Dr. P. Germonpré, Med. Capt., Military Hospital Queen Astrid, Brussels

Introduction

Hyperbaric Oxygen Therapy (HBOT) involves the administration, by inhalation, of 100% oxygen under pressure. Oxygen, when breathed at pressures higher than the atmospheric pressure, takes on all of the properties of a pharmaceutical drug. Just like e.g. for an antibiotic, there needs to be a precise indication for which it is administered. Also, the correct dosis has to be given, and the therapy must be given for a sufficient period of time, in order to obtain therapeutic effects. Unfortunately, administering HBOT involves the use of heavy and potentially dangerous apparatus, whereas oxygen itself, a highly explosive and combusting gas, must be handled with extreme care. However, when practised with the necessary skill and based on scientific evidence, HBOT can be life-saving or a most valuable adjunct in various therapeutic protocols.

Principles of Hyperbaric Oxygen Therapy

In order to breathe a gas under pressure, a person must be completely exposed to that same pressure. This is easily understood as one realises that the resistance of the pulmonary structures to transpleural pressures is of the order of 80cm H2O. When breathing any gas mixture at a pressure of (customarily) 2,5 atmospheres, which is equal to 1500cm H2O, the lung alveoli would instantly burst, causing pneumothorax, pneumomediastinum or arterial gas embolism (AGE). Therefore, the patient has to be submitted to the same external pressure as the one at which the gas is inhaled. He is placed inside a hyperbaric chamber, which is nothing more than a hermetic steel, plexiglas or other compound hull, which is, prior to the administration of the inhaling gas, pressurised at that given pressure.

Different types of hyperbaric chambers exist, based upon size and pressurising gas used. Monoplace hyperbaric chambers are cylinders of about 1m diameter and 3m length, in which only the patient is placed on a stretcher, and that is usually pressurised with 100% oxygen. The advantages are the low cost, and the absence of the need for a special mask or other breathing apparatus. However, in Europe, this type of chamber is no longer advocated, due to the numerous disadvantages. First, only one patient can be treated at the time. When this patient needs to be concomitantly treated with drugs or fluid infusions, a through-the-hull pressure connection must be made. Monitoring of the patient, and possibly ventilatory assistance can only be controlled from the outside. When the patient presents with an acute problem during the treatment (e.g. vomiting, convulsions), the only option is to depressurise the chamber rapidly (i.e. in approx. 2 minutes), thus encurring the risk of lung overpressure syndromes. Finally, and maybe not in the least the less important disadvantage, is the high risk of explosion or fire inside these chambers, which would mean instant carbonisation of all its contents. Therefore, draconian safety measures are necessary to prevent all sparks, static electricity or grease, both on the patient and on any introduced apparatus. This may represent a serious problem when treating patients with greasy ointments applied to wounds (e.g. burn patients). The most currently employed type of HBOT chamber in Europe is the multiplace hyperbaric chamber. At the military Hospital Brussels, a Comex 1500 multiplace treatment chamber was put into use in April 1991, as the first multiplace HBOT chamber in a Belgian hospital. Recently, three more multiplace chambers were installed in different hospitals in Belgium that previously disposed only of a monoplace chamber. Multiplace HBOT chamber are bigger than monoplace chambers, several patients can be treated at the same time. In the HMRA chamber, up to 4 patients can be treated simultaneously. The main advantage however, lies in the fact that these chambers are pressurised with compressed air, thus almost completely eliminating fire and explosion hazard. The patient breathes 100% oxygen via a hermetic nose-mouth mask; via a plastic head-tent or via the endotracheal tube. A personnel transfer lock is always present, via which a nurse or physician can be entered into the treatment chamber without the need for decompressing the patient. All medical treatment and monitoring apparatus, when tested for safety and good functioning under pressure, are placed inside the chamber, and this way, a complete intensive care environment can be created - in other words, the ICU patient can be treated with HBOT without having to interrupt his normal ICU treatment. An inside attendant (nurse or physician) is present at the side of the patient throughout the treatment session (approx. 1.5 hours). This may seem cumbersome, but it offers many advantages. First, the risk of middle ear barotrauma, which is the most common complication, can be reduced to an absolute minimum, by guiding and observing the patient carefully. Secondly, this presence eliminates almost completely any claustrophobic feelings. Thirdly, even in non-critical patients, vomiting or convulsions may (rarely) occur; these can be handled by the inside attendant prior to decompression. The disadvantages of multiplace chambers are mainly related to the higher cost and the need for specialised technical personnel to manipulate the chamber circuits. However, the improved safety and therapeutic possibilities make this the chamber of choice for major health care institutions.

