Feline Osteoarthritis (Degenerative Joint Disease)
Feline degenerative joint disease (DJD), Feline DJD, Arthritis in cats, Feline OA, Cat joint disease
Osteoarthritis means the smooth cartilage cushioning a joint wears down and the bone around it changes shape, which hurts. It is very common in cats, and it is easy to miss, because most arthritic cats never limp. Instead they stop jumping on the counter, take the stairs one at a time, groom less, sleep more, hide, or start missing the litter box. Owners often read this as 'she's just getting older.' BE CLEAR ABOUT WHAT TREATMENT CAN AND CANNOT DO: arthritis does not go away and cannot be cured. It is a permanent, slowly progressive condition. What treatment does is reduce the pain and keep it under control for the rest of your cat's life, which means ongoing medication or monthly injections, recheck visits and repeat bloodwork, with a recurring cost. Improvement is usually real but modest rather than dramatic - in the trial that got the monthly injection approved, about 9 cats had to be treated for 1 extra cat to be rated a success by its owner - so judge it over 8 to 12 weeks against a written baseline, not by whether your cat suddenly acts like a kitten. TWO THINGS THAT KILL CATS, BOTH OF THEM AVOIDABLE. First, NEVER give your cat human pain medicine. Not acetaminophen or paracetamol (Tylenol, Panadol), not ibuprofen (Advil, Motrin), not naproxen (Aleve), not aspirin. Acetaminophen has no safe dose in a cat at all - cats cannot process it the way people do - and the others can cause stomach bleeding and kidney failure. Second, NEVER give your cat medicine that was prescribed for a dog or for a different cat. Dog arthritis drugs are not cat arthritis drugs: the anti-nerve-growth-factor injection for dogs (Librela, bedinvetmab) is a different molecule from the one for cats (Solensia, frunevetmab), and dog anti-inflammatories are not dosed or licensed for cats. If your cat has already been given any of these, call your veterinarian or an animal poison control service immediately - do not wait for symptoms. AND CALL BEFORE YOU CHANGE ANYTHING. Once a plan is running, do not raise or lower a dose, skip a dose, stop a medicine, restart a leftover medicine, or add a supplement or an over-the-counter product without asking your veterinary team first. The good news is that this is a treatable condition: there is a monthly injection that the US FDA has approved specifically to control arthritis pain in cats, plus diet and home changes that help, and your veterinarian can build a plan that fits your cat's kidneys and age.
What you might notice
- Stops jumping onto favorite high spots, or jumps up in two smaller hops instead of one
- Hesitates at stairs, or goes up and down one step at a time
- Grooms less, so the coat looks matted or greasy, especially over the back and hips
- Sleeps more and plays less; seems to have 'slowed down' over months, not days
- Less sociable or more irritable, hides more, or objects to being picked up or brushed
- Starts urinating or defecating outside the litter box (climbing in over a high side hurts)
- Stiff or stilted walk, especially after resting; a shortened stride rather than an obvious limp
- Noticeable limping is uncommon in cats with arthritis, so its absence does not rule the condition out
- A cat can have painful arthritis and still show nothing on a hands-on joint exam - in one 101-cat series no cat with appendicular OA had obvious pain on palpation - so ask for two-view X-rays rather than accepting a normal exam as the answer
Standard of care
Stage the kidneys and blood pressure first, then choose the drug
Approved / guideline-backedBecause chronic kidney disease and DJD co-occur far more often than chance in affected cats, renal staging and blood pressure measurement are part of treating arthritis, not a separate workup. This determines whether an NSAID is appropriate, at what dose, and how often the cat needs rechecking. It is also the step that most often changes the plan. Take a full medication history at the same time: senior cats on this pathway are frequently already on benazepril, telmisartan, amlodipine or a diuretic.
