Comparison of grapiprant and meloxicam for management of postoperative joint pain in dogs: A randomized, double‐blinded, prospective clinical trial

Abstract

Background

Grapiprant is a novel anti‐inflammatory drug approved for the treatment of pain associated with osteoarthritis in dogs.

Objective

Compare the efficacy of grapiprant vs meloxicam for the management of postoperative joint pain in dogs.

Animals

Forty‐eight dogs presented with cranial cruciate ligament disease and treated by tibial plateau leveling osteotomy (TPLO) between May 2020 and May 2022.

Methods

In this randomized, double blinded, prospective clinical trial, client‐owned dogs with naturally occurring unilateral cruciate ligament rupture were enrolled on the day of surgery. The day after surgery, all animals received a subcutaneous injection of 0.2 mg/kg of meloxicam and were randomly assigned to receive either oral grapiprant (2 mg/kg) or meloxicam (0.1 mg/kg), once a day for 14 days, in a blinded manner. The primary endpoint of the study was the pain severity (PSS) and interference (PIS) scores, assessed by the Canine Brief Pain Inventory (CBPI) at day 3, 7, 10 and 15 after the surgery.

Results

Three days after surgery, grapiprant treated dogs had lower PSS compared to meloxicam treated dogs with a mean ± SD of 2.76 ± 0.18 vs 3.25 ± 0.23, respectively (difference of −0.49 [95% CI −0.94 to −0.04], P = .032). Pain Interference Score was also lower in grapiprant group at day 3 (4.11 ± 0.18 vs 4.69 ± 0.16 in meloxicam group [difference of −0.58 {95% CI −1.03 to −0.13}, P = .013]) and at day 10 (2.23 ± 0.13 vs 2.72 ± 0.28 [difference of −0.49 {95% CI −0.92 to −0.01}, P = .049]).

Conclusions and Clinical Importance

Our study supports the use of grapiprant as an alternative analgesic to meloxicam for management of postoperative joint pain in dogs.

Abbreviations

CBPI

Canine Brief Pain Inventory

GI

gastrointestinal

NSAID

nonsteroidal anti‐inflammatory drugs

OVH

ovariohysterectomy

PGE₂

prostaglandin E₂

PIS

Pain Interference Score

PSS

Pain Severity Score

TPLO

tibial plateau leveling osteotomy

VA

veterinary assessment

1. INTRODUCTION

Management of pain of mild to moderate intensity in dogs is mainly based on the administration of nonsteroidal anti‐inflammatory drugs (NSAID). ¹ , ² These medications are used for the management of osteoarthritis, postoperative pain, or for other inflammatory related moderate pain. ³ , ⁴ , ⁵ However, the administration of NSAID is often associated with several adverse effects mostly related to gastrointestinal (GI) injury including vomiting, diarrhea, anorexia, lethargy and GI ulceration, and also kidney and hepatic injury. ⁶

The main mechanism of action of traditional NSAIDs is because of the inhibition of cyclooxygenases (COX) responsible for the synthesis of prostanoids, which are involved in the establishment of inflammation but also in numerous constitutive and physiological functions. ⁷ Among all prostanoids, prostaglandin E₂ (PGE₂) is the main proinflammatory prostaglandin and promotes inflammatory hyperalgesia by binding the G‐protein coupled receptor EP₄. ⁸ This receptor is the most abundant PGE₂ receptor of the osteoarticular system and is involved specifically in the development of inflammatory related acute and chronic pain. ⁹ , ¹⁰ Grapiprant, a potent and specific antagonist of the EP₄ is approved by FDA and EMA for the management of mild or moderate pain associated with osteoarthritis in dogs. ¹¹ In contrast to traditional NSAIDs, the unique mechanism of action of grapiprant does not interfere with homeostatic functions of the prostaglandins and is thought to have an improved margin of safety. ¹² , ¹³

However, clinical evidence of the efficacy of grapiprant remains limited. One clinical trial evaluated the analgesic efficacy of grapiprant vs placebo in the management of pain associated with osteoarthritis in dogs. ¹⁴ In this study, a decrease in pain score in grapiprant treated‐dogs compared to dogs, which received the placebo was observed. There is a similar benefit of grapiprant and carprofen in the perioperative management of pain after ovariohysterectomy (OVH) in female dogs. ¹⁵ , ¹⁶ To our knowledge, no studies have compared clinically the analgesic efficacy and adverse effects of grapiprant vs conventional NSAIDs in the management of pain associated with perioperative joint surgery in dogs.

