Review of Support of Ventilation in the Anesthetized Horse

DRUGS AND ANESTHESIA

John A. E. Hubbell, DVM, MS, Diplomate ACVA

Anesthetized horses hypoventilate. Combined with changes in body position to lateral or dorsal recumbency, this hypoventilation can lead to hypoxemia and insufﬁcient delivery of oxygen to tissues. Healthy horses tolerate respiratory embarrassment for short periods, but most beneﬁt from increased inspired oxygen concentrations and support of ventilation, either manually or through the use of a ventilator. Author’s address: Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Ohio State University, 601 Vernon L. Tharp Street, Columbus, Ohio 43210; e-mail: [email protected]. © 2010 AAEP.

1. Introduction creases in minute alveolar ventilation (the The drugs used to induce and maintain anesthesia in amount of fresh gas that reaches the alveoli in 1 the horse are respiratory depressants. The degree of min) and subsequent retention of carbon dioxide respiratory depression imposed by anesthesia is de- measured by increases in arterial carbon dioxide pendent on the drugs and techniques used, the body tension (PaCO2) from its normal value of 40–45 position chosen (dorsal or lateral recumbency), and the mm Hg. Decreases in minute alveolar ventila- duration of the anesthetic period. Frequently, respi- tion occur because of decreases in respiratory rate, ratory depression in the horse is accompanied by lower decreases in tidal volume (the amount of gas in arterial oxygen tensions (hypoxemia) than those seen one breath), changes in the distribution of venti- in other species. A number of studies have been con- lation (distribution of a breath to alveoli that are ducted to characterize this respiratory depression and not well-perfused or vice versa), or combinations hypoxemia, and some have proposed methods to im- of these changes. Decreases in respiratory rate prove ventilation and oxygenation. The purpose of are easy to detect, but changes in tidal volume are this paper is to describe what is known about ventila- difﬁcult to assess clinically and changes in the tion and oxygenation and to provide practical sugges- distribution of ventilation such as increases in tions for employing ventilatory assistance in the dead-space ventilation (ventilation of areas where anesthetized horse. no gas exchange takes place) are limited to the

laboratory. The relationship between PaCO2 and 2. Overview of Ventilation and Oxygenation in the Horse minute ventilation is curvilinear, with increases

Hypoventilation in PaCO2 to levels in excess of 80 mm Hg indi- Respiratory depression can be assessed in a number cating that minute ventilation has decreased by of ways but fundamentally is determined by de- 50%.

