Materials and Methods

Case selection—Medical records from the Matthew J. Ryan Veterinary Hospital of the University of Pennsylvania were reviewed retrospectively to identify cats for which emesis induction was attempted between 2009 and 2014. Cats were included in the study if emesis induction was attempted and a complete medical record was available for review. Cats were excluded if a complete medical record was not available for review or if emesis induction was attempted but the emetic agent used was not recorded.

Medical record review—From each cat's medical record, the following data were recorded in spreadsheet software^c: signalment; weight; reason for emesis induction; emetic agent (including dose and route administered); whether emesis was successfully induced; and, when noted in the record, elapsed time until emesis. Time until emesis was defined as the interval, in minutes, from administration of the emetic until either vomiting occurred or, if the exact time of vomiting was not noted, the time when a reversal drug was administered to counter the sedative effect of the emetic agent.

Statistical analysis—For statistical analysis, comparisons were made between cats that did and did not vomit for each of the emetics evaluated. Categorical data were reported as percentages (with 95% CIs calculated by means of the exact method) and were compared with a Pearson χ² test. The OR and 95% CI were calculated by means of the exact method. Continuous variables were evaluated for normality by means of the Shapiro-Wilk test. All of those variables underwent nonparametric testing and are reported as median (range). The Wilcoxon rank-sum test was used to compare continuous variables between groups. Statistical analysis was performed with the aid of commercially available statistical software.^d For all tests, values of P < 0.05 were considered significant.

Results

Induction of emesis was attempted in 45 cats during the period of interest. Two cats were excluded from the study owing to incomplete medical records; thus, data for 43 cats were evaluated. The 43 cats included 19 castrated males, 18 spayed females, 4 sexually intact males, and 2 sexually intact females. Breeds included domestic shorthair (n = 35), domestic longhair (4), and Himalayan, Siberian, Bengal, and Munchkin (1 each). Median age was 1.5 years (range, 2 months to 15 years), and median weight was 4.38 kg (9.64 lb; range, 1.00 to 6.65 kg [range, 2.2 to 14.63 lb]).

In most (n = 33) cats, emesis induction was attempted following known or suspected toxicant ingestion, including anticoagulant rodenticide (n = 19), lily plant (8), NSAIDs (4), gabapentin (1), and sago palm (1). In 10 cats, emesis induction was attempted after recent ingestion of a string foreign body and was successful in 5. Endoscopic foreign body retrieval was performed in 2 of the 5 cats that did not vomit; the other 3 cats were discharged from the hospital for at-home monitoring and subsequently lost to follow-up.

In 3 (7%) cats, hydrogen peroxide^e (1.5 to 2.0 mL/kg [0.68 to 0.91 mL/lb]) was administered orally to induce emesis. None of these cats vomited. Xylazine was administered to 25 (58%) cats, including 1 cat that had already been administered hydrogen peroxide. The median dose of xylazine was 0.44 mg/kg (0.2 mg/lb; range, 0.4 to 0.5 mg/kg [0.18 to 0.23 mg/lb]). Xylazine was administered IM in 23 cats and IV in 1 cat. Dexmedetomidine was administered to 16 (37%) cats. The median dose of dexmedetomidine was 7.0 μg/kg (3.2 μg/lb; range, 0.96 to 10.00 μg/kg [0.44 to 4.55 μg/lb]). Dexmedetomidine was administered IM in 7 cats at a median dose of 7.0 μg/kg (range, 7.0 to 10.0 μg/kg). For 8 cats, dexmedetomidine was administered IV at a median dose of 3.5 μg/kg (1.59 μg/lb; range, 0.96 to 10.00 μg/kg). One additional cat received dexmedetomidine IV; however, the dose was not recorded in the medical record.

Emesis was successfully induced in 24 of the 43 (56%) cats. Emesis occurred in 11 of 25 (44%; 95% CI, 24% to 75%) cats that were administered xylazine and in 13 of 16 (81%; 95% CI, 54% to 96%) cats that were administered dexmedetomidine. Compared with xylazine administration, dexmedetomidine administration was significantly (P = 0.018) more likely (OR, 5.5; 95% CI, 1.1 to 36) to result in emesis.

Considering only the cats that were administered dexmedetomidine, the median dose that resulted in emesis was 7.0 μg/kg (range, 0.96 to 10.00 μg/kg). The median dose of dexmedetomidine given to the 3 cats that did not vomit was 2.0 μg/kg (range, 1.0 to 5.0 μg/kg [0.45 to 2.27 μg/lb]. These 2 median doses were not significantly (P = 0.088) different. Emesis was successfully induced in 7 of 7 (100%; 1-sided 95% CI, 60% to 100%) cats that were administered dexmedetomidine IM and in 6 of 9 (67%; 95% CI, 30% to 90%) cats that were administered dexmedetomidine IV. The difference in efficacy of IM versus IV administration of dexmedetomidine was not significant (P = 0.212).

