Retrospective evaluation of maropitant and perioperative factors affecting postoperative appetite in cats

Abstract

The primary goal of this retrospective study was to evaluate the effect of postoperative appetite return in cats premedicated with maropitant citrate. Medical records of 75 cats admitted for gastrointestinal (GI) and urogenital (UG) surgeries were reviewed and analyzed. Buprenorphine analgesia was used with 36 cats (48%) premedicated with maropitant and 39 cats (52%) that were used as a control group. No significant differences in postoperative appetite return were reported with maropitant premedication compared to controls. Age, breed, preoperative weight, surgery type, surgery and anesthesia times, and total hospitalization time were also evaluated and were not reported to be significantly associated with postoperative appetite return. Presenting complaints of hyporexia or anorexia were significantly associated with earlier postoperative appetite return. Results of this study suggest that cats treated with buprenorphine for GI or UG surgeries do not have postoperative appetites return sooner when premedicated with maropitant.

Introduction

Anorexia or hyporexia (decreased appetite) are commonly reported within the immediate postoperative period, and if prolonged, can result in hormonal and metabolic changes that may impact immune function and wound healing (1–3). Offering food earlier (< 24 h) after gastrointestinal surgery has become increasingly accepted in human and veterinary medicine with the potential for reductions in complications (4) and shortened lengths of hospitalization (5,6). Rapid return of voluntary appetite can also be considered as an important response to treatment and positive recovery progress.

Postoperative appetite can be diminished in cats for many reasons including, but not limited to, clinical disease, pain, medications, diet and feeding preferences, and environmental or behavioral factors (7,8). Spontaneous appetite return is ultimately expected with resolution of underlying pathology and inflammation. Therefore, pharmacological appetite stimulants are not often used after surgery (1,7,8).

Perioperative and postoperative nausea and vomiting associated with full mu agonist opioid analgesics, ileus, and intraabdominal inflammation are likely contributing factors to the persistence of postoperative inappetence (9–11). Treatment protocols to control for these factors may improve the speed of postoperative appetite return. Dogs premedicated with maropitant have less perioperative vomiting and decreased signs of nausea (12,13) with secondary improvements in postoperative appetite return (13,14). Buprenorphine analgesia in cats has been associated with fewer gastrointestinal effects than analgesia with full mu agonist opioids (15,16), which may eliminate the need for concurrent antiemetic therapy.

Maropitant citrate is a commonly used antiemetic in dogs and cats, with effective suppression of peripheral and central mediated causes of emesis (17). It is a potent and highly specific centrally acting neurokinin-1 receptor antagonist that blocks the effects of substance P, with at least 24 h of persistent antiemetic activity seen in cats regardless of route of administration (18). Widespread antiemetic use of maropitant in dogs led to expanded labeled usage in 2012 for cats (19) with subsequent approval of intravenous administration in 2016 (20). Extralabel indications for maropitant have been investigated for gastroesophageal reflux (21), adjunct analgesia (22–24), feline asthma (25), adjunct chronic kidney disease management (26), and as an antipruritic agent (27) with variable results.

Current feline-specific studies appear limited to perioperative effects of maropitant and are lacking in postoperative appetite data. The primary goal of this retrospective study was to evaluate whether preanesthetic treatment with maropitant facilitated earlier postoperative appetite return in cats and to secondarily identify associated factors, such as type of surgery, that may significantly affect postoperative appetite return.

Materials and methods

Medical records were reviewed to identify cats admitted for gastrointestinal (GI) or urogenital (UG) surgery at Garden State Veterinary Specialists from May 2016 to December 2017. Cats with complete medical records that received buprenorphine analgesia, with or without preanesthetic maropitant treatment, and were hospitalized for at least 30 h after surgery, were eligible for inclusion in the study. Cats that received maropitant after surgery or were clinically diagnosed with concurrent endocrinopathies were excluded.

Data collected from each patient included age (y), preoperative body weight (kg), 9-point body condition score, onset and duration of hyporexia or anorexia, surgery time (min), anesthesia time (min), surgery type, length of postoperative and total hospitalization time (d), and any significant remarks of aggressive or fractious demeanors.

