Answering your questions: Sedating and anesthetizing patients that have organ system dysfunction

John D. Jacobson, DVM, MS, DACVA

Preanesthetic evaluations can reduce morbidity by alerting a clinician to take action to optimize a patient's preoperative status and perioperative management.

Q: Based on the results of physical examination and routine screening tests, what constitutes dysfunction of such severity to warrant a change in routine sedation or anesthetic protocols?

Preanesthetic evaluations can reduce morbidity by alerting a clinician to take action to optimize a patient's preoperative status and perioperative management. In my opinion, preanesthetic evaluations should initially include only a thorough history and physical examination. When abnormalities are found in the history and during the physical examination, additional diagnostic tests should be performed to confirm and measure these findings. When no abnormalities are identified in the history or during physical examination and a complete blood count, serum chemistry profile, and urinalysis are nevertheless performed, the likelihood of obtaining an abnormal test result is high, but I think the clinical relevance of an abnormal result under these circumstances is questionable.1 We get into trouble if we make erroneous laboratory results the problem.

For example, a threefold increase in liver enzyme activity on a serum chemistry profile may suggest that something (e.g. trauma, inflammation) has caused enzymes to leak out of liver cells; however, this does not tell us much about liver function or the liver's ability to metabolize anesthetic agents. Without supporting evidence, I probably wouldn't change my anesthetic protocol.

However, the decision to alter my protocol is obvious if I am already looking for changes in liver function based on the patient's history and physical examination findings, and the tests reveal decreased albumin and blood urea nitrogen concentrations. Such abnormalities suggest decreased liver function.

Another situation that requires a change in anesthetic management would be any elevation in the serum creatinine concentration. If it is determined that renal azotemia is present, we know that well over half of kidney function is gone. Measures I would take in response include

  • Eliminating any prerenal azotemia component and electrolyte abnormalities (e.g. potassium) before anesthesia

  • Maintaining a diuresis throughout the perioperative period by using a higher fluid administration rate (15 to 20 ml/kg/hr) and monitoring urine output (oliguria is less than 1 to 2 ml/kg/hr)

  • Monitoring anesthetic depth more closely and making appropriate changes more frequently to avoid excessive anesthetic agent administration.

With regard to changes in the use of specific anesthetic agents, ketamine is generally avoided in cats with renal disease since recovery may be unnecessarily prolonged because of decreased renal elimination. I also avoid alpha2 agonists, such as xylazine and medetomidine, in any patient with systemic disease, as cautioned by their labels. Dramatic changes in oxygen delivery to the tissues can occur in healthy patients after administering alpha2 agonists,2-5 and compromised patients should not be challenged when alternatives are available. Instead, I premedicate with opioids intravenously so I can avoid tranquilizers, decrease my induction agent dosages, and decrease vaporizer settings intraoperatively.

Any hepatic, renal, cardiovascular, pulmonary, or central nervous system dysfunction in a patient should elicit a proportional response when making decisions about the patient's anesthetic management. If I can reasonably use a measure that may improve preoperative patient status and perioperative management, I should take that measure.

Q: Is intravenous induction safer than mask induction in dogs or cats with cardiac or respiratory disease?

The concentrations of inhalant anesthetics required for mask induction cause cardiorespiratory depression (i.e. peripheral vasodilation, hypercapnia) similar to or greater than that associated with most intravenous induction agents. Moreover, mask inductions are usually more expensive than injectable inductions, and environmental pollution is unavoidable.

Perceived advantages of mask inductions include the following:

  • Inhalation anesthetics can be easily eliminated from the body with ventilation; they are not nearly as dependent on redistribution and metabolism for recovery as most injectables are.

  • The change in anesthetic depth is typically gradual compared with boluses of intravenous induction agents, giving the patient time to compensate for cardiovascular changes.

  • Intravenous access is not required (though it is ideal before any anesthetic induction).

I have switched from intravenous to mask induction when the premedication effect was much greater than I expected. In this situation, mask induction is possible by using vaporizer settings no higher than 1% isoflurane or 2.5% sevoflurane.

