of HBO patients treated for COM at theVGH for fiscal 1990-91- 5.8%); no cases of ORN were included in the series . The failure rate,' . .includirig death, was 3.7% which was generally lower than for the other conditions in the series:like gas gangrene (22%); ,chronic skin ulcer (10%); and bum injury (23.9%).In general, the Handbook (2), states that the overall success rate of HBO therapy for COM acrossvarious studies (not referenced) was between 68-85%. The range of success rates for the clinicalseries reported here was between 50% (Davis (18\u00bband 100% (Slackt lSj). However, as notedabove, although many of the patients in these case series served as their own controls (given their. .previous treatment history), it is also true that the success rate in patients treated with othermodalities (mainly good surgical debridement and appropriate antibiotics) were on the sameorder of magnitude (9-14). Thus, without valid controls, the issue of whether HBO provides anybenefit beyond clearly agreed upon primary surgical and antibiotic therapy is left unresolved bythese papers since virtually all of these data are derived from weak study designs . The remarksby Davis et. all (quoted above) are therefore particularly pertinent to the role of HBO in thesuccessful management of COM.3.3 TrialsIn terms of COM, a single trial is of interest. The study by Esterhai et. al.(24) was a case\u00admatched trial of 28 consecutive patients with COM uncomplicated by fracture nonunion, septicarthritis, total joint arthroplasty, or major systemic disease (tumors , immune deficiency, renal orhepatic disease, etc.) were treated from January 1980 through December 1985. Patients wereclassified using one of the several current staging systems (Ciemy, Mader, and Perminck (25\u00bb9and were matched by stage, although how this was done is not discussed. Staging and matchingwere carried out after the initial debridement (and presumably after initiation of antibiotictherapy-although this is not explicitly stated). HBO consisted of 100% 02, at 2 ~TA, for one 2hour dive per day, 6 dives per week, in the 14 HBO patients. The other 14 patients received noHBO. Outcomes included: length of hospitalization; rapidity of open wound soft tissuegranulation repair; the initial clinical outcome '(whatever that may mean); and the recurrence ofinfection. No patient was lost to follow-up and the patients were followed monthly for the firstthree months and evaluated at 3 and 6 months. Of the 28 patients. 24 (86%) demonstratedhealing at six months, with no difference between the HBO and non-HBO group: three of thetreatment failures were in the HBO group and one failure occurred in the non-HBO group. Allfailures were reported to have occurred because of insufficient surgical debridement: two patientsrefused a second surgical attempt; one patient had very extensive injuries; and one patient (whoalso had a tumor) refused amputation and preferred to live with his osteomyelitis. The meanlength of hospitalization for the HBO group was 54 days (range 41-143) compared to the non\u00adHBO group of 47 days (range 10-66). However, it should be noted that three patients in the non\u00adHBO group went home on parenteral antibiotics quite early, at 10, 16, and 24 days respectively,thus skewing the length of stay data. Their removal incr:.ased the mean length of stay for thenon-HBO group to 51 days. The large difference in the upper end of the range for length of stay,66 vs. 143 days, suggests either a poor (biased) match or a possible untoward effect of HBO orboth. There were \"significant complications\" for two extremely heavy individuals (330 and 250pounds) but their treatment group was not specified. Three recurrences were reported after anaverage follow-up of 41.1 months (range 11.-77 months): two in the HBO group andone in thenon-HBO group. Thus, in the HBO group, a total of five patients ultimately failed (35%) whilein the non HBO group only 2 (14%) failed: Fischer's exact test, p = .38. It is not clear what thismeans since the protocol as initially described does not indicate continued follow-up past 6months for the purpose of evaluation. Thus, the late follow-up results reported by the authorsmay represent a convenience sample. The authors concluded that HBO did not offer anysignificant additional benefit over standard therapy.Despite many shortcomings, this is the only trial to date directly related to the management ofCOM with HBO. Thus, it is of more than passing interest that the reviews briefly discussedabove, published after this study (the fourth was published prior to it in 1984) either: 1) give thistrial only a brief mention and essentially ignore it (6); or criticize it (7); or ignore it completely(8). Indeed, Mader et. al. (7) raise the issue that, in the four failures, all the patients refusedsecond surgical interventions (either further debridement