ABSTRACT
Hyperbaric oxygen therapy (HBOT) is an established adjunctive treatment for selected diabetic foot ulcers, although evidence regarding its effectiveness remains heterogeneous. We conducted a retrospective, single‐center, observational study of adult diabetic patients receiving HBOT (2.5 ATA, 100% oxygen) for a lower‐limb wound at the only hyperbaric center in Guadeloupe, a French Caribbean territory, between January 2020 and June 2023. Among 257 patients treated with HBOT, 71 diabetic patients met inclusion criteria (mean age 63.1 ± 12.7 years; 69% male). Peripheral arterial disease and osteitis were highly prevalent, while the proportion of current smokers (7.9%) was notably lower than in previously published HBOT cohorts, a distinctive feature of this population. HBOT was considered a last resort, after failure of prolonged conventional therapy, in 63% of patients, likely contributing to the modest six‐month healing rate of 34% (18/53 evaluable patients) and 19% rate of major amputation occurring during or after HBOT. Patients referred through the multidisciplinary Diabetic Foot Care Unit had fewer major amputations despite older age and greater vascular burden, though this comparison may reflect referral‐related selection bias. No candidate factor, including glycemic control, referral pathway, or delay to HBOT, reached statistical significance in univariable analysis. This descriptive, single‐arm finding cannot establish HBOT efficacy but is consistent with a role for HBOT within coordinated multidisciplinary diabetic foot care, highlighting the importance of earlier referral and optimised patient selection.
Highlights
In our cohort, 63% of patients received HBOT only after failure of conventional therapy, with a median delay of 15 weeks, a delay that likely limits its effectiveness.
Patients refered through the Diabetic Care Unit had a heavier clinical profile (older, peripheral arterial disease) yet a markedly lower rate of major amputation than those refered by surgery (1 vs 7, p = 0.044), suggesting that early identification and care coordination drive outcomes more HBOT itself.
Complete healing at 6 months is only 34% of evaluable patients and 19% major amputations; this study cannot establish HBOT efficacy but supports its earlier integration into a structured, multidisciplinary diabetic foot care pathway.
1. Introduction
Diabetic foot ulcers (DFUs) are a major global health concern, affecting up to 25% of individuals with diabetes during their lifetime and frequently leading to infection, amputation, and increased mortality rates [1, 2]. Lower limb complications represent one of the most common causes of hospitalisation in diabetic patients and constitute a substantial burden on healthcare systems, particularly in regions with limited access to multidisciplinary wound care services [3, 4].
Standard management of DFUs includes infection control, vascular assessment and revascularisation when indicated, metabolic optimisation, pressure offloading, and wound debridement [5]. This standard of care, as defined by IWGDF (International Working Group of Diabetic Foot) guidelines [6] assumes timely access to a multidisciplinary team including vascular, infectious disease, and wound care specialists—access that remains uneven across Caribbean territories compared with metropolitan France, where specialist density is higher and referral pathways are more established. Despite optimal care, a significant proportion of ulcers remain refractory to healing, especially in patients with advanced peripheral arterial disease, osteomyelitis, or delayed referral [6, 7]. Wound severity in this population is commonly staged using the Wagner or University of Texas classification systems, both of which incorporate depth, ischemia, and infection, and are strongly associated with the risk of underlying osteomyelitis at higher grades [8].
Hyperbaric oxygen therapy (HBOT), which involves the intermittent administration of 100% oxygen at pressures greater than atmospheric pressure, has been proposed as an adjunctive treatment for chronic diabetic foot ulcers. It is thought to enhance oxygen delivery to ischemic tissues, stimulate angiogenesis, reduce inflammation, and improve leukocyte's function [9].
Several studies have reported positive outcomes with HBOT in terms of wound healing and reduction in major amputations [10, 11]. However, the evidence remains conflicting, with studies differing in design and offering variable levels of reliability [12, 13]. Also, most recent guidelines of the IWGDF states that HBOT ‘may be considered as adjunctive therapy in individuals with diabetes and a foot ulcer (particularly Wagner grade 3 or higher) that has failed to respond to standard therapy, and is complicated by ischemia or infection’, but recommends against routine use due to limited high certainty evidence [6]. This conditional recommendation underscores the need for further real‐world data to better define the appropriate indications and timing for HBOT.
In Guadeloupe, a French overseas department in the Caribbean, the prevalence of diabetes and its complications is among the highest in France [14]. The University Hospital of Guadeloupe is the only facility in the region offering HBOT. Since its implementation in 2019, the use of HBOT in diabetic wound management has gradually increased. We hypothesise that structural barriers to specialist access in this island setting, rather than a lack of clinical indication, may explain why HBOT is frequently initiated only as a last resort, after failure of prolonged conventional therapy, and the patient profile treated under these conditions remains poorly characterised.
2. Methods
2.1. Study Design
We conducted a retrospective, observational, single‐centre study at the University Hospital of Guadeloupe.
2.2. Ethical Considerations and Patient Consent
This non‐interventional study was approved by the Ethics Committee of the University Hospital of Guadeloupe (Approval No. A123_07/11/2023). All patients provided informed non opposition to the use of their anonymised data for research purposes, in accordance with French regulations. All procedures were conducted in accordance with the Declaration of Helsinki.
2.3. Selection Criteria
All patients aged ≥ 18 with diabetes who received at least one session of HBOT for a lower‐limb wound were eligible for inclusion. Patients were identified through a registry provided by the Medical Informatics Department, and individual medical records were reviewed to confirm eligibility criteria.
Although the initial inclusion period spanned from 1 January 2020 to 30 June 2023. Only the first wound treated with HBOT was analysed for patients who underwent multiple courses of therapy.
