Comparison of Propofol and Propofol/Ketamine Anesthesia for Evaluation of Laryngeal Function in Healthy Dogs

Introduction

Laryngeal paralysis is a common acquired disease of older, large-breed dogs.^1,2 Congenital and acquired forms can occur, with idiopathic acquired being the most common presentation.^1–3 Arytenoid paralysis results in partial to complete loss of laryngeal function, resulting in variable degrees of upper airway obstruction. Affected dogs present with clinical signs ranging from mild exercise intolerance to life-threatening respiratory distress. Disease severity is directly correlated with the degree of laryngeal dysfunction, with clinical signs occurring most frequently with bilateral disease. Regardless of the underlying etiology, accurate observation of arytenoid function is critical to determine if laryngeal paralysis is present.

Echolarynography and transnasal laryngoscopy have been reported for laryngeal function evaluation; however, diagnosis of laryngeal paralysis is most commonly made using direct oral laryngoscopy under a light plane of anesthesia.⁴ The ideal anesthetic protocol for accurate laryngeal examination results in a light anesthetic plane with decreased jaw tone, intact laryngeal reflexes, and minimal respiratory depression. Doxapram has been added to variable anesthetic protocols to stimulate respirations in apneic patients and can increase intrinsic laryngeal motion, facilitating laryngeal function examination in dogs.³

Thiopental, propofol, diazepam/ketamine, and gas anesthesia have all been used to provide anesthesia for laryngeal function examination.^5–7 Thiopental has been found to be the best injectable anesthetic for evaluating laryngeal function.^5,6 Unfortunately, thiopental is no longer commercially available. In the absence of thiopental, the ideal anesthetic protocol for the evaluation of laryngeal function has yet to be determined. Studies evaluating propofol for laryngeal examination have revealed adequate laryngeal exposure with weaker laryngeal motion compared with thiopental.^5,6 Unfortunately, significant respiratory depression often occurs with propofol, with apnea related to dose, speed of injection, and concurrent premedications, complicating laryngeal evaluation.⁸ Ketamine has been shown to preserve the laryngeal reflex to a greater degree than thiopental in people, although the anesthetic plane may be inadequate for veterinary patients given previous studies.^5,6,9 Ketamine, in conjunction with propofol, may hypothetically provide an additive effect on relaxation and laryngeal exposure, while minimizing respiratory depression. The objective of this study was to compare the effects of propofol to propofol/ketamine anesthesia on laryngeal visualization and function in healthy dogs. The study authors hypothesized that propofol/ketamine would provide an adequate plane of anesthesia with decreased respiratory depression for laryngeal examination in dogs.

Materials and Methods

Experimental Protocol

Each dog was anesthetized once for elective sterilization with one of two randomly assigned anesthetic protocols (i.e., propofol^a, propofol/ketamine^b). Dogs were premedicated with butorphanol^c (0.5 mg/kg) and glycopyrrolate^d (0.01 mg/kg) injected intramuscularly 20 min prior to anesthetic induction. An IV catheter was placed approximately 15 min after premedication followed by 5 min of flow-by O₂ prior to anesthetic induction.