Oxygen is transported in the blood, via two distinct mechanisms, according to the following formula:

Ca(O2) = (Hb x 1.34 x SaO2) + (PaO2 x 0.003)

In normal (normobaric) circumstances, the transport via haemoglobin is the most important. On this protein, almost 20ml of oxygen can be transported per 100ml of blood, since the saturation (SaO2) of the (normal quantity of ) 15g of Hb per 100ml is near to 100%. Unfortunately, because of this near-100% saturation, this transport mechanism is limited, and virtually nothing can be gained further. The second transport mechanism is in normal circumstances minimal: it is the physical dissolution of oxygen into the blood plasma, according to Henry's Law (Q = a P). Since a is so small (0.003ml/100ml/mmHg), this quantity amounts to 0.3ml oxygen per 100ml blood. When the partial pressure of oxygen is raised however, this quantity augments in a linear fashion, and unlimited by saturation or concentration of other substances. When breathing 100% of oxygen at a pressure or 2.5 atmospheres, almost 5 ml of oxygen can be transported in this way in the blood plasma, which is the entire normal oxygen consumption of a resting human being. Incidentally, this would mean that in these circumstances, we would no longer need the red blood cells to transport oxygen. This has been prove experimentally already in 1960 (Boerema et al.) by exsanguinating a pig while under pressure and replacing the blood with Macrodex solutions until less than 0.5 g Hb was present without any signs of myocardial or cerebral ischaemia.

Pharmacological effects of oxygen under pressure

On a macrovascular level, hyperoxia induces a generalised precapillary vasoconstriction, which, dependent of the tissue considered, varies, but lies around a 20% perfusion reduction. A reflex bradycardia occurs, and the mean arterial blood pressure stays identical. Despite this reduction in perfusion, a fourfold or more increase in peripheral oxygen delivery is achieved. Indeed, when breathed at high pressures, oxygen is present in high concentrations in the plasma. Therefore, the final diffusion distance from the capillary vessel (which is dependent on the pressure difference between capillary and cells who utilise the oxygen) will be more than quadrupled. Oxygen is delivered at higher pressures to the cells and to cells further away from any given capillary. This can be easily demonstrated by means of transcutaneous or intratissular oxymetry, combined with laser Doppler flowmetry. On a cellular level, this enhanced oxygen delivery has important consequences, especially when the local oxygen delivery was impaired by swelling (oedema), acute vascular compromise (embolism or thrombosis) or chronic vascular insufficiency (arteriosclerosis, endarteritis, radionecrosis).

Ischaemia leads to tissue cell death. Ischaemia can be caused by tissue swelling (oedema) alone, especially when the tissue involved in enclosed in a non-expandable fibrous sheath, like the perimuscular fascia in limb muscles. Because of the increased intravascular water content, capillary vessels are compressed, which slows and eventually stops their perfusion. When a fasciotomy is performed the muscle tissue has the possibility to swell, but this does not always prevents tissue ischaemia. The intercapillary distance may have become so important, that cells located "in the middle" are not receiving any oxygen, the oxygen diffusion distance being insufficient. The effects of HBOT are here twofold: on one hand, the tissue perfusion is diminished, limiting further oedema formation and permitting a faster oedema resolution, and on the other hand, the oxygen diffusion distance is greatly augmented, permitting the survival of more tissue cells.

Hypoxic tissues are prone to infection, not only because of the possibility of development of anaerobic infections, but mainly because one of the key cells of the aspecific immune defence, the polynuclear white blood cell (PMN) or granulocyte, depends on a sufficient supply of surrounding oxygen to perform its microbial "killing" function. This so-called "respiratory burst" that occurs after the phagocytosis causes a 20-fold increase in the oxygen consumption of the cell. When an insufficient quantity of molecular oxygen is available, insufficient quantities of reactive oxygen species such as HOCl and H2O can be produced by the PMN's lysosomes, and eventually the PMN succumbs to the bacteria it phagocytized. HBOT restores tissue oxygen tension and the function of the PMN. Moreover, this will prevent the often dramatic evolution of anaerobic tissue infection. Oxygen itself is not bactericidal at the pressures used in clinical HBOT, but it acts as a bacteriostatic agent, permitting or enhancing the action of appropriate antibiotic treatment.