We investigated the analgesic benefit of grapiprant vs meloxicam, which is a NSAID conventionally used for the management of postoperative articular pain, in client‐owned dogs, and the associated adverse‐effects. The primary endpoint of this study was the comparison of the Pain Severity Score (PSS) assessed by the Canine Brief Pain Inventory (CBPI) score.

2. MATERIALS AND METHODS

2.1. Study design and animals

Our study was a prospective, randomized, double‐blinded, placebo controlled clinical trial performed in client‐owned dogs presented at Veterinary School of Alfort (EnvA) from May 2020 to May 2022 for cranial cruciate ligament rupture treated by tibial plateau leveling osteotomy (TPLO). The protocol, including the immediate perioperative analgesia plan, was approved by our local clinical research ethics committee (protocol # 2020‐02‐22).

Inclusion criteria was dogs between 1 and 10 years, weighting more than 5 kg and with a recent (less than 2 months) diagnosis of unilateral cranial cruciate ligament rupture. To be included in this study, a preoperative clinical examination verified that the animals were in good health or had a stabilized chronic disease (ASA score <3) to limit study bias and anesthetic risks to the animal. In order to avoid inclusion of dogs with preexistent signs of GI disturbances, dog with more than 2 episodes of vomiting or diarrhea within the 5 days before surgery were not included in the study. Dogs with known contraindications to the administration of NSAIDs or grapiprant such as renal failure, liver disorders, cardiovascular disorders, GI diseases, as well as gestating or lactating bitches, were not included in the study. Dogs that had received NSAID or grapiprant within 2 weeks preceding surgery or glucocorticoids within 4 weeks preceding surgery were also not included in the study.

2.2. Anesthesia and surgery

The day of the surgery, owners signed an informed consent before enrollment and received a detailed written description of the protocol. Animals then received preanesthetic examination and standard biochemical analyses including hematocrit, total proteins, nitrogen and creatinine blood dosage. All animals were fasted 12 hours before surgery. Just before surgery, an evaluation of osteoarthritis lesion was performed with orthogonal radiographs of the stifle and graded by a radiologist with a 6 grades‐scale (from no osteoarthritis to severe osteoarthritis) based on a validated radiographic scoring system. ¹⁷

Dogs were anesthetized using a standardized protocol. Premedication was ensured by intramuscular administration of methadone (0.2 mg/kg) and medetomidine (10 μg/kg), followed by induction with propofol (1 mg/kg) and ketamine (0.5 mg/kg) IV. Animals were then intubated, and anesthesia was maintained by isoflurane in 100% O₂ throughout surgery. Before incision, an ultrasound guided femoral‐sciatic nerve block was performed by local administration of 0.5 mg/kg of bupivacaine at each site. All animals received an IV administration of amoxicillin before surgery (20 mg/kg).

Tibial plateau leveling osteotomy was performed according to standard procedure. Depending on the presence of meniscal tears, partial or complete meniscectomy was performed and documented. During surgery, the level of isoflurane was continuously adjusted in order to achieve adequate anesthetic plan. If dogs showed signs of pain, such as an increase in mean arterial pressure or heart rate of more than 20% as compared to baseline, fentanyl (1 μg/kg IV) was administered. All animals received a perfusion of 5 mL/kg/h of Ringer's Lactate throughout surgery.

2.3. Postoperative care

After surgery, animals were hospitalized for 24 hours and received IV Ringer's Lactate (2 mL/kg/h) during this period. During hospitalization, postoperative pain was treated by 1 intramuscular administration of methadone (0.2 mg/kg) at the discretion of the clinician. If higher level of analgesia was needed, the animal was excluded from the study. Just before discharge, all animals received an IV injection of meloxicam (Inflacam, 0.2 mg/kg).

2.4. Number of animals, randomization, and study medication

A sample size calculation was performed to allow a detection of a clinically meaningful difference of 20% in PSS in order to demonstrate the superiority of 1 or the other treatment. No studies previously described the values of CBPI after TPLO treated by NSAID administration. However, based on previous studies in other settings, we could expect a meloxicam to provide a PSS close to 5 after TPLO with a SD of approximately 1.7. ¹⁸ Then, the number of animals needed is 61 in each group, with a statistical power of 90% and an alpha risk at .05 (bilateral test). The day after surgery, all included animals were allocated by permuted block randomization to 1 of the 2 treatment groups, in a 1 : 1 ratio.