NOTES

AAEP PROCEEDINGS ր Vol. 56 ր 2010 33 DRUGS AND ANESTHESIA Correcting Hypoventilation cardiac output, perfusion pressures, and physical Horses anesthetized with IV techniques such as xy- factors related to positioning. lazine and ketamine or xylazine-diazepam-ket- Correcting Hypoxemia amine usually maintain PaCO2 within normal limits (40–50 mm Hg) if anesthesia is limited to the induc- Several methods for increasing PaO2 in anesthe- tion doses (20–30 min), but hypoventilation (in- tized horses have been investigated. Supplement- creases in PaCO2) frequently occurs if anesthesia ing ambient air with oxygen-rich gases is the most is extended with guaifenesin recipes.1,2 Hypo- reliable method for increasing oxygenation. In- ventilation occurs more rapidly when horses are creasing cardiac output using an inotropic agent induced with IV techniques and transitioned to in- such as dobutamine should improve the distribution halant anesthesia. Horses induced with xylazine- of perfusion within the lung, but it does not consis- 9 guaifenesin-ketamine or xylazine-diazepam-ketamine tently increase PaO2. Decreasing inspired oxygen have increases in PaCO2 to levels in excess of 50 mm concentrations from 100% to 50% could reduce atel- Hg within 5 min of induction.3 Horses maintained ectasis caused by closure of alveoli (absorption atel- under anesthesia with inhalant anesthetics that are ectasis), but it does not reliably increase oxygen allowed to spontaneously ventilate routinely have content in the arterial blood.10 PaCO2 in excess of 60 mm Hg, and levels may exceed The application of positive-end expiratory pres- 80 mm Hg.4 Hypoventilation can be corrected in al- sure (PEEP) has been used to reduce atelectasis by most all horses by increasing minute ventilation. attempting to hold more terminal airways and alve- Assisted ventilation (delivery of a tidal volume when oli open during the expiratory phase of respiration. the horse initiates a breath) does not normalize venti- Levels of PEEP in the range of 5–15 cm H20 are lation in anesthetized horses, but controlled ventila- variably effective in other species, but values in the tion (operator-determined respiratory rate and tidal upper end of the range must be used in the horse to 11,12 volume) usually allows the anesthetist to establish the increase PaO2. These higher PEEP values im- desired ventilatory values. prove arterial oxygenation but reduce oxygen delivery to tissues, because they lower venous return Hypoxemia (and thus, cardiac output) as a result of the PEEP- Hypoventilation in the anesthetized horse is fre- induced increases in mean thoracic pressure. The quently associated with hypoxemia (arterial oxygen decreases in venous return and cardiac output can tensions below 100 mm Hg), particularly when be partially offset by ﬂuid and inotrope (dobut- 9 horses are breathing ambient oxygen tensions amine) administration. Beta2 adrenergic ago- (20%). Arterial oxygen tensions (PaO2) fall from nists, delivered both intravenously and through the standing levels (approximately 100 mm Hg) to airway, have been used to cause bronchodilation. 60–80 mm Hg within 5 min of the induction of The effects of IV clenbuterol on arterial oxygenation lateral recumbency and fall even further if dorsal are equivocal.13–15 Albuterol,a when aerosolized recumbency is employed.1,5 Hypoventilation con- and delivered through the airway, has been shown to 16 tributes to the decreases in arterial oxygenation, but improve PaO2 in hypoxemic anesthetized horses. physiologic shunting of blood past unventilated al- Selective mechanical ventilation of dependent lung re- veoli and other ventilation/perfusion mismatches gions is a novel technique for the redistribution of 5–7 play larger roles. Decreases in arterial oxygen ventilation that is effective in increasing PaO2, but the tension occur in all laterally and dorsally recumbent technique requires a tracheostomy with placement of horses, and the decreases are exacerbated in horses endotracheal tubes using endoscopy and thus, is not with rounded or distended abdomens.8 practical clinically.17 Ideal levels of arterial oxygenation when increased oxygen concentrations are delivered (95– Assessing Ventilation and Oxygenation 100% of the inspired air) are in excess of 400–450 The adequacy of ventilation and oxygenation is best mm Hg, but these levels are rarely attained. The assessed by obtaining a sample of arterial blood for appropriate target for PaO2 is controversial, but blood-gas analysis. Portable self-calibrating point- most are concerned if levels fall below 50–60 mm of-care blood-gas analyzers are currently available Hg. The critical level for oxygen tension is deter- that are easily used in both the hospital and ambu- mined by the oxyhemoglobin dissociation curve, be- latory setting.18 Normal blood-gas values for anes- cause greater than 90% of the oxygen carried in the thetized horses are presented in Table 1. Other blood is bound to hemoglobin. Oxygen tensions in methods of estimation of the adequacy of ventilation excess of 75 mm Hg are associated with greater than and oxygenation include measuring respiratory 95% saturation of hemoglobin. Even a PaO2 of 50 rate, watching for the degree of thoracic excursion mm Hg is associated with 90% saturation of hemo- during inspiration, visualizing the color of mucous globin. As oxygen tensions fall below 50 mm Hg, membranes, employing monitoring equipment such hemoglobin saturations decrease more linearly to as pulse oximetry, and measuring carbon dioxide the point where only 50% of hemoglobin is saturated tensions in the expired gasses. Pulse oximeters are at a PaO2 in the range of 20–25 mm Hg. Other used noninvasively to measure heart rate and esti- factors governing oxygen delivery to tissues include mate the percent saturation of hemoglobin. As pre-