Cats were primarily categorized into 2 treatment groups based on preanesthetic administration of maropitant citrate (Cerenia; Zoetis, Kalamazoo, Michigan, USA), 1 mg/kg body weight (BW), IV, versus no maropitant administration (control). Maropitant, when given as a preanesthetic treatment, was administered at least 1 h before induction of anesthesia or within 24 h prior to surgery.

All cats included in the study exclusively received premedication with IV buprenorphine (0.5 mg/mL; Wedgewood Pharmacy, Swedesboro, New Jersey, USA), 0.015 to 0.02 mg/kg BW, or SC buprenorphine (Simbadol 1.8 mg/mL; Abbott Laboratories, North Chicago, Illinois, USA), 0.24 mg/kg BW. Postoperative buprenorphine was continued every 8 h or every 24 h (Simbadol) until the day of discharge.

Anesthetic induction protocols included single agent propofol (10 mg/mL; Sagent Pharmaceuticals, Schaumburg, Illinois, USA), 4 to 6 mg/kg BW, IV, or combinations of ketamine (100 mg/mL; Zoetis, Kalamazoo, Michigan, USA), 1 mg/kg BW, IV, midazolam (5 mg/mL; Hospira, Lake Forest, Illinois, USA), 0.2 mg/kg BW, IV, and propofol followed by endotracheal intubation. Isoflurane gas anesthesia was maintained throughout all surgeries. Perioperative cefazolin (100 mg/mL; Hikma Farmaceutica, Terrugem, Portugal), 22 mg/kg BW, IV, was administered to all cats in the study and was continued every 90 min during surgery, then continued every 8 h IV after surgery until appetite returned or the day of discharge. Additional postoperative treatments and monitoring were determined based on individual needs.

Cats were secondarily categorized by type of surgery: gastrointestinal (GI) versus urogenital (UG). All surgeries were performed by American College of Veterinary Surgeons diplomates or surgery residents under the direct supervision of ACVS diplomates.

All cats included in the study were offered food within 6 h of endotracheal extubation, with subsequent meals offered every 6 h until hospital discharge. Subjective assessment of the volume of food ingested was recorded upon observation. A variety of diets was offered including dry and canned adult or kitten feline formulas, low-fat, sensitive stomach, or prescription formulas based on patient needs or to encourage intake. The 6-hour time interval in which food was first ingested after surgery was used to indicate return of postoperative appetite.

Statistical analysis

Data were analyzed using SAS version 9.4 for Windows (SAS, Cary, North Carolina, USA). Frequency tables for categorical variables and descriptive statistics for continuous variables were used to summarize the data. Non-parametric Kaplan-Meier estimators, stratified by maropitant treatment, type of surgery, and preoperative appetite abnormalities were used to estimate time to postoperative appetite return using the expectation-maximization iterative convex minorant (EMICM) algorithm (28) implemented in the SAS procedure PROC ICLIFETEST. Generalized log-rank tests based on the method of multiple imputations developed by Huang et al (29) were used to determine if there was any relationship between postoperative appetite return and maropitant treatment, type of surgery, or preoperative appetite abnormalities. An accelerated failure time (AFT) model with Gamma distribution was fitted using the SAS procedure PROC LIFEREG to assess the effects of the predictors (maropitant treatment, type of surgery, preoperative appetite abnormalities, age, weight, surgery time, anesthesia time, and total time of hospitalization) on postoperative appetite return. The AFT model with Gamma distribution was appropriate for the data based on the results of the comparison of i) empirical Kaplan–Meier and parametric estimators; ii) Akaike information criterion and Baysian information criterion; and iii) the Cox-Snell residual plots, among the AFT models with the following error distributions including normal, logistic, exponential, Weibull, lognormal, log-logistic, and Gamma. A P-value < 0.05 was used to indicate significance.

Results

Eighty-nine medical records were reviewed, and 75 cats met the inclusion criteria for data analysis. Fourteen cats were excluded due to continued maropitant administration after surgery (n = 5), concurrent endocrinopathy (n = 2), and inconsistent or > 6 h withholding of postoperative offering of food (n = 7).

Postoperative appetite return outcomes for the 75 cats are summarized in Table 1. Of the 75 cats analyzed, 52 cats (69.3%) had recorded appetite return within 48 h of their postoperative hospitalization. Twenty-three cats (30.7%) did not have appetite return during their postoperative hospitalization and these cats were considered censored in the modeling process.