The elapsed time until emesis or postemesis administration of a reversal agent (to counter the sedative effects of the emetic agent) was recorded for 15 cats, of which 5 were administered xylazine and 10 were administered dexmedetomidine. The median time until emesis or postemesis reversal agent administration was 10 minutes (range, 1 to 175 minutes). For the xylazine-treated cats, the median time until emesis or reversal agent administration was 10 minutes (range, 5 to 175 minutes); for the 10 dexmedetomidine-treated cats, the median time was 5 minutes (range, 1 to 12 minutes). These elapsed times were not significantly (P = 0.068) different. One cat that received xylazine vomited 175 minutes after administration. Removal of this outlier did not change the median time until emesis.

A reversal agent was administered to 26 cats following an attempt at emesis induction. Yohimbine hydrochloride^f was administered to 11 cats, all of which had emesis induction attempted with xylazine. The median dose of yohimbine was 0.11 mg/kg (0.05 mg/lb; range, 0.04 to 0.35 mg/kg [0.018 to 0.16 mg/lb]). Yohimbine was administered SC, IV, or IM to 1, 3, and 6 cats, respectively; for 1 cat, the route of administration was not recorded. For 1 cat, the same dose of yohimbine (0.05 mg/kg [0.023 mg/lb], IM) was administered twice, although neither the rationale for nor the timing of the second dose was recorded. Atipamezole hydrochloride^g was administered IM to 15 cats, 14 of which had emesis induction attempted with dexmedetomidine and 1 of which had emesis induction attempted with xylazine. The dose of atipamezole used for each cat was inconsistently recorded in the medical records, although standard practice at our institution is to administer a volume of atipamezole equal to that of the dexmedetomidine administered. One cat was reportedly sedated following a single dose of atipamezole, so dose administration was repeated and the sedative effect was successfully reversed.

Two cats that received xylazine and 1 cat that received dexmedetomidine were reportedly sedated following administration. Sedation was successfully reversed in all 3 cats. No other adverse effects were reported following administration of either xylazine or dexmedetomidine.

Discussion

The results of the present study indicated that dexmedetomidine was an effective medication for the induction of emesis in cats. Most cats in this study that were given dexmedetomidine vomited within a short period, whereas emesis induction was successful in fewer than half of the cats that were administered xylazine.

The decision to induce emesis is based on the characteristics of the ingested toxicant, the interval since ingestion, and whether the patient has signs of toxicosis. In general, emesis is contraindicated in clinically affected patients, particularly if they are sedated, have seizures, or are otherwise unable to protect their airway, given their risk for aspiration of gastric contents during vomiting. Other contraindications for emesis induction include ingestion of caustic substances that may injure the esophagus during vomiting, ingestion of hydrocarbons that may be aspirated into the airway during vomiting, or the elapsing of more than a few hours since toxicant ingestion. Additionally, emesis should not be induced in patients with a predisposition to aspiration pneumonia because of an anatomic anomaly or concurrent disease process.⁵ No specific contraindications for emesis induction were found on review of the medical records for the cats in the present study.

Among the 43 cats included in the present study, a common reason for emesis induction was known or suspected ingestion of either an anticoagulant rodenticide or toxic plant (n = 28 [65%]). This finding is consistent with that of a previous study² that assessed patterns of animal poisonings reported to poison centers in Texas. In that study,² exposure to pesticides and plants accounted for 1,141 of 2,735 (41.7%) reports involving cats between 1998 and 2002. Similarly, nearly 30% of 17,023 feline nondrug exposures reported to the American Association of Poison Control Centers during 1993 and 1994 involved rodenticides and plants.³ The higher proportion of rodenticide and toxic plant exposures among cats of the present study likely reflected the fact that we specifically selected cases involving oral exposures for which emesis was deemed appropriate, whereas the previous reports^2,3 also included dermal, ocular, inhalational, and other toxicoses. Additionally, cats of the present study were from an urban environment; consequently, the study population may have had increased risk of exposure to rodenticides, compared with that for cats in a different environment.

For 10 cats in the present study, emesis induction was attempted after recent ingestion of a string foreign body. Although many linear foreign bodies will pass through the gastrointestinal tract uneventfully, some become fixed around the base of the tongue or at the pylorus with subsequent bunching and plication of the small intestine as peristaltic waves occur. A linear object may then embed in the mesenteric border of the intestine, often leading to full-thickness erosion through the intestinal wall with subsequent peritonitis.¹² In some cats, gastrointestinal linear foreign bodies can be managed conservatively, but many require surgical intervention.¹³ In the event of a recent linear foreign body ingestion, successful induction of emesis as described in this report may prevent gastrointestinal tract obstruction, thereby eliminating the need for a surgical procedure. However, emesis should not be attempted if small intestinal obstruction caused by an anchored linear foreign body is suspected.