After randomization, grapiprant (Galliprant, Elanco) or meloxicam (Metacam, Boehringer Ingelheim) tablets were crushed and reconditioned in opaque capsules by the pharmacy department of the EnvA, in order to appear identical between groups. All study personnel and owners, excepted for the investigator in charge of the randomization, were unaware of the treatment group to which each dog was assigned.

All animal then received either 0.1 mg/kg of meloxicam or 2 mg/kg of grapiprant by oral route, for 14 days, starting 24 hours after the postoperative injection of meloxicam (ie, 48 hours after surgery), in a blinded manner. All owners received the instruction to give the medication just before a meal.

2.5. Pain evaluation

Throughout the study period, the evaluator contacted each owner by phone to evaluate pain, using the CBPI, at day 3, 7, 10 and 15 after surgery. This score allowed to assess both pain and quality of life of the dogs. ¹⁹ The CBPI is divided between Pain Severity Score (PSS, mean value of 4 questions on 10 points each) and Pain Interference Score (PIS, mean value of 6 questions on 10 points each), the latter allows evaluation of the pain interference with the dog's general activity, enjoyment of life and locomotive function. In addition, the owner was asked to also rate their overall impression of the dog's quality of life, which was graded as Poor, Moderate, Good, Very Good or Excellent.

Dogs presenting a PSS higher than 5 during the follow‐up were considered as therapeutic failure and were excluded from the study.

2.6. Veterinary assessment

On the day 15, we conducted a veterinary assessment (VA), always by the same investigator, based on an orthopedic score previously validated. ¹⁴ This score was the sum of the result of 7 items: lameness during trot, lameness while walking, support asymmetry, difficulty lifting the contralateral limb, abnormal extension of the limb, pain and edema, each from 0 (absent) to 4 (very severe). The total orthopedic score, that could range from 0 to 28, was the sum of the score of the 7 items.

2.7. Safety, adverse effects, and postoperative complications

Throughout the follow‐up, owners were informed of the risks of most common adverse effects, including vomiting, diarrhea, melena and anorexia and were asked to inform the investigators of any potential adverse effects observed, whether it could be related or not to the medication. All reported adverse effects were documented and analyzed by the investigators. Gastrointestinal disturbance lasting longer than 24 hours or signs with unacceptable severity, such as clinical signs of a surgical site infection, led to the exclusion of the dog from the study at the date of the report.

2.8. Statistical analysis

Statistical analysis was performed using Prism 9.4.0 (GraphPad Software, San Diego, CA). Data were expressed as proportions, mean ± SEM or median [95% confidence interval], as appropriate. Pain Severity Score and PIS were compared by 2‐way ANOVA for time and group and each time point were compared between groups by a post hoc test of Fisher Least Significant Difference (LSD). When differences were statistically significant, differences between means and its 95% confidence interval were reported. Proportions (sex, breed, number of side‐effects, and owner appreciation of the global quality of life), were compared with contingency tables and Fisher's exact tests and effect size was reported as odds ratio and 95% confidence interval. Nonparametric data like orthopedic score and radiographic grade of osteoarthritis were compared using Mann‐Whitney test and parametric data like weight and age were compared between groups using unpaired t test. Cumulative incidence of digestive side effects was compared between groups using a Log‐Rank test (Mantel‐Cox).

For all statistical tests, the significance level was set at P = .05.

3. RESULTS

3.1. Study dogs

Fifty‐four dogs were enrolled and randomized in this study with 6 dogs (3 in each group) prematurely excluded because of either need for rescue analgesia (n = 3) or noncompliance to the protocol because of the preparation of the treatment (n = 3). The final study sample included 24 dogs in each group. Flow chart and evolution of effectives during the study is illustrated in Figure 1. During follow‐up, 4 dogs in the grapiprant group and 6 dogs in the meloxicam group were excluded because of prolonged signs of GI disease, signs of surgical site infection, or a combination of these signs. One animal in the meloxicam group was also excluded at day 3 after surgery because PSS was higher than 5.

