Intensive functional therapy and rehabilitation programs for post stroke patients are considered essential for maximizing the patients' quality of life [1], [2]. Unfortunately, these programs are often just partially successful, and additional therapeutic approaches towards metabolic recovery of affected cerebral tissues are called for. While a considerable amount of preclinical research supports the use of hyperbaric oxygen therapy (HBOT) for post-stroke damaged brain tissue, so far, only 5 articles reported controlled clinical trials of HBOT for stroke patients. These studies, in which the treatment started during the early-acute phase immediately after stroke, yielded non conclusive and somewhat contradicting results [3], [4], [5], [6], [7]. In contrast, a recent phase-I study evaluating the effect of HBOT on chronic neurological deficiencies (due to traumatic brain injury) revealed promising results [8]. However, to date the effects of HBOT on neurological deficiencies due to stroke during the late-chronic phase (the focus of the current report) have not yet been investigated in a prospective randomized trial.

Years of clinical experience revealed that the dramatic spontaneous recovery from stroke occurs mainly within the first 30 days, though moderate and severe stroke survivors continue to improve for at least 90 days [9]. Most of the recovery involves brain regions rendered dysfunctional, but not dead [10]. Accumulated data from visualizations of these non-active (stunned) regions indicates that they may persist alive but dysfunctional for months, even years, after the acute injury [11], [12], [13]. It was proposed that the oxygen supply to these under-active neurons was low due to stroke damage to blood vessels in these regions, leading to oxygen deficiency, anaerobic metabolism and ATP depletion [14], [15]. The decreased oxygen level not only causes reduction in the neuronal activity but also prevents angiogenesis to replace the stroke-damaged blood vessels and the generation of new synaptic connections. Since 1 cm³ of normal brain tissue contains about 1 km of blood vessels, high oxygen supply is essential for repair of the stunned regions. Indeed, as has been demonstrated by previous studies, an increase in dissolved oxygen has several beneficial effects in damaged brain tissues [13], [16], [17], [18], [19], [20]. Transport of oxygen to glial mitochondria, the main sites of oxygen utilization, follows oxygen release from erythrocytes into the plasma and then diffusion of the blood-dissolved oxygen across the Blood-Brain Barrier (BBB). Breathing oxygen under hyperbaric conditions has been shown to be a potent means of increasing arterial oxygen tension and consequently the brain oxygen tension [20], [21], [22]. For example, at 2ATA (atmospheres absolute), the plasma O₂ partial pressure rises above 1,110 mmHg. Hence, it is reasonable to expect that HBOT can be an efficient (and clinically feasible) method for increasing tissue/cellular oxygenation and thus effectively evoking neuroplasticity in the chronically non-active areas during the late post-stroke phase.

Many physiological pathways, each with a different characteristic time, are spontaneously activated following the onset of stroke. Therefore, a challenging question to be addressed considers the optimal time lapse after stroke to start the HBOT procedure. It should also be kept in mind that signals and chemical cues associated with cell death during the acute stage of stroke might, in fact, promote repair during recovery [23] and can be negatively affected by premature application of HBOT. Unlike the case of preclinical animal studies, in clinical practice it is not feasible to apply the HBOT immediately at the stroke onset. Thus, HBOT procedure can practically begin either at the degenerative or at the regenerative stage. One can assume that any added energy during the degenerative stage could further increase the unwanted, post-injury damage. On the other hand, elevated oxygen supply during the regenerative stage would supply the energy needs for the innate brain repair processes. The differences in the time lapse between stroke onset and HBOT application in previous studies are likely to be the reason for the contradictive results obtained for HBOT application during the acute phase after stroke [3], [4], [5], [6], [7]. The aim of the current study was to evaluate the effects of HBOT started at the late-chronic phase after the acute stroke.

The study was performed as a prospective, randomized, controlled, two-group trial. The population included patients of ages 18 years or older, who had either ischemic or hemorrhagic stroke 6–36 months prior to their inclusion. All patients had to have at least one motor dysfunction. Exclusions were based on chest pathology incompatible with HBOT, inner ear disease, claustrophobia and inability to sign informed consent. Additional exclusions were based on dynamic neurologic improvements during the last month (based either on objective measurements by external evaluator or on subjective statement by the patients). Smoking was not allowed during the study. All patients signed written informed consent; the protocol was approved by the local Helsinki committee. The study was conducted in the hyperbaric and research units of Assaf-Harofeh Medical Center, Israel.