Hydrogen peroxide was administered to 3 cats in the present study, none of which vomited. In contrast to xylazine and dexmedetomidine, hydrogen peroxide is not a centrally acting emetic. Rather, it is thought to cause vomiting via direct gastric irritation.^6,14 Although hydrogen peroxide is an effective emetic in dogs,⁶ its use in cats is not currently recommended because of an association with the development of hemorrhagic gastritis and esophagitis.⁵ In the present study, each of the cats that received hydrogen peroxide was reexamined within 48 hours after hydrogen peroxide administration, and none had clinical signs consistent with gastritis or esophagitis. However, given the few cats that were treated with hydrogen peroxide, no conclusions about the safety of this chemical compound as a medication can be made.

Xylazine and dexmedetomidine are potent α₂-adrenergic receptor agonists frequently used in small animal practice as part of a sedation, anesthesia, or analgesia protocol.^11,15 Colby et al⁸ demonstrated experimentally that the emetic effect of xylazine in cats is initiated in the chemoreceptor trigger zone of the CNS. In that study,⁸ the emetic effect of xylazine was mitigated by ablation of the area postrema of the medulla, although the sedative effect was not affected. A subsequent study by Hikasa et al⁷ confirmed that emesis occurs as a result of activation of α₂-adrenergic receptors and that emesis can be prevented by pretreatment with yohimbine, an α₂-adrenergic receptor antagonist.⁷ Stimulation of peripheral α₂-adrenergic receptors leads to vasoconstriction and reflex bradycardia within several minutes after administration of xylazine or dexmedetomidine, although these effects are reversed by administration of an α₂-adrenergic receptor antagonist.¹¹ Heart rate and blood pressure measurements following administration of xylazine or dexmedetomidine were rarely noted in the medical records for cats in the present study. However, most cats that received a reversal agent following an attempt at emesis induction did so within a short period.

In the cats of the present study, vomiting was significantly more likely to occur following administration of dexmedetomidine than it was following administration of xylazine. In a study by Selmi et al,¹⁶ the dose of propofol required for induction of anesthesia was significantly lower for cats premedicated with medetomidine, compared with findings for cats premedicated with xylazine. This was thought to reflect a greater potency and α₂-adrenergic receptor selectivity of medetomidine.¹⁶ Medetomidine exists as a racemic mixture consisting of 2 enantiomers: the d-isomer, dexmedetomidine, and the l-isomer, levomedetomidine. Of the 2 enantiomers, only dexmedetomidine is thought to have a pharmacological effect.¹¹ The effectiveness of dexmedetomidine in the present study may have been related to an increased potency and adrenergic receptor selectivity, compared with those for xylazine.

A reversal agent, either yohimbine or atipamezole, was administered to 26 cats in the present study. Both yohimbine and atipamezole are α₂-adrenergic receptor antagonists and are used primarily to reverse the cardiovascular and sedative effects of α₂-adrenergic receptor agonists.^7,11 Given the retrospective nature of this study, we are unable to determine why some, but not all, cats were administered a reversal drug after emesis induction. Our suspicion is that reversal agent administration occurred primarily in cats that appeared sedated following an attempt at emesis induction.

The present study had several important limitations, including retrospective data collection and a small number of cases. Emesis was more often successful following dexmedetomidine administration, compared with xylazine, although the population of cats included in this study was small. Heart rate and blood pressure were not consistently recorded for these cats. Although no serious adverse events were recorded, it is possible that bradycardia or severe blood pressure changes went undetected in some cats. Finally, for the purpose of statistical analysis, response to emesis induction was considered as a dichotomous phenomenon (ie, emesis vs no emesis), although data suggest that recovery of ingested material is highly variable following successful emesis induction.¹⁷ Consequently, successful emesis induction may not equate to decreased risk of toxicosis in some cases.

Despite limitations of the present study, dexmedetomidine was used successfully to induce emesis in 13 of 16 cats. Although we did not find a significant difference in efficacy between dexmedetomidine following IM or IV administration, prospective research is warranted to determine the most effective dose and route of dexmedetomidine administration to induce emesis in cats. On the basis of the data obtained in the present study, we would suggest a dose of 7.0 μg of dexmedetomidine/kg administered IM or 3.5 μg of dexmedetomidine/kg administered IV for induction of emesis in cats.