Metabolic response of horses to a high soluble carbohydrate diet: effects of low-intensity submaximal exercise and sodium bicarbonate supplementation.

Four mares fed a low fiber, high soluble carbohydrate diet were used in a crossover design to evaluate the effects of dietary sodium bicarbonate (NaHCO3) supplementation during daily low-intensity submaximal working conditions. Mares were fed the diet at 1.7 times the maintenance energy requirement for mature horses at work. The horses tolerated the diet well and had no clinical abnormalities. Resting venous blood bicarbonate (HCO3), standard HCO3, and base excess (BE) concentrations significantly (P less than 0.05) increased with NaHCO3 supplementation, but no significant changes in resting venous blood pH or carbon dioxide tension (PCO2) were recorded. Venous blood HCO3, standard HCO3, BE, hemoglobin, and heart rate were significantly (P less than 0.05) increased and plasma lactate concentration was significantly (P less than 0.05) decreased in the control horses and in the horses given the NaHCO3 supplement during low-intensity submaximal exercise. There were no significant changes in venous blood pH, PCO2, or plasma protein concentration with exercise. Venous blood HCO3, standard HCO3, and BE concentrations were significantly (P less than 0.05) greater during submaximal exercise in horses given the NaHCO3 supplement. There were no significant differences in plasma lactate or total protein concentrations, blood pH, PCO2, or hemoglobin concentration between the 2 groups during exercise.     

Evaluation of peritoneal fluid pH, glucose concentration, and lactate dehydrogenase activity for detection of septic peritonitis in horses

OBJECTIVE: To determine whether peritoneal fluid pH, glucose concentration, and lactate dehydrogenase activity can be used to differentiate horses with septic peritonitis from those with nonseptic peritonitis. DESIGN: Prospective study. ANIMALS: 46 horses, including 10 healthy horses, 15 horses with septic peritonitis, and 21 horses with nonseptic peritonitis. PROCEDURE: Peritoneal fluid and blood samples were analyzed for pH, glucose concentration, and lactate dehydrogenase activity. Complete blood cell counts were performed, and peritoneal fluid samples were submitted for bacterial culture. RESULTS: Horses with septic peritonitis had significantly lower peritoneal fluid pH and glucose concentrations than horses with nonseptic peritonitis and healthy horses. Compared with other tests, serum-to-peritoneal fluid glucose concentration differences > 50 mg/dl had the highest diagnostic use for detection of septic peritonitis. Peritoneal fluid pH < 7.3, glucose concentration < 30 mg/dl, and fibrinogen concentration > 200 mg/dl were also highly indicative of septic peritonitis. CLINICAL IMPLICATIONS: Peritoneal fluid pH and glucose concentration can be used to assist in the identification of horses with septic peritonitis. These measurements can provide an early indication of sepsis, especially if cytologic evaluation of peritoneal fluid is unavailable or results are equivocal and peritoneal fluid bacterial culture results are pending.     

Blood gas and acid-base values in calves, sampled from the brachial and coccygeal arteries

Blood samples were taken from the brachial and coccygeal arteries of young calves and blood gas and acid-base values determined. There was no significant difference in pH, PO2, PCO2 or HCO3- between sites as demonstrated by a paired t-test (P greater than 0.05). Significant correlations between sites existed for individual values of PO2 (P less than 0.001), HCO3- (P less than 0.05) and pH (P less than 0.02), but not for PCO2.     

Changes in gas composition and acid-base values of venous blood samples stored under different conditions in 4 domestic species

BACKGROUND: The effect of storage temperature and time on blood gas and acid-base values has been investigated intensively in cattle and dogs; however, data are lacking in other species. OBJECTIVE: The aim of our study was to evaluate changes in gas composition and acid-base values in venous blood stored at different temperatures and for different times in 4 domestic species in Italy. METHODS: Blood samples from Comisana sheep (n = 10), Maltese goats (n = 10), Ragusana donkeys (n = 10), and Thoroughbred horses (n = 10) were analyzed after storage at 23 degrees C (room temperature) for 15 minutes (group I), 23 degrees C for 1 hour (group II), 37 degrees C for 8 hours (group III), and 4 degrees C for 24 hours (group IV). Results were analyzed using a 1-way repeated measures ANOVA. RESULTS: In all species no statistically significant differences in pH values were present in samples stored at 4 degrees C for 24 hours. This also was true for PCO2 in all species except the horse. Except for HCO3- concentration in the horse, significant changes in PO2, HCO3- concentration, base excess, and the standard bicarbonate concentration were observed for all species in samples stored at 4 degrees C. In samples stored for only 1 hour at room temperature, significant changes in most analytes were detected. CONCLUSIONS: The results of this study underline the need for rapid assessment of acid-base samples, because any delay, even for 1 hour, may affect the results.     