Table 1.

Summary of postoperative appetite return for 52 cats whose appetite returned before discharge, and time of discharge of 23 cats whose appetite had not returned before discharge.

Appetite return (hours after surgery)	Number of cats	%
< 6 h	20	26.7
> 6 to 12 h	9	12.0
> 12 to 18 h	7	9.3
> 18 to 24 h	3	4.0
> 24 to 30 h	3	4.0
> 30 to 36 h	8	10.7
> 36 to 42 h	1	1.3
> 42 to 48 h	1	1.3
> 48 h to discharge	0	0
Discharged between 36 to 42 h	1	1.3
Discharged between 42 to 48 h	2	2.7
Discharged > 48 h	20	26.7

Results of non-parametric Kaplan-Meier estimators and generalized log-rank tests

Preanesthetic maropitant was administered to 36 cats with a control group of 39 cats which did not receive maropitant. Median time of postoperative appetite return was 27 h for cats treated with maropitant and 24 h for the control group (Figure 1). There was no significant difference between the treatment groups [χ²(1) = 0.000012, P = 0.997]. The characteristics of the cats with or without maropitant treatment were similar in terms of type of surgery, clinical signs for appetite abnormalities, age, weight, surgery time, anesthesia time, and total time of hospitalization (Table 2).

Figure 1.

Kaplan-Meier estimator by preanesthetic maropitant treatment. Maropitant treatment — Yes/dotted line (n = 36, median: 27 h after surgery); control — No/solid line (n = 39, median: 24 h after surgery).

Table 2.

Characteristics of the 75 cats in the study.

	Preanesthetic maropitant treatment		Total
	Yes (N = 36)	No (N = 39)
Surgery type
Gastrointestinal (GI)	11 (30.56)	13 (33.33)	24 (32.00)
Urogenital (UG)	25 (69.44)	26 (66.67)	51 (68.00)
Preoperative anorexia or hyporexia
Yes	14 (38.99)	15 (38.46)	29 (38.67)
No	22 (61.11)	24 (61.54)	46 (61.33)
Age (y)	5.50 (0.83 to 14.00)	5 (0.25 to 16.00)	5 (0.25 to 16.00)
Weight (kg)	6.24 (3.22 to 9.53)	5.81 (2.68 to 10.84)	6.12 (2.68 to 10.84)
Surgery time (min)	47.50 (20 to 120)	40 (25 to 100)	45 (20 to 120)
Anesthesia time (min)	72.50 (45 to 180)	60 (45 to 135)	65 (45 to 180)
Total time of hospitalization (d)	3 (2 to 15)	4 (2 to 15)	4 (2 to 15)

N (%) for type of surgery and preoperative appetite abnormalities. Median (range) for age, weight, surgery time, anesthesia time, and total time of hospitalization.

Cats were also categorized into surgery groups with 24 cats in the GI group (32%) and 51 cats in the UG group (68%). Specific GI surgeries included gastrotomy or enterotomy for foreign body retrievals (n = 17), gastrointestinal resection and anastomosis for neoplasm excision (n = 2), abdominal exploration with multiple intra-abdominal biopsies (n = 2), splenectomy (n = 1), cholecystoduodenostomy (n = 1), and subtotal colectomy (n = 1). Specific UG surgeries included perineal urethrostomy (n = 36), cystotomy (n = 9), perineal urethrostomy with cystotomy (n = 3), perineal urethrostomy with bladder tear repair (n = 1), perineal urethrostomy revision (n = 1), and cystotomy with uterine stump biopsy (n = 1). The median time of postoperative appetite return was 33 h for GI surgery and 18 h for UG surgery cats. There was no significant difference between the surgery groups [χ²(1) = 0.0110, P = 0.916] regardless of maropitant treatment status.

At clinical presentation, 38.7% (n = 29) of cats had reported appetite abnormalities of hyporexia or anorexia (Table 2). The median time of postoperative appetite return was 18 h for cats with reported preoperative appetite abnormalities. In comparison, 61.3% (n = 46) of cats did not have reported preoperative appetite abnormalities and these cats had a median time of postoperative appetite return of 27 h. The difference between the median times of postoperative appetite return was statistically significant [χ²(1) = 3.9077, P = 0.048] with a 50% increased time of postoperative return in cats without reported preoperative appetite abnormalities. Therefore, cats with preoperative hyporexia or anorexia were expected to have earlier postoperative appetite return (Figure 2).