34 2010 ր Vol. 56 ր AAEP PROCEEDINGS DRUGS AND ANESTHESIA Table 1. Expected Arterial Blood Gas Values in Awake and Anesthe- demand valve is triggered, and both nostrils are tized Horses occluded. Because the horse is an obligate nasal Awake Anesthetized breather, the thorax will expand, and a breath is delivered as above. If only a compressed gas source pH 7.4 Ϯ 0.2 7.3–7.45 is available, a tube can be attached. The tube is PaCO (mm Hg) 40 Ϯ 3 40–60 2 advanced from the nostrils into the nasal cavity. PaO (mm Hg) 94 Ϯ 3 100–500 (Ͼ95% inspired 2 Occlusion of the nostrils will cause the thorax to oxygen) Base Excess (mEq/l) 0 Ϯ 20Ϯ 4 rise. When the thorax has risen appropriately, the nostrils should be released, and the horse will exhale. The duration of ventilation that can be ac- complished through these methods is limited by the viously discussed, hemoglobin saturation in the capacity of the compressed gas source. horse remains in excess of 90%, even when PaO 2 Enhancing Oxygenation values are in the range of 60 mm Hg. The accuracy of pulse-oximeter values in predicting actual hemo- Oxygenation can be enhanced if a compressed source globin saturation has been questioned, leading to of oxygen is available. As with emergency ventila- the recommendation that any value below 95% sat- tion, oxygen can be supplied with an anesthetic ma- uration should be validated by other methods.19 chine or a demand valve, if available. Insufflation Measurement of end-tidal carbon dioxide tensions (the delivery of an oxygen flow to an open airway) is effective in normalizing PaO2 if flow rates in excess potentially provides a method to continuously and 23 noninvasively assess ventilation. End-tidal ten- of 15 l/min are used. The efficacy of insufflation is sions should match alveolar tensions, because the enhanced if the delivery tube is advanced into the gases exhaled at the end of expiration should be pharynx or trachea. Most insufflation systems do not alveolar in origin. Recent studies have questioned generate sufficient pressure to produce ventilation. the accuracy of this method of assessment of venti- Enhancing Ventilation and Oxygenation During Anesthesia lation.19,20 No current noninvasive method effectively replaces arterial blood-gas analysis. On a Hypoventilation should be assumed when horses are practical basis, horses can be safely and effectively anesthetized, particularly when inhalant anesthet- ventilated in the absence of arterial blood-gas anal- ics are used. Horses anesthetized with injectable ysis if appropriate guidelines are followed and at- agents for brief periods rarely require ventilatory tention is paid to ventilator-induced alterations in support but would benefit from the delivery of en- cardiac function. hanced oxygen tensions. Horses anesthetized through any method for extended periods frequently 3. Practical Ventilatory Assistance in Anesthetized hypoventilate, and this hypoventilation can compro- Horses mise oxygenation. If an extended procedure is con- templated, early ventilation with increased oxygen Emergency Ventilation tensions provides the best chance for maximization 24 The need for emergency ventilation associated with of PaO2. When inhalants are used, controlled anesthesia in the horse has decreased dramatically ventilation provides consistent delivery of the anes- with the increased use of ketamine as the primary thetic gas, helping establish and maintain a stable anesthetic agent. Prior use of more respiratory- anesthetic level. It is the author’s opinion that depressant drugs, such as thiopental or thiamylal, these factors must be balanced with the knowledge was associated with an increased incidence of apnea that the use of mechanical ventilation decreases car- and severe respiratory depression. In a hospital diac output and frequently, arterial blood pressures setting, emergency ventilation can be performed us- and increases the depth of anesthesia more rapidly. ing an anesthetic machine (if available) by either Arterial blood pressures should be monitored manually compressing the rebreathing bag or using closely, particularly in compromised patients. a ventilator. Emergency ventilation can be accom- Ventilation can be assisted in the anesthetized plished in any location if there is a source of com- horse by compressing the rebreathing bag- pressed air or oxygen and a means to deliver the gas . Manual ventilation in the adult horse requires under pressure to the horse’s airway. Emergency considerable expenditure of energy (both arms) to ventilation is most efficiently performed using a de- generate inspiratory pressures in the range of b 21,22 mand valve and an endotracheal tube. The 20–30 cm H2O. Respiratory rates of 6–10 c endotracheal tube is placed into the trachea, and breaths/min will usually produce PaCO2 levels be- the cuff is inflated to seal the airway. The demand tween 45 and 55 mm Hg in patients with normal valve is fitted to the end of the tube and triggered. pulmonary compliance. Inspiratory times (time The triggering mechanism is released when the for the delivery of the breath when there is a chest wall has expanded appropriately, and the positive pressure applied to the airway) should be horse is allowed to exhale. Alternatively, a smaller 1–3 sec in duration. Expiratory time (time when tube (12–14 mm inner diameter) can be inserted in there is not a positive pressure applied to the one nostril and the demand valve attached. The airway) should always equal or preferably, exceed