CKD prevalence was 68.8% in cats recruited to four prior DJD studies at a referral institution and 50% in a randomly selected age-stratified cohort spanning 6 months to 20 years; the authors concluded there is a need for increased CKD screening when selecting DJD treatments (cats). CARRY THE COHORT CAVEAT: the DJD-recruited figure comes from studies that had themselves EXCLUDED azotemic cats (BUN and/or serum creatinine >20% above the upper end of normal in two studies, IRIS stages 3 and 4 in the other two), with stages 1 and 2 dominating - so 68.8% is not a primary-care prevalence and neither figure should be restated as 'most affected cats'. The clinical instruction (stage the kidneys first) needs no inflation to stand. The 2024 ISFM/AAFP consensus guideline provides information on screening prior to prescription and on use in the presence of comorbidities, and notes that management of chronic pain may be complicated by prescription of other drugs with the potential for interactions with NSAIDs. Species: CAT.
Frunevetmab (Solensia) - monthly subcutaneous anti-NGF monoclonal antibody, US FDA-approved for cats
Approved / guideline-backedA felinized monoclonal antibody that binds nerve growth factor, given by subcutaneous injection once a month. As verified on 2026-09-12 it is the only product the US FDA has approved specifically to control the pain of osteoarthritis in cats (this is a time-sensitive exclusivity statement and must be re-verified against the FDA Green Book before each publication cycle). Being a monoclonal antibody, it is cleared by ordinary protein catabolism rather than by hepatic or renal drug-metabolizing pathways - state that as a pharmacological fact and NOTHING MORE. It is not a comparative safety claim against NSAIDs, no head-to-head comparative safety study exists, and the approved label itself lists renal insufficiency as an adverse reaction in 6.6% of treated cats (12/182) versus 4.3% of controls (4/93), which matters on a page built around a CKD-heavy population. TWO THINGS TO SAY BEFORE THE FIRST INJECTION. (1) It cannot be withdrawn. The half-life is roughly 10 days (10.1 +/- 1.9 days after IV dosing; label t1/2 9.8 +/- 3.1 days in laboratory cats and 11.0 +/- 2.5 days in osteoarthritic cats), the dosing interval is monthly, and there is no antidote - so unlike an oral NSAID, an adverse reaction cannot be stopped by discontinuing the drug. (2) Effect sizes are real but modest, so it should be used alongside diet and home modification, and its response should be measured against a written baseline at 8 to 12 weeks, not assumed.
Named products: Frunevetmab (Solensia, Zoetis Inc.) - US FDA NADA 141-546; FDA-approved indication: control of pain associated with osteoarthritis in cats. Say 'FDA-approved in the United States', never 'approved' unqualified. DOSE AS THE LABEL DOSES IT, BY WHOLE VIALS AND WEIGHT BAND, NOT BY mg/kg: 5.5-15.4 lb (2.5-7.0 kg) = the full content of 1 vial; 15.5-30.8 lb (7.1-14.0 kg) = the full content of 2 vials; administered subcutaneously once a month, targeting a MINIMUM of 0.45 mg/lb (1 mg/kg). The whole-vial banding exists to prevent vial splitting and partial draws, so do not publish the trial's mg/kg exposure range as a dose - it belongs in the evidence field only. Cats below 2.5 kg (5.5 lb) or above 14.0 kg (30.8 lb) fall OUTSIDE the labeled weight bands and the label gives no banding for them. LABEL CONTRAINDICATIONS: should not be administered to cats with known hypersensitivity to frunevetmab; should not be used in breeding cats or in pregnant or lactating queens, because it may pass through the placental blood barrier and be excreted in milk (fetal abnormalities and increased stillbirths and postpartum mortality). LABEL HUMAN USER WARNING, verbatim: 'Not for use in humans. Keep out of reach of children. Hypersensitivity reactions, including anaphylaxis, could potentially occur in the case of accidental self-injection. In case of accidental self-injection, seek medical advice immediately and show the package leaflet or the label to the physician. Pregnant women, women trying to conceive, and breastfeeding women should take extreme care to avoid accidental self-injection.' The label's stated basis: 'The importance of NGF in ensuring normal fetal nervous system development is well-established and laboratory studies conducted on nonhuman primates with human anti-NGF antibodies have shown evidence of reproductive and developmental toxicity.' This warning must reach the person holding the syringe, including any owner offered at-home administration. DO NOT SUBSTITUTE THE CANINE PRODUCT. Bedinvetmab (Librela, Zoetis Inc.) is US FDA-approved for the control of pain associated with osteoarthritis in DOGS under NADA 141-562 (verified against the FDA label record, DailyMed setid 61c7b4e0-aab6-4cf5-8471-f94be438c758) and has no feline indication. The reason not to dose-adjust it for a cat is pharmacological, not a claim about the manufacturer's motives: a caninized antibody is a foreign protein to a cat, carrying immunogenicity and hypersensitivity risk, with no feline pharmacokinetic, safety or efficacy data.