2.4. HBOT Characteristic
HBOT was delivered in a multiplace chamber with patients breathing 100% oxygen via mask at 2.5 atm absolute (ATA) for 45 min per session, five days per week. A standard course consisted of 20 sessions, extendable to 40 or more according to clinical response. Time to HBOT was defined as the interval between first wound specialist consultation and HBOT initiation. This variable reflects the referral and treatment‐pathway delay rather than the true duration of the wound, as the exact date of wound onset was not reliably documented in the medical records. HBOT was generally considered after failure of optimised conventional therapy or following an unfavourable surgical outcome.
2.5. Data Collection
Clinical data were extracted from electronic health records, including demographics (age, sex, and body mass index), diabetes characteristics (type, duration, insulin requirement), cardiovascular risk factors (hypertension, smoking), and the presence of microvascular and macrovascular complications. Treatment‐related variables included glycated haemoglobin (HbA1c) levels, ulcer characteristics (size, location, aetiology, and progression), as well as medical therapies and surgical procedures performed. Satisfactory glucose control was defined as achieving the HbA1c target set by the diabetologist, individualised according to the patient's age and comorbidities. Standardised wound severity (Wagner or University of Texas grade) and infection severity (IDSA grade) were not systematically recorded in this retrospective cohort.
2.6. Primary Outcomes Analysis
The primary outcome was complete wound healing at 6 months, defined as complete epithelialisation without need for additional surgical intervention. In patients with multiple wounds, healing required complete closure of all lesions; new ulcers occurring during treatment were not counted against healing. Surgical procedures performed during the index hospitalisation were considered part of initial management and were not classified as HBOT failure.
2.7. Secondary Outcomes Analysis
The principal secondary outcome was major amputation‐free status at 6 months, defined as the absence of any amputation above the ankle during the observation period.
2.8. Statistical Analysis
Continuous variables were analysed using Welch's t‐test or Mann–Whitney U test, and categorical variables using Chi‐squared or Fisher's exact test. Patients referred by the Diabetic Foot Care Unit were compared with those referred by surgical departments across pre‐specified variables (age, delay to HBOT, peripheral arterial disease, osteitis, glycemic control, wound location, completed HBOT sessions, and major amputation), using the tests described above. Univariable logistic regression was used to estimate associations between complete healing and pre‐specified candidate factors (age, peripheral arterial disease, osteitis, glycemic control, and delay to HBOT initiation). Given the limited number of healing events (n = 18), multivariable adjustment was restricted to a parsimonious model including the two covariates with the most complete data (age and delay to HBOT); a fully adjusted model including all candidate confounders was not performed owing to an insufficient events‐per‐variable ratio. Normality of continuous variables was assessed using the Shapiro–Wilk test to guide the choice between Welch's t‐test and the Mann–Whitney U test; all statistical analyses were performed using R software, version 4.3 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was set at p < 0.05.
The authors used Claude for proofreading and consistency review of the manuscript and for appropriate medical English usage. All content proposed by the AI was verified before integration into the manuscript.
3. Results
A total of 257 patients underwent hyperbaric oxygen therapy at the University Hospital of Guadeloupe. Among them, 71 diabetic patients received at least one course of HBOT for a lower limb wound and were included in the present study.
The remaining 186 patients were excluded for the following reasons (Figure 1).
FIGURE 1.
Flowchart showing the selection process of diabetic patients treated with HBOT at the University Hospital of Guadeloupe (n = 257).
3.1. Population Clinical and Demographic Characteristics
The mean age was 63.1 ± 12.7 years, and 69% were male (Table 1).
TABLE 1.
Population characteristics (N = 71).
| Variable | Value |
|---|---|
| Age (years), mean ± SD | 63.1 ± 12.7 |
| Sex (male), n (%) | 49 (69) |
| BMI (kg/m2) (n = 29), mean ± SD | 28.0 ± 4.4 |
| Smoking status (n = 38), n (%) | |
| Smoker | 3 (7.9) |
| Ex‐smoker | 5 (13) |
| Non‐smoker | 30 (79) |
| Diabetes‐related characteristics | |
| Type 2 diabetes (n = 66), n (%) | 64 (97) |
| Diabetes duration (years) (n = 40), median (IQR) | 15.5 (10.0–25.5) |
| Insulin therapy (n = 65), n (%) | 41 (63) |
| HbA1c at baseline (%) (n = 36), mean ± SD | 8.2 ± 2.3 |
| Macrovascular complications (n = 56), n (%) | |
| Peripheral artery disease (PAD) | 40 (71) |
| Stroke or carotid stenosis | 37 (71) |
| Coronary artery disease | 7 (16) |
| Microvascular complications (n = 49), n (%) | |
| Nephropathy | 48 (98) |
| Neuropathy | 22 (44) |
| Retinopathy | 45 (92) |
| Hypertension (n = 56), n (%) | 45 (80) |
| Wound location, n (%) | |
| Foot | 58 (82) |
| Leg | 13 (18) |
| Wound type, n (%) | |
| Post‐surgical wound | 34 (50) |
| Diabetic foot ulcer (DFU) | 13 (19) |
| Necrotic wound | 8 (12) |
| Venous ulcer | 5 (7.4) |
| Other types a | 8 (12) |
| Osteitis (n = 62), n (%) | 53 (85) |
| Osteitis in foot wounds (n = 56), n (%) | 52 (92.9) |
Other wound types include mixed ulcers (n = 3), calciphylaxis (n = 3), ulcerated vasculitis (n = 1), and erysipelas (n = 1).
The diabetic patients referred were predominantly overweight, with a mean body mass index (BMI) of 28.0 ± 4.4 kg/m2 (available for 29/71 patients, 41%), with 41% classified as obese. Hypertension was observed in 80% of patients.