Depending on the patient’s assigned protocol, either propofol (6 mg/kg) or ketamine (2 mg/kg) followed by propofol (2–4 mg/kg) were administered IV for anesthetic induction until the mouth could easily be opened for examination. Propofol was administered slowly over 1 min to effect in both protocols. The amount of propofol required to achieve adequate anesthesia for laryngeal examination was recorded. The percent of the propofol dose administered was calculated to normalize dose for body weight and allow dose comparisons between groups. Ketamine was given as a calculated bolus followed by propofol to effect. The anesthesiologist administered all anesthetic agents with the exception of the intramuscular premedications, which were given by fourth year veterinary students under direct supervision. Laryngeal function observers were blinded to the administered protocol by giving an equal volume of normal saline in lieu of ketamine to patients in the propofol only group. Following anesthetic induction, dogs were placed in sternal recumbency with the head elevated to the level of normal carriage. Laryngeal assessment was performed as soon as the mouth could be opened, and the larynx was visualized by the same two individuals in succession (K.M., M.R.). Laryngeal examination began within 30 min of premedication. Interobserver experience (between a third year resident and a board-certified surgeon with ∼ 20 yr of experience) was normalized by alternating who first evaluated each patient. The evaluations were defined as early if they were the first evaluator and late if they were the second evaluator. All evaluations were completed within 4 min of induction. A similar protocol for laryngeal examination described by Gross et al. (2002) was used.⁵ Breathing score was defined as the score of the breath when laryngeal function determination was made and was assigned by the laryngeal observers. Parameters were outlined as described in Table 1. Swallowing and laryngospasm were graded together as either present or absent. Laryngeal function was characterized as normal (arytenoid cartilages abduct during inspiration), weak (arytenoid cartilages abduct but poorly), or abnormal (one or more arytenoid cartilages do not abduct during inspiration). The observer’s ability to assess the co-ordination between respirations and arytenoid abduction was facilitated by a technician verbalizing the onset of inspiration. Overall respiratory score was characterized by the anesthesiologist following drug administration using the scale described in Table 2 and was a determination of how well the dogs breathed during the entire laryngeal examination.

TABLE 1 Definitions of Laryngeal Examination Responses

View Fullscreen

TABLE 2 Overall Respiratory Function Evaluation

*

0 = no spontaneous respirations; 1 = shallow respirations, slow respiratory rate, weak attempt; 2 = moderate respirations, respiratory rate, and attempt; 3 = deep respirations, normal respiratory rate, strong attempt

View Fullscreen

Anesthetic monitoring, including heart rate, respiratory rate, and saturation of peripheral O₂ (SPO₂) measured by pulse oximetry were performed during anesthetic induction and laryngeal examination. If a patient failed to abduct the arytenoid cartilages, a cotton tipped applicator was used to gently stimulate respiration by direct application of pressure to one arytenoid. If the patient became apneic (no spontaneous respirations ≥ 60 sec) and respirations could not be manually stimulated, a single dose of doxapram^e (1 mg/kg IV) was administered and laryngeal function was subsequently evaluated. Doxapram administration, as well as whether it was administered early or late in the examination, was recorded. Following laryngeal examination, patients were intubated and isoflurane in O₂ administered in preparation for elective sterilization.

Results

Forty-eight primarily mixed-breed dogs undergoing elective sterilization procedures were included in this study. The mean body weight ± standard deviation was 23.1 kg ± 5.1 kg in the propofol group and 22.5 kg ± 5.8 kg in the propofol/ketamine group and was not statistically different between groups.

The mean percentage of total calculated propofol administered in the propofol and propofol/ketamine group were 64.2% ± 20.3% (3.8 mg/kg ± 1.2 mg/kg) and 60.1% ± 18.5% (2.4 mg/kg ± 0.7 mg/kg), respectively, and were not significantly different between the two groups (two tailed t test, P = 0.464).

The summary of the laryngeal examinations between the two observers were presented in Tables 3 and 4. The first evaluator’s results were designated as early and the second evaluator results as late. No significant difference was identified in breathing scores in regards to either treatment group or time of evaluation (γ statistic, −0.0309 and P = 0.8955 for early evaluation; γ statistic, 0.0265 and P = 0.9112 for late evaluation). Evaluation of jaw tone score by the early evaluator was not significantly different between the two treatment groups (γ statistic −0.2484; P = 0.3351). A significant difference was noted when evaluating jaw tone score between the two groups late, with dogs in the propofol/ketamine group having statistically significant lower jaw tone scores than the propofol group (γ statistic −0.5063; P=0.0178). Laryngeal exposure scores were not significantly different between either time of evaluation or treatment received, with all dogs having moderate to excellent exposure. No significant difference was noted in the prevalence of swallowing/laryngospasm with regards to treatment group or time of evaluation (Fisher exact test, P = 0.3715 for early evaluation and P = 0.7702 for late evaluation).