Determinants of oxygen delivery and hemoglobin saturation during incremental exercise in horses

OBJECTIVE: To determine components of the increase in oxygen consumption (VO2) and evaluate determinants of hemoglobin saturation (SO2) during incremental treadmill exercise in unfit horses. ANIMALS: 7 unfit adult mares. PROCEDURES: Horses performed 1 preliminary exercise test (EXT) and 2 experimental EXT. Arterial and mixed venous blood samples and hemodynamic measurements were taken during the last 30 seconds of each step of the GXT to measure PO2, hemoglobin concentration ([Hb]), SO2, and determinants of acid-base state (protein, electrolytes, and PCO2). RESULTS: Increased VO2 during exercise was facilitated by significant increases in cardiac output (CO), [Hb], and widening of the arteriovenous difference in O2. Arterial and venous pH, PaO2, and PvO2 decreased during exercise. Arterial PCO2, bicarbonate ([HCO3-])a, and [HCO3-] decreased significantly, whereas PVCO2 and increased. Arterial and venous sodium concentration, potassium concentration, strong ion difference, and venous lactate concentration all increased significantly during exercise. CONCLUSIONS AND CLINICAL RELEVANCE: Increases in CO, [Hb], and O2 extraction contributed equally to increased VO2 during exercise. Higher PCO2 did not provide an independent contribution to shift in the oxyhemoglobin dissociation curve (OCD) in venous blood. However, lower PaCO2 shifted the curve leftward, facilitating O2 loading. The shift of ODC resulted in minimal effect on O2 extraction because of convergence of the ODC at lower values of PO2. Decreased pH appeared responsible for the rightward shift of the ODC, which may be necessary to allow maximal O2 extraction at high blood flows achieved during exercise.     

Accuracy of a portable blood gas analyzer incorporating optodes for canine blood

The accuracy of a portable blood gas analyzer (OPTI 1) was evaluated using canine blood and aqueous control solutions. Sixty-four arterial blood samples were collected from 11 anesthetized dogs and were analyzed for pH, partial pressure of carbon dioxide (PCO2) partial pressure of oxygen (PO2), and bicarbonate concentration ([HCO3-]) values by the OPTI 1 and a conventional blood gas analyzer (GASTAT 3). The conventional analyzer was considered as a standard against which the OPTI 1 was evaluated. Comparison of OPTI 1 results with those of GASTAT 3 by linear regression analysis revealed a high degree of correlation with the GASTAT 3 (r = .90-.91). The mean +/- SD of the differences between OPTI 1 and GASTAT 3 values was -0.008 +/- 0.017 for pH, -0.88 +/- 3.33 mm Hg for PCO2, 3.71 +/- 6.98 mm Hg for PO2, and -0.34 +/- 1.45 mEq/L for [HCO3-]. No statistically significant difference was found between the OPTI 1 and the GASTAT 3. Agreement between these 2 methods is within clinically acceptable ranges for pH, PCO2, PO2, and [HCO3-]. The coefficients of variation for measured pH, PCO2, and PO2 values of 3 aqueous control solutions (acidic, normal, and alkalotic) analyzed by the OPTI 1 ranged from 0.047 to 0.072% for pH, 0.78 to 1.81% for PCO2, and 0.73 to 2.77% for PO2. The OPTI 1 is concluded to provide canine blood gas analysis with an accuracy that is comparable with that of conventional benchtop blood gas analyzers.     

Blood gas analyses on equine blood: required correction factors [see comment]

Correction factors have been determined to obtain the best estimates of PO2, PCO2 and pH in equine blood with standard blood gas and pH electrodes. There was a significant difference between the PO2 readings for tonometred blood of most horses and the equilibrating gas. Thus, if the PO2 electrode is calibrated with a gas, an electrode correction factor should be obtained by tonometring a blood sample from each horse. This factor was not dependent on packed cell volume. No such correction is required for the PCO2 electrode. If the animal's temperature differs from that of the analyser, the PO2, PCO2 and pH values must be corrected to the animal's body temperature. Temperature correction factors determined for equine blood were similar to those for human blood. Failure to make temperature corrections can result in errors for PO2 and PCO2 of 6 to 7 per cent per degree of temperature difference.     

