Cancer is a foremost cause of global morbidity and mortality (1–4). The International Agency for Research on Cancer predicts a 47% rise in the global cancer burden in 2040 compared to 2020, expecting 28.4 million cancer cases (5). Pelvic organ cancer designates any malignant tumor primarily originating in any pelvic cavity organ (6, 7). Despite the therapeutic avenue innovation, radiotherapy (RT) remains one of the cornerstones for managing malignant tumors, with 50% of patients receiving RT during their course of illness (8–10). Introducing new RT techniques (e.g., three-dimensional conformal radiation therapy and intensity-modulated radiation therapy), it has been possible to eradicate malignant tumor cells with minimal exposure to surrounding tissues (11–13). However, these modern RT approaches are not exempted from side effects, late radiation-induced proctitis (RIP) being one of them (9, 14).

The RIP is defined as an iatrogenic lesion of the mucosa and submucosa of the rectum or rectosigmoid transition following RT for pelvic cancer (8, 10). It can be classified into acute or chronic RIP, considering the time from the end of RT protocol to the onset of RT-related clinical signs and symptoms (8, 15). The incidence rate is estimated from 2 to 39% (16). Proctitis starts with an initial mucosal injury induced by ionizing radiation, and the organism reacts through a chronic inflammatory response leading to fibrosis, obliterative endarteritis, tissue hypoxia and ischemia, and neoangiogenesis (17, 18). Gut microbiota and oxidative stress may also be important in chronic radiation enteritis and colitis (19–21). Different clinical manifestations may develop, including proctalgia, diarrhea, constipation, recurrent ileus, mucus discharge, hematochezia, ulceration/fistulae formation, fecal incontinence/urgency, or a predisposition to a malabsorption syndrome. Patients may also experience other symptoms, such as urinary frequency, dysuria, pelvic pain or discomfort, urinary retention, and hematuria (due to synchronous radiation cystitis [RC]) (22). After a meticulous anamnesis, the clinical diagnostic hypothesis should be confirmed by performing (procto)sigmoidoscopy. Endoscopic images usually reveal an edematous, hyperemic friable mucosa with telangiectasias and possibly areas of ulceration or sloughing (8, 14, 23, 24). Three major therapeutic approaches can be used in RIP – medical, endoscopic, and surgical – and there is no gold standard treatment regimen (8). Among pharmacological regimens, formalin application (4-10%) has been demonstrated to be an effective medical treatment (8, 23, 25–28), as well as sucralfate enemas (6, 23, 29), and metronidazole combined with ciprofloxacin (20, 29). Argon plasma coagulation (APC) is the first-line endoscopic strategy for individuals with hemorrhagic RIP (24, 29–31). Surgery is generally reserved for advanced stages of RIP refractory to optimal medical and endoscopic therapies (8, 30).

Hyperbaric oxygen therapy (HBOT) is another effective and safe intervention for managing RIP (24). It is characterized by inhaling pure oxygen (FiO₂ 100%) inside a hyperbaric chamber frequently pressurized at 2.0 to 2.8 atmospheres absolute (ATA) for 60 to 120 minutes daily (8, 32, 33). This treatment leads to a series of physiological effects due to increased plasma oxygen concentration and delivery to tissues, contributing to better tissue oxygenation, enhancing microvascularization, and promoting angiogenesis. It also exerts a tumor sensitization effect on RT (34). It might be associated with mild and transient side effects due to pressure and oxygen toxicity, mostly middle ear barotrauma (MEB) (35, 36).

Currently, HBOT represents a primary and complementary standard treatment in several clinical scenarios, including late radiation tissue injuries (LRTIs), namely RIP and RC (type 1 recommendation, level B evidence) (36, 37). However, clinical evidence is still scarce. This retrospective study pretends to evaluate the effectiveness of the HBOT in treating RIP. Secondary objectives include assessing the hematochezia resolution rate; estimation of hematochezia recurrence; evaluation of endoscopic improvement; the correlation between response rates and the number of HBOT sessions; and finally, the evaluation of the HBOT safety level.

Consistent guidelines are not currently available for the management of RIP (24). Generally, patients are treated empirically on a tailored basis according to patient comorbidities, the severity of the clinical condition, and medical and institutional expertise (6, 8, 16, 24).

