Effects of alfaxalone, thiopental, or propofol and diazepam on laryngeal motion in healthy dogs

Abstract

Laryngeal function is assessed by direct visualization of the larynx under a light plane of anesthesia. This study compared the effects of 3 anesthetic protocols on arytenoid motion in healthy dogs. Eight dogs were randomly assigned to receive alfaxalone, propofol and diazepam, or thiopental. Videolaryngoscopy was performed and still images at maximum inspiration and expiration were used to measure the area and height of the glottal gap. The normalized glottal gap area (NGGA = area in pixels/height²) was calculated. The NGAA change was defined as the difference between NGAA during inspiration and exhalation. Data were analyzed using Mann-Whitney and Kruskal-Wallis tests, P-values < 0.05 were considered statistically significant. No significant difference among induction protocols was found when comparing NGGA change after induction or before recovery. Alfaxalone and propofol/diazepam are useful for evaluation of laryngeal function when administered to effect and a light plane of anesthesia is maintained.

Introduction

Laryngeal paralysis is a common cause of upper respiratory tract obstruction in dogs (1,2). Clinical diagnosis of laryngeal paralysis is usually made during laryngoscopy under light plane of anesthesia to permit retraction of the jaws so that a laryngoscope or videoendoscope can be inserted to view the larynx (3). Analgesics, sedatives, and anesthesia induction agents used to facilitate restraint for laryngeal examination may inhibit laryngeal motion (3,4). Several anesthetic induction protocols for evaluation of laryngeal function have been assessed in dogs. One study reported that after use of butorphanol and glycopyrrolate as preanesthetic medication, propofol or thiopental was superior to the combination of ketamine-diazepam for the evaluation of laryngeal motion in dogs (5). A second study found that thiopental administered to effect was a better choice for assessment of laryngeal function in unpremedicated dogs than propofol or ketamine-diazepam (4). Recent lack of availability of thiopental makes propofol the anesthesia induction agent most commonly recommended for the evaluation of laryngeal function in dogs (2). Alfaxalone is a short-acting steroid anesthetic induction agent. Like propofol, alfaxalone produces good muscle relaxation and a rapid and smooth induction (6). One recent study used a subjective scoring system to assess laryngeal function and reported that thiopental, propofol, and alfaxalone administration had similar effects on arytenoid motion (7).

The objective of our study was to compare the effects of alfaxalone, propofol co-administered with diazepam, and thiopental on arytenoid motion in normal dogs during tidal breathing by comparing normalized glottal gap areas during inspiration and expiration. We hypothesized that there would be no difference among the 3 anesthetic protocols and that alfaxalone and the combination of propofol and diazepam would be appropriate for laryngeal examination.

Table 1.

Median (range) change in normalized glottal gap area (NGGA) measured from 3 breaths after induction and before recovery for 3 anesthetic induction protocols used for evaluation of arytenoid motion of 8 healthy dogs.

Parameter	Alfaxalone	Propofol and Diazepam (0.4 mg/kg BW)	Thiopental
Change in NGGA after induction (units)	0.025 (0 to 0.132)	0.032 (0 to 0.140)	0.010 (0 to 0.141)
Change in NGGA before recovery (units)	0.012 (0 to 0.163)	0.042 (0 to 0.174)	0.015 (0 to 0.112)
Dose (mg/kg BW)a	2.6 (2 to 3)	4 (2.5 to 7.5)	14.2 (8.1 to 20)
Videolaryngoscopy time (s)b	357 (108 to 603)	381 (69 to 733)	489 (197 to 992)

^aMedian (range) anesthetic dose and videolaryngoscopy time for examination of arytenoid motion.

^bMean (range).

Materials and methods

Animals

This study was approved by the University of Saskatchewan’s Animal Research Ethics Board (protocol 20090080). Eight adult medium- to large-breed client-owned dogs [mean weight 24 ± 11.0 kg standard deviation (SD); range: 8.5 to 39.4 kg] that were scheduled to undergo anesthesia for dental cleaning for another research project were enrolled in the study after obtaining informed and written owner consent. Dogs had no previous history of respiratory dysfunction and were determined to be healthy based on physical examination. Results of a complete blood (cell) count (CBC) and serum chemistry analysis were within accepted normal limits for our laboratory.

