Oxygen Toxicity

Vol. 21, No. 4 April 1999

CE

V

20TH ANNIVERSARY

Refereed Peer Review

Oxygen Toxicity FOCAL POINT ★Although prolonged exposure to high alveolar oxygen (O2 ) concentrations can lead to pulmonary injury, O2 therapy should not be withheld from hypoxemic patients.

KEY FACTS ■ Under normal conditions, O2 supplementation has little benefit in increasing arterial O2 content if the measured partial pressure is higher than 70 mm Hg or the fraction of saturated hemoglobin is greater than 93%. ■ Prolonged use of high fractional inspired O2 concentrations (FiO2) can lead to tracheobronchial irritation, atelectasis, and decreased ability to transport O2 from the environment to the alveoli and from the alveoli to the blood. ■ Positive end-expiratory pressure and continuous positive airway pressure are two ventilatory modalities that can lessen the need for high FiO2. ■ Once changes begin in the lungs, higher O2 concentrations may be required to achieve desired levels in the blood, creating a vicious cycle of O2-induced lung injury.

Tufts University

Dove Lewis Emergency Animal Hospital Portland, Oregon

Steven Mensack, VMD

Robert Murtaugh, DVM, MS

ABSTRACT: Molecular oxygen (O2) manifests its toxic effects through the production of free radicals. If an animal breathes a high fractional inspired O2 concentration (FiO2), the increased production of free radicals can overwhelm the endogenous antioxidant systems. Depending on the atmospheric pressure by which O2 is delivered, clinical signs of toxicity may be exhibited in the lungs or central nervous system. Under conditions of normal atmospheric pressure, the lungs are the primary target for manifestations of toxicity. Clinically, increased work of breathing, decreased tidal volume, and increased arteriovenous shunting are manifestations of O2 toxicity. Although there is no therapy, several methods are being investigated to ameliorate the pathologic changes associated with prolonged exposure to toxic O2 concentrations. Until treatment methods are available, judiciously limiting exposure to high Fio2 is the key to prevention.

S

ince the discovery of oxygen (O2) in the late 18th century, scientists have postulated that although it is necessary for life, too much exposure can be detrimental. As human and veterinary critical care medicine have advanced, the use of supplemental O2 therapy has become more common. With increased O2 use comes the need to recognize possible complications. This article addresses the biochemical and pathophysiologic effects of prolonged use of high O2 concentrations. The primary focus is on the pulmonary effects of O2 delivered at normal atmospheric pressure (normobaric toxicity), although the toxic neurologic effects of hyperbaric (high-pressure) oxygen (HBO) therapy is also addressed. Finally, prevention and therapy for O2 toxicity and tolerance to high O2 concentrations are discussed.

RATIONAL USE OF OXYGEN The use of supplemental O2 therapy has gained widespread acceptance in veterinary medicine. Indications for supplemental O2 include providing supportive care for anesthetized patients; increasing the O2 content in blood during periods of hypoxemia; and aiding in the healing of chronic complicated wounds, acute traumatic soft tissue injuries, and serious skin wound infections (HBO therapy). Oxygen Transport To understand when supplemental therapy is necessary, it is important to understand normal O2 transport. One of the primary purposes of both the respiratory and circulatory systems is the transport of O2 from the outside environment to tissues. In the lungs, O2 diffuses down its concentration gradient from the alveolus into the blood. Once in the blood, O2 is transported to the tissues, where it again diffuses down its concentration gradient to be used for normal cellular metabolism. The amount of O2 delivered to a particular tissue depends

Small Animal/Exotics

20TH ANNIVERSARY

Oxygen Saturation of Hemoglobin (%)

100 90 80 70 60 50 40 30 20 10 0 0

10

20

30

40

50

60

70

80

90

100

Partial Pressure of Oxygen (Po2) (mm Hg) SaO2 (normal) SaO2 (left shift) SaO2 (right shift)

Figure 1—The oxyhemoglobin dissociation curve. A left shift from normal is caused by decreased body temperature, increased blood pH (alkalosis), decreased partial pressure of carbon dioxide (PCO2) in blood, or decreased red blood cell (RBC) concentration of 2,3-diphosphoglycerate (2,3-DPG). A right shift from normal is caused by increased body temperature; decreased blood pH (acidosis), increased PCO2, and increased RBC concentration of 2,3-DPG (O2 = oxygen; SaO2 = fraction of saturated hemoglobin).

on the amount entering the lungs, efficiency of pulmonary gas exchange, blood flow to the tissue, and ability of blood to carry O2. Under standard conditions (barometric pressure [Patm], 760 mm Hg; fractional inspired oxygen concentration [FiO2] in room air, 21%), each 100 ml of arterial blood carries approximately 19.5 ml of O2,1,2 which can be calculated by the following equation: O2 content = (1.34 × [Hb] × SaO2) + 0.003 PaO2 where 1.34 = the amount (in ml) of O2 carried by 1 g of hemoglobin (Hb), Hb = the concentration of hemoglobin in blood (normal, 15 mg/dl), SaO2 = the fraction of saturated Hb (normal, 93% to 97%), and PaO2 = the partial pressure of O2 dissolved in arterial plasma (normal, 85 to 105 mm Hg).1–6 As calculated by this equation, the majority of O2 carried by blood is bound to Hb (19.2 ml) and very little is dissolved in plasma (0.3 ml).1–4 The SaO2 can be measured by cooximetry or estimated by pulse oximetry, the Hb concentration attributable to red blood cells (RBCs) can be measured by a hemoglobinometer or estimated by multiplying the measured hematocrit by one third, and the PaO2