TABLE 3 Summary of Early Subjective Scoring of Laryngeal Examinations Between Protocols

*

0 = no spontaneous respirations; 1 = shallow respirations, slow respiratory rate, weak attempt; 2 = moderate respirations, respiratory rate, and attempt; 3 = deep respirations, normal respiratory rate, strong attempt

†

0 = absent; 1 = slight; 2 = moderate; 3 = strong

View Fullscreen

TABLE 4 Summary of Late Subjective Scoring of Laryngeal Examinations Between Protocols

*

0 = no spontaneous respirations; 1 = shallow respirations, slow respiratory rate, weak attempt; 2 = moderate respirations, respiratory rate, and attempt; 3 = deep respirations, normal respiratory rate, strong attempt

†

0 = absent; 1 = slight; 2 = moderate; 3 = strong

View Fullscreen

Overall respiratory scores between the two treatment groups have been summarized in Table 2, with no significant difference identified (γ statistic, −0.3522; P = 0.0759). Higher mean respiratory rates were noted in the propofol group than the propofol/ketamine group, which were not statistically significant at 1 min following drug administration (28 breaths/min in the propofol group and 16 breaths/min in the propofol/ketamine group; P = 0.0518). Mean SPO₂ values were significantly higher in the propofol group than the dogs receiving propofol/ketamine (94% in the propofol group and 87% in the propofol/ketamine group 1 min following drug administration; P = 0.0008).

The breathing score was significantly related to the potential for a patient to have a normal laryngeal function examination. The breathing score and laryngeal function were taken from each evaluator to give a total of 96 data points for analysis. A significant difference was found in the likelihood of a patient having normal laryngeal function when a higher breathing score (i.e., deeper breath) was used to make that determination compared with lower breathing scores (γ statistic, 0.7610; P < 0.0001).

The two evaluators disagreed in the evaluation of laryngeal function determination 58.3% of the time. When selecting the result (i.e., either early or late) with a higher breathing score, that value decreased to 20.8%, with 80% of the discrepancies existing being between weak and normal laryngeal function determinations. Weak laryngeal examinations were classified as normal for statistical analysis because laryngeal abduction was present. Two dogs were classified as abnormal by one evaluator and normal by the other despite equal breathing scores at the time of evaluation. One dog was determined to have an abnormal laryngeal evaluation by both evaluators due to failure of the right arytenoid to abduct. Six dogs (12.5%) would have been classified as abnormal based on examination and a low breathing score, but were evaluated as normal by the second observer based on a higher breathing score during their evaluation.

Doxapram was administered to 6 of 24 dogs in the propofol group and 10 of 24 dogs in the propofol/ketamine group. The need for doxapram was not statistically different between the two groups (two sided t test, P = 0.3587). Doxapram administration early tended to result in higher breathing scores (γ statistic −0.6279; P = 0.0373) and increased the likelihood of normal laryngeal function determination (γ statistic −0.8125; P = 0.0002).

Discussion

Laryngeal paralysis is a commonly diagnosed acquired disease of older, large-breed dogs.^1,2 Diagnosis is most commonly based on oral laryngoscopy under a light plane of anesthesia in combination with historical and physical examination data. Ideal anesthetic protocols allow adequate depth of anesthesia for laryngeal visualization (while protecting the examiner/equipment), cause minimal respiratory suppression, and maintain the laryngeal reflex. The laryngeal reflex is defined as adduction of the arytenoid cartilages following exhalation and stimulation of the epiglottis.^5,10 The reflex is a protective mechanism against aspiration. That reflex, combined with the observer’s ability to safely and adequately observe the reflex, is typically used as an end point for determining whether an adequate depth of anesthesia has been obtained, while maintaining laryngeal function. The patient is considered too deep for accurate evaluation of laryngeal function once that reflex is lost, increasing the risk of falsely diagnosing laryngeal paralysis.