I usually avoid mask inductions in patients with pulmonary disease because inhalation anesthetics depend on the respiratory tract as the administration route. This is particularly important in patients with upper airway obstruction. In these patients, the goal of rapid induction and intubation is usually accomplished with intravenous induction agents (e.g. thiopental, propofol). Patients experiencing upper airway obstruction or hypoxemia should receive oxygen through a mask or nasal cannula for three to five minutes before induction. Struggling can be avoided by sedation with an opioid; however, excessive sedation may compromise a patient's respiratory drive and should be avoided in dyspneic patients. Patients with cardiac disease may benefit from preoxygenation as well to minimize deficits in oxygen delivery.

In general, inducing anesthesia in patients with cardiac disease usually relies on high doses of opioids (e.g. fentanyl, hydromorphone) that have minimal effects on cardiovascular performance (i.e. blood flow to the tissues). I usually intravenously administer an opioid and a benzodiazepine followed by a low dose of thiopental, propofol, ketamine and diazepam, or etomidate, if necessary. As stated previously, low concentrations of an inhalant may also be used to complete induction. The goal is a slow transition toward a surgical plane of anesthesia that allows the cardiovascular and autonomic nervous systems more time to respond. This also allows the anesthetist more opportunity to prevent overdose.

Q: What are the differences between the cardiovascular effects of isoflurane and sevoflurane?

I am not aware of a single cardiopulmonary characteristic I would use to persuade someone to switch from one of these agents to the other. Sevoflurane has a low blood:gas solubility coefficient, thus changes in anesthetic depth occur more rapidly than with isoflurane. This includes changes during anesthesia as well as inductions and recoveries. However, I think many other factors affect recovery, and I don't use mask inductions frequently enough to desire a more rapid induction. I do like the ability to change anesthetic depth more quickly, but as many practitioners will remember, that characteristic made our change from halothane to isoflurane frustrating because it was difficult to achieve a steady plane of anesthesia.

Q: For patients with organ system dysfunction, I've heard that the best protocol is to induce anesthesia with propofol and maintain with isoflurane. Is there really anything better?

While some fantastic anesthetic agents have become available over the past 20 years, none of them is perfect. It is not the anesthetic agent itself that makes it safe, but how it is used. Knowing the advantages and disadvantages of any anesthetic agent or combination of agents will enhance its value.

Propofol with isoflurane is a valuable option for induction and maintenance of anesthesia. Both can be used to produce rapid changes in anesthetic depth and are associated with short recovery times.

However, remember that neither propofol nor isoflurane provides analgesia. Both propofol and isoflurane are dose-dependent respiratory depressants, and the anesthetist should be prepared to support ventilation. Likewise, both anesthetic agents cause dose-dependent peripheral vasodilation,6-8 and patients predisposed to hypotension will be at risk. Propofol is short-acting, and keeping a patient anesthetized with isoflurane alone can require high vaporizer settings that are more likely to result in these cardiopulmonary consequences. Using opioid premedication for analgesia with this combination will allow the use of less propofol and lower vaporizer settings. This will be particularly important in patients with organ dysfunction. In addition, high-dose or multiple administrations of propofol in cats has caused Heinz body formation.9,10

Typically, there are no advantages to using propofol rather than thiopental when inducing anesthesia before general anesthesia with inhalant anesthetics. Cardiopulmonary effects are similar, and when large doses of thiopental are avoided by using premedication, no differences in recovery times are noted. Along the same lines, isoflurane typically has no advantages over sevoflurane.

Q: What are your recommendations for sedating and anesthetizing dogs that are heartworm-positive?

I would avoid anesthetizing patients with heartworm disease. Anesthesia may be necessary to remove heartworms in patients with caval syndrome, but this situation is not common. If sedation is required for diagnostic evaluation of heartworm-positive dogs, an opioid with or without a benzodiazepine can be used. All elective procedures should be postponed until after treatment.

Q: What are your protocol recommendations for sedating and anesthetizing cats that have hyperthyroidism?

The most important anesthetic considerations associated with surgical treatment of feline hyperthyroidism relate to the cardiovascular status and the corresponding treatment with antithyroid medication or beta-blockers. Minimizing anesthetic risk in these patients depends on medical treatment before surgery. Patients with severe tachycardia or heart failure should be treated medically before surgery, and such treatment should not be discontinued before anesthesia. Anesthetic considerations include monitoring cardiovascular performance, decreased access to the head during neck surgeries, and potential hemorrhage. Postoperative complications include hypothermia, hypocalcemia, and laryngeal paralysis.