Compendium April 1999

can be measured by arterial blood gas analysis. The ability of blood to carry O2 is profoundly affected by the ability of Hb to bind O2. The oxyhemoglobin dissociation curve (Figure 1) is sigmoid shaped and represents the relationship between SaO2 and the partial pressure of O2 (PO2).1,3,4,7 Under normal conditions, the plateau of the curve occurs at a PO2 of approximately 70 mm Hg. An increase in the PO2 above this level causes a minimal increase in the O2 saturation of Hb. However, decreases in PO2 below 60 mm Hg have increasingly negative effects on O2 saturation of Hb.5 The ability of Hb to bind O2 can be altered by several pathophysiologic variables: blood pH; body temperature; the partial pressure of carbon dioxide (P CO 2 ) dissolved in plasma; and in dogs, changes in the concentration of the RBC carbohydrate 2,3-diphosphoglycerate (2,3-DPG).5 A right shift of the curve indicates that an increased PO2 is needed for Hb to bind a specific amount of O2. Increased body temperature, decreased blood pH, and increased P CO 2 all shift the curve to the right.1,3,4,7,8 An increased 2,3-DPG also shifts the curve to the right.1,3,4,7,8 Conditions that increase the RBC concentration of 2,3-DPG include hyperthyroidism, anemia, chronic exercise, and chronic hypoxia.1 The PO2 at tissue level is lower than it is in arterial blood.5 Thus, a right shift favors the off-loading of O2 from the Hb molecule.5 A left shift of the curve represents increased affinity of the Hb molecule for O2 and decreased O2 delivery to tissues. Decreased body temperature, increased blood pH, decreased PCO2, and decreased RBC concentration of 2,3-DPG shift the curve to the left.1,3,4,7,8 Stored RBCs contain decreased amounts of 2,3-DPG.3 This decreased level is not a problem in stored feline blood because the release of O2 is independent of 2,3-DPG.9

Indications for Supplemental Oxygen Therapy The most common nonanesthetic use of supplemental O2 is for treating or preventing hypoxemia. Hypoxemia is a relative deficiency of O2 tension in arterial blood (decreased PaO2 ) and becomes significant when the PaO2 is lower than 70 mm Hg.2,6 Hypoxia, another common but not interchangeable condition, is a relative deficiency of O2 in tissues and can be caused by multiple factors, one of which is hypoxemia.1,10 Multiple known causes of hypoxemia include low FiO2, hypoventilation, diffusion impairment, pulmonary ventila• • tion/perfusion (V/Q) inequity, intrapulmonary shunting of blood, and certain toxins that may inhibit O2 uptake in the lungs (see Causes of Hypoxemia). These causes are not mutually exclusive. Hypoventilation, which decreases O2 delivery from the environment to the lungs, can be caused by drugs

OXYHEMOGLOBIN DISSOCIATION CURVE ■ NONANESTHETIC OXYGEN SUPPLEMENTATION

Compendium April 1999

20TH ANNIVERSARY

(e.g., narcotics and barbiturates that depress the respiratory centers of the brain),3,6,11,12 thoracic wall trauma,3,7 pleural space disease (e.g., pneumothorax, hemothorax, pleural effusion, diaphragmatic hernia with abdominal content displaced into the pleural space),7,11 central nervous system (CNS) trauma,12 upper airway obstruction,7,11 and neuromuscular disease (e.g., polyradiculoneuritis, myasthenia gravis).3,7 Diffusion impairment develops when equilibration cannot occur between alveolar gas and pulmonary capillary blood because of a thickened alveolar wall or decreased contact time between blood and gas in the alveolus.7 Conditions such as pulmonary fibrosis and pulmonary edema can cause diffusion impairment.11 • • Pulmonary V/Q inequity occurs when areas of the lungs are receiving adequate fresh gas from the environment but• blood flow is inadequate to effect gas ex• change (V/Q higher than 1) or when areas of the lungs are not receiving fresh gas from the environment but are receiving a normal supply of fresh blood for gas ex• • /Q lower than 1).1,3 Conditions that lead to change (V • • V/Q mismatching include atelectasis, alveolar pneumonia, pulmonary edema of any cause, and pulmonary thromboembolism.7,11,12

Small Animal/Exotics •

Causes of Hypoxemia ■ Hypoventilation (less oxygen delivered from the environment to the lungs) — Drugs (narcotics and barbiturates) — Thoracic wall trauma (rib fractures) — Pleural space disease (pneumothorax, hemothorax, pleural effusion, diaphragmatic hernia) — Central nervous system trauma — Upper airway obstruction (foreign body, laryngeal paralysis, edema, neoplasia) — Neuromuscular disease (polyradiculoneuritis, myasthenia gravis) — Diffusion impairment (thickened alveolar septa or decreased transit time of blood through pulmonary capillaries prevents equilibration of oxygen between alveolus and blood) — Pulmonary edema — Pulmonary fibrosis •

•

■ Ventilation/perfusion (V/Q) mismatching (excessive • • blood flow to a region of underventilated lung [V/Q

•

Intrapulmonary shunting, an extreme form of V/Q mismatching, occurs if blood in the pulmonary circulation totally bypasses ventilated lung tissue before re• • turning to systemic circulation (V/Q = infinity).1,3,11,12 Conditions that cause intrapulmonary shunting of blood include lung lobe consolidation, arteriovenous fistulas, and right-to-left intracardiac shunts.11,12 O2 therapy can be beneficial to some extent in correcting hypoxemia caused by any of these pathophysiologic processes except intrapulmonary shunting.1,3 Supplemental O2 therapy may also have limited benefit for severe anemia and acute hemorrhage.11 As indicated by the equation presented earlier, approximately 97% of O2 in blood is in the form of oxyhemoglobin, whereas the remainder is dissolved in plasma. With severe anemia or acute hemorrhage, the amount of Hb is reduced and the relative contribution of dissolved O2 to overall O2 content in blood is greater. For each increase of 100 mm Hg in PaO2, the overall increase of dissolved O2 in blood is very small (approximately 0.3 ml O2 per 100 ml blood).1 Ultimately, therapy for severe anemia involves replacing Hb in the form of RBCs and/or an O2-carrying plasma-phase Hb solution (Oxyglobin®; Biopure, Cambridge, MA). In the interim, supplemental O2 therapy may help stabilize a

•

•

lower than 1] or excessive capillary blood low [V/Q higher than 1]) — Atelectasis — Alveolar pneumonia — Pulmonary edema — Pulmonary thromboembolism — Asthma ■ Arteriovenous shunting (blood in pulmonary circulation bypasses ventilated lung tissue before returning to systemic circulation) — Atelectasis and lung lobe consolidation — Arteriovenous fistula — Right-to-left intracardiac shunt ■ Low fractional inspired oxygen concentration — High altitude — Administration of nitrous oxide ■ Toxins — Carbon monoxide — Methemoglobin

HYPOVENTILATION ■ DIFFUSION IMPAIRMENT ■ INTRAPULMONARY SHUNTING

Small Animal/Exotics

patient until additional Hb can be administered. In patients with low cardiac output, hypotension, hyperthermia, or problems with cellular O2 uptake, supplemental therapy may have limited benefit. In these patients, therapeutic considerations should be aimed at correcting the underlying circulatory problem with the provision of intravascular volume expansion, positive inotropes, or pressor agents.