Historically, thiopental has been the drug of choice for laryngoscopy in most practices. Thiopental is an ultrashort-acting injectable barbiturate that has been used for anesthetic induction in dogs. Thiopental was shown to have the least suppressive effect on arytenoid motion prior to anesthetic recovery and allowed accurate determination of laryngeal function in dogs and people.^6,11 Unfortunately, thiopental has become unavailable and the need for alternate protocols has arisen. Previous studies have evaluated propofol, ketamine/diazepam, acepromazine/propofol, and gas anesthesia independently for laryngeal function examinations with variable success.^5–7 Ketamine/diazepam studies have resulted in decreased laryngeal exposure due to excessive jaw tone as well as decreased arytenoid movement.^2,5,6 Propofol has been associated with apnea and weak arytenoid motion when used either alone or in combination with acepromazine.⁶

Propofol is a highly lipid soluble alkylphenol derivative that has been used for sedation, anesthetic induction, and maintenance of anesthesia.¹² It provides a rapid and smooth induction of anesthesia by enhancing the effects of γ-aminobutyric acid.^12–14 Although propofol provides excellent muscle relaxation, propofol can result in apnea and significant respiratory depression following bolus administration, making laryngeal examination difficult.

Ketamine is a phencyclidine derivative that acts as a dissociative anesthetic through noncompetitive binding of the N-methyl-d-aspartate receptor causing depressed thalamocortical and limbic system activity.^12–14 A dose-related effect is seen on the respiratory system with minimal depression at lower dosages, although additive effects have been reported when ketamine is combined with other central nervous system depressants.^13–15 Although ketamine is more apt to maintain normal airway reflexes and respiratory drive, muscle relaxation is poor. In this study, the authors compared propofol to propofol/ketamine anesthesia with the hope that the combination protocol would accentuate the favorable effects of both drugs (i.e., better muscle relaxation with the propofol and using ketamine to reduce the dose of propofol required for anesthesia, thereby causing less respiratory depression and maintaining airway reflexes).

The percentage of total calculated propofol administered was not significantly different between the two groups. That was evaluated to determine whether ketamine resulted in propofol-sparing effects on the amount required to facilitate laryngeal examination. Ketamine did not reduce the propofol requirement for laryngeal examination. That was similar to findings described by Mair et al. (2009) evaluating the propofol-sparing effects of ketamine in coinduction of anesthesia in dogs, where ketamine did not change the blood propofol targets required for endotracheal intubation.¹⁵

Evaluation of laryngeal function requires adequate laryngeal exposure, which has been a disadvantage of many anesthetic protocols. In this study, no significant difference in the prevalence of either swallowing or laryngospasm was observed in either treatment group. Jaw tone score was significantly decreased during the late observer’s evaluation in patients who received ketamine. That was likely due to the continued action of the ketamine following rapid redistribution of the propofol, which was in contrast to another study that found increased laryngospasm and jaw tone in patients receiving diazepam/ketamine.⁵ The additive effects of the propofol may have resulted in enough sedation to overcome the poor muscle relaxation achieved with ketamine alone, resulting in better laryngeal exposure scores in the current study.

The study authors evaluated each patient at two separate time points to determine if the ability to evaluate the larynx changed as the patients neared recovery compared with initial induction (i.e., late versus early time points). No significant difference was present in laryngeal exposure scores between the two time points and no difference between treatment groups was noted, indicating that both anesthetic protocols provided prolonged and adequate exposure for evaluation. Clinically, that translates to being able to evaluate those patients through several respirations, with final evaluation made just prior to recovery if function is unclear.

