Introduction

Traumatic crush injuries to the lower limb impose substantial burdens, comprising in psychosocial, physical and emotional implications [1]. These injuries encompass damage to various structures within the lower limb, including soft tissue, bone and neurovascular structures. The extent of lower limb soft tissue injury and contamination leads to an increased risk of complications, such as wound infection, osteomyelitis, wound necrosis and non-union [2,3,4]. Managing these complex injuries requires a comprehensive approach that addresses both immediate interventions and long-term functional and psychosocial outcomes.

Hyperbaric oxygen therapy (HBOT) has emerged as a potential adjunctive intervention in the management of limb injuries to diminish injury complications and improve outcome [5]. HBOT is administered in a monoplace or multiplace hyperbaric chamber, in which a patient breathes 100% oxygen supplied at a pressure greater than that at sea level. As a result of this hyperoxygenation, oxygen diffusion to tissues increases, contributing to the wound healing process [6,7,8]. Furthermore, HBOT reduces edema formation through vasoconstriction, diminishes inflammatory processes, increases collagen synthesis and angiogenesis, and inhibits the biochemical events in ischemia–reperfusion injury [9]. HBOT is regarded as a safe treatment modality with mostly avoidable or self-limiting side-effects. The most common side-effect of HBOT is middle ear barotrauma and the most feared adverse event is oxygen toxicity seizure [10].

Although prior research has predominantly focused on the role of HBOT in the treatment of chronic wounds, there is growing evidence supporting the beneficial results of HBOT in managing acute traumatic ischemias as well [7, 8, 11, 12]. Given the above-mentioned pathophysiological mechanisms HBOT has on wound healing, it is plausible that HBOT holds potential in addressing acute traumatic ischemia in the context of severe lower limb injuries. Therefore, the aim of this study is to present an overview of the available literature on the addition of HBOT in the management of crush-associated severe lower limb soft tissue injuries and to evaluate its effectiveness.

Methods

Literature search

The methods and results of this systematic review are written in accordance to the Preferred Reporting Items of Systematic Reviews and Meta-Analyses (PRISMA) statement [13,14,15]. The electronic bibliographic databases Medline, Embase and Cochrane Library were searched from inception up to November 21st 2022 by a clinical librarian. The search employed Mesh/Emtree search terms associated with ‘oxygen inhalation therapy’, ‘hyperbaric oxygen therapy’, ‘therapy, lower extremities, lower extremities’, ‘lower limbs’ and ‘wound healing’. The full electronic search strategy is detailed in Appendix A.

Study selection

After duplicate removal, two reviewers (EK, GG) independently screened the titles and abstracts of the articles in the web-tool Rayyan [16]. The full texts of the studies were requested, and two reviewers (EK, GG) performed analysis of these articles to identify the final selection. In case the full text of an article could not be obtained, abstracts were allowed to be included. Whenever there was no consensus between the reviewers, a third reviewer (MB) assisted in the discussion.

During the selection, all studies on the addition of HBOT to standard trauma care in patients with crush-associated severe lower limb soft tissue injuries were included. Reviews, animal studies, studies in children, conference abstracts, poster presentations and non-English articles were excluded.

From the included articles more relevant studies were identified through reference list screening.

Data extraction and quality scoring

During the data collection process, two reviewers (EK, GG) analyzed the final articles in detail. The following data was extracted: age, sex, type and severity of injury, time to surgery after injury, time to HBOT, HBOT protocols and follow-up time. All primary outcome measures on wound healing were extracted from the data, as well as secondary outcome measures: time until complete wound healing, incidence of wound complications and need for additional surgical interventions. The study characteristics, patient characteristics and results were entered into an electronic spreadsheet. Again, if consensus was not reached, a third reviewer (MB) assisted in the discussion.

Two reviewers (EK, GG) independently screened the included articles for risk of bias using the ROBINS-I tool for non-randomized studies and the Cochrane risk of bias tool for randomized controlled studies [17, 18]. Any discrepancies were resolved by consulting a third senior author (MB).

Data synthesis

A descriptive overview was provided on the outcomes of the studies. The quantitative results were described separately for each study. If we required to calculate P values that were not reported in the original publications, this was done with a Z-score test for two proportions. P < 0.05 was regarded as statistically significant.