20TH ANNIVERSARY

Compendium April 1999

TABLE I Approximate Flow Rates of Oxygen Administered by Nasal Cathetera Oxygen Flow Rate (L/min) Patient 30% to 50% 50% to 75% 75% to 90% Weight (kg) FiO2 FiO2 FiO2

a

0–10

0.5–1.0

1.0–2.0

3.0–5.0

10–20

1.0–2.0

3.0–5.0

> 5.0

20–40

3.0–5.0

> 5.0

Unknown

From Court MH: Respiratory support of the critically ill small animal patient, in Murtaugh RJ, Kaplan PM (eds): Veterinary Emergency and Critical Care Medicine. St. Louis, Mosby Year Book, 1992, p 577 from data derived from Fitzpatrick RK, Crowe DT: Nasal oxygen administration in dogs and cats: Experimental and clinical investigations. JAAHA 22:293–300, 1986. Reprinted with permission.) FiO2 = fractional inspired oxygen concentration.

aid in healing chronic complicated wounds, acute traumatic soft tissue injuries, and serious skin wound infections.16,19 With these types of injuries, sufficient O2 may not reach areas of reduced blood circulation. HBO therapy allows increased O2 delivery to tissues by increasing the amount dissolved in plasma (up to 20 times normal),16,17,19–21 which is beneficial because O2 dissolved in plasma diffuses into tissues more readily. Increased O2 supply to damaged tissue leads to increased antimicrobial effects, induction and propagation of angiogenesis, vasoconstriction without O2 loss to tissues (which helps prevent edema), and fewer tissue bubbles.16,17,19–21 HBO therapy may also prove to be useful treatment of certain toxicoses (e.g., cyanide or carbon monoxide poisoning).17,19–21

Oxygen Delivery Oxygen delivery should be based on the FiO2 required, the patient’s condition, and available equipment. Endotracheal intubation (or tracheostomy tube) connected to Cytochrome Oxidase an O2 source with or without concomitant use of any – + e– e– 2H+ e– H+ mechanical ventilation can –e H – O O H O OH H2O 11 2 2 2 2 achieve 100% FiO2. O2 delivered via mask or ElizaH2O bethan collar fronted with plastic wrap can attain 60% FiO2,2,7 although nearly 100% Figure 2—Metabolism of oxygen (O2 ) to water. The sequen- OXYGEN METABOLISM tial addition of single electrons to the O2 molecule yields toxAND PRODUCTION can be achieved with a well- ic free radicals. Cytochrome oxidase within the mitochondria fitted mask that contains a prevents these intermediates from escaping and causing dam- OF TOXIC OXYGEN reservoir bag.2,6,11,13 age to cells (e – = electron; H + = hydrogen atom; H2O = water; METABOLITES Oxygen can also be deliv- H2O2 = hydrogen peroxide; O2 – = superoxide radical; OH – Oxygen normally exists as ered by nasal cannulation; = hydroxyl radical). (From Rochat MC: An introduction to a molecule of two O2 atoms depending on the flow rate, reperfusion injury. Compend Contin Educ Pract Vet with two unpaired electrons in the outer orbital. These patient tolerance, and 13(6):925, 1991.) electrons prevent the whether one or two cannulas molecule from being highly reactive because of their are used, up to 90% Fi O 2 can be achieved (Table I).7,11,14,15 Intratracheal catheterization has been reported parallel spins, a concept called spin restriction.2,5,22 Norto attain 80% to 90% FiO2 while using lower flow rates mal cellular O2 metabolism involves the stepwise addithan that used for nasal cannulation.15 Commercial tion of four single electrons (Figure 2), a process called reduction.2,5,22–24 During reduction, each addition of an cages typically do not exceed 50% FiO2.7,10,11 electron to O2 creates a reactive O2 species; these reacHyperbaric Oxygen tions normally occur under controlled conditions withHyperbaric oxygenation is just beginning to be used in the cell in the presence of catalysts, usually cyin veterinary medicine. This therapy involves placing tochrome oxidase.5,24 Under conditions of normal PaO2 (100 mm Hg), 95% of the molecules are completely repatients in a chamber and supplying 100% O2 at pressures greater than 1 atm at sea level (760 mm Hg).16–21 duced to water2,22,24 and the remaining molecules are A common treatment protocol involves administering partially reduced, generating toxic metabolites.24 These O2 at 2.0 to 2.4 atm (1520 to 1800 mm Hg) for 40 to reactive molecules have a high affinity for other elec60 minutes once or twice daily.16 trons within surrounding molecules, potentially resultThe most common indications for HBO therapy are to ing in oxidative damage to the molecules. ACHIEVING OXYGEN REQUIREMENTS ■ HYPERBARIC OXYGEN THERAPY ■ OXYGEN METABOLISM