or amputation) concluding that all fourwere doomed to L;: anyway which begs the question since three of the failures were in the HBO10group. Mader et. al. also complain that the recurrences (two which were in HBO treated patients)were due to patient refusal of funher surgical intervention after HBO. Thus. they claim the trialto be inconclusive because the patients were non-compliant. However, one could suggest thatHBb should reduce the need for second or third debridements (and recurrences). Moreover, theEsterhai et. al. trial found that the failure rate in the HBO group was just over double that in thecontrol group, suggesting a potentially clinically significant negative effect of HBO in thesepatients. Thus the failure to demonstrate any statistically significant difference (even in the\"wrong\" direction) between HBO and control groups was not because of bias. but because of its'low power.3.4 CostsInterms of costs , Davis & Heckman (5) cite an example of a patient with COM whose totalmanagement costs were $240,000 over a 7 year period. Of this total cost, they estimated thatHBO costs added an average of $11,600, based on their own experience. HBO costs, however,varied across studies with some estimates ranging from $10,000 to $12,500 or about 5% ofoverall management costs.3.5 SummaryIn summary, case series data all reIX>rt fairly high recovery rates of COM with the use of HB9 asan adjunct to standard therapy, and the only trial in a small number of patients (n=28) showed nobeneficial (and possibly a deleterious) effect. Thus one is left with conflicting data regarding theefficacy and effectiveness of this intervention and little can be concluded until larger trials areundertaken. Moreover, COM would appear, fortunately! to be a fairly rare phenomenon basedboth on VGH and Chinese data. Since success rates (whether measured in cures or arresteddisease) are highly variable and do not seem to be consistently higher with HBO than without it,and given that the expense ofHBO has been estimated by some authors to be as high as $12,500(whether this includes all chamber and physician costs or even the total HBO costs for acomplete course of therapy which may occur intermittently over many months is not clear), HBOmay not provide value for money in the management of COM. More compelling evidence of itseffectiveness and greater understanding of its costs is needed for informed policy decisions.114.0 REVIEW OF THE LITERATURE: OSTEORADIONECROSISORN is a complex clinical entity involving bone, skin and teeth. As noted above, radiationtherapy for cancer produces significant necrosis and resultant hypoxia. Bone, because it is acontinuously growing tissue, is particularly .susceptible to radiation injury. The mandible,because of its idiosyncratic blood supply and its particular proximity to potentially pathogenicbacteria (i.e. the teeth and gums the integrity of which is compromised by the cancer beingtreated), is especially sensitive (2,3). Thus the literature available for review or cited either in theHandbook (2) or Davis and Hunt (3) refers almost exclusively to ORN of the mandible. Theseare mostly basic science or animal studies and case reports (excluded from consideration), and. -case series. A single randomized controlled trial on the effect of HBO vs. penicillin in theprevention of post-radiation osteonecrosis in the context of tooth extraction was reviewed.4.1 ReviewsMyers and Marx' (26) provide the most recent comprehensive discussion of thepathophysiological and clinical aspects of ORN. They cite several early studies in which HBOwas used without adequate surgical management demonstrating that HBO alone was of littletherapeutic value. In the late 1970's the basic pathophysiological problem of hypoxia wasclarified and led to the same type of management approach used in COM: good surgical care,appropriate antibiotics, and HBO as an adjunct maneuver. This reference, essentially a review,(much of it of Marx's own work) provides little detail on the selection of patients or othermethodological issues. Considerable space is devoted to the specifics of HBO treatment such asthe management protocol developed by Marx for the several stages of ORN (Stage I, Stage II,Stage III, and Stage IIIR). It is claimed that ORN was resolved in all 268 patients (38 in Stage I,48 in Stage 11.182 in Stage III). In Stage I, the HBO treatment consisted of 40 dives at 2.4 ATAfor 90 minutes each. If there was insufficient resolution, patients were considered to be in StageII and given an additional 10 dives and if there was still no resolution patients were considered tobe in Stage III and given a further 10 dives. Stage I1IR patients were those who, followingresection and adjunct HBO. were given more radiation. Resolution and cost data are providedfor this large case series. Among the non-HBO patients (N=65) the average one-year