Overall respiratory scores during the examination were not statistically different between the two groups. Patients in the propofol/ketamine group had lower overall respiratory scores compared with the propofol only group, although that difference did not reach statistical significance. That is likely due to inadequate patient numbers. Clinically, decreased respiratory rates translated to statistically significant decreases in SPO₂ readings in dogs receiving propofol/ketamine. That was supported by previous studies with variable ketamine doses (0.5–2 mg/kg) where increased respiratory depression was observed with propofol/ketamine compared with propofol alone.^13–15 The true impact of the propofol/ketamine combination was difficult to ascertain in those studies due to confounding factors such as premedication with acepromazine and morphine, acepromazine and meperidine, and medetomidine, all of which have the potential for significant additive respiratory suppression.^13–15 In this study, the authors elected to use butorphanol due to its very mild sedative and respiratory depressive effects. Given the finding of decreased SPO₂ readings in patients receiving propofol/ketamine despite the light premedication, it appears that ketamine may exacerbate the respiratory depressant effects of propofol in dogs. Although increased relaxation may occur with the propofol/ketamine combination, respiratory depression limits its use in laryngeal function examination.

In addition to adequate laryngeal exposure, inspiratory depth and phase of respiration at laryngeal evaluation are important. Paradoxical laryngeal motion with paralyzed arytenoid cartilages being passively abducted upon exhalation can be mistaken for normal arytenoid abduction if the phase of the respiratory cycle is not recognized. This was addressed in the current study by having an assistant verbally identify each inspiratory phase. Shallow inspirations can also make evaluation difficult with weak inspiration resulting in weak arytenoid abduction, and an increased potential to falsely diagnose laryngeal paralysis. Final determination of laryngeal function should always be made during the deepest inspirations possible, as found in this study. Accordingly, because observing arytenoid function, depth, and timing of inspiration simultaneously is difficult, if not impossible, for a single observer, an assistant should be present to assist with those determinations during laryngeal examination. Patients with a higher breathing score at final laryngeal function evaluation were significantly more likely to be considered normal. That conclusion was strengthened when looking at overall laryngeal function. The majority (58.3%) of the laryngeal function determinations between the first and second observer were in overall disagreement. When considering the breathing score during arytenoid evaluation, and removing the patients with low scores (i.e., shallow breaths) at both evaluation periods, only 20.8% of the patients had differences in their final function evaluation between the two observers, 80% of which were between weak and normal determinations. The difference between weak and normal determinations of laryngeal function between observers could have been because of the subjective nature of the assessment and was clinically unlikely to be significant as most clinicians would classify those together as normal. The clinical significance of weak laryngeal abduction is unknown and must be considered in light of clinical signs. Those patients may either have early laryngeal dysfunction or weak abduction may be a product of the anesthetic protocol. While the true laryngeal function status in those dogs was unknown, given the lack of respiratory signs and young age of the dogs evaluated in this study, weak arytenoid abduction in this population of dogs was most likely a function of the anesthetic protocol, depth of inspiration at evaluation, and subjectivity of the evaluator. Electromyography could have been considered for those dogs to ensure normal arytenoid function; however, that is rarely available clinically and would have added to the invasiveness of this study.

Two dogs were classified as abnormal by one evaluator and normal by the other with equal breathing scores (scores of 1 and 2, respectively). The true status of those dogs is unknown but believed to be normal given the young age and abduction seen by one observer. That variability was likely due to the depth of breath in the first dog (the breathing score was low) as well as effects of anesthesia and variable anesthetic depths at evaluations between the two observers. Observer experience level may also have played a role, with the more experienced observer labeling both of those dogs as abnormal. Clinically, the findings would be taken in conjunction with historical and physical examination findings (normal in both dogs with no historical respiratory disease) when determining therapeutic interventions required. One dog was considered abnormal by both observers, with the right arytenoid failing to abduct for either observer. That finding was believed to be truly abnormal, giving either an incidence of 2% for laryngeal paralysis in this population or an incidence of 6% if the other two dogs above were included, compared with a 25% prevalence of laryngeal paralysis in a group of dogs undergoing general anesthesia.¹⁶ While the number of affected dogs in the current study was much less than previously reported, direct comparisons are difficult as the population of dogs in the aforementioned study were significantly older than the study population included in this study. An increased prevalence of laryngeal disease would be expected in the older population.