FIGURE 1.

Flowchart of the inclusion protocol.

The follow‐up visit was delayed or canceled for 2 dogs in the grapiprant group and for 3 dogs in the meloxicam group; in addition, 1 owner in the meloxicam group chose to do the follow‐up visit in another facility. This led to a final veterinary assessment and orthopedic score documented in only 18 dogs for the grapiprant group and 13 dogs in the meloxicam group.

3.2. Signalment and animal characteristics

The grapiprant and meloxicam groups had similar characteristics regarding age, weight and sex ratio as shown in Table 1. The median age was 4.4 years [3.2‐5.8] and 5.4 years [3.5‐6.1] in grapiprant and meloxicam groups respectively. The mean weight was 31.7 ± 1.9 kg and 29.6 ± 2.9 in grapiprant and meloxicam groups respectively. Male represented 42% and 58% of the grapiprant and meloxicam groups, respectively and sprayed or neutered dogs were 50% of the animals in the grapiprant group and 37.5% in the meloxicam group. No statistical differences were noted regarding the breed repartition between the different groups. Grapiprant group had 6 Labradors, 4 American Staffordshire terriers, and 14 other purebred dogs, and meloxicam group had 2 American Staffordshire terriers, 2 Boxers, 16 other purebred dogs and 4 mixed breed dogs.

TABLE 1.

Demographic and baseline characteristics.

	Groups
	Grapiprant (n = 24)	Meloxicam (n = 24)	P‐value
Animals characteristics
Median age in years [95% CI]	4.4 [3.2–5.8]	5.4 [3.5–6.1]	.51
Mean weight in kg ± SD	31.7 ± 9.4	29.6 ± 14.6	.56
Male (intact, neutered)	42% (4, 6)	58% (4, 10)	.62
Female (intact, sprayed)	58% (6, 8)	42% (5)
Operative characteristics
Surgery of the left stifle	62.5%	58.3%	.99
Surgery of the right stifle	37.5%	42.7%
Partial or total meniscectomy during surgery	37.5%	33.3%	.99
Median osteoarthritis severity score before surgery (95% CI)	4 [3.1‐3.8]	4 [3.0‐4.0]	.64
Methadone injection during surgery	19	12

Baseline characteristics regarding the lateralization of the surgical site (left stifle for 62.5% and 58.3% in Grapiprant and Meloxicam groups, respectively) and severity of osteoarthritis radiographic lesions (median score of 4 in each group) were comparable between groups. Only 1 animal in each group was graded with no osteoarthritis. The same proportion of dogs received partial or total meniscectomy during surgery (37.5% and 33.3% in grapiprant and meloxicam groups, respectively).

3.3. Canine Brief Pain Inventory

Twenty‐four animals have been evaluated in Grapiprant and Meloxicam groups but, because of losses to follow‐up or exclusion (see Safety paragraph), only 23 and 22 animals were evaluated at day 7, 23 and 19 at day 10 and 20 and 17 at day 15 in Grapiprant and Meloxicam, respectively. As illustrated in Figure 2A, there was a significant decrease in pain over time, as assessed by PSS after surgery in all groups. Additionally, 3 days after surgery, grapiprant treatment was associated with a significantly (P = .032) lower value of PSS (2.76 ± 0.18; n = 24) as compared to meloxicam treatment (3.25 ± 0.23; n = 24). No other statistically significant differences between both treatments were noted during the follow‐up for PSS.

FIGURE 2.

Evolution of Pain Severity Score (PSS) and Pain Interference Score (PIS) reported by the owner during the follow‐up after surgery (A) and individual (dots) orthopedic score and median (horizontal line) assessed by a veterinarian 15 days after surgery (B). *P < .05 between groups.

Pain Interference Score was also significantly decreased in both groups throughout the follow‐up (Figure 2A). Grapiprant treatment lead to a significant lower PIS at day 3 and day 10 as compared to meloxicam, with a value of 4.11 ± 0.18 vs 4.69 ± 0.16 at day 3 (−0.58 for grapiprant [−1.03 to −0.13], P = .013) and 2.23 ± 0.13 vs 2.72 ± 0.28 at day 10 (−0.49 for grapiprant [−0.92 to −0.01], P = .049) in grapiprant and meloxicam, respectively.