Hypoxic tissues present deficient wound healing. Like the PMN, healing wounds have typically a greatly augmented oxygen need. In hypoxic tissues, several of the normal wound healing mechanisms are impaired. Wound healing typically occurs at the outer edges of the wound, where a steep oxygen gradient exists - the centre of the wound having a very low oxygen tension. This oxygen gradient stimulates the formation and excretion of Macrophage Derived Growth Factor (MDGF), which in turn stimulates the multiplication of fibroblasts. Fibroblasts need sufficient oxygen tensions, not for the secretion of pro-collagen (which has been shown to occur at oxygen tensions of almost zero), but for their own multiplication. Furthermore, the extracellular polymerisation of the single pro-collagen chains to a helicoïdal triple collagen fibre, is dependent on oxygen. In other words, the both quantity and quality of the collagen deposited are dependent on sufficient oxygen tensions. Clinically, hypoxic wounds present with fragile and insufficiently irrigated granulation tissue, if any at all. Without a proper wound granulation, epithelialisation will not occur, and the open wound surface is very infection-prone.

Indications

Many of the effects of HBOT are subtle and only adjutant to normal disease repair processes in the human body. Therefore, it is not easy to determine the exacts therapeutic role of HBOT in many diseases. HBOT as a clinical therapeutic modality is only performed and studied since about 1960, when Prof. Boerema started to use HBOT in the treatment of Gas Gangrene. About 130 different indications have been proposed, and sometimes accepted for some time. However, with the conducting of more well-designed clinical studies, most of these indications have been abandoned. An international consensus withholds now some twelve different diseases where HBOT provides a major benefit and often is the only treatment available. It lies beyond the scope of this presentation to give a detailed discussion of each of these indications, but those indications that in Belgium are most regularly treated, are briefly discussed below.

1. Carbon Monoxide (CO) intoxication

CO is a gas that is potentially produced by all appliances that combust carbon chains, be it gas, fuel, wood, coal... It is formed in greater quantity as the burning conditions are unfavourable, i.e. when little oxygen reaches the flame. This can be caused by insufficient evacuation of burn exhaust fume, or by insufficient oxygen availability in the environment (small, hermetically isolated or closed spaces), typically when the flame is "set to a low level". When inhaled, CO fixes itself to the haemoglobin molecule, thus preventing the fixation of oxygen for transport. Since the fixation of CO is much more specific than that of oxygen, only little CO (about 100ppm) needs to be present in the inspired air to cause a severe oxygen transport problem. When less CO is present, the victim usually is exposed for a longer time, permitting the transport of CO via the plasma (see Henry's Law) to the tissue cells, where it can bind to other haeme proteins in the mitochondria; in this way, CO "blocks" the cellular respiration much like cyanide does. This explains why some CO intoxicated patients present with serious neurological or cardiological symptoms despite a low carboxy-haemoglobin level. This also explains why a foetal intoxication must always be regarded as serious: not only is the foetus already in relative hypoxic conditions in normal circumstances, but the elimination of CO from the foetus' blood and tissue cells will always be retarded by 4 hours or more to that of the mother. 4 absolute indications are withheld for HBOT treatment of CO intoxicated patients: a) when the patient has been unconscious (even if he regained consciousness after oxygen administration) b) when the patient presents with objective neurological signs at the emergency room (vomiting, photophobia, impairment of consciousness, altered peripheral tendon reflexes,...) c) when the patient presents with cardiac complaints d) when the patient is pregnant

2. Clostridial gas gangrene

This is a soft tissue infection caused by a strict anaerobic micro-organism, Clostridium Perfringens. "Classical" gas gangrene involves only the muscle tissue; however, there are numerous mixed bacterial and soft tissue infections that involve Clostridia and other anaerobic and even aerobic bacteria, in a process called "synergism" where one bacteria facilitates the growth of another by creating the correct hypoxic tissue conditions. Although HBOT has no bactericidal effect, it permits the arrest of the rapid progression in a clostridial infection, by causing the bacteria to sporulate in these normal or high oxygen tensions. In the sporulated state, no tissue-destructing toxins are produced, and the infection is effectively halted for the time of the treatment. In these infections, where the evolution may be dramatic - death may incur over a 36 hours period - several HBOT treatments are given per day, and typically alternated with surgical debridements. Antibiotic therapy must always be large and early, but the addition of HBOT permits the saving of lives and high amputations, by an early control of infection.