Figure 2.

Kaplan-Meier estimator by preoperative appetite abnormalities. Preoperatively reported clinical signs of hyporexia or anorexia — Yes/dotted line (n = 29, median: 18 h after surgery), none reported — No/solid line (n = 46, median: 27 h after surgery).

Sub-analysis of the 29 cats that had preoperative complaints of hyporexia or anorexia was also conducted using the generalized log-rank test to determine if there was a relationship between maropitant treatment status and postoperative appetite return within this group. Fourteen cats (48.3%) were administered preanesthetic maropitant and 15 cats (51.7%) did not receive maropitant. The median time of postoperative appetite return was 12 h for cats treated with maropitant and 24 h for cats that did not receive maropitant. Statistical significance was not found between these 2 sub-groups [χ²(1) = 0.1372, P = 0.711].

Results of the accelerated failure time model

Further analysis of the 75 total cats was performed to identify other associated factors that may affect appetite after surgery. The median age at the time of surgery was 5 y (range: 0.25 to 16 y). Gender was not analyzed due to a marked bias toward male cats admitted for urogenital procedures. Median weight at the time of surgery was 6.12 kg (range: 2.68 to 10.84 kg). Median body condition score at the time of surgery was 6.00 (range: 3.00 to 9.00) with results recorded from only 56 cats, therefore, the data were not further analyzed. The median surgery and anesthesia times were 45 min (range: 20 to 120 min) and 65 min (range: 45 to 180 min), respectively. The median total hospitalization time was 4 d (range: 2 to 15 d). No statistically significant associations were found between postoperative appetite return and maropitant treatment status, type of surgery, age, preoperative weight, surgery or anesthesia time, and total time of hospitalization (Table 3).

Table 3.

Assessment of effects of perioperative factors on postoperative appetite return (accelerated failure time model).

Parameter	β	Standard error	Exp(β)c	95% Confidence interval of exp (β)	χ2	P-value
Maropitant treatment
Noa	−0.12	0.35	0.89	(0.45, 1.76)	0.12	0.7321
Yesb
Surgery
GIa	0.47	0.41	1.60	(0.72, 3.54)	1.34	0.2464
UGb
Clinical signs for appetite abnormalities
Noa	1.26	0.57	3.51	(1.15, 10.76)	4.84	0.0278
Yesb
Age (y)	0.01	0.06	1.01	(0.88, 1.14)	0.01	0.9323
Weight (kg)	0.21	0.13	1.24	(0.96, 1.60)	2.60	0.1071
Surgery time (min)	−0.04	0.02	0.96	(0.91, 1.01)	2.62	0.1054
Anesthesia time (min)	0.04	0.02	1.04	(0.99, 1.09)	2.72	0.0991
Total time of hospitalization (d)	0.02	0.13	1.02	(0.78, 1.32)	0.02	0.8947

^aDegrees of freedom = 1.

^bRef = reference group (a category of comparison for the other categories).

^cExp(β) represents the estimated ratio of the expected (mean) times for postoperative appetite return for the 2 groups being compared. For example, for type of surgery, the coefficient beta = 0.47. Thus, exp(0.47) = 1.60; therefore, the expected time for postoperative appetite return for the GI group was 60% greater than for that of the UG group. For continuous variables, such as age (and weight, surgery time, anesthesia time, and total time of hospitalization), the transformation 100(exp(β) – 1) is used, which gives the percent change in the expected time of postoperative appetite return for each 1-unit increase in the variable. Thus, for age, 100(exp(0.01) – 1) = 1.01; therefore, each additional increase in age is then associated with a 1.01% increase in expected time of postoperative appetite return.

β — Parameter estimate, χ² — Chi-square statistic.

Five cats were deemed to be amenable to only minimal handling without sedation after surgery. Only one of these cats remained inappetent throughout their postoperative hospitalization. Three of these cats received preanesthetic maropitant, including the patient which was inappetent after surgery.