Experimental design

Each dog was administered 3 anesthetic induction protocols in a random order with a minimum rest period of 14 d between treatments. Each anesthetic protocol consisted of an initial dose administered to effect over 1 min and possible top-up boluses: Thiopental 2.5% (Vétoquinol Canada, Lavatrie, Quebec), 10 mg/kg body weight (BW), top-up bolus 2.5 mg/kg BW; Alfaxalone 1% (Alfaxan-CD RTU; Jurox Pty, Rutherford, NSW, Australia), 2 mg/kg BW, top-up bolus 0.5 mg/kg BW; and Propofol/Diazepam: propofol 1% (Rapinovet; Schering-Plough Animal Health, Kirkland, Quebec), 2 mg/kg BW administered first, followed by diazepam (5 mg/mL, Diazepam Injection USP; Sandoz Canada, Boucherville, Quebec), 0.4 mg/kg BW, then propofol, 1 mg/kg BW, given to effect, top-up bolus of 0.5 mg/kg BW if needed.

Thirty minutes after the topical application of lidocaine/prilocaine cream (EMLA Cream; AstraZeneca, Mississauga, Ontario) a 20-gauge over-the-needle catheter (BD Insyte-W; Becton Dickinson, Mississauga, Ontario) was placed aseptically in a cephalic vein. All dogs were pre-oxygenated with a tightly fitting mask for 5 min. The same individual administered the initial dose of all anesthesia induction agents slowly over 1 min to effect and evaluated depth of anesthesia. Endpoint of administration was the achievement of a light plane of anesthesia, defined as relaxation of jaw tone sufficient to visualize the larynx, with absence of a palpebral reflex. If the endpoint was achieved before the end of 1 min, the drug administration was stopped. If the dog was deemed inadequately anesthetized at the end of 1 min, top-up boluses over 10 s were administered until the depth of anesthesia was adequate. Once adequately anesthetized, dogs were positioned in sternal recumbency in a manufactured device to hold the head in the same position for each evaluation.

The holding device consisted of a frame with 2 perpendicular support pillars on each side of the dog’s head. Each of the pillars had a series of holes oriented in a vertical line through which bars were passed to suspend the dog’s head above the table with its mouth in an open position. Two parallel bars passed just behind the canine teeth on the lower and upper jaw, holding the mouth loosely open at a standardized distance of 5 or 7.5 cm. The distance between lower and upper jaw and the distance between the bottom of the device and the lower bar was recorded for each dog in order to standarize the body position for each dog for subsequent treatments.

A 5-mm flexible bronchoscope (Olympus BF-P180 Evis Exera II; Olympus Medical Systems, Tokyo, Japan) connected to a videocassette recorder, was inserted into the mouth and over the tip of the epiglottis to a point where the entire laryngeal ostium was visible on the monitor. To standardize endoscope position, the distance from the tip of the videoendoscope to the caudal border of the left maxillary canine tooth was measured, marked with tape, and used for the entire examination and for each subsequent examination. An assistant observing the respiratory pattern marked the beginning of inspiration with an “x” on the videorecording to help correlate arytenoid movement with the respiratory cycle during subsequent analysis of the video segments. The laryngoscopic examination was recorded from the time the dog was properly positioned until the dog could no longer be safely restrained. At that point the videoendoscope was removed from the oropharynx and the dog was re-anesthetized with the same induction agent and intubated for the dental procedure.