The HBOT is used in several clinical conditions as well as in professional and military training (8, 37, 38). As a result of the 10^th European Consensus Conference on Hyperbaric Medicine, HBOT was recommended for the treatment of soft tissue radionecrosis, namely, RIP and RC (type 1 recommendation, level B evidence) (36). The first cutting-edge case report in this setting was conducted by Charneau J et al. (1991). It revealed the utility of HBOT in treating a 74-year-old male patient with a history of transfusion-dependent hemorrhagic RIP refractory to medical treatment. This patient achieved a complete response after 82 HBOT sessions (2.5 ATA, for 90 min, twice daily), with the clinical response remaining for at least 9 months, avoiding colostomy (39). Afterward, it was the motto for several case reports and small series, retrospective studies, clinical trials, reviews, and meta-analyses published in the last three decades (40–68).

In a Cochrane systematic review, Bennett M et al. (2016) reviewed 14 studies that showed moderate-quality evidence of HBOT in the management of LRTIs, including RIP (40). Another earlier systematic review conducted by Feldmeier J and Hampson NB (2002) considered 74 publications (including 14 on RIP and radiation enteritis), in which all but seven reported a positive outcome. Thus, based on the consistency of these findings, HBOT was considered “Likely to be Beneficial” for the treatment of patients with RIP and radiation enteritis (41).

We report real-world evidence of HBOT effectiveness in a large cohort of patients (n= 88) with RIP over a significant period of time (10 years).

Achieving an overall clinical response and hematochezia resolution rates of 94.3% and 93.7%, respectively, we can state that HBOT proved effective in managing RIP. These numbers reinforce the importance of this study compared to others, although positive, have less statistical robustness (40–46). The HOT2 clinical trial was an exception, which controversially demonstrated that HBOT was ineffective in treating chronic bowel dysfunction following pelvic RT (44). However, several limitations were pointed out for this study, such as the small sample size for a phase 3 trial, the predefined follow-up duration of only 12 months or even the exclusion of patients with grade 3 symptoms of LENT-SOMA score (including severe fecal incontinence and transfusion-dependent RIP) (8).

Generally, there is an underestimation of RIP mild symptoms, either by patients or physicians (8, 46). As expected, hematochezia represented our series’ most common clinical manifestation of RIP. The reason for this is possibly related to the fact that hematochezia is the most appreciated symptom and the primary manifestation of RIP, as it is an alarming symptom for the doctor-patient binomial. This information is consistent with data from previous studies, including that by Clarke RH et al. (2008), with the majority of patients presenting lower GI hemorrhage and other chronic symptoms such as diarrhea, constipation, ulceration, and pain (47). Additionally, our study showed a higher hematochezia resolution rate (complete and partial) when compared to other studies, such as that by Dall’Era MA et al. (2006) and Oliai C et al. (2012) with 76% and 75%, respectively (49, 50).

Furthermore, some patients have experienced symptoms other than GI as a result of RT damage to surrounding healthy tissues (22). Tolerance to irradiation of pelvic structures varies from specific organ to another. Normal tissues of the vagina, cervix, and uterus can withstand high radiation doses and recover quickly from RT lesions (51, 52). On the opposite side, the bladder, sigmoid, rectosigmoid, and rectum are more prone to radiation toxicity. Utmost point doses of 75 and 70 Gy are acceptable for the bladder and rectum, respectively (51, 53, 54). This supports that RIP and RC are two commonly reported LRTIs, possibly arising concurrently (51, 54–56). In fact, the combination of RIP and RC was our second most common LRTI diagnosis. A previous study conducted by Ribeiro de Oliveira T et al. (2015) evaluating hemorrhagic RC response to HBOT pointed to 23.9% of patients with LRTIs other than RC (namely RIP and enteritis) (55). Our data showed that individuals with at least two concurrent LRTIs were associated with complete response to HBOT, probably because other LRTIs require fewer HBOT sessions to achieve a complete clinical resolution and, consequently, are easier to treat when compared to RIP. Furthermore, previous studies investigating the efficacy of HBOT in treating RC, such as that by Oscarsson N et al. (2019), demonstrated that 30 HBOT sessions were associated with clinically relevant improvement in RC (57). Ribeiro de Oliveira T et al. (2015) also showed that 89.6% of patients achieved complete resolution of hematuria after an average of 37 sessions (55). Similarly, Degener S et al. (2012) presented an 80% complete resolution of hematuria after 30 sessions for a 35-month follow-up (58). Therefore, we believe that this treatment modality should be specially considered as a first-line treatment for patients who develop more than one LRTI (55, 59).

