Controlled CMS Data Demonstrates a Cost and Clinical Advantage for Hyperbaric Oxygen for Radiation Cystitis

UNDERSEA & HYPERBARIC MEDICINE

Controlled CMS Data Demonstrates a Cost and

Clinical Advantage for Hyperbaric Oxygen for

Radiation Cystitis

John J. Feldmeier, DO1; John P. Kirby, MD2; Helen B. Gelly, MD3; Marc Robins, DO4; John Peters, FACHE5; Peter Gruhn, MA6; Sarmistha Pal, PhD6

1 Professor Emeritus and Past Chairman, Radiation Oncology, University of Toledo Medical Center 2 Associate Professor of Surgery, Washington University School of Medicine, St Louis, MO
3 President, Regenerative and Hyperbaric Medicine, Marietta, Georgia
4 Senior Medical Director, Intermountain Health

5 Executive Director, Undersea and Hyperbaric Medical Society 6 Dobson DaVanzo and Associates, LLC

CORRESPONDING AUTHOR: John J. Feldmeier – jfeldmeier@aol.com

RESEARCH ARTICLE | Copyright ©2024 Undersea & Hyperbaric Medical Society, Inc.

145

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

INTRODUCTION

Cancer survivorship has increased dramatically over recent decades in part due to the increased appli- cation and technological improvements of radia- tion therapy in treatment protocols, which are often multidisciplinary [1]. The American Cancer Society estimates that about 714,000 cases of pelvic malig- nancy occur annually in the United States [1]. More than 50% of these are likely to receive radiation as part of their cancer management. Modern radia- tion techniques and advanced technologies have allowed dose escalation, which has surely contribut- ed to improved tumor control, especially in prostate cancer. Many in the radiation oncology community had predicted superior tumor control and reduced serious toxicity with the newer targeted techniques, prominently including intensity-modulated radia- tion (IMRT). [2,3]

However, late toxicity involving normal tissues still occurs despite new modern targeted techniques. In part, the failure to achieve decreased toxicity may be due to the increased radiation dose that has become the standard of care in many tumors, especially for prostate cancer. Pelvic radiation, in many instances curative, still can cause damage to adjacent normal tissues in some patients. Several reviews do indeed show a reduction in rectal complications when using newer techniques, but for bladder toxicity, the re- sults are not clearly improved. Recent papers report nearly identical incidences of bladder complications with conformal radiation (an older tumor-targeting technique) and IMRT [4]. The reported incidence of bladder late radiation tissue injury (LRTI) in recent studies varies from about 7% to 11 % [5-8] and to perhaps as high as 13%.

A very telling statistic is reported in the Ma paper [5], which prospectively reviewed 1,198 consecu- tive patients admitted to a tertiary hospital urology service over six months. Twenty-three percent of all emergency urologic admissions in this review were attributable to radiation therapy complications. Radiation cystitis can be a chronic problem that re- quires multiple hospitalizations, several endoscopic and invasive procedures, multiple transfusions, di- minished quality of life, and even death.

The review of 709 patients by Martin et al. [7]. provides additional perspective here, reporting that radiation cystitis patients require an average of 2.5 admissions and an average hospital stay of 7.6 days (range 1-42 days). Fifty-two percent require a blood transfusion of an average of 4.3 units. These patients underwent procedures including cystos- copy with fulguration +/-clot evacuation in 86% of cases. Four of their 709 subjects died due to radia- tion cystitis. In another publication, Linder et al. [9] reported a 16% mortality for patients undergoing urinary diversion and cystectomy when the blad- der is deemed to be unsalvageable due to the per- sistence and severity of late effects of radiation injury.

The National Cancer Institute (NCI) has developed a widely employed reporting and grading system for cancer treatment adverse effects and continues to update it to grade the severity of complications. This system is called the Common Terminology Cri- teria for Adverse Events (CTCAE) [10]. Many pub- lications emphasize Grades 3 and 4 CTCAE dam- age in their studies of cystitis because, at these levels, the severity of symptoms requires interven- tion to restore quality of life and/or preserve life.

Our understanding of the pathophysiology of late radiation effects on normal tissue has been historical- ly attributed to vascular injury of the irradiated tissues characterized by endarteritis with resultant tissue hy- poxia and attendant tissue damage [11-13]. More re- cently, some experts have proposed a model known as the Fibroatrophic Effect [14]. In this model, delayed radiation injury evolves due to apoptotic cellular de- pletion of parenchymal cells and stem cells within tis- sues and organ systems. Actively functioning cells are replaced by fibrous amorphic, avascular, and acellular tissue fields. The two models are not mutually exclu- sive. Inadequate vascularity can result from fibrosis replacing normal functional tissues [15], and chronic tissue hypoxia caused by vascular compromise can result in tissue fibrosis [16]. The described mecha- nisms of both models result in a positive feedback loop that promotes and perpetuates chronic delayed damage. Hyperbaric oxygen has been demonstrated to benefit all three elements of the established patho- physiology of LRTI. It enhances angiogenesis, stimu- lates stem cell production, and reduces fibrosis [17].

146

Feldmeier J, Kirby J, Gelly H, et al.

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

Pascoe and colleagues [18], after reviewing 23 pa- pers on the topic, published a manuscript providing guidelines and an algorithm for managing radiation cystitis. They discuss cystoscopy with clot extraction, irrigation, chemical, laser, argon coagulation, chem- ical cautery, and hyperbaric oxygen, although hy- perbaric oxygen is not introduced until after more conservative interventions have failed. They also admitted that there is a dearth of Level 1 evidence supporting any of these interventions.

Hyperbaric oxygen has been used to treat radia- tion cystitis for several decades [19-25]. These pub- lications consist mostly of fairly small clinical series. Results have been consistently positive, but the level of evidence is low. Oscarsson and co-authors have recently [26] reported a multicenter phase 2-3 randomized controlled trial in 2019. The study was conducted employing the Expanded Prostate Index Composite Score (EPIC), a patient self-accomplished quality of life questionnaire. The hyperbaric group improved by 10.1 points on the serial EPIC assess- ment compared to 7.7 points for the non-hyperbaric group. The comparison of these changes was statis- tically significant.