The study included 74 patients (August 2008-October 2010). 7 patients from the treated group and 8 patients from the cross group were excluded: 8 refused the SPECT, 3 had no measurable paresis, 1 had a medical problem, 1 had a stroke during the control period, and 2 refused to quit smoking (Figure 1).

Twenty four patients (80%) from the treated group had a history of ischemic stroke; of those, 17(71%), 3(13%), 2(8%) and 2(8%) patients were classified as TOAST 1, 2, 3 and 4, respectively. Twenty five patients (86%) from the cross group had ischemic stroke; of those, 18(72%), 3(12%), 2(8%) and 2(8%) patients were classified as TOAST 1, 2, 3 and 4, respectively; p = 0.8 for comparison of the TOAST classification between the groups. Of the 6 patients (20%) in the treated group that had hemorrhagic stroke, 5(83%) had anterior circulation stroke; and of the 4 patients with hemorrhagic stroke in the cross group, 3(75%) had anterior stroke. Baseline patients’ characteristics are summarized in Table 1.

In the current study, the effect of HBOT on chronic neurological deficiency due to stroke was evaluated in a prospective, randomized controlled study. Statistically significant improvements were obtained following treatment for almost all treated patients from both the HBOT-treated group and the HBOT-treated cross group (with no false negative), as was evaluated by NIHSS, ADL, brain SPECT and life quality. The significance of the improvements in this chronically debilitated population of patients is further noticeable when compared to the lack of improvement during the control (no-treatment) period of the cross group (with no false positive).

This is the first prospective, randomized clinical study evaluating the effect of HBOT in the late post-stroke period (6 months to 3 years after the acute event). There are two major reasons for selecting this study population. First, by carefully selecting patients with chronic stable neurological deficiency we were able to avoid unexpected changes in their condition. In this regard, the selection proved very useful since the control group demonstrated neurological stability with no outliers. The second reason was, as discussed in the introduction, to test our hypothesis that the optimal time for the HBOT procedure should be during the regenerative and not during the degenerative stage. While it is not possible to mark a clear line between the regenerative and the degenerative phases [23], it is quite clear that 6 months after the acute event in a stable patient the degenerative process has ended. As mentioned in the introduction, the differences in initiation times and protocols of HBOT may explain contradictive results in previous studies, where HBOT was used in the early phase after stroke [3], [4], [5], [6], [7]. The recent publication by Harch et al., evaluating the effect of HBOT on chronic neurological deficiencies due to traumatic brain injury, also supports the use of HBOT in the late stage after the acute insult [8].

The issue of “how to handle the control group” was discussed by a multidisciplinary team including physicians specializing in hyperbaric medicine, physicists specializing in neuronal-glia interactions and the ethics committee. The patients can tell if pressure is increased or not, so the pressure must be increased also in the control group. The only way to administer “placebo” of HBOT is to bring the patients to the hyperbaric chamber and to increase the environmental pressure to an extent that the patients will feel it in their ears. The minimal pressure needed to gain such a feeling should be 1.3 ATM. Henry’s law states: “the amount of a given gas dissolved in a given type and volume of liquid is directly proportional to the pressure of that gas in equilibrium with that liquid”. Thus, hyperbaric environment significantly increases the dissolved oxygen pressure even if a person holding his breath [29]. Compressed air at 1.3 ATA increases the plasma oxygen tension by at least 50% and that is certainly notable. There are many case reports illustrating significant effects following small increase in air pressure [30], [31], [32]. Moreover, even a slight increase in partial pressure, such as, for example, to 1.05 ATM at altitude 402 m below sea level (the Dead Sea), can lead to noticeable physiological effects [33], [34], [35], [36], [37]. However, it should be kept in mind that oxygen is not a drug, and because it is metabolized mainly in the mitochondria, there is no simple dose-response curve.