Patients eligible for our study tried at least one last conservative treatment modality (medical or endoscopic) that proved to be ineffective in treating RIP-associated symptomatology. Besides only 35 clinical records confirm the attempt to treat RIP initially with medical treatment, we believe that some pharmacological treatment has been performed globally as it is classically the first-line treatment to be applied in the management of RIP (8, 23, 29, 69). Sucralfate and mesalamine enemas were the most frequently prescribed treatments; less common options included metronidazole, vitamin A, and corticosteroid enemas. About 50% of patients underwent APC treatment before starting HBOT. Álvaro-Villegas JC et al. (2011) found that APC and HBOT were equally effective in treating RIP with regard to controlling rectal bleeding, the need for transfusions, and the impact on tissue toxicity. However, APC was associated with faster and higher responses at baseline (60). One of the relevant aspects of our series is that we have a heavily treated population in which HBOT played an important role in its subsequent control.

Additionally, our data showed that the time between RT and the onset of hematochezia was inversely associated with the number of HBOT sessions required. Therefore, it can be inferred that patients with a longer time until the clinical manifestation of hematochezia may have a better prognosis. After completing a median of 60 sessions, the overall response to HBOT tended to be inversely associated with the number of sessions, possibly because patients with a better response to HBOT needed fewer sessions. However, there was a trend for hematochezia resolution after completing at least 70 HBOT sessions. In general, the number of sessions performed in our study was higher than in other studies that present a full course treatment with an average of 40 sessions (42, 49, 61). This higher number of prescribed sessions may be explained by the fact that we treat more severe cases of RIP, as patients come to us in later phases and, consequently, with more refractory symptoms. Hence, due to our institutional experience, we anticipate the need for a more prolonged treatment, frequently surpassing 40 sessions.

Although rare and usually non-severe, HBOT has known potential adverse events. In a retrospective analysis of 1.5 million treatments, only 0.68% were associated with a side effect, being barotrauma and confinement anxiety the most reported events (62). The incidence of MEB varies according to the series and ranges from 0.37 to 84% in non-ventilated patients versus 94% in ventilated patients (38, 63, 64). Our data align with published data and showed that the number of HBOT sessions was a predictor of side effects, mainly MEB.

Lastly, our retrospective study is not exempted from limitations. The current analysis is limited by several factors, including its retrospective nature, the single institution source of data, and the quality of our medical and supervisor records compromised by the lack of complete clinical information. Once patients turned asymptomatic (mainly in relation to hematochezia) and completed the treatment course, they were referred to the attending physician. One of the limitations is the short follow-up period and the loss of follow-up of treated patients after referral to the attending physician. These facts may have prevented more reliable monitoring of symptoms or even their relapse. Moreover, it may even influence the evaluation of the hematochezia recurrence rate. The lack of grading of adverse events based on a validated classification did not allow for drawing more robust conclusions, configuring another important limitation of our study. The missed information about the different RT protocols administered (total dose and fractionation) and comprehensive information related to pre-HBOT medical treatment conditioned our analysis of the potential relationship between different variables and the clinical response to HBOT. Also, the inexistence of a more detailed characterization of the post-HBOT clinical picture limited us from assessing the remaining symptoms.

In this cohort of HBOT-treated refractory RIP patients (the largest reported to date), the overall clinical response rate was 94.3% (including 62.5% complete responses) after a median of 60 HBOT sessions. Furthermore, HBOT proved to be a safe treatment when patients are previously studied and the correct protocol is applied. This real-world evidence study adds value to published data on the management of RIP with HBOT.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by Armed Forces Hospital Ethics Committee - Armed Forces Hospital, Lisbon, Portugal. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

The present manuscript is the result of the original work by the authors. AMM and DAC conceived and designed the study. AMM performed the consultation of medical records. AMM, DAC, and VM performed the analysis and interpretation of the data; all the authors approved manuscript draft. DAC, VM, and CEA performed the manuscript supervision.