The present study utilizes Medicare claims data to compare the clinical outcomes as reconstruct- ed from billing codes and the cost-effectiveness of using hyperbaric oxygen therapy to treat radiation cystitis versus conventional care. Using de-identi- fied billing records and without access to the clini- cal records, we reconstructed the clinical course and cost to Medicare for treating 3,309 patients between 2014 and 2019. In this fashion, a large experience contrasting treatment with and without hyperbaric oxygen was made available for analysis. Note where billing quotes are presented in this paper, they in- clude both the Medicare payout and the patient co- pays.

MATERIAL AND METHODS
Database construction
This study employed Medicare Part A and Part B fee-for-service (FFS) claims for patients with radia- tion-induced cystitis. Claims included data for inpa- tient, outpatient, durable medical equipment (DME), home health agency (HHA), hospice, and skilled

nursing facility (SNF), as well as carrier administra- tive data, including Medicare beneficiary demo- graphic and enrollment-related data from the Mas- ter Beneficiary Summary File (MBSF) for the period from October 1, 2014, through December 31, 2019. These data were used to construct a study group of beneficiaries who received hyperbaric oxygen treat- ment (HBO2) and a control group who received only non-hyperbaric care, which included primarily con- servative care with irrigation, clot extraction, and cautery for their radiation cystitis. A lack of accu- rate coding resulted in some cases not being iden- tified, and we felt that it was reasonable to include patients with diagnoses of both chronic cystitis and hematuria, along with a history of pelvic malignan- cy. All aspects of patient privacy were protected per HIPPA regulations by virtue of the study design, and all information was de-identified without access to clinical records. For inclusion, subjects had to have at least two non-laboratory or non-imaging billings consistent with radiation cystitis. The vast majority of both groups and essentially all of the study group, had been coded with a diagnosis of radiation-in- duced cystitis. We know that the subjects receiving hyperbaric oxygen were almost certainly coded properly because only those with late effects of radi- ation would be covered by Medicare.

Beneficiaries were excluded from the analytic file if they had radiation proctitis or had received active cancer treatment less than three months before their enrollment. In instances where a beneficiary had more than one eligible episode of HBO2 over the study period, the longer episode was included in the analysis. In instances where more than one course of hyperbaric oxygen was completed, almost always the second course was the longer of the two, and consistently, the first course was very short, typically consisting of one or two treatments.

The focus was on beneficiaries with chronic de- layed radiation cystitis for which standard treat- ments had not resulted in remission. To be consid- ered in remission after treatment, there needed to be six months or more without cystitis-specific ad- ditional therapeutic interventions, including trans- fusion.

Feldmeier J, Kirby J, Gelly H, et al.

147

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

For both the study group and the control group, the analysis consisted of 3 distinct time periods or phases. Initially, all Medicare records available to us were scanned to identify procedures and tests consistent with radiation cystitis. These included bladder endoscopic procedures or surgical inter- ventions, clot retraction, and bladder irrigation and transfusions. Once the diagnosis of cystitis was made, a one-year pre-treatment review was done to establish baseline Medicare expenditures for all en- rolled patients. This pre-treatment phase constitut- ed Phase 1 for both groups. The second phase was the treatment phase, wherein for both the study group and the control group, the costs specifically related to cystitis treatment were identified. Phase 3 was the one-year follow-up for both the study and control groups. In both Phase 1 and 3, total Medi- care outlays were reviewed and compared. Figure 1 provides a quick visual depiction of the study design and sequence.

Our inclusion of all Medicare payments in Phase 1 and Phase 3 incorporated the cost of comorbidities and included all therapeutic interventions. For the HBO2 study group, the treatment phase (Phase 2) began on the first day, ended on the last day of hy- perbaric treatment, and was 81 days on average. The hyperbaric oxygen phase was considered complete if there was a four or more months break between HBO2 treatments.

For the non-hyperbaric group, Phase 2 began when billings were also consistent with the onset of radiation cystitis. When modeling was done, the length of the treatment group, Phase 2, for the con- trol subjects was initially set at 97 days as an esti- mate based on preliminary findings. Because of a wide variation in clinical response to treatment, it was decided not to apply this information in calcu- lating differences in the cost of treatment.

Other interventions for radiation cystitis would continue for both groups and be quite variable for each patient until and if there was a remission. Be- cause some patients would continue to have cystitis symptoms beyond the defined and modeled Phase 2 definitions of the treatment period and well into the follow-up period, we decided that the most ap- propriate and relevant financial comparison of treat-

Schema for Phases of Study

Entry to Study

Diagnosis Of Cystitis

Treatment Phase 2

One Year Follow-Up Phase 3

148

Feldmeier J, Kirby J, Gelly H, et al.

One Year Look back Prior to First Treatment of HBO2 for Study Group and One Year Prior to the first Pseudo Treatment in ControlGroup Phase 1

Figure 1. This Schema provides a pictorial depiction of the Study Activities and Timing of Various Phases.

ment expenses would be the cost of the hyperbaric treatments, which was unique to the study group. Certainly, for the study group, these would have the largest impact on the expenditures during that phase. Both groups would continue to have oth- er expenses, including cystoscopies, fulgurations, transfusions, etc. However, since the study group ultimately showed significant clinical improvements comparatively, these interventions and their atten- dant costs in this group would decrease more on average over time than the costs of those interven- tions for the control group. A previously cited review paper had recommended 40 hyperbaric treatments routinely for radiation injuries.[17] In order to calcu- late the Medicare-authorized reimbursement for a typical course of hyperbaric treatment, the number 40 was multiplied by the daily authorized Medicare

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

reimbursement for a single hyperbaric treatment, including the physician’s fee. The total expenditure for 40 hyperbaric treatments would be just about $25,000.