Acid-base balance parameters and a value of anion gap of arterial and venous blood in Małopolski horses

The comparative study of the acid-base balance (ABB) parameters has been performed on 20 clinically healthy mature Ma?opolski horses. An arterial blood sample from the facial artery and a sample of venous blood from the external cervical vein were colected from each animal. In the samples tested, the blood pH, pCO2, tCO2, HCO3-, concentration of Na+, K+, Cl-, and a value of the anion gap were determined. The difference among pCO2, tCO2, and HCO3- in both samples tested was statistically significant, whereas the pH of the arterial blood and the pH of the venous blood did not differ significantly. The anion gap in both types of blood did not differ significantly. Conclusions: 1) ABB parameters such as pCO2, HCO3-, and tCO2 determined in the arterial and venous blood of the Ma?opolski horses differ from each other significantly. 2) In spite of the lack of the differences between pH of the arterial and venous blood, the ABB parameters in horses should be determined in the arterial blood, because the comparative study performed proves that the analysis of the ABB parameters determined for the venous blood of a healthy horse may lead to a wrong diagnosis of the compensated respiratory acidosis. 3) The mean value of anion gap in horses aged 8-12 years amounts to 20.9 mmol/l for the arterial blood and 19.93 for the venous blood; the difference between the two values is not statistically significant.     

Comparison of peritoneal fluid and peripheral blood pH,bicarbonate, glucose,and lactate concentration as a diagnostic tool for septic peritonitis in dogs and cats

OBJECTIVE: To establish a reliable diagnostic tool for septic peritonitis in dogs and cats using pH, bicarbonate, lactate, and glucose concentrations in peritoneal fluid and venous blood. STUDY DESIGN: Prospective clinical study. ANIMALS: Eighteen dogs and 12 cats with peritoneal effusion. METHODS: pH, bicarbonate, electrolyte, lactate, and glucose concentrations were measured on 1- to 2-mL samples of venous blood and peritoneal fluid collected at admission. The concentration difference between blood and peritoneal fluid for pH, bicarbonate, glucose, and lactate concentrations were calculated by subtracting the peritoneal fluid concentration from the blood concentration. Peritoneal fluid was submitted for cytologic examination and bacterial culture. Peritonitis was classified as septic or nonseptic based on cytology and bacterial culture results. RESULTS: In dogs, with septic effusion, peritoneal fluid glucose concentration was always lower than the blood glucose concentration. A blood-to-fluid glucose (BFG) difference > 20 mg/dL was 100% sensitive and 100% specific for the diagnosis of septic peritoneal effusion in dogs. In 7 dogs in which it was evaluated, a blood-to-fluid lactate (BFL) difference < -2.0 mmol/L was also 100% sensitive and specific for a diagnosis of septic peritoneal effusion. In cats, the BFG difference was 86% sensitive and 100% specific for a diagnosis of septic peritonitis. In dogs and cats, the BFG difference was more accurate for a diagnosis of septic peritonitis than peritoneal fluid glucose concentration alone. CONCLUSIONS: A concentration difference > 20 mg/dL between blood and peritoneal fluid glucose concentration provides a rapid and reliable means to differentiate a septic peritoneal effusion from a nonseptic peritoneal effusion in dogs and cats. CLINICAL RELEVANCE: The difference between blood and peritoneal fluid glucose concentrations should be used as a more reliable diagnostic indicator of septic peritoneal effusion than peritoneal fluid glucose concentration alone.     

Effects of enalaprilat on cardiorespiratory, hemodynamic, and hematologic variables in exercising horses.

OBJECTIVE: To determine the effects of IV administration of enalaprilat on cardiorespiratory and hematologic variables as well as inhibition of angiotensin converting enzyme (ACE) activity in exercising horses. ANIMALS: 6 adult horses. PROCEDURE: Horses were trained by running on a treadmill for 5 weeks. Training was continued throughout the study period, and each horse also ran 2 simulated races at 120% of maximum oxygen consumption. Three horses were randomly selected to receive treatment 1 (saline [0.9% NaCl] solution), and the remaining 3 horses received treatment 2 (enalaprilat; 0.5 mg/kg of body weight, IV) before each simulated race. Treatment groups were reversed for the second simulated race. Cardiorespiratory and hematologic data were obtained before, during, and throughout the 1-hour period after each simulated race. Inhibition of ACE activity was determined during and after each race in each horse. RESULTS: Exercise resulted in significant increases in all hemodynamic variables and respiratory rate. The pH and PO2 of arterial blood decreased during simulated races, whereas PCO2 remained unchanged. Systemic and pulmonary blood pressure measurements and arterial pH, PO2, and Pco2 returned to baseline values by 60 minutes after simulated races. Enalaprilat inhibited ACE activity to < 25% of baseline activity without changing cardiorespiratory or blood gas values, compared with horses administered saline solution. CONCLUSIONS AND CLINICAL RELEVANCE: Enalaprilat administration almost completely inhibited ACE activity in horses without changing the hemodynamic responses to intense exercise and is unlikely to be of value in preventing exercise-induced pulmonary hemorrhage.     