cost was$31,000 with total average costs of $104,000 and a resolution rate of 8%. In contrast, among the51 patients with HBO without surgery, the mean one-year cost was $20,000 and the mean totalcost was $82.000, with a resolution rate of 17%. For the Marx protocol 130 patients had a mean12one-year cost of $35,000 and a mean total cost of $35,000 with a 100% resolution rate. All costswere estimated using 1985 dollars. Whether these costs include physician fees is not clear.Del Core et.al. (27) have written a recent review ( in Italian), but it adds little to the informationalready provided.4.2 Case SeriesOther case series reviewed or cited by Myers and Marx, and Marx and Johnson (28), reportessentially the same results and will only be mentioned briefly.Hart and Strauss (29) reported a large series (N=378) in 1986, with 336 patients completing HBOtreatment and receiving 2 ATA for two hours daily (outpatients) or 1.5 hours per day (inpatients)for a total of approximately 120 hours. HBO was started no sooner than 2 months followingradiation because of known complications with. early HBO. The overall success rate is notreported in-the Handbook but data are provided on the 235 patients with mandibular ORN whichaccounted for 62.2% of all the patients in this series. Of these 235 patients, 206 completed RBOtreatment and 72% were said to have \"excellent results\", 10% \"good results\", 15% \"a fairresponse\", and 3% were \"failures\". How these outcomes were defined is not stated . .Similarly.Mainous et. al. (30) reported on 11 cases of mandibular ORN and 22 cases of mandibular COMall of which were said to have healed completely . No specifics are provided regarding treatmentregimen, patient selection, blindness. etc. Finally, Tobey et. al (31) reported in 1977 that 2 ATAappeared to be more effective in mandibular ORN than 1.2 ATA but no specifics are provided.Other case series report essentially the same results (32-35).4.3 Trials.Marx et, al. (36), report the only randomized controlled trial of patients with ORN. There were74 \"high risk irradiated patients\" requiring dental surgery which, as noted above, is acomplicating factor in ORN of the mandible. The focus of this study was on the prevention ofORN and the contrast was between penicillin and .HBO. The results revealed that in thepenicillin control group, 11 patients (29%) developed ORN, while in the HBO group only 2. .patients (5.4%) developed ORN. However, although this was a randomized trial it was not madeclear whether the observers were blinded. As well, the exclusion criteria were severe (e.g. nosignificant co-morbidity etc.), which reduces generalizability.13Myers and Marx also cite an early study by Greenwood and Gilchrist (37) which was said to berandomized trial of 104 patients with ORN who received 6400 Grays. The treatment group wasgiven HBO at 2.4 ATA for 90 minutes daily for twenty days compared with a control groupwhich received only air. The success rate in the HBO group was 92% (with a complication rateof 11%) compared to a non HBO group success rate of 66% (with a complication rate of 26%).No comments are made regarding the methodology employed (how patients were randomized.whether the groups were comparable at baseline or whether blinded assessment was used, etc.),These results seem impressive with a relative increase in success of almost 40% (an absolutedifference of 26%) but no statistical assessment is reported. 'Assuming an equal numbers ofpatients in each group (N=52), the power may have been limited. Unfortunately, the referencecited by Myers and Marx is not a randomized controlled trial but a case series of 4 patients withadditional animal data. Further searches have failed to find the randomized controlled trial of104 patients they describe\".4.4 SummaryIn summary, HBO for ORN of the mandible would seem to have some support, particularly as anadjunct to surgery post irradiation of the mandible. However, the single controlled trial is smallalthough its focus is on an important aspect of this disease: the prevention of ORN. Data fromcase series, although roughly consistent with regard to success rates, are poorly described interms of methodology. Moreover, it is clear that for ORN of the mandible (and perhaps othersites) costs of HBO are substantial, although according to Myers and Marx (26), the costs are lessexpensive than alternative therapeutic regimens. Given the complexity of post radiation ORNfrom a physiological and clinical perspective, and since there would seem to be little else one canoffer the patient, HBO may very well be a useful adjunct in this situation. However, as notedabove, the costs are not trivial, even if less than the alternatives, and further research providingappropriate comparisons of treatment with and without HBO should be undertaken in order toprovide more conclusive evidence than is available to date.