Statistical design and analysis

For analysis, we used Difference-in-Differences (DiD) modeling to evaluate the effect of HBO2 on Medi- care spending as well as for other outcome mea- sures, including the incidence of bleeding, the need for transfusion, the number of transfusions, and the record of urologic endoscopic procedures. For other metrics, the DiD model was applied separately. Fi- nally, using a linear probability model, we examined whether there was any difference in mortality rate between the study and control groups.

The main statistical challenge arises from the fact that patient assignment to treatment or control group is not random. Self-selection could be expect- ed to increase entry into the HBO2 study group. Oth- er factors, observed (e.g., health status or comorbid- ities) and unobserved (e.g., mental stress or family background), could have also affected the number of patients seeking hyperbaric treatment.

There may also be a physician bias in referring only more severely affected patients for hyperbaric oxy- gen due to the perceived increase in expense when HBO2 is employed. Several review articles, including clinical practice guidelines, do not include consid- eration of hyperbaric oxygen until most or all other common interventions have failed. We used an in- verse probability of treatment weighting (IPTW) propensity score model to address the non-random- ness issue. [27].

The IPTW method has several advantages [27-30]. The over-expressed weight of the control group is weighted down [28]. The IPTW approach uses the entire cohort and can address a very large number of confounding variables, and in doing so, it provides increased precision and representativeness [28,29]. Additionally, IPTW requires fewer distributional as- sumptions about the underlying data and avoids the potential residual confounding that arises from stratification on a fixed number of strata [30].

To address the potential self-selection issues, we estimated the Propensity Score model to assign ap-

propriate weights to each of the comparison group and treatment group beneficiaries. [31-33] This method allows observational studies to be designed similarly to randomized experiments. We have also performed the Balanced Diagnostic Test to ensure that the observed baseline covariates are similar be- tween the treatment and comparison groups. [34] We used the Standardized Mean Difference statistic to confirm that the covariates are balanced between these two groups at the baseline period. The bal- anced diagnostic test result suggests that all the co- variates satisfy this diagnostic test in the base year since the absolute value of the standardized mean difference of all the explanatory variables is less than 0.25. Please reference Table 1.

RESULTS

Results are included below in two major catego- ries and summarized in Tables 2 and 5. The first is a cost comparison of Medicare payments in the three phases. In Phases 1 and 3, both one-year periods, the comparison was made for average total Medicare outlays for both groups. See Table 2. For Phase 2, it was felt that the most pronounced difference in bill- ings was the cost of the hyperbaric treatments and that the other cystitis-specific interventions would be comparable or reduced in the study group as more hyperbaric treatments were completed. Such a reduction would offset some of the expenses in- curred by the hyperbaric treatments. A more specific comparison of these relative outlays is not possible due to the ongoing interventions for radiation cys- titis in those who were not brought into remission in either group. Our modeling was not designed to provide this information. We note that the costs of continued specific interventions for unresolved cys- titis are included in the total expenditures incurred during the one-year follow-up period. As noted ear- lier, these payments would include cystitis-specific interventions.

We found that the hyperbaric group was more likely to carry a diagnosis of hematuria and more likely to require transfusion in Phase 1. These find- ings suggest that the hyperbaric group included pa- tients with more severe radiation cystitis. See Tables 3 and 4. Notably, the HBO2 group also had higher

Feldmeier J, Kirby J, Gelly H, et al.

149

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

Table 1. Summary of Covariates and Diagnostic Test in Pre-Period for HBO2 and Comparison

without weight

with IPTW weighting

hyperbaric group (1,030)

control group (2,279)

hyperbaric group (1,030)

control group (2,279)

explanatory variables

age

female

dual

white

black

asian

hispanic

north american native

others CCI

mean

76.59 8.06% 5.15% 85.92% 9.51% 0.78% 0.97%

0.19%

2.62% 0.46

explanatory variables

age

female

dual

white

black

asian

hispanic

north american native

others CCI

mean

76.90 25.81% 13.08% 89.30% 6.54% 1.00% 0.82%

0.17%

2.17% 0.47

explanatory variables

age

female

dual

white

black

asian

hispanic

north american native

others CCI

mean

76.76 21.33% 12.98% 88.08% 7.27% 0.86% 1.05%

0.31%

2.42% 0.45

explanatory variables

age female dual white

mean

76.92 31.41% 15.50% 89.98%

balanced diagnostic test

-0.02 -0.23 -0.07 -0.06 0.05 -0.02 0.03

0.02

0.02 -0.06

black 5.91%

asian 1.06% hispanic 0.80%

north
american 0.18%

native
others 2.07%

CCI 0.48

This table compares pre-treatment demographics and the Charlson Co-morbidity Index for both unweighted and weight- ed study and control groups. Both groups are well-matched in each category, as demonstrated by the Balanced Diagnos- tic Test, which reflects the validity of the weighting and subsequent statistical determinations.

Table 2. Average Total Annual Patient Medicare Authorized Payments

total spending

study (HBO2) control

pre-treatment (phase 1)

$31,293 $29,130

follow-up at one year follow-up (phase 3)

$26,234 $31,496

difference post-pre-treament

-$5,059 (any hyperbaric treatment)

+$2,366

difference post-pre-treament for 40 treatments

-$11,548 (40 or more hyperbaric treatments

+$2,366

This table presents total Medicare-authorized payments for both groups, comparing post-treatment to pre-treatment (phase 3-phase 1) The entries suggest a more severe initial level of disease in the study vs control group with $2163 additional charges. At the one-year follow-up, the average savings for the study group at all hyperbaric dose levels equals $5059. Note in the table a minus sign indicates a savings and a + sign represents an increased Medicare- authorized payment.