Since increasing the pressure even without adding oxygen can also increases the dissolved oxygen partial pressure, the only way to maintain normal (placebo) levels of dissolved oxygen is to supply air with lower than normal level of oxygen, which we deemed unethical. To partially compensate for this inherent limitation, the patients in the cross group started with a two-month control period of no treatment, at the end of which they were crossed to two months of HBOT sessions. To gain better validity of the results, we used the scatter plot analysis of the changes in the combined neurological evaluations. The scatter plots (figure 3) show changes in the NIHSS and ADL scores in terms of the scaled relative differences, as is defined in the methods section. In that analysis, summarized in figures 2 & 3, the correlation between changes NIHSS & ADL after HBOT is clearly demonstrated. Moreover, the analysis evidently demonstrates that the effect of HBOT on these “not completely blinded evolution” was the same in the treated group and the control group after blind randomization. The correlation between the improvement in NIHSS and ADL and the improvement in the brain SPECT results, which was done in a completely blinded fashion, further substantiates the clinical findings. Moreover, the consistency between the anatomical locations of the changes in the brain metabolism, as demonstrated by the SPECT, with the finding in the neurological evaluation provides important validation of the neurological evaluation.

During most of the 20^th century, there was an ongoing debate about the time window available for induction of neuroplasticity. The improvements in the chronic late stage reported here support the view that neuroplasticity can be activated months to years after the acute event when a proper brain stimulation (such as HBOT) is applied. More specifically, the current study included patients that underwent stroke more than 6 months prior to treatment and after their condition reached a steady state (no improvements were monitored for at least a month). These important and unexpected findings arein agreement with other recent findings revealing that many aspects of the brain remain plastic even at adulthood [38]. They are also consistent with several other studies in post stroke patients [39], [40], [41]. In the current study, patients were treated only with HBOT without any additional guided training and/or practice. This was done in order to demonstrate the therapeutic potential of this treatment. It is reasonable to expect that exploiting the HBOT in conjunction with other rehabilitation intervention can lead to even better results leading to optimal future practice. The current study paved the way for future investigations of this promising direction, which should be one of the aims of upcoming more elaborated clinical studies.

Current imaging technologies reveal that the stunned brain areas (regions of high anatomy-physiology mismatch) may persist for months and years after an acute brain event [11], [12], [13]. The changes in SPECT images after treatment demonstrate that the HBOT procedure led to reactivations of neuronal activity in the stunned areas. While SPECT imaging has limited spatial resolutions (in comparison, for example, to fMRI), the changes in activity were sufficiently robust to be clearly detected by the SPECT images. However, a future, more detailed study using fMRI (along with direct observations on animal model) will be able to provide additional valuable insights, in particular regarding the operative underlying mechanisms that activate the neuroplasticity (e.g. the putative role of glial cells). We note that patients were not selected based on their anatomical and functional brain imaging evaluation. It might be possible that the results would have been even better had the study included only patients with high SPECT/CT mismatch. This issue of the preferred population for HBOT should be further investigated in future clinical trials. We also note that in the current pioneering study aimed at “proof of concept”, all patients underwent 40 HBOT sessions. Based on our current clinical experience, more sessions of HBOT may be needed, at least for some patients, in order to obtain the maximal improvement effect.

In any case, the observed reactivation of neuronal activity in the stunned areas imply that increasing the plasma oxygen concentration with hyperbaric oxygenation is a potent means of delivering to the brain sufficient oxygen for tissue repair: HBOT might initiate a cellular and vascular repair mechanism and improve cerebral vascular flow [8], [13], [16], [17]. At the cellular level, HBOT can improve mitochondrial function (in both neurons and glial cells) and cellular metabolism; improve BBB and inflammatory reactions; reduce apoptosis; alleviate oxidative stress; increase levels of neurotrophins and nitric oxide, and up-regulate axon guidance agents [13], [16], [17], [20]. Moreover, the effects of HBOT on neurons can be mediated indirectly by glial cells, including asrocytes [18]. HBOT may also promote neurogenesis of the endogenous neural stem cells [19]. The major limitation of the above-mentioned data is that it has been tested in different types of models and includes different protocols of HBOT. However, it is well noticed that there is at least one common denominator to all repair/regeneration mechanisms: they are all energy/oxygen dependent. It might be possible that HBOT enables the metabolic change simply by supplying the missing energy/oxygen needed for those regeneration processes.

To conclude, in this study we provide, for the first time, convincing results demonstrating that HBOT can induce significant neurological improvement in post stroke patients. The neurological improvements in a chronic late stage demonstrate that neuroplasticity can be operative and activated by HBOT even long after acute brain insult. Thus, the findings have important implications that can be of general relevance and interest in neurobiology. Although this study focused on stroke patients, the findings bear the promise that HBOT may serve as a valuable therapeutic practice in other neurological disorders exhibiting discrepancy between the anatomical and functional evaluation of the brain.