Effects of limb tourniquet ischemia on local and systemic acid-base and blood gases of cattle.

Effects of forelimb tourniquet ischemia of 90 minute duration were investigated in six bulls aged two to three years. Studies were also conducted up to 150 minutes after release of the tourniquet. Parameters investigated were pH, PCO2, PO2, oxygen saturation and HCO3. In systemic circulation no variations in different parameters were observed during 90 minutes of ischemia. However, significant increase in arterial and venous pH were observed after 30 and 45 minutes of the release of tourniquet, respectively. These increases were accompanied by an increase in HCO3. In the affected limb, ischemia resulted in severe acidosis with a significant increase in PCO2 and a nonsignificant decrease of HCO3. There was a significant fall in PO2 and oxygen saturation. After release of the tourniquet, limb venous pH increased significantly due to a significant fall in PCO2 and a nonsignificant increase in HCO3. A significant increase in the limb venous PO2 and oxygen saturation post tourniquet was observed up to the end of the experiments. There was evidence of very poor oxygen exchange and utilization up to 150 minutes after release of the tourniquet. These results demonstrated that tourniquet ischemia of 90 minutes duration of the limb of cattle may not be safe.     

Blood-gas and acid-base parameters in nontranquilized Arabian oryx (Oryx leucoryx) in the United Arab Emirates.

Arterial and venous blood-gas and acid-base values were established from a herd (n = 19; 14 male, 5 female) of semi-free-ranging Arabian oryx (Oryx leucoryx) in the United Arab Emirates. The animals were restrained with the use of a modified raceway incorporating a commercially available handling crate. Statistically significant differences were found between arterial and venous values for PO2 (p < 0.001), PCO2 (p = 0.0141), SO2 (p < 0.001), pH (p = 0.0494), and glucose (p < 0.0001). The results are similar to those reported for the same species under field anesthetic conditions, and to those reported from other species of wild bovidae, both tranquilized and nontranquilized, established under similar methods of restraint. In addition, Bland and Altman plots suggest adequate levels of clinical agreement between venous and arterial pH but not between arterial and venous PCO2.     

Effects of hyperosmolar ionic and low osmolar non-ionic contrast media on acid base, blood gas and electrolyte status in dogs

The dogs in groups I, II and III in equal numbers received diatrizoate, iohexol and ioxilan at a dose of 700 mgI/kg intravenously (i.v.) as a bolus, respectively. Blood samples were collected prior to contrast media (CM) administration and thereafter at 3, 15, 30, 60, 90 and 180 min to evaluate acid-base, venous blood gas status (pH, PCO2, PO2, HCO, BE, O2) and electrolytes (Na+, Ca++, K+). Values of pH, PCO2, BE, HCO, Na+ and K+ remained unchanged or within non-significant fluctuations compared with the baseline values. PO2 was significantly different from the baseline values in group 1 up to 90 min after administration, significant alterations were found for O2 saturation in group 1 up to 90 min, and in group II at 3, 60 and 180 min; and for Ca++ in group 1 at all time points except at 90 min, and groups II and II at 3 and 15 min post administration. It was concluded that none of the CM are considered to cause long-lasting and major effects on acid-base, blood gas and electrolyte status.     