Table 3. Incidence of radiation cystitis with bleeding for each group during Phase 1

Table 4. Percent of patients receiving blood transfusion during Phase 1 (Pre-treatment)

radiation cystitis with urinary bleeding

study group
control group
difference (study-comparison)

phase 1. pre-treatment

63.6% 14.9% 48.7%

percentage requiring blood transfusion

study group
control
difference (study-comparison)

phase 1. pre-treatment

10.4% 5.1% 5.3%

This table shows that the occurrence of hemorrhagic cysti- tis is more than four times more likely to occur in the study group than in the control group during the pretreatment phase strongly suggesting more severe disease.

This table demonstrates that transfusion during phase 1 is twice as likely in the hyperbaric group compared to control suggesting more severe disease.

150

Feldmeier J, Kirby J, Gelly H, et al.

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

billings than the control during Phase 1 (pre-treat- ment). These higher payments for the hyperbaric group during Phase 1 also suggest that the study group started with more serious radiation complica- tions on average at entry.

Table 5 shows several important comparisons. Control patients are compared to hyperbaric pa- tients treated at four increasing hyperbaric dose points: one or more treatments, more than or equal to 20 treatments, more than or equal to 30 treat- ments, and more than or equal to 40 treatments. Analyses at these dose points show better results as the number of treatments is increased sequen- tially to 40 or more. The figures in the parentheses indicate the percent change relative to the mean value of the respective dependent variables in the pre-treatment period (except for mortality). The mortality analysis was based on a linear probability model, so the figures reflect the percent change rel-

ative to the mean value of the comparison group in the post-treatment period.

As shown in Table 5, total Medicare spending in the post-treatment period relative to the mean value in Phase 1 (pre-period) among patients in the HBO2 study group who received at least 40 treatments was $11,548 lower than the change in spending for the control group patients during the same period, a reduction in spending of about 37%. This finding was significant at the 1% level. The spending reduc- tion was primarily driven by lower inpatient hospital spending of $9,222 and lower physician services-re- lated spending of about $1,149, a reduction of 60% and 15%, respectively. These reductions in spending were also significant at the 1% level. As shown in the analyses of the subgroups divided according to the number of hyperbaric oxygen treatments, the sav- ings to Medicare increased as the number of treat- ments increased in almost all clinical outcomes.

Table 5. Overview of Additional Clinical Results with Increasing Number of Treatments

hyperbaric study group (N)

comparison group (N)

dependent variables

total spending

inpatient spending

physician spending

endoscopic treatment [Y/N]

number of endo- scopic procedures

urinary bleeding (with radiation cystitis dx) [Y/N]

blood transfusion [Y/N]

mortality [Y/N]

base model hyperbaric treatments ≥ 1

1,030

2,279

percentage difference

-$3,267 (10% ↓ *)

-$3,188 (22% ↓ **)

-$1,114 (14% ↓ **)

(30% ↓ ***) (44% ↓ ***)

(27% ↓ ***) -

(not significant) (18% ↓ **)

hyperbaric treatments ≥ 20

695

2,279

percentage difference

-$8,274 (27% ↓ ***)

-$6,686 (46% ↓ ***)

-$1,358 (17% ↓ ***)

(33% ↓ ***) (37% ↓ ***)

(31% ↓ ***) (57% ↓ ***)

(46% ↓ **)

hyperbaric treatments ≥ 30

693

2,279

percentage difference

-$10,868 (34% ↓ ***)

-$8,221 (55% ↓ ***)

-$1,760 (22% ↓ ***)

(31% ↓ ***) (33% ↓ ***)

(31% ↓ ***) (91% ↓ ***)

(56% ↓ ***)

note: *** indicates significant at 1%, ** indicates significant at 5%, * indicates significant at 10%

This table shows specific financial and clinical results comparing the results for the hyperbaric group overall to the control group for the same time period and at additional intervals of 20 or more hyperbaric treatments, 30 or more treatments, and 40 or more treatments. Even at 1 to 20 treatments, there was a clinical and financial advantage. All dollar amounts are for a 12 month period per individual patient.

Feldmeier J, Kirby J, Gelly H, et al.

151

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

DISCUSSION

The study provides clinical support and econom- ic justification for using hyperbaric oxygen to treat chronic radiation-induced cystitis. All clinical out- come parameters compared to standard care (Table 5) show that hyperbaric oxygen is more effective than traditional interventions. Analyses in Table 5 also support a clear dose response, demonstrat- ing that patients had better clinical outcomes at 40 treatments or more than at fewer treatments. Thirty to 60 hyperbaric treatments for radiation in- juries have been previously recommended by an extensive review article in 2012.[17]. Because of its claims-based design, our study does not allow us to compare specific hyperbaric protocols, including treatment pressures or duration of individual treat- ments. We note that fewer than our study patients actually received 40 or more treatments.

Reductions in Medicare outlays for the Study Group across all doses equal the $5,059 decrease in the Study Group. This advantage increases with in- creasing numbers of hyperbaric treatments. For the group receiving at least 40 treatments, this advan- tage has increased to $11,548 in the first year. With- out additional years of follow-up, we cannot project the savings in the second year or beyond. However, it is notable that nearly one-half (46%) of the cost of the hyperbaric treatments was recouped in the first year.

Smart and Wallington [35] have demonstrated cost savings with a 2.5-year follow-up that are consistent with our findings and suggest a durable hyperbaric effect. Admittedly, the results of the current study would be stronger if we had a more extended fol- low-up period to analyze them. Further analysis of the data with a longer post-treatment study period would help clarify the longer-term financial impact. Even so, by one-year post-treatment, the advantage of hyperbaric treatments in comparative billings has recovered over the cost of hyperbaric oxygen.