The lack of spontaneous respirations can be a tremendous problem clinically in the diagnosis of laryngeal paralysis. Either direct stimulation of the arytenoids or administration of a respiratory stimulant have been proposed as corrective measures.^3–7 Doxapram is a respiratory stimulant that acts through direct stimulation of the medullary respiratory centers as well as the carotid body chemoreceptors.^12,17 Its effects (increased respiratory effort and intrinsic laryngeal motion) are transient, with the primary clinical application for doxapram being for patients at an excessively deep plane of anesthesia and for differentiating dogs with normal laryngeal function.^3,7 Doxapram was used in this study as a rescue protocol for dogs that would not breathe spontaneously, even with direct stimulation of the arytenoids. One-third of the patients in this study required doxapram for laryngeal function evaluation, with no difference between treatment groups. Laryngeal examination would not have been possible in those patients without doxapram. Doxapram resulted in subjectively higher breathing scores for evaluation of laryngeal function and significantly increased the likelihood of a declaration of normal laryngeal function. Doxapram is considered to be generally safe and was well tolerated in the young, healthy population of dogs included in this study, with no adverse effects identified. Reported side effects include tachycardia, arrhythmias, muscle rigidity, seizures, hyperthermia, hypertension, and hypoxia.⁷ Given the low dose administered, the authors felt additive cardiovascular or seizure activity was unlikely in patients receiving ketamine concurrently. Caution should be used in patients with laryngeal paralysis as significant paradoxical movement and possible upper airway obstruction can result, necessitating rapid endotracheal intubation.⁷

Limitations of this study included the lack of a control group, the inability to compare the study results to laryngoscopy using thiopental, and the subjectivity of laryngeal assessment with different breaths used for laryngeal evaluation by each observer. The lack of a control group (i.e., each dog receiving each treatment with an appropriate washout period) was addressed by having larger numbers of dogs in each treatment group and by evaluating a population expected to have normal laryngeal function. The study authors attempted to locate thiopental for comparison in this study without success because it is not currently manufactured. By using the same premedication and similar laryngeal assessment criteria as Gross et al. (2002), the authors felt that some comparisons could be made between the two studies.⁵ The subjectivity of the laryngeal assessment between observers was considered to be a minor concern in this study and similar to that which faces clinicians on a daily basis in the diagnosis of this disease. Although more objective criteria such as electromyography and rima glottis area measurements have been evaluated for the diagnosis of laryngeal paralysis, the diagnosis in the clinical arena remains primarily based on subjective laryngoscopic evaluation coupled with additional clinical and historical information.^3,18 The study authors believe that this study closely mirrored the realities of clinical practice, which was the authors’ intent. Endoscopic videography would have enabled each observer to evaluate the same breath for laryngeal function examination and could have eliminated some of the effect of respiratory character on laryngeal function determination. Although that approach would have allowed more conclusions regarding the effects of interobserver experience on characterization of laryngeal function, the authors’ approach did demonstrate the critical nature of evaluating laryngeal function in conjunction with phase and depth of respiration.

Conclusion

Propofol and propofol/ketamine provided adequate visualization of the larynx in all dogs. Propofol/ketamine failed to have a dose-sparing effect on propofol and resulted in increased respiratory depression evidenced by reduced SPO₂ values making propofol/ketamine a poor choice for the evaluation of laryngeal function in dogs. It is critical that the clinician evaluate laryngeal function in conjunction with depth of inspiration to avoid a false diagnosis of laryngeal paralysis when drug-related apnea results in poor arytenoid function. Doxapram can facilitate deeper inspiration aiding proper laryngeal function examination and should always be administered when arytenoid abduction is not observed under propofol anesthesia to eliminate the possibility of a false positive diagnosis.