Compendium April 1999

20TH ANNIVERSARY

Cellular O2 metabolism involves the addition of single electrons to molecular O2, yielding sequential reactive O2 intermediates.5,23,24 The addition of the first electron yields the superoxide anion (O2–), a very reactive product. The addition of a second electron (in the presence of hydrogen atoms) yields hydrogen peroxide (H2O2 ), a less reactive intermediate. The addition of a third electron yields the very highly reactive hydroxyl radical (OH–) and a molecule of water; through the addition of the fourth electron, OH– is finally reduced to water. Within the mitochondria, cytochrome oxidases that contain copper or iron bind most of the molecules involved in the reduction reactions, preventing the escape of reactive O2 species generated during the reactions.2,22,23 Under normal cellular conditions, approximately 5% of the O2 metabolized within the mitochondria is only partially reduced, generating toxic free radicals (FRs)2,22 that leak into the cytosol and out from the cell, possibly through anion channels.2,22 O2-derived FRs in the cytosol or outside the cell undergo further reactions. Two superoxide anions can spontaneously react to form H2O2. In the presence of iron or copper, O2– may react with existing H2O2 to produce OH– and singlet oxygen (1O2), a reaction known as the Fenton reaction.2,24,25 These FRs (O2–, OH–, 1O2) are believed to cause much of the cellular damage observed with O 2 toxicity. 25 H2O2 is a less reactive toxic intermediate because it contains no unpaired electrons. Thus, it is not an O2derived FR in the truest sense. However, it is highly reactive and does have the capacity to cross biologic membranes, enabling the more damaging FRs to form at sites away from where the toxic intermediates are initially produced.26,27 Because FRs are products of normal cellular metabolism, biochemical defense mechanisms protect the organism from excessive FR damage.

THE ANTIOXIDANT SYSTEMS Two types of defense mechanisms protect cells from damage by O2-derived FRs: intracellular enzymatic systems and FR scavengers. The primary enzymatic defenses consist of superoxide dismutase (SOD), catalase, and glutathione peroxidase (GPO).2,13,22–24 These enzymes eliminate O2– and H2O2 and prevent FR chain reactions by decreasing the available levels of FRs that initiate the process.2,24,25 SODs act on the superoxide anion to produce H2O2.22,28 This system is found in both the mitochondria and cytosol of cells. Catalase acts on H2O2 to form water and an O2 molecule without the creation of OH–.22,23 The GPO system also acts on H2O2 to directly form water and molecular O2.22,23 Unlike catalase, GPO can recycle itself through an energy-dependent mechanism that involves glutathione reductase.22 Free-radical scavengers include α-tocopherol (vitamin

Small Animal/Exotics

E), ascorbic acid (vitamin C), niacin (vitamin B6), riboflavin (vitamin B 2), vitamin A, and plasma proteins.2,22,25 These antioxidants, which act to stop FR chain reactions by accepting electrons, are mainly found in extracellular fluid and plasma and are primarily effective at stopping lipid peroxidation of cellular membranes.24 When exposed to increased O2 concentrations, both intracellular and extracellular defense systems are overwhelmed by the increased production of O2-derived FRs and increased cellular injury occurs.

CELLULAR OXYGEN INJURY Inside cells and tissues, O2-derived FRs disrupt numerous cellular processes.2,22,24,29 The targets of oxidative attacks include lipids, proteins, and nucleic acids. Lipids (e.g., pulmonary surfactant and those found in biologic membranes) react with these FRs to produce lipid peroxides.5,22,24,25,29 These products lead to increased membrane permeability, inactivation of surfactant, and inhibition of normal cellular enzyme processes and can damage proteins and intracellular membranes.5,22,30 Proteins are also subject to direct damage. Protein synthesis may be decreased through inhibition of ribosomal translation, or formed proteins may be destroyed through oxidative processes.22,24,30 Consequently, inactivation of intracellular enzymes and transport proteins occurs, leading to impaired cellular metabolism and accumulation of cellular waste products.30 O2-derived FRs can cause breaks in DNA and inhibit the enzyme systems involved in repairing or replicating DNA.22,24,30 All this damage culminates in cellular death. PULMONARY OXYGEN TOXICITY The lungs are the primary organ affected by high FiO2 because they act as a barrier that prevents the remainder of the body from experiencing high O2 concentrations. The amount and type of damage that occurs in the lungs depend on the FiO2 and the duration of exposure to high O2 concentrations. Based on current data, the sequence of morphologic changes that occur in the lungs in response to toxic O2 concentrations is apparently conserved across species.31–35 However, the severity and time to onset of pulmonary changes show both species and individual variation.36 These data need to be interpreted cautiously because most of the research has been conducted in laboratories on rats, primates, and healthy humans and on neonatal or terminal human patients in clinical settings. Morphologic changes in lungs exposed to hyperoxic conditions follow a sequence of phases: initiation, inflammation, destruction, proliferation, and fibrosis (see Phases of Pulmonary Oxygen Toxicity). During initiation, the

CELLULAR METABOLISM ■ TOXIC FREE RADICALS ■ ANTIOXIDANTS

Small Animal/Exotics

20TH ANNIVERSARY

Compendium April 1999

Phases of Pulmonary Oxygen Toxicity ■ Initiation — Increased production of toxic oxygen (O2) metabolites — Depleted antioxidant stores — No evidence of lung injury ■ Inflammation — Destruction of pulmonary endothelial lining — Increased inflammatory mediators to the site of injury — Development of pulmonary edema ■ Destruction — Release of soluble inflammatory mediators — Amplification of destruction of pulmonary endothelial lining — Phase of O2 toxicity most associated with mortality

production rate of toxic FRs in lung tissue increases considerably37 because of increased O2 metabolism in the cells,38 which depletes cellular storage of enzymes, vitamins, and other substances involved in combating FR-derived injury. However, there is no morphologic evidence of pulmonary injury.25 During this time, the flow of tracheal mucus decreases, which may impair the clearance of debris from the lower airway and predispose the patient to respiratory infection. The length of the initiation phase varies inversely with the concentration of O2 delivered; in rats, the time course is 24 hours at an FiO2 of 100%, 72 hours at 85% FiO2, and up to 7 days at 60% FiO2.31,32 The remaining phases follow in sequential order until the patient either dies from respiratory compromise or the hyperoxic conditions are resolved. Morphologic changes in the lungs represent a continuum. The first change is accumulation of plasma in the pericapillary spaces secondary to damage of the endothelial lining.31,33 Accumulation of platelets and then of neutrophils in the pulmonary vasculature and interstitium follows (inflammation phase).13,30,31,33,39 These cells, along with pulmonary macrophages and damaged capillary endothelial cells, release soluble mediators of inflammation, thereby amplifying the destruction of endothelial cells (destruction phase).13,30,31,33,39 If a patient survives the destruction phase, the remaining capillary endothelial cells hypertrophy and other cell lines proliferate. The number of monocytes and type II (surfactant-secreting) alveolar epithelial cells increases (proliferation phase; (Figure 3).13,30,31,33,39 Finally, collagen deposition in the lung interstitium, thickness of the alveolar interstitial space,