Most patients had type 2 diabetes (97%), and 63.1% of those with available data (n = 65) were undergoing insulin therapy. The majority of patients had poorly controlled diabetes (61%).
Peripheral artery disease (PAD) and stroke or carotid stenosis were each reported in 71% of patients with macrovascular complication (n = 56), while 98% of patients with microvascular complication (n = 49) had nephropathy. Neuropathy (44%) and retinopathy (92%) were also frequent.
Among the 38 patients with known smoking status, 79% were non‐smokers and 13% were ex‐smokers, all of whom had ceased smoking for more than 2 years prior to treatment.
Wound localisation was predominantly on the foot (82%). The most frequent aetiologies were post‐surgical wounds (50%) and diabetic foot ulcers (19%). Necrotic lesions (12%) and venous ulcers (7.4%) were also observed, along with less common causes such as calciphylaxis, vasculitis, and erysipelas. Notably, 11.8% of cases were classified as ‘other wound types’, encompassing burns, pressure injuries and traumatic wounds. For most patients, an investigation for the presence of osteitis was performed; 85% were found to have osteitis, including 92.9% of foot‐located wounds with known osteitis status (52/56, renamed here from ‘diabetic foot ulcers’ to avoid ambiguity with the wound‐type classification below).
3.2. Evaluation of Patient Care Pathways
Patients completed a median of 36.0 HBOT sessions (IQR 20.0–51.5), as session count was not normally distributed (Table 2). The mean frequency of treatment was 4.5 sessions per week. Patients were mainly referred by orthopaedic surgery (37%) and the Diabetic Foot Care Unit (35%). The remaining 4% of patients were referred from the nephrology department, while 24% were referred from outside the University Hospital, either by private general practitioners or by wound care consultations in private clinics.
TABLE 2.
Characteristics of HBOT and clinical outcomes in diabetic patients.
| Variable | Value |
|---|---|
| Referral department, n (%) | |
| Surgery (mainly orthopaedic) | 26 (37) |
| Diabetic foot care unit | 25 (35) |
| External referrals | 17 (24) |
| Nephrology | 3 (4.2) |
| Use as last resort (n = 51), n (%) | 41 (63) |
| Completed HBOT sessions, median (Q1–Q3) | 36.0 (20.0–51.5) |
| Mean treatment frequency (sessions/week) | 4.5 |
| Reported side effects, n (%) | 4 (5.8) |
| Satisfactory glucose control (n = 53), n (%) | 28 (39) |
| Limb angioplasty required (n = 54), n (%) | 19 (35) |
| Minor amputation (n = 61), n (%) | |
| None | 28 (46) |
| Before HBOT | 24 (39) |
| During HBOT | 8 (13) |
| After HBOT | 2 (3) |
| Major amputation (n = 59), n (%) | |
| None | 47 (79.6) |
| Before HBOT | 1 (1.7) |
| During HBOT | 3 (5) |
| After HBOT | 8 (13) |
Abbreviations: DFCU: diabetic foot care unit; HBOT: hyperbaric oxygen therapy.
HBOT was considered a last resort in 63% of patients with available data (n = 51). Side effects were rare (5.8%).
Limb angioplasty was necessary in 35% of patients (n = 19). For most patients, vascular surgeons at the institution considered revascularisation unnecessary or too delayed.
Among the 61 patients assessed for minor amputation, 46% did not require amputation, while 39% underwent amputation before HBOT, 13% during therapy, and 3% after completion. Regarding major amputations (n = 59), 79.6% of patients did not undergo amputation, whereas 1.7% had a major amputation before HBOT, 5% during treatment, and 13% after HBOT.
Overall, major amputation occurred during or after HBOT in 11/59 patients (19%); one additional patient (1.7%) had undergone major amputation prior to HBOT initiation and was excluded from this outcome rate.
3.3. Comparative Analysis of Hyperbaric Oxygen Therapy Use by Medical (DFCU) and Surgical Teams
Patients referred from the DFCU were older (68.1 ± 12.6 vs. 55.5 ± 11.8 years, p < 0.001) and had a longer median delay to HBOT initiation (15.0 vs. 3.8 weeks, p < 0.01). These patients also had higher rates of PAD (83% vs. 59%) and lower rates of osteitis (75% vs. 100%, p < 0.01).
Notably, satisfactory glucose control was more frequent in the DFCU group (65% vs. 19%, p < 0.001) (Table 3).
TABLE 3.
Comparison of patients referred by DFCU versus surgery.
| Variable | DFCU (n = 25) | Surgery (n = 26) | p |
|---|---|---|---|
| Age (years), mean ± SD | 68.1 ± 12.6 | 55.5 ± 11.8 | < 0.001 a |
| Time to HBOT (weeks), median (Q1–Q3) | 15.0 (8.43–39.3) | 3.79 (1.89–7.82) | < 0.01 b |
| PAD, n (%) | 19 (83) | 10 (59) | 0.15 c |
| Osteitis, n (%) | 18 (75) | 26 (100) | < 0.01 c |
| Wound located on foot, n (%) | 20 (77) | 26 (100) | 0.023 c |
| Completed HBOT sessions, median (Q1–Q3) | 36.0 (20.0–41.8) | 27.0 (14.8–49.0) | 0.48 b |
| Satisfactory glucose control, n (%) | 17 (65) | 5 (19) | < 0.001 d |
| Major amputation, n (%) | |||
| During HBOT | 0 (0) | 2 (8.7) | |
| After HBOT | 1 (4.2) | 5 (22) | 0.044 c |
Abbreviations: DFCU: diabetic foot care unit; HBOT: hyperbaric oxygen therapy; PAD: peripheral artery disease.