Results

The number of articles found was 1528, following duplicate removal 1004 articles remained. After analyzing the abstracts and full texts of the studies, a final selection of seven articles was made [19,20,21,22,23,24,25]. One article could not be obtained in full text but was included in the review based on the high relevance as judged from the abstract [22]. A flow chart of the inclusion process is shown in Fig. 1.

Fig. 1
figure 1

PRISMA flow chart of the included articles

Full size image

The study characteristics of the seven included studies are outlined in Table 1. The assessment of potential bias is detailed in Tables 2 and 3. The review included two randomized clinical trials [21, 25], one retrospective cohort study [23], three case series [19, 20, 22] and one case report [24]. The total study population size consisted of 229 patients, with 138 patients in the HBOT group and 91 patients in the control group. The mean age varied between 23.3 and 55 years. HBOT protocols exhibited variations across the studies, with pressure levels ranging between 2.3 and 2.8 ATA. Most studies used a HBOT protocol of two HBOT sessions a day with a duration of 90 min per session.

Table 1 Study Characteristics
Full size table
Table 2 ROBINS-I for non-randomised studies
Full size table
Table 3 Quality assessment using Cochrane risk of bias tool
Full size table

Three studies were comparative studies, hence incorporating a control group [21, 23, 25]. Among these, Bouachour et al. and Miller et al. conducted randomized clinical trials. The study by Bouachour et al. was a placebo-controlled clinical trial, where sham hyperbaric treatment was applied in the control group, using a similar protocol as the intervention group but with a slightly increased pressure (1.1 ATA), and air instead of oxygen. The trial by Miller et al. followed a non-placebo-controlled design.

The results of the individual studies are presented in Table 4, including the following study outcomes: complete wound healing, incidence of necrosis, incidence of infection, additional surgical interventions, hospitalization and time to wound healing.

Table 4 Study Outcomes
Full size table

Complete wound healing

Wound healing was the primary outcome measure in most studies. The results are demonstrated in Fig. 2. Bouachour et al. defined complete wound healing as wound healing without tissue necrosis requiring surgical excision and demonstrated significantly higher rates of wound healing in the HBOT group when compared to the control group (94% vs. 56%, P < 0.01). Matos et al. achieved complete wound healing in 20 out of 23 patients (87%). Monies-Chass et al. reported complete wound healing in six out of seven patients (86%), with one patient requiring toe amputation due to residual ischemia distally despite the beneficial effects of HBOT. Shupak et al. reported complete wound healing in eight out of 13 patients (62%), a reduction of ischemia distally in four patients and no improvement in one patient. Stefanidou et al. reported complete wound healing in their case report.

Fig. 2
figure 2

Complete Wound Healing

Full size image

Incidence of necrosis

Bouachour et al. reported significantly lower rates of necrosis in the HBOT group when compared to the control group (5.6% vs. 44%, P = 0.007). Millar et al. demonstrated reduced necrosis in the HBOT group when compared to the control group (29% vs. 53%, P = 0.01) at 14-day assessment.

Incidence of infection

Yamada et al. reported an infection rate of zero percent in the HBOT group versus 46% of the patients in the control group (P = 0.003). Millar et al. did not find statistically significant differences in acute infection rates between the HBOT group (22%) and the control group (32%) at 14-day assessment, and similar results were observed at 12-month assessment for the incidence of deep infections (8% vs. 15%, respectively).

Additional surgical interventions

Bouachour et al. reported a significantly reduced need for additional surgical interventions in the HBOT group when compared to the sham HBOT group (5.6% vs. 33.3%, P < 0.05). Yamada et al. reported no need for additional surgical interventions in the HBOT group, while 38% of the patients in the control group required additional surgical interventions (P = 0.013). Millar et al. found that 67% in the HBOT group required additional surgical interventions, compared to 56% in the control group, with the difference being non-significant.

Hospitalization

Length of hospital stay was assessed in three studies, no significant differences in outcomes between the HBOT group and the control group were observed.

Time to wound healing

Bouachour et al. did not find any significant differences between the HBOT group and the control group regarding the time to wound healing.