■ Proliferation — Evident after prolonged exposure to less than 100% O2 concentrations (60% to 85% fractional inspired oxygen concentration) — Increased monocytes — Increased type II alveolar epithelial cells (surfactant secreting) ■ Fibrosis — Permanent lung damage — Collagen deposition in lung interstitium — Increase in the thickness of pulmonary interstitial space — Increase in interstitial fibrosis

and interstitial fibrosis all increase (fibrosis phase).13,31,35 These changes occur in rodents and nonhuman primates after 48 to 96 hours of 100% O2 exposure.23 Because these changes are very similar to those described for patients with acute respiratory distress syndrome and because of the uniform use of supplemental O2 during respiratory failure, the histopathologic diagnosis of purely O2-induced pulmonary injury is very difficult. Clinically, O2 toxicity may be more difficult to define in veterinary patients because pulmonary function tests are difficult to perform. In healthy human volunteers exposed to 100% O2, clinical signs of substernal soreness, cough, sore throat, nasal congestion, and painful inspiration developed within 12 to 14 hours.5,23,38,40,41 Mucociliary clearance was shown to decrease in humans after 3 to 6 hours of 100% O2 exposure23,30,41 and in healthy dogs42 and cats43 after 72 hours, which can lead to an increased risk for pneumonia (a common complication in mechanically ventilated patients). During the initial 6- to 12-hour period in healthy humans, no changes in vital capacity (total amount of gas exhaled after a full inspiration), alveolar–arterial O2 gradient, pulmonary artery pressure, or total lung water were noted.25,44 The first clinically measured change in human volunteers was decreased vital capacity, which occurred within 24 hours after beginning to breathe 100% O2.23,30,40 This decrease became progressively more severe after 60 hours of exposure.5,23 In normal baboons, this decrease became evident after breathing 100% O2 for 48 hours, and total lung capacity was less than 50% of normal by day 6.33,45 After healthy human subjects breathed 100%

OXIDATIVE ATTACKS ■ SEQUENCE OF TOXIC PHASES

Small Animal/Exotics

20TH ANNIVERSARY

Compendium April 1999

O 2 for approximately 30 ple enzyme systems is dehours, the gas-diffusing capressed by an undefined pacity (a measure of gas exmechanism not involving change in the lungs) deFR generation. Decreased creased.46 enzymatic activity ultimateThese changes in lung ly leads to seizure activity. function are postulated to Seizures induced by occur from progressive atHBO therapy are rare, with electasis attributed to reports of seizures occurring washout of normal lung niin 0.03% 48 to 0.21% 49 of treatments in humans. When trogen, decreased surfactant seizure activity is noted, paactivity, and edema sectients should be removed ondary to endothelial cell damage. 5 The effects and Figure 3—This photomicrograph from a 3-year-old spayed from the chamber but not clinical time course of O2- female German shepherd with suspected oxygen (O2) toxicity until the seizure activity has induced functional changes demonstrates the proliferative phase of toxicity. The dog had ceased. Decompression of a in diseased lungs have not a history of chronic pneumonia, had been on a ventilator on patient during the tonic been elucidated because of two separate occasions, and had received high-concentration phase of the seizure can put O2 therapy for extended periods before being euthanatized. the confounding influence Note the similarity to changes associated with acute respira- the patient at risk for air or of various disease processes tory distress syndrome. Thin arrows = type II alveolar epithe- oxygen embolus. 20 There that occur in lung tissue. lial cells; thick arrow = hyaline membrane deposition. (Cour- has been no report of longterm clinical sequelae to Lower FiO2 (60% to 85%) tesy of Michael Hawes, DVM, Tufts University) has also been implicated in HBO-induced seizures. 21 However, pathologic CNS causing pulmonary damage changes have been noted in experimental animals,17 inin rodents, primates, and humans, although to a lesser cluding white matter necrosis with either pyknosis and extent and over a longer period (days to weeks).25,31,32,39,47 Although these lower concentrations can cause permahyperchromatosis of the neurons, vacuolization of the nent lung damage, they have not been shown to lead dicytoplasm, and simultaneous swelling of the perineural rectly to death. glial processes or lysis within the nerve cell’s cytoplasm and karryorhexis. NEUROLOGIC OXYGEN TOXICITY Several factors have been shown to predispose a paThe potential toxic effects of O2 are not limited to tient to seizure activity while receiving HBO therapy. the lungs. The CNS, primarily the brain, has been Candidates for this therapy should not be febrile or acishown to respond adversely to abnormally high PaO2, dotic because both states have been associated with a which is normally achieved using HBO therapy. One higher prevalence of seizure activity17,18; should not receive high concentrations of supplemental O2 just beof the major clinical manifestations of neurologic O2 toxicity is generalized seizures.3,17,18,20,21,48,49 Although fore HBO therapy because seizure activity and pulthe exact process leading to hyperoxia-induced seizure monary damage are more likely; and should not have a activity is unknown, three mechanisms may be inhistory of previous seizure activity.18 Seizure activity as17 sociated with HBO therapy has not been reported in volved. The first is decreased cerebral metabolism, leading to a decreased level of γ-aminobutyric acid the veterinary literature. (GABA), an inhibitory neurotransmitter. This premTHERAPY, PREVENTION, AND TOLERANCE ise is based on the fact that a decrease in the amount The only fully effective way of managing O2 toxicity of GABA in the brain occurs before the onset of seizure is to avoid it. Veterinarians can manage patients at risk activity and that the critical atmospheric pressures for for pulmonary toxicity by maintaining the lowest FiO2 onset of seizure activity and reduction in GABA occompatible with achieving adequate systemic and tissue cur concomitantly.17 The second postulated mechanism involves the generation of toxic FRs, which can oxygenation. There are, however, occasions when the lead to cellular membrane lipid peroxidation, inactivaneed for prolonged ventilation with an FiO2 of 60% or higher is required to maintain sufficient systemic oxytion of enzyme systems, and DNA denaturation. The genation. In these patients, it is important to attempt to changes induced by this damage lead to seizure activireduce the FiO2 by small increments to below this critity. The third postulated mechanism involves enzyme cal level as soon as is safely possible and to monitor for inhibition. With HBO therapy, the activity of multiHISTOPATHOLOGIC DIAGNOSIS ■ HYPEROXIA-INDUCED SEIZURES