Welch's test.
Mann–Whitney U test.
Fisher's exact test.
Chi‐squared test.
Major amputations occurred exclusively in the surgical group during HBOT (8.7%) and after HBOT (22%), whereas only one patient (4.2%) in the DFCU group underwent a major amputation after HBOT. This difference was statistically significant (p = 0.044).
3.4. Factors Favouring Wound Healing at 6 Months
Among the 71 patients included in the study, 53 had available follow‐up data at 6 months; the remaining 18 patients (25%) were lost to follow‐up and were compared with the 53 evaluable patients (Table 4). Baseline characteristics were similar between groups for age, sex and peripheral arterial disease. Patients lost to follow‐up were significantly less often referred through the DFCU (17% vs. 44%, p = 0.049), suggesting attrition was not entirely random and may have biased the reported healing rate toward outcomes from the better‐followed DFCU pathway.
TABLE 4.
Baseline characteristics of patients lost to follow‐up versus patients with available 6‐month outcome data.
| Variable | Lost to follow‐up (n = 18) | Evaluable at 6 months (n = 53) | p |
|---|---|---|---|
| Age (years), mean ± SD | 66.0 ± 12.9 | 62.8 ± 12.6 | 0.38 a |
| Male sex, n (%) | 14 (78) | 35 (66) | 0.40 b |
| Peripheral arterial disease, n (%) | 8/10 (80) | 29/39 (74) | 1.00 b |
| Osteitis, n (%) | 15/15 (100) | 38/47 (81) | 0.098 b |
| Last‐resort HBOT use, n (%) | 6/14 (43) | 36/51 (71) | 0.066 b |
| Referred through the DFCU, n (%) | 3/18 (17) | 23/52 (44) | 0.049 b |
Note: Denominators for peripheral arterial disease, osteitis, and last‐resort use vary owing to incomplete data availability in the source records; percentages are calculated on available data for each variable. Baseline characteristics were similar between patients lost to follow‐up and evaluable patients for age, sex, and peripheral arterial disease. Patients lost to follow‐up were significantly less often referred through the DFCU (17% vs. 44%, p = 0.049) and showed non‐significant trends toward a higher prevalence of osteitis and a lower rate of last‐resort designation, suggesting that attrition was not entirely random and may have been associated with referral pathway.
Abbreviation: DFCU: diabetic foot care unit.
Welch's t‐test.
Fisher's exact test.
Complete wound healing at 6 months was observed in 18 patients (34%). Factors favouring wound closure though not statistically significant included absence of last resort designation, care by the DFCU, and satisfactory glucose control (Table 5). Among rare etiologies, both analysable cases of calciphylaxis achieved healing, while mixed ulcers and vasculitis did not.
TABLE 5.
Comparison of clinical characteristics between healed and non‐healed patients.
| Variable | Healed (n = 18) | Not healed (n = 35) | p |
|---|---|---|---|
| Last resort use, n (%) | 10 (56) | 26 (79) | 0.082 a |
| Foot wounds only, n (%) | 10 (62) | 19 (73) | 0.51 b |
| Satisfactory glucose control, n (%) | 11 (61) | 15 (43) | 0.21 a |
| Foot wounds with glucose control | 10 (62) | 10 (37) | 0.11 a |
| DFCU care, n (%) | 11 (61) | 15 (43) | 0.21 a |
| Foot wounds with DFCU care | 11 (69) | 11 (41) | 0.076 a |
| Wound aetiology, n (%) | 0.19 b | ||
| Post‐surgical | 12 (67) | 12 (36) | |
| DFU | 3 (17) | 8 (24) | |
| Necrotic | 1 (5.6) | 5 (15) | |
| Venous ulcer | 0 (0) | 5 (15) | |
| PAD, n (%) | 10 (67) | 19 (70) | 1.00 b |
| Smoking status, n (%) | 0.32 b | ||
| Smoker | 0 (0) | 3 (14) | |
| Ex‐smoker | 1 (7.1) | 3 (14) | |
| Non‐smoker | 13 (93) | 15 (71) | |
| Leg angioplasty, n (%) | 0.31 b | ||
| Not needed | 12 (80) | 16 (55) | |
| Performed | 2 (13) | 5 (17) | |
| Needed but not performed | 0 (0) | 5 (17) | |
| Not realizable | 0 (0) | 2 (6.9) | |
| Performed after HBOT | 1 (6.7) | 1 (3.4) |
Note: Denominators vary across variables owing to incomplete data availability, as specified for each table.
Abbreviations: DFU: diabetic foot ulcer; HBOT: hyperbaric oxygen therapy; PAD: peripheral artery disease.
Chi‐squared test.
Fisher's exact test.
HBOT was more frequently used as a last resort among non‐healed patients (79%) compared with healed patients (56%), although this difference did not reach statistical significance (p = 0.082).
Foot wounds were common in both groups (62% vs. 73%, p = 0.51). Adequate glucose control was observed in 61% of healed patients compared with 43% of non‐healed patients (p = 0.21).
Similarly, patients followed by the DFCU tended to have better healing outcomes (61% vs. 43%, p = 0.21), with a trend toward significance when combining DFCU care and foot wound management (p = 0.076).
Regarding wound aetiology, post‐surgical wounds were more frequent among healed patients (67%), whereas diabetic foot ulcers (24%), necrotic wounds (15%), and venous ulcers (15%) were more common in the non‐healed group (p = 0.19). Peripheral arterial disease (PAD) was equally prevalent in both groups (67% vs. 70%, p = 1.00).