Objective glottal gap measurement

An evaluator unaware of the anesthetic protocol used, performed objective rima glottis measurement using the digitized video segments (Quick Time Player 7.6.3 Pro; Apple Canada, Toronto, Ontario). The first 30 s of videotape immediately after induction that included 3 breaths and the last 30 s immediately before termination of recording of each dog were used for evaluation. The evaluator selected 3 breaths from each defined period that appeared to have the greatest amount of arytenoid motion. Still images of maximal inspiration and of maximal expiration were converted and imported into an image processing program (Adobe Photoshop CS5 extended, Version 12.0 × 64; Adobe Systems, San Jose, California, USA). Height and area of the glottal gap from the 3 still images were measured in pixels 3 times and averaged. The height was measured from the center of the dorsal connection between the arytenoid cartilages to a central point at the base of the vocal cords (Figure 1). The glottal gap area was traced around the arytenoid cartilages to a central point at the base of the vocal cords (Figure 2). Mean height and mean area of the glottal gap from each set of 3 images, measured at inspiration and expiration, were calculated. Mean area measurements from each set of images were normalized against the height [normalized glottal gap area (NGGA) = area/height²] to correct for variation in size of the dog and the distance between the larygneal ostium and the tip of the endoscope (8,9). The range of arytenoid motion (NGGA change) was determined by subtracting expiratory NGGA from inspiratory NGGA.

Figure 1.

The height of the glottal gap area was measured from the base of the vocal cords to the center of dorsal connection between arytenoid cartilages during maximal inspiration.

Figure 2.

Glottal gap area was traced around vocal cords and arytenoid cartilages during maximal inspiration.

Statistical analysis

A commercial software package (GraphPad Prism 6.0; GraphPad Software, La Jolla, California, USA) was used for statistical analyses. Data were assessed for normality through the D’Agostino & Pearson omnibus normality test. The range of arytenoid motion was compared within groups using a Mann-Whitney test and between groups using a Kruskal-Wallis test. Times of videolaryngoscopy were compared among induction protocols using Friedman test. Statistical significance was set at P < 0.05 and power level of ≥ 80%.

Results

Apnea (defined as no respirations for > 60 s) after the administration of anesthetic induction drugs was observed in 1, 3, and 2 dogs after alfaxalone, propofol/diazepam, and thiopental administration, respectively. Arytenoid movement was not detected during the entire laryngeal examination in 1 dog anesthetized with alfaxalone and in a different dog anesthetized with thiopental, despite the presence of strong respiratory movements. After propofol/diazepam administration 2 dogs had no laryngeal movement after induction but regained laryngeal movement before recovery. In a different dog no laryngeal movement was detected in the second examination period.

Data were not normally distributed and are presented as median and range. Median dosages of anesthetic drugs to perform videolaryngoscopy were: alfaxalone 2.6 mg/kg BW (range: 2 to 3 mg/kg BW), propofol 4 mg/kg BW (range: 2.5 to 7.5 mg/kg BW), and thiopental 14.2 mg/kg BW (range: 8.1 to 20 mg/kg BW). Arytenoid motion (NGGA change) after induction and before recovery for each induction protocol did not differ (alfaxalone P = 0.5949; propofol/diazepam P = 0.8775; thiopental P = 0.3935). No significant differences were observed when comparing NGGA change for all induction protocols after induction or before recovery (P = 0.7013). The examination times (time that videolaryngoscopy could be performed) of 357 s (range: 108 to 603 s), 381 s (range: 69 to 733 s), and 489 s (range: 197 to 992 s) for alfaxalone, propofol/diazepam, and thiopental, respectively were not significantly different.