Compendium April 1999

20TH ANNIVERSARY

S
M’

hypoxemia by serial arterial blood gas measurements. Several strategies have been proven useful in decreasing the need for toxic FiO2. The most common are positive end-expiratory pressure (PEEP) and continuous positive airway pressure (CPAP) delivered while the patient is receiving O2 from a ventilatory circuit.50 PEEP is supplied while the ventilator is performing the work of breathing, whereas CPAP is supplied while the patient is breathing spontaneously. With either mode, positive pressure is being maintained to the airways throughout the duration of the respiratory cycle, thereby preventing complete expiration. By preventing complete expiration, small airways remain open, alveolar size increases, and more alveoli are available for gas exchange. Because the majority of gas exchange between the blood and alveolus occurs during expiration, these • • measures help improve V/Q matching and decrease intrapulmonary shunting, leading to improved arterial oxygenation. Sound medical practice can also lower the FiO2 required. This includes maintaining proper sedation, correcting anemia, optimizing cardiac output to improve O2 delivery to tissues, treating fever or hyperthermia to decrease tissue O2 demands, diagnosing and treating infections, supplying proper nutrition, and providing good nursing care.51 ENDIU Nutrition is very imporMP tant in helping to prevent the toxic effects of O2, esANNIVERSARY pecially in ventilator-dependent patients.52 Vitamin or trace mineral deficiencies increase the susceptibility of organs to FR The discovery of sequential damage secondary to destages of oxygen-induced pleted stores of SODs or damage has helped to fuel FR scavengers. Protein dedramatic advances in the field ficiency potentiates toxiciof oxygen toxicity research over ty from hyperoxic expothe past 20 years. Delineation sure because of a lack of of these stages has allowed sulfur-containing amino researchers to make comparisons acids, which are necessary across species, evaluate the for glutathione synthesis. efficacy of various therapeutic Because patients that receive high FiO2 are usually interventions, and assess the on mechanical breathing ability of lung tissue to recover circuits, total or partial from oxygen-derived damage. parenteral nutrition must be provided intravenously or enteral nutrition through a feeding tube.53 No therapy currently exists to prevent or treat changes
CO

20th 9 9 9 9 - 1 1 9 7

A LookBack

Small Animal/Exotics

associated with O2 toxicity in lung or CNS tissue. Multiple strategies to prevent or ameliorate the toxic effects of acute pulmonary toxicity are currently under laboratory investigation. Investigations in rats and mice have shown that pentoxifylline (a methylxanthine derivative),54 high doses of magnesium sulfate,55 acetylcysteine, 56,57 liposomal-encapsulated antioxidant enzymes,58,59 deferoxamine (an iron chelating agent),29,60 and recombinant human tumor necrosis factor61 all provide some protection against pathologic changes in the lungs of rats or mice exposed to lethal O2 concentrations. Exogenous surfactant administration has been shown to prevent rabbits from developing pulmonary damage.62 The value of these drugs in the clinical setting awaits further investigation. Tolerance to high O2 concentrations has been reported. Rats have been able to increase the antioxidant defenses in their lungs after being exposed to an FiO2 of 85%.31 On subsequent exposure to lethal doses (100% FiO2), many survived prolonged exposures.31 However, lower toxic doses (60% FiO2) did not increase antioxidant defenses or prolong survival with subsequent exposure to lethal concentrations.25,63 A threshold of FR exposure is possibly needed to induce tolerance. Tolerance has not been shown to develop in dogs, mice, and guinea pigs, apparently making this a species-specific phenomenon.64

CONCLUSION The need for supplemental O2 therapy should always take precedence over concerns regarding toxic effects in the acute management of veterinary patients. The lowest FiO2 necessary to achieve adequate arterial oxygenation should be used. Arterial blood gas and pulse oximetry monitoring should be used to guide O2 administration. If prolonged periods of high FiO2 are anticipated, ventilatory adjuncts (e.g., CPAP or PEEP) should be initiated, even though toxic changes to the lungs sometimes occur. Once changes begin in the lungs, higher O2 concentrations may be required to achieve desired levels in the blood, creating a vicious cycle of O2-induced lung injury. Because O2 toxicity studies seldom use dogs and cats, guidelines for safe administration have been extrapolated from studies on other species. Current recommendations for safe administration are up to 24 hours with 100% O2 and up to 48 hours with 60% O2. These recommendations are based on current data that suggest hyperoxia-induced injury does not occur before these time intervals.27,45,52,64 The recovery time for pets with clinical changes attributed to O2 toxicity remains undefined. Once safe levels (60% or less FiO2) are being administered, any

SOUND MEDICAL PRACTICE ■ NUTRITION ■ OXYGEN TOLERANCE

Small Animal/Exotics

20TH ANNIVERSARY

lung injury may take several weeks to partially resolve.45,65–67 Even when administering lower levels of O2, some permanent pulmonary damage may remain. Because there is no therapy to prevent or reverse the toxic pulmonary changes, avoidance is essential. Several strategies and drugs that may eventually allow higher concentrations of O2 to be delivered for longer periods are being investigated.