No significant differences were observed regarding smoking status (p = 0.32) or leg angioplasty (p = 0.31). However, angioplasty was more often deemed unnecessary among healed patients (80%), whereas a higher proportion of non‐healed patients required but did not undergo revascularisation (17%).
4. Discussion
This study revealed several distinctive features in the profile of patients receiving hyperbaric oxygen therapy. Notably, the mean age in our cohort was 63 years, higher than the 56 years reported in a similar Moroccan study conducted in 2018, which also observed a predominance of male patients [15]. Our cohort provides detailed clinical data, including smoking status and microvascular complications, offering a comprehensive patient profile.
A particularly noteworthy finding in our population is the significantly lower proportion of smokers compared to other published cohorts. This is clinically relevant, as chronic smoking (especially > 10 pack‐years) is known to impair HBOT efficacy, requiring more sessions and increasing therapeutic costs [16]. Few patients were eligible for lower limb revascularisation, and HBOT therefore represented the last available option to deliver oxygen to the wound. This aligns with existing evidence suggesting that HBOT may improve tissue perfusion, mobilise endothelial progenitor cells, and reduce the risk of major amputations incases of chronic ischemia [17, 18, 19]. Our data also highlight a high prevalence of microvascular complications, particularly diabetic nephropathy and retinopathy. Interestingly, the prevalence of diabetic neuropathy was lower than expected, despite its established role in the pathogenesis of diabetic foot ulcers [20]. This finding may suggest a referral bias toward ischemic or neuro‐ischemic ulcers, which are more likely to be considered for HBOT [21].
Alternatively, underdiagnosis of neuropathy cannot be ruled out due to the lack of systematic neurological evaluation or electrophysiological testing [22].
In our study, post‐surgical wounds represented the most frequent indication for HBOT, accounting for 50% of treated cases. This high prevalence may reflect the increasing use of HBOT as an adjuvant strategy in complex surgical site infections or dehisced incisions, especially in patients with comorbidities that impair wound healing, such as diabetes or peripheral arterial disease [23, 24]. Non chirurgical diabetic foot ulcers comprised 19.1% of the cases, an expected proportion given that DFUs remain one of the most common and challenging indications for HBOT, particularly in neuroischemic forms [25]. Necrotic wounds and venous ulcers were less represented, at 11.8% and 7.4%, respectively, consistent with data suggesting variable referral patterns and access to HBOT in patients with chronic wounds of vascular origin [26].
Interestingly, 11.8% of the treated wounds fell under the category of ‘other types’, which included burns, pressure injuries, and traumatic wounds. This diversity highlights the broadening indications of HBOT beyond diabetic ulcers, particularly for acute wounds at high risk of infection, ischemia, or delayed healing [27]. The distribution of wound types in our population reinforces the evolving role of HBOT in multidisciplinary wound care pathways, though prospective data are still needed to better stratify which wound etiologies derive the greatest benefit from hyperbaric treatment.
We observed a 34% complete wound healing rate at 6 months. While these figures may appear modest, they are consistent with prior studies reporting healing rates between 30% and 40% in similar high‐risk populations [28, 29]. Patients followed by a DFCU and those with satisfactory glycemic control tended to achieve better outcomes, although statistical significance was not always attained. The association between tight glycemic control and improved wound healing is well established in the literature [30]. The lack of significance in our study likely reflects the small proportion of patients who achieved complete healing, as well as the clinical severity of those selected for HBOT, often representing a last‐resort intervention. A key finding was the distinct difference in clinical profiles and outcomes between patients referred from surgical departments and those referred by the DFCU. Interestingly, patients from the DFCU were older and more likely to present with peripheral artery disease, yet experienced better glycemic control and significantly fewer amputations. This paradox underscores the critical role of coordinated multidisciplinary care in managing complex diabetic wounds. Multiple studies have shown that integrated care models involving early vascular assessment, podiatric follow‐up, infectious disease management, and endocrinology input are associated with reduced complication rates. Meloni et al., in a prospective cohort, demonstrated that patients managed within a structured diabetic foot unit had lower rates of major amputations and ulcer recurrence compared to those receiving standard care [31]. Similarly, early multidisciplinary referral has been shown to reduce healing time and healthcare costs [32].
These observations, along with our findings, suggest that HBOT is most effective when implemented as part of a coordinated therapeutic strategy, rather than as an isolated or delayed intervention. Proactive monitoring and early metabolic and vascular optimisation may enhance wound stabilisation and treatment adherence, reinforcing current international guidelines such as those from the IWGDF [6]. Notably, in our cohort, HBOT was frequently used as a last‐resort intervention, with a median initiation delay of 15 weeks, particularly among patients referred by DFCU. This finding underscores a persistent gap between evidence‐based recommendations and real‐world clinical practice. According to the IWGDF and other expert consensus statements, HBOT should be considered in non‐healing, ischemic diabetic foot ulcers, especially in patients who fail to improve with standard wound care, ideally before the onset of irreversible tissue loss or critical infection [6]. Recent studies support this recommendation.