Perhaps the most notable outcome of our study was a decrease in mortality in the hyperbaric-treated pa- tients by more than 50% compared to control at one year. A review of the literature shows that the trend to- ward a decrease in mortality after hyperbaric oxygen treatments is not unique to our study. Löndahl [36]

has previously reported an increase in survival as- sociated with hyperbaric oxygen in his randomized controlled trial of hyperbaric oxygen for patients with diabetic foot ulcers. At six years follow-up, patients in this trial receiving HBO2 had a survival of 63.2%, while non-hyperbaric controls had a survival of 40.5%. A more recent, non-randomized trial reports increased survival in patients receiving hyperbaric treatment of chronic osteomyelitis. In this cohort comparison study of 5,312 patients admitted to the hospital for chronic osteomyelitis authored by Thai and associates [37], overall survival was better for the hyperbaric treated group at one year compared to the non-randomized control group (3.8 % mortality vs 7.6%).

The Charlson comorbidity index (CCI) is used to predict one-year and ten-year mortality in study populations.[38] The CCI calculated for both groups before treatment demonstrated an excellent match for the presence of intercurrent diseases and pre- dicted comparable mortality. Some might argue that those receiving hyperbaric oxygen in the interven- tion for chronic radiation cystitis represent a group not well-matched to non-hyperbaric patients and that patients referred for hyperbaric oxygen might represent a more favorable group. In fact, clinicians often employ hyperbaric oxygen as a last resort due to its perceived expense and inconvenience. The previous discussion in the results section provides evidence suggesting that hyperbaric oxygen is fre- quently reserved for more seriously injured patients. In our study, the likelihood of delayed referrals is borne out by the fact that approximately two-thirds of the patients we studied did not have hyperbar- ic treatment. Several publications recommend de- ferring hyperbaric oxygen application until all oth- er interventions have failed. [39-43] As previously discussed, clinical practice guidelines frequently require “conventional” care failure before coverage of hyperbaric oxygen therapy can be provided. The National Coverage Determination 20.29, adopted by the Centers for Medicare and Medicaid Services, explicitly requires a lack of response to conven- tional therapy prior to adding hyperbaric oxygen treatments to the clinical protocol.[44] Our patient population was limited to those who were Medi- care-eligible. Therefore, they likely failed multiple

Feldmeier J, Kirby J, Gelly H, et al.

153

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

rounds of conventional care. The cost of their con- ventional care is shown in Table 4 for the time limits of the study, and the treatment arm has higher costs as compared to the control arm, arguing for more complexity and interventions. However, Chong et al. [45] have shown that early intervention with hy- perbaric oxygen for radiation cystitis improves clin- ical outcomes. Pursuing ineffective treatments only adds to the cost of management.

This study employed a novel methodology for re- constructing medical interventions and comparing the Medicare-authorized payment for beneficiaries with chronic radiation cystitis. Employing the key- words “cost-effectiveness” and “Medicare payment records” in a MEDLINE search, eight publications were discovered that address Medicare payment and a variety of clinical interventions in various dis- orders. [46-53] These study designs were compared to the methodology of this study. As opposed to the present study, these publications either had access to clinical records or used a simulated population. Many did not present a clinical outcome. In con- trast, our access to Medicare payment data was em- ployed to identify the actual clinical interventions in real patients and the reimbursed amounts for treat- ing chronic radiation-induced cystitis.

Using this methodology, a total of 3,309 patients were identified and studied for the years in which Medicare payment records were available to us. Of these, 1,030 patients received hyperbaric oxygen, and 2,279 received traditional treatments only. Our study's analyzed cases far exceed even comprehen- sive review studies comparing the results of several common interventions at many clinical centers. By comparison, two comprehensive review papers had pooled patients of 985 and 1194 respectively. [18,54] Despite the novel study design, the methodology employed in the current study was straightforward, and validity is expected to be high. Patient self-se- lection or managing physician preference might have impacted the random nature of patients in the hyperbaric group. Previous discussion in the Mate- rials and Methods section presents some statistical techniques employed to correct this possible bias. By combining all Medicare beneficiaries treated for chronic radiation cystitis during the period of the study, most, if not all, biases were excluded.

While some may question its evidentiary value be- cause it is not an RCT, this study permits a compara- tive analysis of more than 3,000 patients with chron- ic radiation cystitis who met eligibility requirements treated over a period of 63 months without identifi- able bias in selecting a control group or study group. While randomized controlled trials have been the gold standard for evidence-based medicine for sev- eral decades, they, too, have inherent weaknesses in their study design. In their paper, Saturni et al. [55] compared the advantages and disadvantages of both randomized controlled trials and studies based on real-world data. They note that RCTs limit their enrollment to a select group of eligible subjects that may not represent the population at large. Ad- ditionally, they observe that there are ethical con- siderations when control groups are randomized to no therapy or therapies that are clearly less effec- tive. Very few studies address the outcomes that are achieved in the “real world,” where practice patterns may vary according to availability and no patient is excluded based on their existing comorbidities unless those co-morbidities would be a contraindi- cation to hyperbaric exposure. Our study accepted all identified patients except for those having had recent cancer treatment or those who had radia- tion-induced proctitis in addition to cystitis.

Recommendations

In summary, this paper utilized a large-scale public database of Medicare payments. This study showed both a clinical and economic benefit to post-irradi- ation cystitis patients who were referred for hyper- baric oxygen treatments and who received the rec- ommended forty treatment protocols. There were demonstrable reductions in endoscopy transfusions and decreases in hospitalizations when adjunctive HBO treatment was added to conservative therapies. The use of this novel statistical modeling may con- tribute to valuable comparative clinical effectiveness research. Such methods may complement tradition- al clinical research efforts, such as RCTs, as they help uncover clinical improvements and future research targeted at lower costs. Moore and colleagues have reported the costs of pivotal randomized controlled trials. [56] Based on their investigation, a controlled

154

Feldmeier J, Kirby J, Gelly H, et al.