AAEP PROCEEDINGS ր Vol. 56 ր 2010 35 DRUGS AND ANESTHESIA

Table 2. Recommended Ventilatory Settings for Horses 3. Brock N, Hildebrand SV. A comparison of xylazine-diazepam-ketamine and xylazine-guaifenesin-ketamine in equine Adults Foals anesthesia. Vet Surg 1990;19:468–474. 4. Grosenbaugh DA, Muir WW. Cardiorespiratory effects of Tidal Volume (ml/kg) 14–18 12–16 sevoflurane, isoflurane, and halothane anesthesia in horses. Respiratory Rate (breaths/min) 6–8 8–12 Am J Vet Res 1998;59:101–106. Inspiratory to Expiratory Ratio 1:2–1:4 1:4–1:5 5. Marntell S, Nyman G, Funkquist P, et al. Effects of aceproma- Peak Inspiratory Pressure (cm H2O) 25–40 15–25 zine on pulmonary gas exchange and circulation during sedation and dissociative anaesthesia in horses. Vet Anaesth Analg 2005;32:83–93. 6. Nyman G, Hedenstierna G. Ventilation-perfusion relationships in the anaesthetized horse. Equine Vet J 1989;21:274–281. inspiratory time. The application of manual ven- 7. Nyman G, Funkquist B, Kvart C, et al. Atelectasis causes gas exchange impairment in the anaesthetized horse. tilation in the adult horse for periods in excess of Equine Vet J 1990;22:317–324. 10–15 min is exhausting and precludes the anes- 8. Moens Y, Lagerweij E, Gootjes P, et al. Distribution of in- thetist from completing other activities, such as spired gas to each lung in the anaesthetized horse and influ- monitoring the patient. Manual ventilation of ence of body shape. Equine Vet J 1995;27:110–116. the anesthetized foal requires less energy expen- 9. Swanson CR, Muir WW. Hemodynamic and respiratory re- sponses in halothane-anesthetized horses exposed to positive diture. The respiratory rates employed are in the end-expiratory pressure alone and with dobutamine. Am J Vet range of 8–12 breaths/min using similar inspira- Res 1988;49:539–542. tory pressures to those used in the adult. The 10. Cuveliez SG, Eicker SW, McLauchlan C, et al. Cardiovascular indicated inspiratory pressures usually result in and respiratory effects of inspired oxygen fraction in halothane- anesthetized horses. Am J Vet Res 1990;51:1226–1230. the delivery of normal or slightly reduced tidal 11. Wilson DV, McFeely AM. Cardiopulmonary effects of posi- volumes. tive end-expiratory pressure in anesthetized, mechanically Ventilation is easily supported through the use ventilated ponies. Am J Vet Res 1990;51:734–739. of commercially available anesthetic ventilators. 12. Wilson DV, McFeely AM. Positive end-expiratory pressure dur- The ventilators function by compressing a re- ing colic surgery in horses: 74 cases (1986–1988). J Am Vet Med Assoc 1991;199:917–921. breathing bag or bellows to deliver a mixture of 13. Keegan RD, Gleed Rd, Sanders EA, et al. Treatment of low oxygen and anesthetic gas. Most use compressed arterial oxygen tension in anesthetized horses with clenbuterol. gases to compress the bellows, but the newest Vet Surg 1991;20:148–152. ventilatord substitutes a linear actuator (similar 14. Lee YH, Clarke KW, Alibhai HI. The cardiopulmonary ef- to a piston) to deliver a tidal volume. Controlled fects of clenbuterol when administered to dorsally recumbent halothane-anesthetized ponies—failure to increase arterial ventilation rather than assisted ventilation oxygenation. Res Vet Sci 1998;65:227–232. should be used in the anesthetized horse, because 15. Dodam JR, Moon RE, Olson NC, et al. Effects of clenbuterol horses do not consistently trigger the ventilator hydrochloride on pulmonary gas exchange and hemodynam- ics in anesthetized horses. Am J Vet Res 1993;54:776–782. frequently enough to ensure normal PaCO2 tensions. The primary determinants of appropriate 