3. Arterial Gas Embolism

The introduction of large amounts of gas in the arterial bloodstream, be it by lung rupture, traumatic or iatrogenic causes, obstructs the flow of blood to large areas of (typically) the brain. HBOT is the only treatment that can at once rapidly remove air or gas bubbles and reoxygenate the tissue, avoiding or attenuating ischaemia-reperfusion phenomena. Unfortunately, HBOT has to be started within 6 hours to have a significant chance of success. Often, ignorance or haemodynamic instability prohibit this early HBOT treatment, compromising life or physical integrity. However, spectacular results have been obtained by late HBOT, even after several days.

4. Decompression Sickness

SCUBA divers, after surfacing, sometimes suffer from venous or arterial nitrogen bubble emboli, causing a syndrome known as Decompression Sickness (or Illness). The cause is actually related to nitrogen being dissolved in the body tissues (according to Henry's Law) during the dive, and being "washed out" while ascending and surfacing. If for any reason the diver does not permit sufficient time for a gradual wash out of nitrogen, bubbles will be formed, obstructing the blood flow. Symptoms are those of tissue ischaemia, typically in the central nervous system. However, all body tissues may be affected, and the diagnosis is therefor sometimes extremely difficult to make for a non experienced diving physician. The treatment is recompression while breathing oxygen, according to specialised "treatment tables" that may take 6 or more hours to complete.

5. "Problem Wounds"

These wounds may be skin ulcers due to arterial insufficiency, diabetic arteriopathy, or radiation induced skin or bone radionecrosis. The hypoxic nature of these wounds can be demonstrated by means of transcutaneous oxygen tension measurements, and the application, once or twice daily, of HBOT to permit normal oxygen tensions in and around the wound, in many cases permits normal wound healing to resume. These treatments are however quite tedious and ask for great motivation both from the patient's as from the physician's side. Often, thirty or more HBOT sessions are necessary. Skin grafting or other surgical procedures can be necessary, and optimisation of the "classical" treatment (including surgical revascularisation where possible) is mandatory.

In the case of soft tissue or osteoradionecrosis, three gradations can be distinguished. Grade I presents only with pain (presumed to be of ischaemic origin). HBOT, when started at this stage, can probably cure these patients. Unfortunately, they are most often not referred, and progress gradually to Grade II, where soft tissue defects and infection is present. Grade III patients develop tissue necrosis and bone sequesters, and will always necessitate surgical correction. However, thirty HBOT sessions are given prior to surgery, followed by 10 post-operative daily sessions, to optimise woundhealing and the "take" of grafts or flaps.

6. Sudden Deafness

Sudden Idiopathic Sensorineural Deafness is a syndrome with a great psychological and functional impact. It often strikes young patients, in good general health. Hearing loss is most often unilateral and severe. Over the course of 2 to 3 days, about 60% of the patients recover their hearing function spontaneously, and the cause is then presumed to be a limited cochlear vasospasm. Many medical treatments have been proposed, but the only therapeutic scheme that seems to stand the test of time consists of a high-dose intravenous steroid therapy, combined with normovolemic haemodilution if the haematocrit is above 45%. Animal experiments have shown that the hearing loss is accompanied by a profound hypoxia in the cochlear fluid, the endolymph, and that HBOT, but not normobaric oxygen, is capable of restoring almost normal oxygen tensions. Clinical studies are extremely difficult, since there is no way of predicting which patients will recover spontaneously and which patients will not. We are currently conducting a randomised study where patients are assigned to either no further therapy or HBOT after a 7 days steroid treatment course, if no amelioration has been noted. Preliminary results indicate still a very substantial and sometimes spectacular benefit from HBOT, even after sometimes more than 10 days interval.

7. Other indications

Other indications, less often treated in the Belgian setting - because of the relative lack of suitable HBOT chambers and (mainly) because of the lack of awareness of the Belgian physicians - are chronic refractory osteomyelitis (where HBOT could optimise the action of the administered antibiotics), thermal burns (where HBOT could, if applied early, prevent the conversion of 2nd to 3rd degree burn), and the salvage of compromised tissue grafts and flaps (by direct oxygen supply and resolution of pedicular oedema). Finally, the use of HBOT in severe anaemia by blood loss, in cases where either no suitable blood is available or is refused by the patient because of religious reasons, can be considered.

Side effects

Molecular oxygen has toxic effects, and these should be known. HBOT is aimed at administering a maximal effective dose, while staying under the toxic threshold. Perhaps the most evident sign of the toxicity of oxygen is the protective vasoconstriction that takes place during HBOT. It is probably mediated via the inactivation of Nitric Oxide (NO, a vascular tone regulator with vasodilating action) by Superoxide Anion, an oxygen free radical (OFR) produced at the endothelial cell surface.