3.4. Overall appreciation of the quality of life

During CBPI evaluation, we asked the owners to rate the overall quality of life of their dog as “Poor,” “Moderate,” “Good,” “Very good,” or “Excellent.” As illustrated in Figure 3, there was a global increase in the perceived quality of life of the animal throughout the follow‐up in all groups. Yet, in the grapiprant group, 10 owners (41.7%) rated their dog with “Poor,” and 14 owners (58.3%) with “Moderate” (n = 7) or “Good” (n = 7) quality of life at day 3, whereas in the meloxicam group, 20 dogs (83.3%) were rated with “Poor,” and 4 (16.7%) with “Moderate” (n = 2) and “Good” (n = 2) quality of life (odds ratio of 6.97 [1.70‐22.53] in favor of grapiprant). The same results were evidenced at day 7, where most of animals were graded with “Poor” and “Moderate” quality of life (72.7%) in the meloxicam group vs the majority of dogs (73.9%) had “Good,” “Very good,” or “Excellent” grade in the grapiprant group (odds ratio of 7.56 [1.85‐27.20] in favor of grapiprant). No significant differences were observed between groups at day 10 or 15 after surgery regarding the appreciation of their quality of life.

FIGURE 3.

Overall impression of the dog's quality of life rated by the owner throughout the follow‐up. *P < .05 between groups.

3.5. Orthopedic score

In dogs in which the final veterinary assessment was performed, there was a statistically significant difference in orthopedic score assessed by 1 surgeon between grapiprant and meloxicam, 15 days after surgery. Among the 18 dogs analyzed in the grapiprant group, the median value of the orthopedic score was 6 [5.5‐8.16] whereas, the 13 dogs in the meloxicam group had a median value of 9 [7.50‐9.57] (median difference of −3 [−4 to 0] with grapiprant, P = .031, Figure 2B).

3.6. Safety, adverse effects, and postoperative complication

Reported adverse effects were mostly from GI origin with vomiting, diarrhea and anorexia observed in 8 animals (33%) in the grapiprant group and 9 animals (38%) in the meloxicam group. As illustrated in Figure 4, 65% of these GI adverse effects were reported to appear during the first 6 days after surgery. No significant differences between grapiprant and meloxicam treatment were noted, neither for the total percentage of animals (odds ratio of 1.2 [0.36‐3.92], P = .76) nor in the time course of incidence of these adverse effects (P = .740). In 2 (8%) dogs in the grapiprant group and 3 (12%) dogs in the meloxicam group, GI adverse effects lasted more than 24 hours and led to the exclusion of the animals from the study.

FIGURE 4.

Cumulative incidence of reported gastrointestinal adverse effects.

Superficial surgical site infection was suspected in 7 dogs, 5 dogs in the meloxicam group, and 2 dogs in the grapiprant group. These differences were not statistically different (odds ratio of 2.89, [0.50‐16.67], P = .23). Because wound infection could alter the analysis of PSS, PIS and orthopedic scores, dogs presenting with wound infection were withdrawn from the study at the time of the onset of clinical signs. In the meloxicam group, 2 dogs with wound infection also had signs lasting more than 24 hours.

All these reported adverse effects are summarized in Table 2.

TABLE 2.

Number of dogs with reported adverse effects throughout follow‐up.

	Groups
	Grapiprant	Meloxicam
Minor GI adverse effects lasting less than 24 hours
Vomiting	3	2
Diarrhea	3	2
Anorexia	0	1
More than 1 GI symptoms	0	1
Mild or severe GI adverse effects lasting more than 24 hours
Vomiting	0	0
Diarrhea	0	1
Anorexia	0	0
More than 1 GI symptoms	2	2
Total number of animals with GI signs	8	9
Other adverse effects
Surgical site infection suspicion	2	5 a
Exclusion because of long lasting GI adverse effects or wound infection	4	6

Abbreviation: GI, gastrointestinal.

^a Two animals with suspicions of surgical site infection also had severe GI signs.

4. DISCUSSION

The present study is a randomized, double‐blinded clinical trial comparing grapiprant to conventional NSAID in the management of postsurgical articular pain in dogs. We demonstrated that grapiprant and meloxicam are both associated with decreasing postoperative pain after TPLO in dogs. Because of the limited number of animals included in our study and the lack of baseline pain score in our study, we cannot conclude that grapiprant significantly reduce PSS as compared to meloxicam. ²⁰ However, we incidentally demonstrated a significant improvement in overall quality of life assessed by the owners, as well as the orthopedic score assessed by veterinarians, with grapiprant as compared to meloxicam after TPLO.