16. Robertson SA, Bailey JE. Aerosolized salbutamol (albuterol) improves PaO2 in hypoxemic anaesthetized horses—a controlled ventilation are tidal volume (12–15 ml/ prospective clinical trial in 81 horses. Vet Anaesth Analg kg) and respiratory rate (6–8 breaths/min for 2002;29:212–218. adults and 8–10 breaths/min for foals).25 Inspira- 17. Nyman G, Frostell C, Hedenstierna G, et al. Selective me- tory times are somewhat dependent on the capa- chanical ventilation of dependent lung regions in the anaesthetized horse in dorsal recumbency. Br J Anaesth bility of the ventilator, but they should never 1987;59:1027–1034. exceed 50% of any respiratory cycle (time from the 18. Matthews NS, Carroll GL. Evaluation of a portable blood- beginning of one breath to the beginning of the gas analyzer in horses, in Proceedings. 51st Annual Amer- next breath) and preferably, should be 20–30% of ican Association of Equine Practitioners Convention the cycle. Conventionally, this is noted as the 2005;47–48. 19. Koenig J, McDonell W, Valverde A. Accuracy of pulse oximetry ratio of inspiratory to expiratory time (I:E ratio). and capnography in healthy and compromised horses during The lower the I:E ratio (1:4 being lower than 1:2 or spontaneous and controlled ventilation. Can J Vet Res 1:1), the less time that there is a positive pressure 2003;67:169–174. within the thorax. Excessively fast delivery of a 20. Rainger JE, Dart CM, Perkins NR. Factors affecting the Ͻ relationship between arterial and end-tidal carbon dioxide breath ( 1 sec) can change the distribution of pressures in the anaesthetized horse. Aust Vet J ventilation because of differences in the compli- 2010;88:13–19. 26 ance of lung segments and should be avoided. 21. Riebold TW, Evans AT, Robinson NE. Emergency ventilation Generally, lower respiratory rates and higher in the horse, in Proceedings. 25th Annual American Associa- tidal volumes should be used. Basic ventilatory tion of Equine Practitioners Convention 1980;113–122. 22. Riebold TW, Evans AT, Robinson NE. Evaluation of the de- guidelines are presented in Table 2. mand valve for resuscitation of horses. J Am Vet Med Assoc 1980;176:623–626. References and Footnotes 23. Mason DE, Muir WW, Wade A. Arterial blood gas tensions in 1. Muir WW, Skarda RT, Milne DW. Evaluation of xylazine the horse during recovery from anesthesia. J Am Vet Med and ketamine hydrochloride for anesthesia in horses. Am J Assoc 1987;190:989–994. Vet Res 1977;38:195–201. 24. Day TK, Gaynor JS, Muir WW, et al. Blood gas values 2. Greene SA, Thurmon JC, Tranquilli WJ, et al. Cardiopulmo- during intermittent positive pressure ventilation and spon- nary effects of continuous infusion of guaifenesin, ketamine, and taneous ventilation in 160 anesthetized horses positioned in xylazine in ponies. Am J Vet Res 1986;47:2364–2367. lateral or dorsal recumbency. Vet Surg 1995;24:266–276.

36 2010 ր Vol. 56 ր AAEP PROCEEDINGS DRUGS AND ANESTHESIA

25. Kerr CL, McDonell WN. Oxygen supplementation and ven- aTorpex, Boehringer Ingelheim Vetmedica, St. Joseph, MO tilatory support. In: Muir WW, Hubbell JAE, eds. Equine 64506-2046. anesthesia monitoring and emergency therapy, 2nd ed. St. bEquine Demand Valve, JD Medical Distributing Co, Inc., Louis, MO: Saunders Elsevier, 2009;332–352. Phoenix, AZ 85029-4914. 26. Robinson NE. The respiratory system. In: Muir WW, Hubbell cEquine endotracheal tube, Jorgensen Laboratories, Loveland, JAE, eds. Equine anesthesia monitoring and emergency therapy, CO 80538-3683. 2nd ed. St. Louis, MO: Saunders Elsevier, 2009;11–36. dTafonius, Hallowell EMC, Pittsﬁeld, MA 01201-4714.

AAEP PROCEEDINGS ր Vol. 56 ր 2010 37