Discussion

Methods for evaluating the range of arytenoid motion and changes of the glottal gap area during tidal breathing include the use of a subjective scoring system (5,7) or calculation of the normalized glottal gap area from digitized images (9,10). The results of this study support the hypothesis that there is no difference in arytenoid motion, defined as change in NGGA, after the administration of alfaxalone, propofol/diazepam, or thiopental. Our results are in agreement with a recent study (7) comparing arytenoid motion after alfaxalone, propofol, and thiopental using a subjective scoring system. Comparison of the NGGA after different induction protocols required certain standardizations during the examination to reduce variability of endoscope distance to the larynx in the individual patient. Dogs in our study were placed in a purpose-built device to achieve the same body position and the tip of the endoscope was maintained at the same distance from the larynx while each anesthetic protocol was evaluated. The initial bolus of each induction agent in this study was administered slowly over 1 min to achieve a light plane of anesthesia. Positioning of the dogs in the purpose-built device, however, might have required more muscle relaxation and a higher depth of anesthesia than ideal for routine assessment of laryngeal motion. An excessively deep plane of anesthesia can cause respiratory depression and may impair laryngeal function. Our dose of thiopental was comparable to the dose of thiopental (14 ± 2.26 mg/kg BW) used in nonpremedicated dogs in a previous study, which concluded that thiopental given to effect is the best choice for assessment of laryngeal motion (4). The propofol dose (5.6 ± 1.14 mg/kg BW) used in the aforementioned study was higher than the dose used in our study in which propofol was co-administered with diazepam. Administration of diazepam (0.4 mg/kg BW) reduced the amount of propofol (3.8 ± 0.9 mg/kg BW) required to induce anesthesia in nonpremedicated dogs (11). When a subjective scoring system was used to evaluate arytenoid motion and dogs were only manually restrained, lower doses of alfaxalone [1.2 (1.2 to 1.2) mg/kg BW], thiopental [6.3 (6.0 to 6.6 mg/kg BW] and propofol [2.4 (2.4 to 2.4) mg/kg BW] were sufficient to facilitate retraction of the jaws and achieve oral laryngeal examination (7). We did not perform a pre-anesthetic scoring of the dogs’ excitement but we used client-owned dogs in our study and different temperament and higher excitement levels of our dogs might be another reason for the overall higher doses of anesthetic induction agents used in this study.

The higher doses of induction agent in the current study are also reflected in longer examination periods than in previous studies (7). Longer examination periods might not be beneficial since laryngeal function is usually evaluated both immediately after induction and before recovery. Depth of anesthesia is difficult to standardize at induction and a lighter plane of anesthesia just before recovery might allow for a more accurate assessment of arytenoid motion. However, no significant difference in arytenoid function was detected between these 2 time periods in our study. Another study with similar study design showed no difference in arytenoid motion between the induction and recovery period of 6 dogs (4).

The failure to find a difference between time periods or induction protocols could be due to the small sample size. A prospective power calculation suggested that a sample size of 8 dogs per treatment was adequate and the limitation of a small sample size was reduced by using a randomized, crossover trial. The variability in the current study as demonstrated by a wide range of NGGA changes resulted in lack of statistical significance and in inadequate statistical power.

Another limitation of our study is the lack of a propofolalone treatment group. Diazepam (0.4 mg/kg BW) reduced the amount of propofol used to induce anesthesia and maintained blood pressure but failed to ameliorate the respiratory depressive effects of propofol (11). In the present study post-induction apnea was observed in more dogs after propofol/diazepam administration (3/8) than after alfaxalone administration (1/8). In contrast, a similar incidence (25%) of post-induction apnea after slow administration of alfaxalone or propofol was reported in a previous study (12). When propofol was combined with the benzodiazepine midazolam, apnea occurred in 4/9 dogs compared with 1/8 dogs after propofol alone (12). Thus, it is possible that diazepam exacerbated the respiratory depressant effects of propofol herein. It is unclear if the co-administration of diazepam had any effect on arytenoid motion. To the authors’ knowledge no studies on the effect of diazepam on arytenoid motion in dogs exist. Further studies are warranted to determine if there is benefit in the co-administration of diazepam with propofol during laryngeal examination.

Our study only used dogs with normal laryngeal function and it is possible that the evaluated anesthetic drug protocols would have a different effect in dogs with laryngeal dysfunction.

Depth of anesthesia required to position dogs in a purpose-built device might be greater than ideal for laryngeal function evaluation and manual positioning for laryngeal function examination should be recommended. Alfaxalone and propofol/diazepam are acceptable alternatives to thiopental for assessment of arytenoid motion in dogs.