Pharmacological targets of grapiprant, EP₄ and EP₂ receptor, are overexpressed during chronic inflammation associated with arthritic conditions like osteoarthritis or rheumatoid arthritis. ²¹ Grapiprant is currently indicated in Europe for the management of light to moderate pain associated with osteoarthritis in dogs. However, the EP₄ receptor also plays a constitutive role in articular and bones tissues and PGE₂ is also involved in acute inflammation. ¹⁰ Additionally, grapiprant inhibits PGE₂ binding to EP₄ as soon as 2 to 4 hours after oral administration. ²² Its rapidity of action suggests that grapiprant could represent an interesting therapeutic option for the management of acute or subacute pain, in addition to its indication in chronic pathophysiological affections. However, the use of grapiprant in indications other than light to moderate osteoarthritis is currently under debate. For example, experimental induction of synovitis by sodium urate injection in dogs is associated with worse lameness scores and peak vertical force when treated with grapiprant as compared to carpofen. ²³ In another study, in a model of experimental acute arthritis in dogs, grapiprant was also inferior to firocoxib for the management of acute analgesia. ²⁴ Conversely, we demonstrate here that grapiprant is at least comparable to meloxicam for management postoperative articular pain in dogs, which represent a new finding in the management of postoperative pain. All experimental studies comparing grapiprant to NSAID are measuring pain within the first hours after arthritis induction and we can hypothesis that the absence of benefit in experimental setting as compared to the present study could be related to different time course and pathophysiology between experimental induced arthritis and postoperative inflammatory join. For example, the number of EP₄ receptors is progressively increased in the dorsal root ganglion within the first hour after peripheral injury. ²⁵ This suggests a mechanism of peripheral sensitization of sensory neurons during inflammatory pain. Therefore, it is probable that grapiprant could exert a more potent benefit when EP₄ are the most abundant, that is, if the central sensitization is already settled. In our study, all animals had preexistent osteoarthritis and therefore probable peripheral sensitization, which could explain that we had a higher effect than in experimental trials. Osteoarthritis is commonly encountered in dogs with cranial cruciate ligament rupture as observed in the present study in which 95% of dogs exhibited preoperative radiographic osteoarthritis. ²⁶ Here, we did not document the pain before surgery because all animals presented a high degree of lameness because of the crucitate cruciate ligament rupture. Therefore, the value of CBPI (especially the PIS component) could have been biased by this lameness independent of the pain, as the pain is higher before TPLO surgery as compared to postoperative period. ²⁷

Assuming that dogs in both groups had comparable pain scores at baseline, the apparent benefit of grapiprant over meloxicam on the CBPI and overall quality of life as compared to meloxicam was observed immediately after TPLO surgery, which suggests that grapiprant does not act solely on osteoarthritis associated with cranial cruciate ligament rupture but also on the inflammatory component of the postsurgical pain. In line with our results, a recent experimental study demonstrated that grapiprant and carprofen are equivalent for the management of inflammatory pain after ovariohysterectomy in beagles. ¹⁶ This has then been confirmed by a clinical trial also where administration of a single oral dose of grapiprant 2 hours prior OVH, was comparable to carprofen in terms of perioperative analgesia in dogs. ¹⁵ In this study, the analgesic effect of single grapiprant administration was detectable after 2 hours after the awakening of the dogs and lasted during 18 hours. Altogether, this suggests that grapiprant could represent a suitable alternative to conventional NSAID in the context of postoperative pain.