Margolis et al. [32] similarly note in their discussion that earlier initiation of HBOT relative to ulcer non‐response may be associated with better outcomes, even though their own cohort data did not demonstrate an overall benefit of HBOT. Similarly, Löndahl et al. demonstrated that patients receiving HBOT within the first 6–8 weeks of documented ischemic non‐healing ulcers had significantly higher healing rates and limb preservation compared to those treated [33]. In our study, the median delay of 15 weeks suggests that HBOT was often initiated only after the failure of all other therapeutic interventions, a strategy that may compromise its overall efficacy. This delay likely reflects multiple factors, including the absence of structured early referral pathways, the limited availability of tools for tissue oxygen assessment, and a degree of hesitation among clinicians to escalate to HBOT. Moreover, several patient‐related constraints may also contribute to delayed initiation or suboptimal adherence. These include the need for frequent and time‐consuming treatment sessions (typically 5 days per week over several weeks), as well as the physical tolerance required to withstand repeated pressure changes within the hyperbaric chamber. Such logistical and physiological burdens may further limit the timely and consistent use of HBOT in routine clinical practice. Moreover, the literature emphasises the importance of early tissue oxygenation assessment to identify patients who may benefit from HBOT before wound deterioration progresses. One study highlighted that TcPO2‐guided decision‐making improved both selection for HBOT and outcomes, with healing rates above 60% in appropriately selected patients [34]. In light of these findings, our results support the need for structured referral criteria and early evaluation protocols, including vascular and oxygenation status assessment, to guide timely HBOT implementation. Delays in initiating therapy may partially explain the suboptimal healing rates and the non‐negligible incidence of amputations observed in our cohort, despite the use of advanced wound care modalities.
Beyond timing, the HBOT protocol itself may have influenced outcomes. Sessions in our unit lasted 45 min at 2.5 ATA, at the shorter end of session durations commonly reported in the literature, where protocols of 90–120 min are frequently used. Whether shorter, more frequent sessions provide comparable tissue oxygenation and clinical benefit to longer sessions remains uncertain, and this difference in dosing could plausibly have contributed to the modest healing rate observed. Future studies directly comparing session duration and pressure protocols in similar populations would help clarify optimal dosing.
Beyond delaying referral, the insular geography of Guadeloupe may also have affected treatment adherence once HBOT was initiated: with a single hyperbaric facility serving the entire territory and a median of 36 sessions required per patient, repeated travel, transportation costs, and the economic impact of extended sick leave may have constituted additional barriers to completing the full treatment course, independent of any effect of HBOT itself. These logistical factors were not systematically captured in this retrospective cohort but represent a plausible contributor to incomplete treatment courses and warrant attention in future prospective studies.
Comparable challenges—delayed referral, advanced disease at presentation, and limited access to timely specialist care—have been reported in other high‐risk, resource‐constrained settings. Rather than reflecting population‐specific biological factors, these parallels point to structural determinants of delayed diabetic foot care: geographic isolation, limited specialist density, and socioeconomic barriers to access, which our findings suggest could be addressed through earlier referral pathways.
4.1. Study Limitations
Our study has several limitations. The retrospective, single‐arm design without a control group precludes causal inference: no conclusion regarding the added benefit of HBOT over conventional wound care can be drawn, and reported associations should be interpreted as exploratory and hypothesis‐generating. The small sample size limited statistical power; multivariable adjustment was restricted to a parsimonious two‐covariate model (age and delay to HBOT) given the limited number of healing events (n = 18), and a fully adjusted model was not statistically justifiable. Selection bias is likely, as patients referred for HBOT may have had more severe or treatment‐resistant wounds. Six‐month outcome data were missing for 18 patients (25%); those lost to follow‐up were significantly less often referred through the DFCU, suggesting that attrition was not random and may have inflated the reported healing rate. Standardised wound severity (Wagner or University of Texas grade) and infection severity (IDSA grade) were not systematically recorded, precluding comparison with studies using these classifications—a meaningful gap, as HBOT outcomes have been shown to vary significantly by Wagner grade [35]. Transcutaneous oxygen pressure was available for only 8/71 patients, reflecting selective rather than random use, and was therefore not analysed quantitatively. Neuropathy status was ascertained from routine clinical documentation rather than a systematic, standardised neurological examination, and may therefore be underestimated relative to its true prevalence. Detailed historical data on wound duration and prior management were not consistently available, particularly for the 24% of patients referred from outside the University Hospital, limiting our ability to characterise their full pre‐HBOT clinical trajectory.
5. Conclusion
In this Caribbean real‐world cohort, complete healing was observed in about one‐third of patients with advanced diabetic lower‐limb wounds, comparable to high‐risk populations reported elsewhere, but limited by delayed HBOT initiation and severe baseline disease. As a descriptive, single‐arm study, these findings cannot establish HBOT efficacy; they are nonetheless consistent with a role for HBOT as an adjunct within coordinated, multidisciplinary diabetic foot care, highlighting the importance of earlier referral and objective, oxygenation‐based patient selection. Prospective studies incorporating standardised severity grading and a comparator arm are needed to define the added benefit of HBOT in this setting.
Ethics Statement
This non‐interventional study was approved by the Ethics Committee of the University Hospital of Guadeloupe (Approval No. A123_07/11/2023). All patients provided informed non‐opposition to the use of their anonymised data for research purposes, in accordance with French regulations. All procedures were conducted in accordance with the Declaration of Helsinki.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgements
The authors thanks Professor Ariane SULTAN for kindly reviewing the manuscript and providing helpful comments.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Armstrong D. G., Boulton A. J. M., and Bus S. A., “Diabetic Foot Ulcers and Their Recurrence,” New England Journal of Medicine 376, no. 24 (2017): 2367–2375, 10.1056/NEJMra1615439. [DOI] [PubMed] [Google Scholar]
- 2. Zhang P., Lu J., Jing Y., Tang S., Zhu D., and Bi Y., “Global Epidemiology of Diabetic Foot Ulceration: A Systematic Review and Meta‐Analysis,” Annals of Medicine 49, no. 2 (2017): 106–116, 10.1080/07853890.2016.1231932. [DOI] [PubMed] [Google Scholar]
- 3. Prompers L., Huijberts M., Schaper N., et al., “Resource Utilisation and Costs Associated With the Treatment of Diabetic Foot Ulcers. Prospective Data From the Eurodiale Study,” Diabetologia 51, no. 10 (2008): 1826–1834, 10.1007/s00125-008-1089-6. [DOI] [PubMed] [Google Scholar]
- 4. Lazzarini P. A., Cramb S. M., Golledge J., Morton J. I., Magliano D. J., and van Netten J. J., “Global Trends in the Incidence of Hospital Admissions for Diabetes‐Related Foot Disease and Amputations: A Review of National Rates in the 21st Century,” Diabetologia 66, no. 2 (2023): 267–287, 10.1007/s00125-022-05845-9. [DOI] [PubMed] [Google Scholar]
- 5. Schaper N. C., van Netten J. J., Apelqvist J., et al., “Practical Guidelines on the Prevention and Management of Diabetic Foot Disease (IWGDF 2019 Update),” Diabetes/Metabolism Research and Reviews 36, no. Suppl 1 (2020): e3266, 10.1002/dmrr.3266. [DOI] [PubMed] [Google Scholar]
- 6. Game F. and Jeffcoate W., “Guidelines on Diabetic Foot Care 2023,” International Working Group on the Diabetic Foot (IWGDF) [Internet] (2023), https://iwgdfguidelines.org/guidelines/.