REFERENCES 1. Ganong WF: Review of Medical Physiology. Norwalk, CT, Appleton & Lange, 1995, pp 608–611, 627–635. 2. Marino PL: The ICU Book. Baltimore, Williams & Wilkins, 1998, pp 19–22, 32–50, 388–400. 3. West JB: Respiratory Physiology. Baltimore, Williams & Wilkins, 1995, pp 51–88. 4. Bryan-Brown CW, Gutierrez G: Pulmonary gas exchange, transport, and delivery, in Shoemaker WC, Ayers SM, Grenvik A, Holbrook PR (eds): Textbook of Critical Care. Philadelphia, WB Saunders Co, 1995, pp 776–784. 5. Jenkinson SG: Oxygen toxicity. New Horizons 1(4):504–511, 1993. 6. O’Connor BS, Vender JS: Oxygen therapy. Crit Care Clinics 11(1):67–78, 1995. 7. Drobatz KJ, Hackner S, Powell S: Oxygen supplementation, in Bonagura JD (ed): Kirk’s Current Veterinary Therapy XII. Philadelphia, WB Saunders Co, 1995, pp 175–179. 8. Shappel SD: Adaptive, genetic, and iatrogenic alterations of the oxyhemoglobin dissociation curve. Anesthesiology 37: 178–206, 1972. 9. Harvey JW: Erythrocyte metabolism, in Kaneko JJ (ed): Clinical Biochemistry of Domestic Animals. San Diego, Academic Press, 1989, pp 185–234. 10. Court MH: Respiratory support of the critically ill small animal patient, in Murtaugh RJ, Kaplan PM (eds): Veterinary Emergency and Critical Care Medicine. St. Louis, Mosby Year Book, 1992, pp 575–584. 11. Court MH, Dodman NH, Seeler DC: Inhalation therapy: Oxygen administration, humidification, and aerosol therapy. Vet Clin North Am Small Anim Pract 15(5):1041–1059, 1985. 12. Filuk RB, Anthonisen NR: Administering oxygen effectively to critically ill patients. J Crit Illness 1(2):21–27, 1986. 13. Coalson JJ: Pathophysiologic features of infant and adult respiratory distress syndromes, in Shoemaker WC, Ayers SM, Grenvik A, Holbrook PR (eds): Textbook of Critical Care. Philadelphia, WB Saunders Co, 1995, pp 797–805. 14. Fitzpatrick RK, Crowe DT: Nasal oxygen administration in dogs and cats: Experimental and clinical investigations. JAAHA 22:293–300, 1986. 15. Mann FA, Wagner-Mann C, Allert JA, Smith J: Comparison of intranasal and intratracheal oxygen administration in healthy awake dogs. Am J Vet Res 53(5):856–860, 1992. 16. Elkins AD: Hyperbaric oxygen therapy: Potential veterinary applications. Compend Contin Educ Pract Vet 19 (5):607– 612, 1997. 17. Fischer B, Jain KK, Braun E, Lehrl S: Handbook of Hyperbaric Oxygen Therapy. Berlin, Springer-Verlag, 1988, pp 4–58. 18. Foster JH: Hyperbaric oxygen therapy: Contraindications and complications. J Oral Maxillofac Surg 50:1081–1086, 1992.

Compendium April 1999

19. Hosgood G, Elkins AD, Hill RK: Hyperbaric oxygen therapy: Mechanism and potential applications. Compend Contin Educ Pract Vet 12(11):1589–1594, 1990. 20. Schafer SE: Fundamentals of hyperbaric oxygen therapy. Orthop Nurs 11(6):9–15, 1992. 21. Tibbles PM, Edelsberg JS: Hyperbaric–oxygen therapy. N Engl J Med 334(25):1642–1648, 1996. 22. Hitt ME: Oxygen-derived free radicals: Pathophysiology and implications. Compend Contin Educ Pract Vet 10(8):939– 946, 1988. 23. Jackson RM: Oxygen therapy and toxicity, in Shoemaker WC, Ayers SM, Grenvik A, Holbrook PR (eds): Textbook of Critical Care. Philadelphia, WB Saunders Co, 1995, pp 784– 789. 24. Rochat MC: An introduction to reperfusion injury. Compend Contin Educ Pract Vet 13(6):35–41, 1991. 25. Klein J: Normobaric pulmonary oxygen toxicity. Anesth Analg 70:195–207, 1990. 26. Rubanyl GM: Vascular effects of oxygen-derived free radicals. Free Radic Biol Med 4:107–120, 1988. 27. Stogner SW, Payne DK: Oxygen toxicity. Ann Pharmacother 26:1554–1561, 1992. 28. Freeman BA, Crapo JD: Biology of disease. Free radicals and tissue injury. Lab Invest 47:412–426, 1982. 29. Ward PA, Johnson KJ, Till GO: Animal models of oxidant lung injury. Respiration 50:1, 5–12, 1986. 30. Fulmer JD, Snider GL: ACCP-NHLBI national conference on oxygen therapy. Chest 86(2):240–241, 243–244, 1984. 31. Crapo JD, Barry BE, Foscue HA, Shelburne J: Structural and biochemical changes in rat lungs occurring during exposures to lethal and adaptive doses of oxygen. Am Rev Respir Dis 122:123–143, 1980. 32. Crapo JD, Peters-Golden M, Marsh-Salin J, et al: Pathologic changes in the lungs of oxygen adapted rats: A morphometric analysis. Lab Invest 39(6):640–653, 1978. 33. De las Santos R, Seidenfeld JJ, Anzueto A, et al: One hundred percent oxygen lung injury in adult baboons. Am Rev Respir Dis 136:657–661, 1987. 34. Kamponei Y, Tosco R, Eggerman J, et al: Oxygen pneumonitis in man. Chest 62:162–169, 1972. 35. Thet LA, Parra SC, Shelburne JD: Sequential changes in lung morphology during the repair of acute oxygen-induced lung injury in adult rats. Exp Lung Res 11:209–228, 1986. 36. Frank L, Bucher JR, Roberts RJ: Oxygen toxicity in neonatal and adult animals of various species. J Appl Physiol 45(5): 699–704, 1978. 37. Jamieson D, Chance B, Cadenas E, Boveris A: The relationship of free radical production to hyperoxia. Ann Rev Physiol 48:703–719, 1986. 38. Freeman BA, Toppolsky MK, Crapo JD: Hyperoxia increases oxygen radical production in rat lung homogenates. Arch Biochem Biophys 216:477–484, 1982. 39. Barry BE, Crapo JD: Pattern of accumulation of platelets and neutrophils in rat lungs during exposure to 100% and 85% oxygen. Am Rev Respir Dis 132:548–555, 1985. 40. Comroe JH, Dripps RD, Dumke PR, Deming M: Oxygen toxicity. The effect of inhalation of high concentrations of oxygen for twenty-four hours on normal men at sea level and at a simulated altitude of 18,000 feet. JAMA 128:710– 717, 1945. 41. Sackner M, Landa J, Hirsch J, et al: Pulmonary effects of oxygen breathing. A six-hour study in normal men. Ann Intern Med 82:40–43, 1975.