Clinical oxygen toxicity presents itself under different forms.

The most spectacular side effect is the neurological toxicity, where the patient suffers from a sudden epileptic insult, grand-mal type, during oxygen breathing. Although there is a considerable inter- and intra-individual variation as to the sensitivity to oxygen convulsions, they never occur at a pO2 of less than 2 atmospheres (absolute pressure). They become very likely above 3 atmospheres. HBOT usually is performed at 2.5 atmospheres pO2, and oxygen convulsions occur once in every 2.000 to 5.000 treatments. Fever and stress seem to render individuals more sensitive, although it can occur in perfectly healthy people. There is no treatment necessary: the convulsions last usually less than 1 minute and no neurological damage whatsoever occurs. There may be lesions resulting from tongue-bite or by falling or hitting objects during the convulsions. Oxygen administration is discontinued and the patient is slowly decompressed after the convulsions have stopped, to prevent lung overexpansion syndromes (closed glottis during the tonic and clonic phases). The treatments may be resumed normally afterwards, although usually a slightly lower pressure is used.

Pulmonary oxygen toxicity is even less frequent in the therapeutic schedules used. Although this does occur at lower pO2s - from 0.5 atmosphere onwards, the onset of toxicity is much more gradual and the time before any significant toxicity occurs is usually much longer (6-10 hours) than the duration of the treatment sessions (typically 1.5 hours). In addition, the hyperoxic exposures do not seem to have an additive effects, and no pulmonary oxygen toxicity is seen in patients who have undergone 100 or more HBOT sessions. However, during long treatments (e.g. for decompression sickness) or when the patient is placed on high-flow oxygen (FiO2 > 0.5 atm) in between HBOT sessions, this toxicity may become clinically evident. First, a dry cough is noted, followed over time by a progressive reduction in functional ventilatory capacity (Vital Capacity). When allowed to go on, irreversible lung tissue fibrosis may result.

Ocular lens manifestations are frequent in elder patients who are treated with daily HBOT sessions. A progressive myopia occurs usually after more than 20 sessions. This is due to a swelling of the ocular lens, probably by involvement of oxygen free radicals (OFR), although the precise mechanism is not known. This myopia is almost completely reversible in the weeks that follow the completion of HBOT. Cataract progression is a more serious complication, that may occur in elder patients who already have a light form of nuclear cataract. In these cases, the benefits of long term HBOT must be outweighed against the possibility of an early cataract operation.

Treatment in a pressurised environment exposes the patient to possible complications due to pressure changes. The most common side effect is middle ear barotrauma or ear squeeze. This is due to an insufficient middle ear pressure equalisation during compression, usually because the patient has not sufficiently (well) performed the Valsalva manoeuvre or other manoeuvres to equalise middle ear pressure through the Eustachian tube. These manoeuvres are taught to all patients before the start of the treatment, and the inside attendant observes every patient carefully during the first phase of the treatment (where the pressure in the HBOT chamber is gradually increased). When necessary, the pressure increase can be temporarily halted until the patient is able to "clear the ears". The incidence of this complication lies below 2%, provided the chamber personnel (both inside attendant and chamber operator) are attentive and motivated. In monoplace chambers, this incidence is much higher, on one hand because the patient is isolated and can be less carefully observed, and because usually the patient is in a horizontal position, where the blood shift causes a relative swelling of the nasopharynx mucosa, thus impeding the easy opening of the Eustachian tube. Other possible pressure-related complications are sinus squeeze, dental squeeze, and pulmonary overinflation (a possible complication in patients with lung bullae or bronchial stenoses). Patients are screened for the possibility of these complications before the treatment.

A rare but severe complication can occur in patients with chronic bronchitis, who have developed a so-called "hypoxic ventilatory drive". Whereas in normal individuals the "drive" to breathe is directed by the rise in pCO2, these patients, because of their low pulmonary gas exchange capacity (emphysema) and subsequently chronic hypercapnia and hypoxia, have become "insensible" to these high CO2 tensions, and have become "hypoxia-driven". If these patients are given oxygen, even in low quantities, they lose this ventilatory drive force and literally stop breathing. Of course, when HBOT is necessary for such patients, often preventive intubation and artificial ventilation is necessary.

Conclusions

Hyperbaric Oxygen Therapy is a therapeutic modality that presents substantial benefits, when applied for the correct indications and in a correct manner. Further study is necessary and is currently conducted both with regard to the mechanisms of action and to the precise place of HBOT in these and other diseases. Falsely high expectations are equally wrong as a denial of its therapeutic possibilities.