\u2022 In addition, the BeOHTA has contacted Marx to obtain the correct reference, but he has been unableto locate it, even after discussion with-Myers.145.0 CONCLUSIONS AND COST CONSIDERATIONSCOM and ORN are infrequent but difficult clinical conditions to manage. The literaturereviewed (mostly case series in humans) suggests that HBQ is a useful adjunct to sound surgicaland antibiotic therapy. However, the overall failure rate in COM patients with HBO in the onlytrial to date was twice the failure rate in those without HBO. Moreover. success rates in caseseries with and without HBO were quite variable and on the same order of magnitude .Although the evidence may be more convincing for ORN, it is still far from satisfactory . Datafrom the YGH indicate that local costs are far lower than those reported by Myers and Marx (26)(but the former do not include physician fees or other costs of care-only HBO chamber costs (seebelow)). Since the proportion of cases seen at the YGH for COM and ORN account for only20% of all patients treated (at least in fiscal 1990-91). and although it is not clear howrepresentative these utilization data are in terms of annual case mix over the last several years, itwould seem that HBO for these conditions represents a relatively marginal cost.There remains the uncertainty of whether HBO does more good than harm or significantlyimproves patient outcomes. Thus, the data at this stage are insufficient to provide cleardirections for policy.' Given the insufficient data regarding COM, it is possible that little is to begained by expanding the utilization of the YGH chamber to include more cases. Indeed, if the1991 data are representative of provincial need (and no argument is put forward by the DivisionHead that there is unmet need), there may be no unmet need. Conversely, since there are severalother conditions for which HBO has been shown effective and is indeed the accepted treatment ofchoice, and given the substantial capital investment, it seems clear that the chamber at the YGHshould continue to operate, and that small changes in the number of cases of COM either up ordown would have a marginal effect on hospital costs although the effect on physician costswould be more dramatic.The situation regarding ORN is less ambiguous but the fact remains that for both conditionspopulation based data of potential provincial caseloads are currently unavailable , thus potentialutilization rates are unknown.5.1 CostsThe issue of costs for the management of these two conditions although hard to estimate warrantsfurther discussion. As noted above, the costs identified in the YGH report are for the operation15of the HBO chamber only and do not include the costs of other management components such assurgery, antibiotics, general hospital care, diagnostic tests (e.g, CT scans) etc. In addition,physician costs were not included. Fees (according to the current BCMA fee schedule, p. 23) are$65.8? for the first hour and $33.86 for each additional fifteen minutes if the physician is insidethe chamber, and $44.86 for the first 30 minutes and $23.86 for each additional 15 minutes if thephysician is outside the chamber. In the Esterhai et. al. trial (23), the HBO group received onedive per day of two hours duration , six dives per week. Although the total number of weeks isnot given, the physician fees for a week of therapy in BC would have be as high as $1178 .58 ifthe physician was inside the chamber, assuming for each day's dive the first hour was billed atthe first hour rate and the second hour at the 15 minute rate x 4, or as low as $1110.24 if thephysician was outside the chamber. According to Davis and Heckman (5), the average numberof treatments for COM is between 30 and 60 (i.e. 5-12 weeks) depending on the rapidity ofresponse, thus the total physician costs per patient would range between $5891 and $14,143using inside chamber fees. However, as noted above, the number of treatment sessions reportedin. the literature may be much higher .so that the figures estimated here may be conseryative and,as reported in the literature, cost data may range widely.With regard to ORN, Marx and Johnson (27) report that with appropriate staging the minimalcost would be for stage I patients who require at least of 30 dives (90 minutes a dive) at anapproximate HB0 cost of at least $4000 for these patients with the best prognosis.While it is noted by many authors that HBO treatment may be carried out on an outpatient b~is,thus shortening hospital length of stay and thereby reducing costs, it is also clear that the overallcosts of managing patients with COM or ORN ate not trivial and much higher than the averagefixed chamber costs calculated from the VGH data. These estimates clearly do not represent aformal cost analysis but give a rough idea of what the minimal management costs of thesepatients are likely to