Compendium April 1999

20TH ANNIVERSARY

42. Wolfe WG, Eeret PA, Sabiston DC Jr: Effect of high oxygen tension on mucociliary function. Surgery 72(2):246–252, 1972. 43. Lauenzi GA, Yin S, Guarneri JJ: Adverse effect of oxygen on tracheal mucus flow. N Engl J Med 279(7):333–339,1968. 44. Van der Water JN, Kagey KS, Miller IT, et al: Response of the lung to six to 12 hours of 100 percent oxygen inhalation in normal man. N Engl J Med 283:621–626, 1970. 45. Jenkinson SG: Pulmonary oxygen toxicity. Clin Chest Med 3(1):109–119,1982. 46. Caldwell P, Lee W, Schildkraut H, et al: Changes in lung volume, diffusing capacity, and blood gases in men breathing oxygen. J Appl Physiol 21:1477–1483, 1966. 47. Hayatdavoudi G, O’Neil JJ, Barry BE, et al: Pulmonary injury in rats following continuous exposure to 60% O2 for 7 days. J Appl Physiol 51:1220–1231, 1981. 48. Grim PS, Gottlieb LJ, Boddie A, et al: Hyperbaric oxygen therapy. JAMA 263:2216–2220, 1990. 49. Rettenmaier PA, Whittaker B, Myers RAM: Acute oxygen toxicity during hyperbaric oxygen therapy. Crit Care Med 14(4):394, 1986. 50. Hendricks JC: Respiratory conditions in critical patients. Vet Clin North Am 19(6):1167–1188, 1989. 51. Brown LH: Pulmonary oxygen toxicity. Focus Crit Care 17(1): 68–75, 1990. 52. Jenkinson SG: Oxygen toxicity. J Intensive Care Med 3(3): 137–152, 1988. 53. Davenport DJ: Enteral and parenteral nutritional support in Ettinger SJ, Feldman EC (eds): Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat, ed 4. Philadelphia, WB Saunders Co, 1995, pp 244–252. 54. Lindsey HJ, Kisala JM, Ayala A, et al: Pentoxifylline attenuates oxygen-induced lung injury. J Surg Res 56:543–548, 1994. 55. Flink EB, Dedhia HV, Dinsmore J, et al: High-dose magnesium sulfate attenuates pulmonary oxygen toxicity. Crit Care Med 20(12):1692–1698, 1992. 56. Patterson CE, Butler JA, Byrne FD, Rhodes ML: Oxidant lung injury: Intervention with sulfhydryl reagents. Lung 163: 23–32, 1985. 57. Sarnstrand B, Tunek A, Sjodin K, Hallberg A: Effects of Nacetylcysteine stereoisomers on oxygen-induced lung injury in rats. Chem-Biol Interact 94:157–164, 1995.

Small Animal/Exotics

58. Padmanabhan R, Gudapaty R, Liener I, et al: Protection against pulmonary oxygen toxicity in rats by the administration of liposome encapsulated superoxide dismutase or catalase. Am Rev Respir Dis 132(1):164–167, 1985. 59. Turrens J, Crapo JD, Freeman BA: Protection against oxygen toxicity by intravenous injection of liposome-encapsulated catalase and superoxide dismutase. J Clin Invest 73:87– 95, 1984. 60. Boyce NW, Campbell D, Holdsworth SR: Modulation of normobaric pulmonary oxygen toxicity by hydroxyl radical inhibition. Clin Invest Med 10:316–320, 1987. 61. Jensen JC, Pogrebniak HW, Pass HI, et al: Role of tumor necrosis factor in oxygen toxicity. J Appl Physiol 72(5): 1902–1907, 1992. 62. Matalon S, Holme B, Notter R: Modification of pulmonary hyperoxic injury by administration of exogenous surfactant. J Appl Physiol 62(2):756–761, 1987. 63. Frank L: Endotoxin reverses the decreased tolerance of rats to greater than 95% O2 after pre-exposure to lower O2. J Appl Physiol 51:577–583, 1981. 64. Jenkinson SG: Oxygen toxicity. J Intensive Care Med 3:137– 152, 1988. 65. Davis WB, Rennard SI, Bitterman PB, Crystal RG: Pulmonary oxygen toxicity: Early reversible changes in human alveolar structures induced by hyperoxia. N Engl J Med 309(15):878–883, 1983. 66. Durr R, Dubayba B, Thet LA: Repair of chronic hyperoxic lung injury: Changes in lung ultrastructure and matrix. Exp Mol Pathol 47:219–240, 1987. 67. Jackson RM: Molecular, pharmacologic and clinical aspects of oxygen-induced lung injury. Clin Chest Med 11(1):73–86, 1990.

About the Authors Dr. Mensack is affiliated with the Department of Clinical Science, Tufts University School of Veterinary Medicine, North Grafton, Massachusetts. Dr. Murtaugh is Director of Critical Care at the Dove Lewis Emergency Animal Hospital in Portland, Oregon, and is a Diplomate of the American College of Veterinary Internal Medicine and of the American College of Veterinary Emergency and Critical Care.