Q: What special type of anesthetic monitoring should be done in patients with organ system dysfunction?

The noninvasive monitoring equipment that is becoming increasingly prevalent in hospitals (e.g. pulse oximetry, capnography, electrocardiography, temperature monitor, blood pressure monitor) can be used in all patients, not just those with organ system dysfunction. In patients with organ system dysfunction, one type of monitoring may be more valuable than another, depending on the patient's problems or the type of procedure being performed. Pulse oximetry is more important to me during bronchoscopy or thoracostomy tube placement. Capnography is difficult to monitor unless the patient is intubated, and displayed values are more difficult to interpret in spontaneously breathing patients under general anesthesia. Electrocardiography will be more important in patients predisposed to arrhythmias (e.g. posttrauma patients, patients with splenic tumors or gastric dilatation volvulus).

More invasive monitoring (e.g. direct arterial blood pressure measurement, central venous pressure monitoring, closed urine collection system, arterial blood gas evaluation) may be reserved for more critically ill patients that may be experiencing organ failure.

Many practitioners think they cannot afford to have all the monitoring equipment now available. I strongly encourage every practice to consider purchasing this equipment. However, there are no good electrical substitutes for frequent auscultation with an esophageal stethoscope and frequent evaluation of peripheral pulse quality, palpebral reflex, jaw tone, mucous membrane color, and capillary refill time. No electrical monitor is as versatile as the anesthetist. So every practice should routinely use a trained technician devoted to monitoring a patient's depth of anesthesia and physiologic responses to the anesthetic agents throughout the procedure and into recovery. In some cases, it may be practical to refer patients to practices where they can be adequately monitored.

Q: Take the case of a young whippet presented 20 minutes after it ingested over 2 lb of chocolate. An arrhythmia was auscultated that had not been previously noted in the medical record. Sedation before gastric lavage was needed. What is the best way to proceed when immediate electrocardiography is not practical?

This situation is frequently encountered. The patient's condition will not always wait for the clinician to collect all the desired information. The goal in this case is to immediately limit absorption of a large amount of theobromine. I suggest placing an intravenous catheter and administering an intravenous bolus of a balanced electrolyte solution (10 to 20 ml/kg). Then perform a rapid induction and intubation sequence (to avoid vomiting and aspiration during induction) with an opioid (e.g. hydromorphone, fentanyl) followed by propofol and an inhalant anesthetic. If electrocardiography is unavailable, frequent evaluation of the peripheral pulse quality and rhythm may be the best substitute. I would not attempt to treat the arrhythmia without a diagnosis, but it would be negligent to anesthetize a patient with ventricular tachycardia without making initial attempts to treat it. If an electrocardiograph is available, I would delay anesthesia for a diagnosis.

If ventricular tachycardia was identified in this or any patient, treatment should be guided by the ECG findings as well as the patient's pulse rate and quality, mentation, respiratory rate, mucous membrane color, capillary refill time, and, perhaps, blood pressure to determine the clinical effect of the dysrhythmia. Because anesthetic agents can compromise cardiovascular performance, ventricular tachycardia may have a greater effect in anesthetized patients. Lidocaine (1 to 2 mg/kg intravenously) can be administered up to a total dose of 8 mg/kg in a 10-minute period in an awake patient. Eliminating all ventricular ectopy is desired but not essential. The effect on ventricular tachycardia can be evaluated within two minutes of lidocaine administration, but the duration of action is short-lived, and readministration is likely to be necessary. Anesthetized patients should receive no more than about 0.5 mg/kg of lidocaine intravenously at one time because in my experience lidocaine has caused asystole in susceptible anesthetized patients.

Q: Are additional supportive measures needed in anesthetized patients with organ system dysfunction?

