Osteoarthritis (OA) is the most common joint disease and cause of chronic pain in dogs. OA can affect any joint, including hips, elbows, and stifles, but also vertebral facet joints, carpal joints, tarsal joints, and metacarpophalangeal and metatarsophalangeal joints (1, 2). Estimates from North America report age specific prevalence values ranging from 20% in dogs older than 1 year up to 80% in dogs older than 8 years, based on radiographic and clinical data from referral settings (3). A recent study suggested that OA and associated clinical signs may have a prevalence of ~37% in the canine population (4).

Non-steroidal anti-inflammatory drugs (NSAIDs) are currently the primary pharmaceutical therapy recommended for dogs with OA (5–7). Carprofen (Rimadyl^®) is one of the leading NSAIDs used globally (6, 8–10). NSAIDs may cause gastrointestinal ulceration as an adverse effect and are contraindicated in the presence of renal insufficiency (2, 11). However, the true incidence of adverse events is currently unknown (12).

Various combinations of glucosamine hydrochloride and chondroitin sulfate are probably the most common nutraceuticals used in dogs with OA (6). Based on the available literature (5, 13–18), the evidence for a benefit of glucosamine and chondroitin in is mixed.

PCSO-524 is a marine based fatty acid compound comprising of a patented extract stabilized marine lipids from the New Zealand green lipped mussel (Perna canaliculus). In dogs, it has been reported that the PCSO-524 is associated with beneficial effect outcomes in clinical OA cases (19–22).

Krill oil is an edible oil extracted from Euphausia superba; a small, red-colored crustacean found in the Antarctic. Bioavailability of krill oil has been suggested to be higher than fish oil as much of the EPA and DHA in krill oil are bound to phospholipids. A recent study in humans found that krill oil improved the subjective symptoms of knee pain in adults with mild knee pain (23). However, the efficacy of krill oil has not been studied in OA dogs. EAB-277 is a combination of phospholipids extracted from krill oil together with lipid fractions from the Green Lipped Mussel.

We hypothesized that Glucosamine/chondroitin sulfate supplements, PCSO-524, EAB-277 and carprofen would result in a superior therapeutic effect in treating OA in dogs compared to placebo and that PCSO-524 and EAB-277 would be more effective than Glucosamine/chondroitin sulfate supplementation.

In total, 155 dogs were screened, and 85 dogs met the inclusion criteria and were enrolled into the study. During the study, 10 dogs dropped out of the study due to loss of contact (n = 3), signs of neurological disease (n = 1), excessive pain (n = 2), vehicular accident (n = 2), and not being able to trot across the force platform (n = 2). At the end of the study, 75 dogs were used in the statistical analyses: 14 dogs in PCSO-524 group, 16 dogs in Glucosamine/chondroitin group, 15 dogs in EAB-277 group, 15 dogs in carprofen group, and 15 dogs in placebo group.

There were 46 male and 29 female dogs. The average (mean ± SD) of age, body weight, and BCS were 5.17 ± 2.76 years, 32.83 ± 9.55 kg, and 3.07 ± 0.66, respectively. Breeds represented are tabulated in Table 3. Clinically, 20 dogs were predominately affected on the right hind limb and 55 dogs predominately affected on the left hind limb. Radiographically, radiographic OA was documented bilaterally in 56 dogs and unilaterally in 19 dogs. Dogs were classified as mild OA (OAS = 2) in 45 dogs and moderate OA (OAS = 3) for 30 dogs.

Study subject characteristics, including gender, body condition score, breed, side of affected limb, uni- or bi-laterally affected, radiographic score, and OAS are detailed in Table 4. There was no significant difference between five treatment groups (p > 0.05) for any variable.

Overall, the results of this study indicate that there appear to be benefits of carprofen, PCSO-524, and EAB-277 for the treatment of OA-pain in dogs based on the measurement of PVF. Glucosamine/chondroitin and placebo (sunflower oil) did not appear to be associated with positive treatment effects based on the measurement of PVF. At 4 and 6 weeks after treatment, the change in PVF of both PCSO-524 and EAB-277 were similar to that of the carprofen group.

Currently, a mainstay of the assessment of OA-pain are clinical metrology instruments (CMIs). These are a sequence of questions that are scored based on the observations of the caregiver, and if constructed correctly, can be a valid measure of the impact of OA-pain on the dog. There are several CMIs that have been developed, validated, and reported for measuring the severity of OA in dogs such as the Liverpool Osteoarthritis in Dogs (LOAD) instrument (26), the Canine Brief Pain Inventory (CBPI) (10), the Helsinki Chronic Pain Index (HCPI) (27). The owner has to complete questionnaire and therefore has to comprehend the questions, and the questions need to be relevant to the contextual environment which includes the local culture. A recent study in Thailand (22) that used the CBPI suggested that a translated version of the CBPI may not have been fully understood, and our own pilot experience with the LOAD indicated that even with translation, the questions may not have been appropriate for the Thai culture. Ideally, each CMI should be validated in each new language and culture. Therefore, no CMIs were used in this study as none have been validated in the Thai language and culture. Future work should be directed at developing valid CMIs for use in Thailand.