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

urologic trial with a similar enrollment would cost at least $12 million dollars. This study was conducted at a fraction of the cost of an RCT enrolling this many patients. This study demonstrates a dose-response curve effect for adjunctive HBO2 in patients with post-irradiation cystitis where more complete HBO2 treatment protocols (greater than equal to 40 HBO treatments) have better outcomes and lower Medi- care billings at least during short-term follow-up. See a graphic plot of these savings in Figure 2. The authors recommend that patients be authorized to receive the full forty treatment course of hyperbaric oxygen.

Earlier intervention might reflect even greater clinical and economic impact, although these data reflect the current practice standard of failing con- ventional care prior to initiation of hyperbaric oxy- gen treatments. The authors recommend the utiliza- tion of hyperbaric oxygen earlier in the treatment of radiation cystitis treatment algorithms. In addition, more widespread acceptance of hyperbaric oxygen

as part of the treatment protocol will impact the clinical trajectory of radiation cystitis patients, many of whom are prisoners of their bladders and have a diminished quality of life. Unsuccessful treatment of radiation cystitis is a costly and sometimes lethal complication of pelvic radiation.

ACKNOWLEDGMENTS:
The Project was supported by an unrestricted grant from Healogics, LLC.

The authors also wish to express their thanks to the following individuals for input and editorial review:
• Joan E. DaVanzo, Ph.D., M.S.W.,

Chief Executive Officer Dobson DaVanzo and

Associates, LLC (deceased)
• John A. Gelly, MD, Retired Physician, Marietta, GA • Costantino Balestra Ph.D., Professor of Physiology,

Vrije Universiteit Brussel ( V.U.B.)

VP. Research & Education DAN Europe
• Renee Duncan, BA, UHMS Consulting Copy Editor

n

1. American Cancer Society. Cancer Facts & Figures 2023. Atlanta: American Cancer Society; 2023.

2. Mangar SA, Huddart RA, Parker CC, Dearnaley DP, Khoo VS, Horwich A. Technological advances in radiotherapy for the treatment of localised prostate cancer. Eur J Cancer. 2005 Apr;41(6):908-21. doi: 10.1016/j.ejca.2004.12.028. PMID: 15808957.

3. Lemanska A, Byford RC, Correa A, Cruickshank C, Dear- naley DP, Griffin C, Hall E, de Lusignan S, Faithfull S. Linking CHHiP prostate cancer RCT with GP records: A study proposal to investigate the effect of co-morbidities and medications on long-term symptoms and radiotherapy-related toxicity. Tech Innov Patient Support Radiat Oncol. 2017 Jun 27;2:5-12. doi: 10.1016/j.tipsro.2017.06.001. PMID: 32095558; PMCID: PMC7033766.

4. Bekelman JE, Mitra N, Efstathiou J, Liao K, Sunderland R, Yeboa DN, Armstrong K. Outcomes after intensity-modulated versus conformal radiotherapy in older men with nonmeta- static prostate cancer. Int J Radiat Oncol Biol Phys. 2011 Nov 15;81(4):e325-334. doi: 10.1016/j.ijrobp.2011.02.006. Epub 2011 Apr 16. PMID: 21498008; PMCID: PMC4265571.

5. Ma JL, Hennessey DB, Newell BP, Bolton DM, Lawrent- schuk N. Radiotherapy-related complications presenting to a urology department: a more common problem than pre- viously thought? BJU Int. 2018 May;121 Suppl 3:28-32. doi: 10.1111/bju.14145. Epub 2018 Feb 27. PMID: 29360286.

6. Mohammed N, Kestin L, Ghilezan M, et al. Comparison of acute and late toxicities for three modern high-dose radiation treatment techniques for localized prostate cancer. Int J Radiat Oncol Biol Phys. 2012 Jan 1;82(1):204-12. doi: 10.1016/j.ijrobp.2010.10.009. Epub 2010 Dec 16. PMID: 21167653.

7. Martin SE, Begun EM, Samir E, Azaiza MT, Allegro S, Abdelhady M. Incidence and morbidity of radiation-induced hemorrhagic cystitis in prostate cancer. Urology. 2019 Sep;131:190-195. doi: 10.1016/j.urology.2019.05.034. Epub 2019 Jun 12. PMID: 31201826.

8. Zelefsky MJCH, Hunt M, Yamanda Y, Shippy A, Amold H. Long-term outcome of high dose intensity modulated radi- ation therapy for patients with clinically localised prostate cancer. J Urol 2006; 176: 1415-1419.

9. Linder BJ, Tarrell RF, Boorjian SA. Cystectomy for refrac- tory hemorrhagic cystitis: contemporary etiology, presen- tation and outcomes. J Urol. 2014 Dec;192(6):1687-1692. doi: 10.1016/j.juro.2014.06.030. Epub 2014 Jun 14. PMID: 24936722.

Feldmeier J, Kirby J, Gelly H, et al.

155

REFERENCES

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

10. National Institutes of Health. U.S. Department of Health and Human Services. Common Terminology Criteria for Adverse Events (CTCAE). v5.0. 2017 [cited 2017 Nov 27]. Report no. 2. Available online: https://ctep.cancer.gov/proto- coldevelopment/electronic_applications/ctc.htm

11. Rubin P, Casarrett GW. Clinical Radiation Pathology, Vol I, 38-62, Philadelphia Pa: WB Saunders, 1968.

12. Stone HB, Coleman CN, Anscher MS, McBride WH. Ef- fects of radiation on normal tissue: consequences and mech- anisms. Lancet Oncol. 2003 Sep;4(9):529-36. doi: 10.1016/ s1470-2045(03)01191-4. PMID: 12965273.

13. Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283-288. 14. Delanian S, Lefaix JL. The radiation-induced fibro-

atrophic process: therapeutic perspective via the antioxidant pathway. Radiother Oncol. 2004 Nov;73(2):119-131. doi: 10.1016/j.radonc.2004.08.021. PMID: 15542158.