All patients can benefit from supportive measures such as providing intravenous access, fluid therapy, temperature homeostasis, and positive pressure ventilation. Patients with organ system dysfunction have decreased physiologic reserve, meaning they have a more limited capacity to tolerate the principal causes of morbidity and mortality we are trying to prevent (i.e. hypotension, hypovolemia, hypercapnia, hypoxemia, hypothermia, electrolyte and fluid imbalances). Ventilatory support (manual or mechanical) helps prevent respiratory acidosis, helps efficiently deliver inhalant anesthetics, and may improve airway monitoring. Intravenous fluid administration will help prevent hypovolemia, maintain a diuresis, and provide intravenous access for anesthetic agents or emergency medications. Warm-water or warm-air blankets can prevent hypothermia, excessive anesthetic depth, and prolonged recoveries. These types of support measures can make the difference necessary to prevent perioperative complications in patients.

As with monitoring practices, some support measures may be more valuable than others, and decisions can be tailored to the patient's presentation. The best supportive measure in all patients is providing continuous monitoring of the anesthetic equipment and anesthetic depth to avoid deep planes of anesthesia associated with decreased oxygen delivery to vital organs, hypothermia, and prolonged recoveries. This can only be accomplished by having an anesthetist devoted to monitoring the patient. This supportive measure determines the difference between being able to recognize that your patient is decompensating and being able to prevent it from happening in the first place. This additional supportive measure is needed in anesthetized patients with organ dysfunction.

It is common to forego special monitoring and support in patients that are only sedated and not anesthetized. Patients with organ system dysfunction will be particularly susceptible to hypoxemia, hypoventilation, and upper airway obstruction. In sedated patients, it is just as important to monitor the airway to make sure that the patient is moving air adequately. In general, if a patient can be intubated, the patient usually should be intubated. Oxygen supplementation should be considered whether or not the patient can be intubated.

Q: What anesthetic protocol and supportive measures do you recommend for a dog with a portosystemic shunt?

Patients with portosystemic shunts can be expected to have decreased ability to metabolize anesthetics dependent on liver function. Hypoalbuminemia may be present in these patients, and care should be taken to avoid excessive dilution of plasma proteins with crystalloid fluids. Colloids may be used. Coagulopathies and hypoglycemia are also potential concerns. Some of these patients will have hepatoencephalopathy and exaggerated responses to anesthetic agents (e.g. benzodiazepines). Acepromazine is usually avoided or used in low doses because it requires liver metabolism and may predispose a patient to hypotension and seizures. I usually administer only an opioid (often intravenously to obtain a greater effect) followed by mask induction with an inhalant to avoid using other medications that require hepatic metabolism. Monitoring blood pressure is essential to evaluate a patient's response to surgical correction. Ideally, blood pressure should be monitored invasively by using an arterial catheter. Hypothermia is a common complication; vaporizer settings should be appropriately turned down, warm-water or warm-air blankets should be used, and cold fluid administration should be avoided.

Q: How do you manage the delicate balance of fluid therapy during anesthesia in an animal with both renal and cardiac disease?

Fluid administration goals should include avoiding overhydration and maintaining urine output. Electrolyte abnormalities should be corrected before the procedure, if possible. Monitoring central venous pressure and performing chest auscultation to detect signs of pulmonary edema can help practitioners avoid excessive fluid administration in patients predisposed to volume overload. Some clinicians would prefer administering fluids with a lower sodium concentration rather than a balanced electrolyte solution in these patients. Urine output can be grossly evaluated by monitoring changes in urinary bladder size; however, a continuous closed monitoring system will provide a quantitative evaluation. Measuring the fluids administered and eliminated as well as monitoring body weight perioperatively may help maintain fluid homeostasis.

Using an anesthetic agent protocol that will minimize changes in cardiovascular performance (e.g. highly dependent on opioids for analgesia) will help prevent both pulmonary edema and inadequate perfusion of the kidneys. Maintaining mean arterial blood pressure above 70 mm Hg by using lower vaporizer settings will also help ensure renal perfusion. Continued monitoring well after recovery is prudent.

Q: How should we manage diabetic patients before surgery?

The goals of anesthesia in a patient with diabetes mellitus include avoiding both hypoglycemia and hyperglycemia. The patient's glucose concentration should be well-regulated before surgery. Eliminate ketoacidosis, dehydration, and electrolyte abnormalities before anesthesia.