In contrast to CMIs, ground reaction forces (GRFs) measured using a force plate are an objective measure, and have been used as a proxy estimate of joint pain in dogs with appendicular joint OA (8, 28–32). In the present study, both a positive control (the NSAID carprofen) and a negative control (placebo) were included in order to put PVF changes in the other groups into context. The PVF of placebo group remained unchanged (−0.33 ± 3.65) as expected after study completed (6 weeks). The changes in PVF (significant when compared to baseline) in this study with carprofen after 2, 4, and 6 weeks were 3.14 ± 5.33, 3.08 ± 5.87, and 4.21 ± 6.52, respectively. These values are similar to those of a recent study (28) which found an increase in PVF of 3.2 ± 0.8 after 2 weeks treatment of carprofen. In that study, most enrolled dogs had hip OA, and the baseline PVF for index limb was similar in both studies (62.1 ± 13.5 and 60.7 ± 13.5). Our result is also similar to another OA study (22) using the NSAID treatment firocoxib where the change in PVF of the index limb was reported to be 3.03 ± 4.67 and 3.25 ± 4.13 at 2 and 4 weeks treatment, respectively. However, all these values are lower than the 5%−10% change discussed in the literature as being “unlikely to have occurred by chance” (32, 33). Those data and conclusions were drawn based on treatment of cruciate ligament disease, and these may not be relevant to a situation where one is assessing the change in limb use in dogs with bilateral OA. We included a placebo treatment group, and so while the change in PVF may not be within the 5%−10% range suggested as clinically significant, the change we found in the PCSO-524, EAB-277, and carprofen groups was significantly different from the placebo group. While further research needs to be done to ascertain what a meaningful change in PVF would be in dogs with bilateral OA, we believe that the change seen in the PCSO-524, EAB-277, and carprofen groups is clinically meaningful. Indeed, the standardized effect size (SES) for PCSO4-524 was 1.1 at 6 weeks.

To our knowledge, this was the first clinical trial of EAB-277 nutraceutical in OA dogs. EAB-277 appeared to produce similar results to carprofen, and PCSO4-524. The main difference between EAB-277 from PCSO-524 is that it contains with high concentrations of phospholipids extracted from Euphausia superba in addition to the components of PCSO4-524.

The results for the glucosamine/chondroitin group were similar to previous studies that measured GRFs at ~1 month following initiation of treatment (5, 21). However, previous studies have reported positive effects of glucosamine/chondroitin for the treatment of OA-pain after 70 days of treatment (15), and 90 days of treatment (16). It is possible that positive effects of glucosamine/chondroitin would have been seen with a longer duration of treatment. The relatively short duration of treatments is certainly a limitation of the study.

Symmetry indices (across opposite limbs, or across all limbs) have been suggested as outcome measures. Indeed, they have been used as a measure of efficacy of locally applied therapeutics (34), and this seems appropriate. We did not use symmetry indices in this study as it is not known how symmetry indices change in dogs with bilateral OA-pain that are administered systemic analgesics. Therefore, we used the scientifically defendable approach of measuring the change in PVF of the index limb (8).

Regarding the subjective veterinarian-based evaluation of lameness score, pain score, joint mobility score and bearing score, there were no differences between groups. The OAS we used has not been validated, and indeed, nor has any other subjective veterinarian assessment of lameness or pain. Indeed, veterinarian assessment of lameness has been shown to be unreliable. (30) That said, the approval studies for carprofen used veterinarian (as well as owner) assessments, and found significant improvements based on veterinarian assessments (9, 10).

The goals of OA treatment are to reduce pain, maintain joint function and overall mobility. In this study, the NSAID carprofen demonstrated the fastest improvement in PVF PCSO-524 and EAB-277, but not glucosamine/chondroitin, demonstrated significant improvements in PVF from baseline after 4 weeks, and 6 weeks, and of a similar degree to that seen with carprofen. Future research should evaluate the combination of PCSO-524 and EAB-277 with an NSAID to test the synergistic effects as the multimodal therapy management.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Institutional Animal Care and Use Committee of Khon Kaen University (IACUC-KKU-77/60). Written informed consent was obtained from the owners for the participation of their animals in this study.

BL and NK designed the study. DK, SJ, TS, SH, KY, PK, and PT collected data. NK analyzed data. SS, BL, and NK contributed to statistical analysis. NK, DK, SJ, TS, SH, KY, PK, PT, SS, and BL participated in drafting and revising the manuscript. All authors contributed to the article and approved the submitted version.