15. O'Sullivan B, Levin W. Late radiation-related fibrosis: pathogenesis, manifestations, and current management. Semin Radiat Oncol. 2003 Jul;13(3):274-289. doi: 10.1016/ S1053-4296(03)00037-7. PMID: 12903016.

16. Darby IA, Hewitson TD. Hypoxia in tissue repair and fi- brosis. Cell Tissue Res. 2016 Sep;365(3):553-562. doi: 10.1007/ s00441-016-2461-3. Epub 2016 Jul 16. PMID: 27423661.

17. Feldmeier JJ. Hyperbaric oxygen therapy and delayed radiation injuries (soft tissue and bony necrosis): 2012 up- date. Undersea Hyperb Med. 2012 Nov-Dec;39(6):1121-1139. PMID: 23342770.

18. Pascoe C, Duncan C, Lamb BW, et al. Current manage- ment of radiation cystitis: a review and practical guide to clinical management. BJU Int. 2019 Apr;123(4):585-594. doi: 10.1111/bju.14516. Epub 2018 Nov 28. PMID: 30113758.

19. Bevers RF, Bakker DJ, Kurth KH. Hyperbaric oxy- gen treatment for haemorrhagic radiation cystitis. Lan- cet:1995;346:803-805.

20. Neheman A, Nativ O, Moskovitz B, Melamed Y, Stein A. hyperbaric oxygen therapy for radiation-induced haemor- rhagic cystitis. BJU Int. 2005;96:107-109.

21. Corman JM, McClure D, Pritchett R, Kozlowski P, Hamp- son NB. Treatment of radiation induced hemorrhagic cystitis with hyperbaric oxygen. J Urol. 2003;160:2200-2202.

22. Norkool DM, Hampson NB, Gibbons RP, Weissman RM. Hyperbaric oxygen therapy for radiation-induced hemor- rhagic cystitis. J Urol. 1993 Aug;150(2 Pt 1):332-334. doi: 10.1016/s0022-5347(17)35476-9. PMID: 8326555.

23. Cardinal J, Slade A, McFarland M et al. Scoping review and meta-analysis of hyperbaric oxygen therapy for radia- tion-induced hemorrhagic cystitis. Current Urology Reports. 2018.;19:38 published on line.

24. Hampson NB, Holm JR, Wreford-Brown CE, Feldmei- er J. Prospective assessment of outcomes in 411 patients treated with hyperbaric oxygen for chronic radiation tissue injury. Cancer. 2012 Aug 1;118(15):3860-3868. doi: 10.1002/ cncr.26637. Epub 2011 Dec 2. PMID: 22139864

25. Villeirs L, Tailly T, Ost P, et al. Hyperbaric oxygen therapy for radiation cystitis after pelvic radiotherapy: Systematic review of the recent literature. Int J Urol. 2020 Feb;27(2):98-107. doi: 10.1111/iju.14130. Epub 2019 Oct 15. PMID: 31617263.

26. Oscarsson N, Müller B, Rosén A, et al. Radiation-in- duced cystitis treated with hyperbaric oxygen therapy (RICH-ART): a randomised, controlled, phase 2-3 trial. Lancet Oncol. 2019 Nov;20(11):1602-1614. doi: 10.1016/S1470- 2045(19)30494-2. Epub 2019 Sep 16. Erratum in: Lancet Oncol. 2019 Sep 23; PMID: 31537473.

27. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70.1:41-55.

28. Raad H, Cornelius V, Chan S, Williamson E, Cro S. An evaluation of inverse probability weighting using the pro- pensity score for baseline covariate adjustment in smaller population randomised controlled trials with a continuous outcome. BMC Med Res Methodol. 2020 Mar 23;20(1):70. doi: 10.1186/s12874-020-00947-7. PMID: 32293286; PMCID: PMC7092449.

29. Brookhart MA, Schneeweiss S. Preference-based in- strumental variable methods for the estimation of treatment effects: assessing validity and interpreting results. Int J Biostat. 2007;3(1): Article 14. doi: 10.2202/1557-4679.1072. PMID: 19655038; PMCID: PMC2719903.

30. Curtis LH, Hammill BG, Eisenstein EL, Kramer JM, Anstrom KJ. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Med Care. 2007 Oct;45(10 Supl 2):S103-107. doi: 10.1097/MLR.0b013e31806518ac. PMID: 17909367.

31. Rubin, D.B., Using Propensity Scores to Help Design Observational Studies: Application to the Tobacco Litigation, 2001, Health Services & Outcome Research Methodology, 169-188

32. Stuart E. A., Matching Methods for Causal Inference: A Review and a Look Forward, Statistical Science, 2010, 25(1).

33. Austin P.C., An Introduction to Propensity Score Meth- ods for Reducing the Effects of Confounding in Observa- tional Studies, Multivariate Behavioral Research, 2011, 46(3), 399- 424.

34. Boonstra PS, Bondarenko I, Park SK, Vokonas PS, Mukherjee B. Propensity score-based diagnostics for cat- egorical response regression models. Stat Med. 2014 Feb 10;33(3):455-469. doi: 10.1002/sim.5940. Epub 2013 Aug 12. PMID: 23934948; PMCID: PMC3911784.

35. Smart D, Wallington M. A cost-analysis case study of radiation cystitis treatment including hyperbaric oxygen therapy. Diving Hyperb Med. 2012 Jun;42(2):92-7. PMID: 22828818.

156

Feldmeier J, Kirby J, Gelly H, et al.

HYPERBARIC OXYGEN FOR RADIATION CYSTITIS – UHM 2024 VOL 51 NO 2

36. Löndahl M, Katzman P, Nilsson A, Hammarlund C. Hyperbaric oxygen therapy facilitates healing of chron- ic foot ulcers in patients with diabetes. Diabetes Care. 2010 May;33(5):998-1003. doi: 10.2337/dc09-1754. PMID: 20427683; PMCID: PMC2858204.