The effect of colic on oxygen extraction in horses

Blood oxygen transport and oxygen extraction were assessed in horses with colic. A gravity score (GS) ranging from 1 to 3 was attributed to each colic case with healthy horses used as controls. Jugular venous and carotid arterial blood samples were collected and concentrations of 2,3-diphosphoglycerate, adenosine triphosphate, inorganic phosphate and chloride were determined. pH and partial pressures of carbon dioxide (PCO(2)), and oxygen (PO(2)) were also measured. Oxygen equilibrium curves (OEC) were constructed under standard conditions and oxygen extraction ratios calculated. Haemoglobin oxygen affinity measured under standard conditions (P50(std)) was unchanged in colic horses compared with healthy controls. Horses with the highest GS, i.e. 3 had lower blood pH values than healthy animals. Arterial and venous partial pressures of oxygen at 50% haemoglobin saturation (P50(a) and P50(v)) were significantly higher in horses suffering from colic (GS=3) than in healthy horses. The oxygen extraction ratio was also significantly increased in colic horses with a GS of 3. A rise in the oxygen extraction ratio detected in the most severely affected animals seemed to reflect the compensatory properties of the oxygen transport system where extraction of oxygen from the blood increases when systemic oxygen delivery decreases, as might be anticipated in horses with colic.     

Cardiopulmonary effects of prostacyclin infusion in anesthetized horses

Prostacyclin was infused IV into 6 horses anesthetized with halothane. Three dosage rates (10, 30, and 100 ng/kg of body weight/min) were evaluated in each horse. Facial and pulmonary artery pressures, heart rate, cardiac output, blood temperature, and arterial and mixed venous pH, PCO2, and PO2 were measured. Arterial blood was collected for determination of glucose, lactate, and PCV. Mixed venous blood was sampled for assay of 6-keto-prostaglandin F1 alpha and catecholamines. Infusion of prostacyclin at 10 ng/kg/min had no effect on the variables measured, whereas the 30 ng/kg/min dosage decreased diastolic and mean arterial pressure at 15 and 30 minutes and PaO2 at 15 minutes (P less than 0.05). Prostacyclin infusion at 100 ng/kg/min significantly decreased arterial pressure, total vascular resistance, and total pulmonary resistance. Heart rate increased slightly, and cardiac output increased by 44%. Arterial PO2 decreased from 311 mm of Hg to 137 and 135 mm of Hg at 15 and 30 minutes, respectively. Blood glucose was increased. Prostacyclin infusions of 30 and 100 ng/kg/min increased blood concentrations of 6-keto-prostaglandin F1 alpha by factors of 5 and 40, respectively. Significant changes in catecholamine concentrations did not occur.     

氨氟醚吸入麻醉妊娠犬及其胎儿动脉血药浓度和血气分析

选用 10只妊娠犬 ,实施母体及胎儿股动脉血管插管后 ,测定了氨氟醚麻醉期间母犬及胎儿的动脉血药浓度和血液 p H、PO2 (动脉氧分压 )、PCO2 (动脉 CO2 分压 )、T- CO2 (血浆 CO2 总量 )、HCO- 3 (实际碳酸氢盐 )、SB(标准碳酸氢盐 )、BEb(全血碱超 )、Sat.O2 (血氧饱和度 )。结果 :氨氟醚可透过胎盘进入胎儿血液 ,胎儿血药浓度低于母犬 ,但两者上升和消除变化趋势接近 ;麻醉期间 ,母犬及胎儿血液 p H、BEb下降 (P<0 .0 1或 P<0 .0 5 ) ,PO2 、PCO2 、Sat.O2 升高(P<0 .0 1或 P<0 .0 5 ) ,HCO- 3 、T- CO2 表现升高趋势 (P>0 .0 5 ) ,SB表现下降趋势 (P>0 .0 5 )。结果表明 ,氨氟醚吸入麻醉期间 ,母犬及其胎儿呈现轻度呼吸性酸中毒和代谢性酸中毒并存 ,并随氨氟醚血药浓度的降低而逐渐恢复     

A Comparison of Simultaneously Collected Arterial, Mixed Venous, Jugular Venous and Cephalic Venous Blood Samples in the Assessment of Blood-Gas and Acid-Base Status in the Dog

Blood samples were collected simultaneously from the pulmonary artery, jugular vein, cephalic vein, and carotid artery in awake dogs. Blood-gas and acid-base values were measured from these blood samples in normal dogs and in dogs after production of metabolic acidosis and metabolic alkalosis. The values obtained from each of the venous sites were compared with those obtained from arterial blood to determine if venous blood from various sites accurately reflected acid-base balance and could therefore be used in the clinical patient. The results of this study demonstrated significant differences between the blood from various venous sites and the arterial site for PCO2 and pH in all acid-base states. Significant differences for standard bicarbonate (SHCO3) were found only when jugular and cephalic venous blood were compared with arterial blood in dogs with a metabolic acidosis. No significant differences were found for BE when blood from the venous sites was compared with arterial blood. The values for pH, HCO3, TCO2, BE, and SHCO3 measured on blood collected at the various venous sites were found to correlate well with those obtained from arterial blood, with a correlation coefficient of 0.99 for HCO3, TCO2, BE, and SHCO3. These correlation coefficients, together with similar values in BE at all collection sites, indicate that, in the dog with normal circulatory status, blood from any venous site will accurately reflect the acid-base status of the patient.     