be. A thorough cost analysis should be undertaken since, despite thesuggestion of significant benefit in numerous case series , effectiveness data are poor and costsare high.5.2 ConclusionThus, it is concluded that until a more convincing case can be made both for efficacy andeffectiveness, and overall provincial rates of these two conditions is clarified, expandedutilization of HBO for COM (and probably also for ORN) is unwarranted. The uncertaintyregarding the recent data on effectiveness might even suggest that current utilization may be of16little value for these conditions and should be discouraged, although the overall cost savingsassociated with a reduction of such utilization are likely to be marginal (except for physicianfees) given the smail number of cases, high fixed chamber operating costs, and the utility ofHBO for other conditions. However, the costing data provided in the VGH report are insufficientto obtain a clear picture of the total costs of treating COM or ORN. Notwithstanding this lack ofdata, evidence from the literature suggests that total management costs are substantial, possiblyas high as $250,000 to $300,000 per patient with COM (the higher figure for patients withORN), with 10% of these figures representing HBO costs. Thus there is an urgent need for datato support, clearly and unambiguously, this therapeutic modality for these conditions.What is perhaps most striking (and sobering) about the evaluation of HBO is that after 30 yearsof intensive research including several thousand published papers, at least two decades ofsymposia of various.kinds, and development of unquestioned clinical expertise by a number ofkey groups, the efficacy and effectiveness. of this therapeutic intervention remains unclear. Onlythe fact that it is an adjunct to other forms of management is unambiguous and represents aconsensus among workers in this field. The lack of appropriate randomized controlled trials(despite the view of many authors that such trials cannot be undertaken- a view based oninsufficient evidence) and the absence of good economic evaluations would appear to leave us inthe clinical and policy equivalent of purgatory, where we shall remain until proper evidence isproduced.17REFERENCES1. Jacobson JH, Marsch JHC, Randall-Baker L: Historical perspectives of hyperberic oxygen _therapy. Ann NY AcadSci, 1965;117:651 cited in Handbook. See Ref 2.2. Fischer B, Jain KK, Braun E, Lehrl S: Handbook of Hyperbaric Oxygen Therapy. Springer\u00adVerlag. New York 1988.3. Problem Wounds: The Role of Oxygen. (IC Davis and TK Hunt, eds). Elsevier. New York. 1988.4. Lepawsky M: Maximizing cost effectiveness in the hyperbaric unit. Vancouver General HospitalHyperbaric Unit Annual Report. 1990-1991.5. Davis JC, Heckman JD: Refractory Osteomyelitis. Problem Wounds. Op cit Ref 3. Chapter 5. pp125-142.6. Grim PS, Gottlieb LJ,\"Boddie A, Batson E: Hyperbaric oxygen therapy. JAMA 1990;263:2216\u00ad20.7. Mader JT, Adams KR, WallaCe WR: Calhoun JH. Hyperbaric oxygen as adjunctive therapy forosteomyelitis. Inf Dis Oinic of N.A. 1990;4:433-40.8. Calhoun JH, Cobos JA, Mader JT: DOes hyperbaric oxygen have a place in the treatment ofosteomyelitis. Orthopedic Infection 199i; 22:467-71.* 9. Kelly PJ, Martin WJ, Coentry MB: Chronic osteomyelitis: Treatment with closed irrigation andsuction. JAMA 1970;213:1843-48.*10. Papineau LF, Alfrange A, Dalcourt J, et al: Osteomylite chronique: Excision et graffe despongieuz a l'air libre apres mises a plat extensives . Int Orthop 1979;3:165.*11. May JW, Gallico GG, Lukash FN: Microvascular transfer of free tissue for closure of bonewounds of the distal lower extremity. N Engl J Med 1982;306:253-57.\u2022 Reference read by citation only - because unavailable locally18* 12. May JW, Gallico GG, Jupiter J. et al: Free latissimus dorsi flap with skin graft for treatment oftraumatic chronic bony wounds. Plast Reconstr Surg 1984;73:641-49.*13. Mathis SJ, Alpert BS, Cheng N: Use of the muscle flap in chronic osteomyelitis: Experimentaland clinical correlations. Plast Reconstr Surg 1982;69:815-28.*14. Mathis SJ, Feng LI, Hunt TK: Coverage of the infected wound. Ann Surg 1983;198:420-29.15. Slack WK, Thomas DA, Perrins DJD: Hyperbaric oxygen in chronic osteomyelitis. Lancet1965;1:1093-4.