- 7. Yang L., Rong G. C., and Wu Q. N., “Diabetic Foot Ulcer: Challenges and Future,” World Journal of Diabetes 13, no. 12 (2022): 1014–1034, 10.4239/wjd.v13.i12.1014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Lavery L. A., Armstrong D. G., Murdoch D. P., Peters E. J. G., and Lipsky B. A., “Validation of the Infectious Diseases Society of America's Diabetic Foot Infection Classification System,” Clinical Infectious Diseases 44, no. 4 (2007): 562–565, 10.1086/511036. [DOI] [PubMed] [Google Scholar]
- 9. Goldman R. J., “Hyperbaric Oxygen Therapy for Wound Healing and Limb Salvage: A Systematic Review,” PM & R: The Journal of Injury, Function, and Rehabilitation 1, no. 5 (2009): 471–489, 10.1016/j.pmrj.2009.03.012. [DOI] [PubMed] [Google Scholar]
- 10. Vinkel J., Holm N. F. R., Jakobsen J. C., and Hyldegaard O., “Effects of Adding Adjunctive Hyperbaric Oxygen Therapy to Standard Wound Care for Diabetic Foot Ulcers: A Protocol for a Systematic Review With Meta‐Analysis and Trial Sequential Analysis,” BMJ Open 10, no. 6 (2020): e031708, 10.1136/bmjopen-2019-031708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wadee A. N., Aref M. H. F., Nassar A. A., Aboughaleb I. H., and Fahmy S. M., “The Influence of Low‐ Intensity Laser Irradiation Versus Hyperbaric Oxygen Therapy on Transcutaneous Oxygen Tension in Chronic Diabetic Foot Ulcers: A Controlled Randomized Trial,” Journal of Diabetes & Metabolic Disorders 20, no. 2 (2021): 1489–1497, 10.1007/s40200-021-00891-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Zhang Z., Zhang W., Xu Y., and Liu D., “Efficacy of Hyperbaric Oxygen Therapy for Diabetic Foot Ulcers: An Updated Systematic Review and Meta‐Analysis,” Asian Journal of Surgery 45, no. 1 (2022): 68–78, 10.1016/j.asjsur.2021.07.047. [DOI] [PubMed] [Google Scholar]
- 13. Fedorko L., Bowen J. M., Jones W., et al., “Hyperbaric Oxygen Therapy Does Not Reduce Indications for Amputation in Patients With Diabetes With Nonhealing Ulcers of the Lower Limb: A Prospective, Double‐Blind, Randomized Controlled Clinical Trial,” Diabetes Care 39, no. 3 (2016): 392–399, 10.2337/dc15-2001. [DOI] [PubMed] [Google Scholar]
- 14. Santé publique France , Surveillance du diabète en Guadeloupe: données 2022 (Santé publique France [Internet], 2022), https://www.santepubliquefrance.fr/. [Google Scholar]
- 15. Jira M., El Omri N., Sekkach Y., Mekouar F., and Amezyane T., “Hyperbaric Oxygen Therapy in the Treatment of Diabetic Foot: Experience in the Management of 80 Cases at a Department of Internal Medicine,” Pan African Medical Journal 30 (2018): 100, 10.11604/pamj.2018.30.100.14826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Otto G. H., Buyukcakir C., and Fife C. E., “Effects of Smoking on Cost and Duration of Hyperbaric Oxygen Therapy for Diabetic Patients With Non‐Healing Wounds,” Undersea & Hyperbaric Medicine 27, no. 2 (2000): 83–89. [PubMed] [Google Scholar]
- 17. Lin P. Y., Sung P. H., Chung S. Y., et al., “Hyperbaric Oxygen Therapy Enhanced Circulating Levels of Endothelial Progenitor Cells and Angiogenesis Biomarkers, Blood Flow, in Ischemic Areas in Patients With Peripheral Arterial Occlusive Disease,” Journal of Clinical Medicine 7, no. 12 (2018): 548, 10.3390/jcm7120548. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Begić A. and Dilić M., “Evaluation of the Occlusive Arterial Disease and Diabetic Angiopathy Treatment Effects by Hyperbaric Oxygenation,” Medical Archives 73, no. 4 (2019): 244–248, 10.5455/medarh.2019.73.244-248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Batinac T., Batičić L., Kršek A., et al., “Endothelial Dysfunction and Cardiovascular Disease: Hyperbaric Oxygen Therapy as an Emerging Therapeutic Modality?,” Journal of Cardiovascular Development and Disease 11, no. 12 (2024): 408, 10.3390/jcdd11120408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Boulton A. J. M., Armstrong D. G., Albert S. F., et al., “Comprehensive Foot Examination and Risk Assessment,” Diabetes Care 31, no. 8 (2008): 1679–1685, 10.2337/dc08-9021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Faglia E., Favales F., Aldeghi A., et al., “Change in Major Amputation Rate in a Center Dedicated to Diabetic Foot Care During the 1980s: Prognostic Determinants for Major Amputation,” Journal of Diabetes and its Complications 12, no. 2 (1998): 96–102, 10.1016/s1056-8727(97)98004-1. [DOI] [PubMed] [Google Scholar]