Discussion

This review summarizes the literature on the effectiveness of HBOT on outcome in traumatic lower limb soft tissue injuries. Based on our results, we can conclude that HBOT when added to standard trauma care holds potential to positively influence wound healing.

Past studies have mainly researched the effects of HBOT in chronic wounds rather than acute injuries, even though evidence on the role of HBOT in the management of acute wounds does exist. The systematic review conducted by Garcia et al. studied the addition of HBOT in the management of crush injuries and traumatic ischemias, including studies published between 1966 and 2003 [26]. The review encompassed diverse injuries, such as upper and lower limb injuries as well as finger traumas and compartment syndromes, with varying mechanisms of injury. In an attempt to address the issue of heterogeneity, we decided to restrict our search to the inclusion of crush-associated lower limb soft tissue injuries. By doing so, we intended to steer towards more specific recommendations on its management.

The positive effects of HBOT on the pathophysiological processes in acute traumatic ischemias, especially on the ischemia, edema and hypoxia triad, are well described in current literature. Despite these findings, HBOT has not yet found its place in guidelines for the management of traumatic injuries, including lower limb injuries. One of the reasons for this, is the lack in large randomized clinical trials considering this research topic. Only two of the studies included in this review are randomized clinical trials. It is therefore tempting to call for more trials, and indeed the highest grade of evidence should always be sought. However, several issues preclude conduction of large trials on this subject. Firstly, severe lower limb injury is a heterogeneous clinical entity, and design of a clinical trial that includes a well-defined patient population while at the same time ensuring an adequate inclusion rate is challenging. Secondly, the application of multiple sessions of HBOT may entail quite a large placebo effect. Inclusion of a control group that undergoes sham HBOT may help to limit bias in this regard. Lansdorp et al. described the design of sham HBOT protocols and the study by Bouachour et al. showed that it is possible [27]. However, for many hyperbaric centers’ treatment with sham HBOT means that their only hyperbaric chamber is blocked for significant amounts of time, reducing availability of ‘real’ HBOT and increasing costs. Also, as Millar et al. pointed out, sham HBOT could be seen as a negative intervention, because it removes patients from regular clinical care for considerable amounts of time [25]. Moreover, the scarcity of hyperbaric chambers in general adds to the difficulty of conducting large-scale trials with HBOT. Moon et al. reported that less than ten percent of all hyperbaric treatment facilities in the United States provide continuous availability for emergency HBOT [28].

We believe that in the field of hyperbaric medicine it is important to make decisions based on all available evidence, and not only rely on large trials, which – for the reasons stated above – are difficult to perform. Evidence from pathophysiology, case reports, observational studies, and randomized clinical trials should be synthetized. Even though our study includes only seven studies in total, of which two were clinical trials, we still felt an important need to perform this review and to draw conclusions from it.

Future studies should prioritize the development of standardized protocols for HBOT in the management of severe lower limb injuries, enhancing the reproducibility of the treatment and facilitating comparisons between the studies. Combing these standardized protocols with data from snapshot audits could advance the understanding of HBOT’s efficacy in severe lower limb injuries [29].

Limitations

This review holds several limitations. Firstly, the included studies lack comprehensive data, notably regarding follow-up duration, thereby complicating the evaluation of long-term functional and psychosocial outcomes after HBOT and after standard care alone. Also, the unavailability of the full text of Matos et al. resulted in the inclusion of an abstract, compromising the quality of this study due to the inability to access all information. Moreover, the majority of included studies were small case reports and series including a relatively limited number of patients, potentially influencing observed effects and restricting the generalizability of findings. Furthermore, variations in time from injury until surgery, the initiation of the first HBOT session, as well as differences in HBOT protocols and number of HBOT sessions among the studies, contributed to existing heterogeneity. At last, it is important to note that, except for Millar et al., the remaining studies span a decade or more, during which standard trauma care significantly evolved. This evolution potentially reduces the relevance of older findings, posing a challenge in weighing their outcomes. Despite these limitations, this systematic review provides novel insights into the treatment of acute lower limb injuries with HBOT.

Conclusion

Based on the seven studies included in this review, the addition of HBOT to standard trauma care seems to have a beneficial effect on wound healing in crush-associated severe lower limb soft tissue injuries.