Withhold food eight hours before anesthesia, but be sure to schedule the procedure for early in the morning. Administer half the typical insulin dose in the morning, and obtain a glucose sample before the procedure. If the value is less than 80 mg/dl, start a 5% dextrose intravenous infusion before surgery. If the value is greater than 300 mg/dl, administer regular insulin (0.2 IU/kg subcutaneously). Dextrose administration during the procedure is often not necessary, but a balanced electrolyte solution with 2.5% dextrose (5 to 10 ml/kg/hr intravenously) may be used.

I avoid alpha agonists because of reports of hyperglycemia,11 and mask inductions may cause unnecessary stress. Obtain a glucose sample after the procedure or during a lengthy procedure. Return the patient to small portions of food as soon as is reasonable after the patient has recovered.

Q: Should practitioners incorporate nitrous oxide more often in their anesthetic protocols?

I have found nitrous oxide useful during mask inductions in attempts to avoid an excitement stage and during general anesthesia as one means of providing additional analgesia in patients that are too light. Nitrous oxide provides additional analgesia and can produce more rapid changes in anesthetic depth without the characteristic cardiovascular effects of other inhalant anesthetics (e.g. isoflurane, sevoflurane). Nitrous oxide can also be used for economical reasons.

While people are frequently anesthetized with nitrous oxide, I don't expect the popularity of nitrous oxide to change in veterinary medicine. The advantages of using nitrous oxide are often over shadowed by the desire to avoid complex anesthetic protocols. Administering a hypoxic mixture of gases is much more likely when using nitrous oxide, and I wouldn't recommend its use in practices that do not have technicians dedicated to anesthesia services.

Nitrous oxide should be avoided or used with extreme caution in patients that are hypoxemic preoperatively or that have pneumothorax. Nitrous oxide is delivered in concentrations of 50% to 70%. This decreases the maximum percentage of oxygen that can be delivered, making some patients more susceptible to hypoxemia. Nitrous oxide crosses membranes faster than nitrogen found in air-filled spaces in the body (e.g. pneumothorax, gastric dilatation); nitrous oxide can expand these gas-filled spaces and create additional problems.

John D. Jacobson, DVM, MS, DACVA

College of Veterinary Medicine

Western University of Health Sciences

Pomona, CA 91766

*Current address: 243 Earhart Circle, Lawrence, KS 66049


1. Roizen MF. More preoperative assessment by physicians and less by laboratory tests. N Engl J Med 2000;342:204-205.

2. Haskins SC, Patz JD, Farver TB. Xylazine and xylazine-ketamine in dogs. Am J Vet Res 1986;47:636-641.

3. Pypendop BH, Verstegen JP. Hemodynamic effects of medetomidine in the dog: a dose titration study. Vet Surg 1998;27:612-622.

4. Pypendop B, Serteyn D, Verstegen J. Hemodynamic effects of medetomidine-midazolam-butorphanol and medetomidine-midazolam-buprenorphine combinations and reversibility by atipamezole in dogs. Am J Vet Res 1996;57:724-730.

5. Jacobson JD, McGrath CJ, Ko JC, et al. Am J Vet Res 1994;55:835-841.

6. Steffey EP, Howland D Jr. Isoflurane potency in the dog and cat. Am J Vet Res 1977;38:1833-1836.

7. Klide AM. Cardiopulmonary effects of enflurane and isoflurane in the dog. Am J Vet Res 1976;37:127-131.

8. Quandt JE, Robinson EP, Rivers WJ, et al. Cardiorespiratory and anesthetic effects of propofol and thiopental in dogs. Am J Vet Res 1998;59:1137-1143.

9. Andress JL, Day TK, Day DG. The effects of consecutive day propofol anesthesia on feline red blood cells. Vet Surg 1995;24:277-282.

10. Matthews NS, Brown RM, Barling KS, et al. Repetitive propofol administration in dogs and cats. J Am Anim Hosp Assoc 2004;40:255-260.

11. Thurmon JC, Steffey EP, Zinkl JG, et al. Xylazine causes transient dose-related hyperglycemia and increased urine volume in mares. Am J Vet Res 1984;45:224-227.