Changes in blood gas and acid-base values of bovine venous blood during storage

The stability of blood gas and acid-base values in bovine venous blood samples (n = 22) stored on ice for 3, 6, 9, or 24 hours was studied. Values studied include pH, PO2 and PCO2 tensions, base excess, standard base excess, bicarbonate concentration, standard bicarbonate concentration, total carbon dioxide content, oxygen saturation, and hemoglobin. The results indicate that, except for PCO2, changes in blood gas and acid-base values during 24 hours of storage and differences between cattle of differing ages, rectal temperatures, and acid-base status were too small to be of clinical significance. Therefore, bovine venous blood samples stored up to 24 hours on ice are of diagnostic utility.     

Comparison of ventral coccygeal arterial and jugular venous blood samples for pH, pCO2, HCO3, Be(ecf) and ctCO2 values in calves with pulmonary diseases

The aim of this study was to determine whether venous blood samples can be used as an alternative to arterial samples in calves with respiratory problems and healthy calves. Jugular vein and ventral coccygeal artery were used to compare blood gas values. Sampling of the jugular vein followed soon after sampling of the ventral coccygeal artery in healthy calves (group I) and calves with respiratory problems (group II). Mean values of arterial blood for pH, pCO2, HCO3act in healthy calves were 7.475 +/- 0.004, 4.84 +/- 0.2 kPa, 28.45 +/- 1.30 mmol/L compared with venous samples, 7.442 +/- 0.006, 6 +/- 0.3 kPa, 30.93 +/- 1.36 mmol/L, respectively. In group II, these parameters were 7.414 +/- 0.01, 5.93 +/- 0.3, 27.73 +/- 1.96 mmol/L for arterial blood and 7.398 +/- 0.008, 6.85 +/- 0.2 kPa, 29.77 +/- 1.91 mmol/L for venous blood, respectively. There were no statistically significant differences between arterial and venous pH, HCO3act, Be(ecf), ctCO2 values with the exception of pCO2 (P = 0.001) in group II. In group I, correlation (r2) between arterial and venous blood pH, pCO2, HCO3act were 84.5%, 87.5%, 95.7%, respectively compared with the same parameters in group II, 80.8%, 77.1%, 70.3%. In conclusion, venous blood gas values can predict arterial blood gas values of pH, pCO2 and HCO3ecf, Be(ecf) and ctCO2- for healthy calves but only pH values in calves with acute respiratory problems (r2 value>80%).     

Evaluation of clinical signs of disease, bronchoalveolar and tracheal wash analysis, and arterial blood gas tensions in 13 horses with chronic obstructive pulmonary disease treated with prednisone, methyl sulfonmethane, and clenbuterol hydrochloride.

We evaluated the efficacy of 3 treatments for chronic obstructive pulmonary disease in horses: prednisone (400 mg/horse, PO, daily; n = 7), methyl sulfonmethane (10 g/horse, PO, q 12 h; n = 6), and clenbuterol hydrochloride (0.4 mg/horse, PO, q 12 h; n = 7). A fourth group acted as controls (n = 6) and was not treated. The treatment period lasted 10 days. Each horse was a member of 2 different groups for 10 days, separated by an 18-day interval of no treatment. All horses were housed together in an outdoor pen without bedding. Horses were fed alfalfa/grass hay mix ad libitum from a large feeder. The same batch of hay was fed throughout the study. Multiple physical and laboratory variables were monitored prior to, during, and at the end of each 10-day trial period. Changes in lung sounds, respiratory effort, degree of anal movement, nasal discharge, temperature, respiratory rate, or heart rate were not significant. Changes in arterial blood gas tensions, tracheal wash or bronchoalveolar lavage cytologic findings, or phagocyte function were not significant. All horses were tachypneic and most were tachycardic. The median value for PaO2 was below normal for all horses. All tracheal wash and most bronchoalveolar lavage cytologic findings represented a suppurative response. Negative linear correlation was observed between PaO2 and degree of respiratory effort in these horses (eg, as PaO2 decreased, the degree of respiratory effort increased).