Safety of long‐term administration of 2 mg/kg grapiprant by oral route has been tested in 2 studies, including in collies Homozygous for MDR1‐1Δ. ¹⁴ , ²⁸ In all these trials, grapiprant was well tolerated with an incidence of digestive adverse effects ranging from 17% to 25% in the treated animals. Here, we document a slighter higher incidence of digestives side effects (33%) than previously. However, an important part of the observed GI adverse effects (65%) were documented within the first 5 days after surgery, which could suggest that those adverse effects are mostly related to anesthesia than NSAID administration. Indeed, it is unlikely that grapiprant adverse effect would appears shortly after the beginning of the treatment as several reports demonstrated no safety concerns with short‐term administration of grapiprant, even at higher dosage than recommended. ¹¹ , ²⁹ However, we cannot rule out the possibility that the use of meloxicam prior to grapiprant in the grapiprant group would have increase the risk of adverse effects. Based on the manufacturer recommendation, meloxicam needs to be injected IV with a loading dose of 0.2 mg/kg, before starting the 1‐daily oral administration at the dose of 0.1 mg/kg. ³⁰ As grapiprant is only available by oral route, and in order to maintain the blinding of the animals, we then decided also to administer this loading dose of 0.2 mg/kg of meloxicam also in dogs receiving the grapiprant. This could have been responsible for a worsening of the incidence of adverse effects as compared to animals receiving only grapiprant. Perioperative injection of meloxicam causes GI signs after surgery in 25% of the treated animals. ³¹ It is generally contra‐indicated to switch from 1 NSAID to another without sufficient washout period. Even if not described with grapiprant, this could represent a limit in the interpretation of our data. Additionally, this could also have increased the analgesic benefit of grapiprant documented within the first period of the study. However, it is unlikely that the effect of such unique initial administration would have been seen at the first pain evaluation (3 days) based on previous pharmacokinetics/pharmacodynamics data on meloxicam. ³²

Another limitation of our study is that, because of blinding reasons, both oral treatments were administered just before meal. This is recommended for meloxicam treatment in order to reduce the risk of adverse‐effects, however, grapiprant has a higher bioavailability, C_max and shorter T_max when administered in fasted animals. ²⁹ Yet, in a study comparing the pharmacokinetics of grapiprant in fasted and fed dogs, the authors concluded that despite modifications in pharmacokinetics, the administration in nonfed dogs minimally prolongs the time when concentration of grapiprant exceed minimal effective concentration (~6 hours) and therefore on its efficacy. ²⁹ The fact that we demonstrated here a higher benefit of grapiprant as compared to meloxicam support this hypothesis. Finally, pain was only evaluated based on CBPI and orthopedic scores, which can represent a major limitation as compared to more objective measurements.

It is also important to note that CBPI baseline score was not measured before surgery in our study. This makes it impossible to assess change from baseline for each dog. While the randomization method was designed so that dogs had the same probability to be allocated to each groups, and the 2 groups were comparable in terms of osteoarthritic severity lesions before surgery, it is possible that the lower pain scores noted with grapiprant were because of unintentional selection bias in this small sample size.

The choice of the pain scoring system (CBPI) instead of other scores specifically designed for postoperative pain evaluation like Glasgow Composite Measure Pain Scale (GCMPS). Contrary to GCMPS, CBPI score is designed to be performed by the owner without specific medical skills and therefore it reduced the need for follow‐up visits and allowed us to have a better follow‐up of the pain after surgery in our study. Additionally, this score also allowed us to distinguish the pain component from its effect on the quality of life of the animals. CBPI is reliable in conditions other than osteoarthritis such as TPLO or bone cancer. ²⁷ , ³³ We also conducted the CBPI evaluation here by phone interviews with a veterinarian which was blinded to the treatment. CBPI is not specifically designed to be done over the phone but such solution, even if not ideal, seemed to be the best compromise for minimizing the risk of bias and ensure regular follow‐up.

Despite our small sample size and lack of baseline data, our study suggests that grapiprant, when given daily for 14 days at 2 mg/kg by oral route, is not associated with higher CBPI score than meloxicam in dogs with cranial cruciate ligament disease treated by TPLO. We incidentally demonstrated a better overall quality of life as assessed by the owner and a significant improvement of the orthopedic recovery assessed by a veterinarian after 15 days. Even if we failed to demonstrate an improvement of safety with grapiprant, we hypothesize that several cofounding factors including anesthesia and postoperative injection of meloxicam, could have biased these observed incidence of adverse effects.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approved by our local clinical research ethics committee (protocol # 2020‐02‐22).

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Cassemiche A, Schoffit S, Manassero M, Kohlhauer M. Comparison of grapiprant and meloxicam for management of postoperative joint pain in dogs: A randomized, double‐blinded, prospective clinical trial. J Vet Intern Med. 2024;38(4):2324‐2332. doi: 10.1111/jvim.17136