- 22. Tesfaye S., Boulton A. J. M., Dyck P. J., et al., “Diabetic Neuropathies: Update on Definitions, Diagnostic Criteria, Estimation of Severity, and Treatments,” Diabetes Care 33, no. 10 (2010): 2285–2293, 10.2337/dc10-1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Zhou D., Fu D., Yan L., and Xie L., “The Role of Hyperbaric Oxygen Therapy in the Treatment of Surgical Site Infections: A Narrative Review,” Medicina (Kaunas, Lithuania) 59, no. 4 (2023): 762, 10.3390/medicina59040762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bakker D. J., “Hyperbaric Oxygen Therapy and the Diabetic Foot,” Diabetes/Metabolism Research and Reviews 16, no. Suppl 1 (2000): S55–S58, 10.1002/1520-7560(200009/10)16:1+<::aid-dmrr132>3.0.co;2-t. [DOI] [PubMed] [Google Scholar]
- 25. Stoekenbroek R. M., Santema T. B., Legemate D. A., Ubbink D. T., van den Brink A., and Koelemay M. J. W., “Hyperbaric Oxygen for the Treatment of Diabetic Foot Ulcers: A Systematic Review,” European Journal of Vascular and Endovascular Surgery 47, no. 6 (2014): 647–655, 10.1016/j.ejvs.2014.03.005. [DOI] [PubMed] [Google Scholar]
- 26. Kranke P., Bennett M., Roeckl‐Wiedmann I., and Debus S., “Hyperbaric Oxygen Therapy for Chronic Wounds,” Cochrane Database of Systematic Reviews 2 (2004): CD004123, 10.1002/14651858.CD004123.pub2. [DOI] [PubMed] [Google Scholar]
- 27. Singh D., Mohanty C., Verma R., and Kumar S., “Challenges in the Administration of Hyperbaric Oxygen Therapy (HBOT) for Complicated Cases in a Tertiary Care Setting,” Diving and Hyperbaric Medicine 55, no. 2 (2025): 211–212, 10.28920/dhm55.2.211-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lalieu R. C., Mulder W., Raap R. D. B., et al., “Hyperbaric Oxygen Treatment for University of Texas Grade 3 Diabetic Foot Ulcers: A Retrospective Cohort Study,” Journal of Wound Care 30, no. 9 (2021): 722–728, 10.12968/jowc.2021.30.9.722. [DOI] [PubMed] [Google Scholar]
- 29. Salama S. E., Eldeeb A. E., Elbarbary A. H., and Abdelghany S. E., “Adjuvant Hyperbaric Oxygen Therapy Enhances Healing of Nonischemic Diabetic Foot Ulcers Compared With Standard Wound Care Alone,” International Journal of Lower Extremity Wounds 18, no. 1 (2019): 75–80, 10.1177/1534734619829939. [DOI] [PubMed] [Google Scholar]
- 30. Dutta A., Bhansali A., and Rastogi A., “Early and Intensive Glycemic Control for Diabetic Foot Ulcer Healing: A Prospective Observational Nested Cohort Study,” International Journal of Lower Extremity Wounds 22, no. 3 (2023): 578–587, 10.1177/15347346211033458. [DOI] [PubMed] [Google Scholar]
- 31. Meloni M., Giurato L., Monge L., et al., “Effect of a Multidisciplinary Team Approach in Patients With Diabetic Foot Ulcers on Major Adverse Limb Events (MALEs): Systematic Review and Meta‐Analysis for the Development of the Italian Guidelines for the Treatment of Diabetic Foot Syndrome,” Acta Diabetologica 61, no. 5 (2024): 543–553, 10.1007/s00592-024-02246-9. [DOI] [PubMed] [Google Scholar]
- 32. Margolis D. J., Gupta J., Hoffstad O., et al., “Lack of Effectiveness of Hyperbaric Oxygen Therapy for the Treatment of Diabetic Foot Ulcer and the Prevention of Amputation: A Cohort Study,” Diabetes Care 36, no. 7 (2013): 1961–1966, 10.2337/dc12-2160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Löndahl M., Katzman P., Nilsson A., and Hammarlund C., “Hyperbaric Oxygen Therapy Facilitates Healing of Chronic Foot Ulcers in Patients With Diabetes,” Diabetes Care 33, no. 5 (2010): 998–1003, 10.2337/dc09-1754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Niinikoski J., “Hyperbaric Oxygen Therapy of Diabetic Foot Ulcers, Transcutaneous Oxymetry in Clinical Decision Making,” Wound Repair and Regeneration 11, no. 6 (2003): 458–461, 10.1046/j.1524-475x.2003.11610.x. [DOI] [PubMed] [Google Scholar]
- 35. Oley M. H., Oley M. C., Kepel B. J., et al., “Hyperbaric Oxygen Therapy for Diabetic Foot Ulcers Based on Wagner Grading: A Systematic Review and Meta‐Analysis,” Plastic and Reconstructive Surgery. Global Open 12, no. 3 (2024): e5692, 10.1097/GOX.0000000000005692. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.