37. Tai CJ, Lu CK, Lee CY, Lee SS, Yang YH. Real-world evidence of hyperbaric oxygen therapy on cardiovascular outcomes in patients with chronic osteomyelitis. J Infect Public Health. 2023 May;16(5):705-712. doi: 10.1016/j. jiph.2023.03.006. Epub 2023 Mar 7. PMID: 36940497.

38. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis. 1987 40(5): 373-383. doi:10.1016/0021-9681(87)90171-8. PMID 3558716.

39. Mendenhall WM, Henderson RH, Costa JA, et al. Hemor- rhagic radiation cystitis. Am J Clin Oncol. 2015 Jun;38(3):331- 336. doi: 10.1097/COC.0000000000000016. PMID: 24322335.

40. 27. D'Amico MJ, Foss H, Uhr A, Rudnick B, Kloniecke E, Gomella LG. Hemorrhagic cystitis: a review of the literature and treatment options. Can J Urol. 2022 Oct;29(5):11276- 11283. PMID: 36245196.

41. Goucher G, Saad F, Lukka H, Kapoor A. Canadian Urological Association Best Practice Report: Diagnosis and management of radiation-induced hemorrhagic cystitis. Can Urol Assoc J. 2019 Feb;13(2): 15-23.

42. Thompson A, Adamson A, Payne H. Guidelines for the diagnosis, prevention and management of chemical- and radiation-induced cystitis. J Clin Urol. 2014 7(1): 25-35.

43. Vanneste Ben GL, Van Limbergen EJ, Marcelissen TA, et al. Development of a management algorithm for acute and chronic radiation urethritis and cystitis. Urol Int 2022 106(1)63-74.

44. https://www.cms.gov/medicare-coverage-database/ view/ncd.aspx?ncdid=12 accessed 11/26/2023

45. Chong KT, Hampson NB, Corman JM. Early hyperbaric oxygen therapy improves outcome for radiation-induced hemorrhagic cystitis. Urology. 2005 Apr;65(4):649-653. doi: 10.1016/j.urology.2004.10.050. PMID: 15833500.

46. Crall TS, Bishop JA, Guttman D, Kocher M, Bozic
K, Lubowitz JH. Cost-effectiveness analysis of primary arthroscopic stabilization versus nonoperative treat- ment for first-time anterior glenohumeral dislocations. Arthroscopy. 2012 Dec;28(12):1755-1765. doi: 10.1016/j. arthro.2012.05.885. Epub 2012 Oct 5. PMID: 23040837.

47. Cohen JR, Bradley AT, Lieberman JR. Preoperative interventions charges before knee arthroplasty. J Ar- throplasty2016 Dec;31(12):2730-2735.e7.doi: 10.1016/j. arth.2016.05.048. Epub 2016 May 31.

48. Jones PE, Kissenberth MJ, Brooks JM, Thigpen CA, Shanley E, Pill SG. Unanticipated costs associated with interscalene nerve catheters for shoulder surgery. J Shoul- der Elbow Surg. 2023 Jun;32(6S):S118-S122. doi: 10.1016/j. jse.2023.02.010. Epub 2023 Feb 23.

49. Zpancic JA, Richardson DK, O'Brien BJ, Eichenwald
EC, Weinstein MC. Cost-effectiveness analysis of predis- charge monitoring for apnea of prematurity. Pediatrics. 2003 Jan;111(1):146-152. doi: 10.1542/peds.111.1.146. PMID: 12509568.

50. Mark DB, Nelson CL, Anstrom KJ, et al. Cost-effective- ness of defibrillator therapy or amiodarone in chronic stable heart failure: results from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Circulation. 2006 Jul 11;114(2):135- 142. doi: 10.1161/CIRCULATIONAHA.105.581884. Epub 2006 Jul 3. PMID: 16818817.

51. Mills MD, Spanos WJ, Jose BO, Kelly BA, Brill JP. Preparing a cost analysis for the section of medical phys- ics-guidelines and methods. J Appl Clin Med Phys. 2000 Spring;1(2):76-85. doi: 10.1120/jacmp.v1i2.2648. PMID: 11674821; PMCID: PMC5726149.

52. Mesel E, Wirtschafter DD, Ramsey-Klee DM. Eco- nomic analysis of an automated billing system for physi- cians' services. Med Care. 1976 Dec;14(12):1037-1051. doi: 10.1097/00005650-197612000-00007. PMID: 794599.

53. Enestvedt CK, Mayo SC, Diggs BS, et al. Diagnostic lap- aroscopy for patients with potentially resectable pancreatic adenocarcinoma: is it cost-effective in the current era? J Gas- trointest Surg. 2008 Jul;12(7):1177-84. doi: 10.1007/s11605- 008-0514-y. Epub 2008 May 10. PMID: 18470572.

54. Marchioni M, DE Francesco P, Campi R, et al. Current management of radiation cystitis after pelvic radiotherapy: a systematic review. Minerva Urol Nephrol. 2022 Jun;74(3):281- 291. doi: 10.23736/S2724-6051.21.04539-0. Epub 2021 Oct 29. PMID: 34714035.

55. Saturni S, Bellini F, Braido F, et al. Randomized con- trolled trials and real life studies. Approaches and method- ologies: a clinical point of view. Pulm Pharmacol Ther. 2014 Apr;27(2):129-138. doi: 10.1016/j.pupt.2014.01.005. Epub 2014 Jan 24. PMID: 24468677.

56. Moore TJ, Zhang H, Anderson G, Alexander GC. Esti- mated costs of pivotal trials for novel therapeutic agents approved by the US Food and Drug Administration, 2015- 2016. JAMA Intern Med. 2018 Nov 1;178(11):1451-1457. doi: 10.1001/jamainternmed.2018.3931. PMID: 30264133; PMCID: PMC6248200.

n

Feldmeier J, Kirby J, Gelly H, et al.

157

DOI: 10,22462/704