Postoperative vomiting or signs of nausea were not reported from any of the included study patients.

Discussion

The results of this study indicate that, in contrast to previous findings in dogs, preanesthetic maropitant treatment does not facilitate earlier postoperative appetite return in cats (13,14). Maropitant was also previously reported to decrease signs of vomiting but did not significantly affect signs of nausea in cats (30). Buprenorphine analgesia was likely a primary reason for lack of postoperative nausea and vomiting in our studied groups. Ramsey et al (13) reported that in maropitant pretreated dogs, premedication with morphine resulted in a quicker postoperative appetite return (< 6 h) than did premedication with buprenorphine (> 6 h). Although all buprenorphine treated dogs in the aforementioned study did not have appetite return until after 6 h after surgery (13), the highest percentage of appetent cats in this study had appetite return within the initial 6 h period after surgery.

Although not statistically significant, the GI surgery group was associated with a longer time to postoperative appetite return with a median of 30 h, compared to 18 h in the UG group. Intra-abdominal pathology can often cause postoperative ileus and inappetence (6). However, when appetent, early nutrition following anesthetic recovery appeared to be well-tolerated (14,31) and was associated with improved postoperative recovery (6), despite traditional fears of interference with intestinal healing. To the authors’ knowledge, an acceptable time for delayed postoperative appetite return has not been established; however, supplemental routes of nutritional support to cats are advised after 48 to 72 h of inadequate caloric intake (1,3,32). Future research should be directed at determining the ideal timing of initiating enteral nutrition based on patient nutritional status.

The only significant variable associated with postoperative appetite return in cats was appetite abnormalities that were reported at clinical presentation. Appetite return after surgery occurred sooner when presenting complaints of hyporexia or anorexia were reported than with preoperative conditions that did not affect appetite. Approximately 70% (17/24) of cats in the GI group and 20% (11/51) in the UG group reported a decreased appetite or anorexia. Greater numbers of GI group cats afflicted with appetite abnormalities may be suggestive of increased severity of illness compared to more localized conditions typically representative of the UG group. Although not statistically significant, a trend towards earlier postoperative appetite return was noted from the maropitant treated cats (12 h versus 24 h) that had preoperative appetite abnormalities. However, without quantifying daily food intake, clinical hyporexia was not likely resolved during postoperative hospitalizations despite appetite return. Consequently, in both groups, warranted surgical intervention of underlying pathology would eventually be expected to resolve these presenting complaints without adjunctive appetite stimulation.

Induction drug combinations of propofol, ketamine, and midazolam that were used in the studied cats were not analyzed but may have had potential contributing effects of postoperative appetite enhancement. Propofol and midazolam have been shown to stimulate appetite in cats (33,34) within 15 to 30 min and 2 h, respectively. Anesthesia times of the studied cats ranged from 45 to 180 min, and to the authors’ knowledge, the postoperative appetite effects of these drugs over longer durations have not been investigated.

Accommodating individual feline temperaments and preferences presents challenges within an observational study and hospital setting. Buisman et al (35) concluded that temperament had a significant effect on postoperative appetite in cats, with those having friendly and less aggressive demeanors being more likely to eat after surgery. Inappetence of shy or more aggressive cats, however, was challenging to determine attributions from stress, pain, or both. Other compounding effects from specific hospital ward lighting and noises to physical impacts of an incision and Elizabethan collars are also likely to exacerbate avoidance behaviors in select patients (2). Behavioral support and environmental modifications may benefit these patients during the postoperative period of hospitalization.

Due to the retrospective and clinical setting of these cases, interval observations did not allow for more specific time points of appetite return and lacked standardization of diets, amount offered, and quantitative consumption data. Lack of quantitative measures of food intake was also limiting for additional trending evaluation of continuous postoperative appetite or caloric intake. Validity of the quicker trends in postoperative appetite return seen in maropitant premedicated cats with appetite affected before surgery and in non-GI surgeries should be further assessed with larger sample sizes and randomized prospective studies.

Preanesthetic treatment with maropitant was not associated with earlier return of postoperative appetite in cats. Only cats with decreased preoperative appetite began eating sooner after surgery. Preanesthetic or perioperative maropitant, in combination with other premedication, may prove beneficial for postoperative appetite in cats and warrants further investigation.