*16. Perrins JD, Maudsley RH, Colwill MW et al: OHP in the management of Chronic Osteomyelitis.In Proceedings of the Third International Conference on Hyperbaric Medicine, (Brown IW, CoxBG eds) Washington D.C. National Acad of Sciences National Research Counci11966:pp. 578-84.17. Dupenbusch FI, Thompson RE, Hart GB: Use of Hyperbaric oxygen in the treatment of refractoryosteomyelitis: A preliminary report. J Trauma 1972;12:763-68.18. Refractory osteomyelitis of the extremeties arid axial spine: In: Hyperbaric Oxygen Therapy. (ICDavis and TK Hunt eds.) Undersea Medical Society, Inc. Bethesda MD. 1977; Chapter 15; pp 217\u00ad227.19. Davis JC: Adjunctive hyperbaric oxygen in chronic refractory osteomyelitis: Long term follow-upresults. Letterto the editor. Clinical Orthop 1986;205:310.20. Morrey BF, Dunn JM, Heimbach RD, et al: Hyperbaric oxygen and chronic osteomyelitis. ClinOrthop 1979;144:121-27.21. Eltorai I,Hart GB, Strauss MB: Osteomyelitis in the spinal cord injured: A review and apreliminary report on the use of hyperbaric oxygen therapy. Paraplegia 1984; 22:17-24.22. Davis, JC, Heckman JD, De Lee J, Buckwold FJ: Chronic non-hematogenous osteomyelitistreated with adjunct hyperbaric oxygen. J Bone and Joint Surg 1986;68A:1210-17.\u2022 Reference read by citation only - because unavailable locally1923. Lee Hui-chieh, Niv, Ko-chi, Chen Shing-Han, et al: Hyperbaric oxygen therapy in clinicalaplication. A report of 12 years' experience. Chung Hua I Hseueh Tsa Chili 1989;43:307-16.24. Esterhai JL, Pisarello J, Brighton CT et al: Adjunctive hyperbaric oxygen therapy in the treatmentof chronic refractory osteomyelitis. J Trauma 1987;77:763-68.* 25. Cierny G, Mader JT, Pennicle 11: A clinical staging system for adult osteomyelitis. ContempOrthop 1985;10:17-37.26. Myers RAM, Marx RE: Use of hyperbaric oxygen in post radiation. Head and Neck Surgery. NCIoMonographs 1990;9:151-157.27. Del Core G, Serpio R, Laino G: L'ossigenoteerapia iperbarica nel tattamentodell'osteoradionecrosia: una revisione del suo uso ed efficacia. Archivo Stomatologico1989;30:421-8.28. Marx RE, Johnson RP: Problem wounds in oral and maxillofacial surgery. Problem Wounds. Opcit Ref 3. Chapter 4, pp 65-123.*29. Hart GB, Strauss MB: Hyperbaric Oxygen in management of radiation injury. In: Proceedings ofthe 1st Swiss Symposium on Hyperbaric Medicine. Foundation for Hyperbaric Medicine, Basel.1986; pp. 19-39.30. Mainous EG, Boyne PJ, Hart GB: Hyperbaric oxygen trea~ent of mandibular osteomyelitis:report of three cases. JAMA 1973;87:1426-30.31. Tobey RE,Kelly JF,Vinton JR, et. al.: Hyperbaric oxygen therapy for chronic osteoradionecrosisof mandible. In: Proceedings of the VI International Congress of Hyperbaric Medicine. Universityof Aberdeen Press. 1977: pp 217-278.32. Teixeria W, Muller F, Villemin T, Meyer E: Hyperbaric oxygen in the treatment ofosteoradionecrosis of the mandible. Laryngo-Rhino-Otologie 1991;70:380-3. In german, abstractreviewed.\u2022 Reference read by citation only - because unavailable locally20* 33. Beumer J, Harrison R, Sanders B, et al.: Osteoradionecrosis: predisposing factors and outcomes oftherapy. Head Neck Surg 1984;6:819-27.*34. Kaufman-T, Hirshowitz B, Monies-Chass 1: Hyperbaric oxygen for postradiation osteomyelitis ofthe chest wall. Harefuah 1979; 97: 220-222.35. Marx RE, Ames JR: The use of hyperbaric oxygen therapy in bony reconstruction of the irradiatedand tissue deficient patient J Oral Maxillofacial Surgery 1982;40:412-420.36. Marx RE, Johnson RP, Kline SN: Prevention of osteoradionecrosis: A randomized prospectiveclinical trial of hyperbaric oxygen. J Am Dental Assoc. 1985;III :49-54.37. Greenwood TW, Gilchrist AG: The effect of hyperbaric oxygen on wound .healing followingionizing radiation. In Proceedings of the Fifth International Congress on Hyperbaric Medicine VolI (Trapp WL, Bonnister E'W, Davidson AJ et al. eds), Burnaby Canada : Simon Fraser Univ. 1973.pp253-263.* Reference read by citation only - because unavailable locally21","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/hasType":[{"value":"Report","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt":[{"value":"10.14288\/1.0045255","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/language":[{"value":"eng","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#peerReviewStatus":[{"value":"Unreviewed","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/provider":[{"value":"Vancouver : University of British Columbia Library","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/publisher":[{"value":"British Columbia Office of Health Technology Assessment (BCOHTA)","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/rights":[{"value":"Attribution-NonCommercial-NoDerivs 2.5 Canada","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#rightsURI":[{"value":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/2.5\/ca\/","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#scholarLevel":[{"value":"Faculty","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/subject":[{"value":"Health technology","type":"literal","lang":"en"},{"value":"Medical technology","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/title":[{"value":"Hyperbaric oxygen for osteomyelitis and osteoradionecrosis","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/type":[{"value":"Text","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#identifierURI":[{"value":"http:\/\/hdl.handle.net\/2429\/50610","type":"literal","lang":"en"}]}}