Randomized, placebo-controlled, double-blind, parallel-group superiority trial in 275 client-owned cats with naturally occurring OA pain (182 frunevetmab / 93 placebo; nominal 1.0 mg/kg SC, effective exposure range 1.0-2.8 mg/kg, days 0, 28, 56). Significant improvement over placebo on client-specific outcome measures at days 28 and 56 (number needed to treat 9; standardized effect size 0.3 at day 56) and on veterinarian-assessed joint pain at days 56 and 84 (effect size 0.18 at day 56). In that trial adverse events did not differ between groups except skin disorders, which were significantly more frequent with frunevetmab (32/182 vs 8/93). DO NOT LET ONE TRIAL'S AE COMPARISON STAND IN FOR THE PRODUCT'S SAFETY PROFILE - quote the approved US label's adverse-reaction table, frunevetmab (n=182) vs vehicle control (n=93): vomiting 13.2% vs 10.8%; injection site pain 10.9% vs 14.0%; diarrhea 6.6% vs 5.4%; abnormal behavior/behavioral disorders 6.6% vs 5.4%; RENAL INSUFFICIENCY 6.6% vs 4.3%; anorexia 6.6% vs 4.3%; lethargy 6.0% vs 3.2%; dermatitis 6.0% vs 1.1%; alopecia 5.5% vs 2.2%; dehydration 4.4% vs 0.0%; lameness 4.4% vs 2.2%; pruritus 3.8% vs 0.0%; weight loss 3.3% vs 5.4%; scabbing on head/neck 3.3% vs 1.1%; gingival disorder 2.7% vs 0.0%; bacterial skin infection 2.2% vs 1.1%; otitis externa 2.2% vs 0.0%. A supporting multisite pilot in 126 cats corroborated the owner-questionnaire result with accelerometry: weekly activity fell 0.9% from baseline in treated cats vs 9.3% in placebo cats. An independent evidence appraisal of three RCTs graded the overall strength of evidence as moderate and called for more objectively measured outcomes to reduce reliance on subjective measures. Immunogenicity is low: treatment-emergent anti-drug antibodies in 4/259 (1.5%) treated cats versus 3/131 (2.3%) placebo cats, with no safety findings attributed to immunogenicity. Species: CAT.
NSAID therapy - meloxicam or robenacoxib, prescribed to the feline label of the jurisdiction, and NEVER carprofen for chronic feline dosing
Approved / guideline-backedNSAIDs remain effective analgesics in cats and are the most-studied drug class for feline OA, but the labeled use differs sharply by country and cats are not small dogs. NAME THE REGULATOR EVERY TIME. In the UNITED STATES, no NSAID is FDA-labeled for chronic osteoarthritis in cats: the METACAM injectable label restricts cats to a single one-time dose for post-operative pain and carries a boxed warning against repeat dosing, and ONSIOR (robenacoxib) tablets are FDA-labeled for post-operative pain for a maximum of 3 days. Outside the US the position may differ, but we could NOT verify the exact feline indication wording of any non-US authorization for long-term oral meloxicam for this page, so we do not assert one - never tell an owner a drug is 'approved for long-term use in cats' in their country without reading that country's own label (EMA for the EU, VMD for the UK, APVMA for Australia, ICA/INVIMA for Colombia). Any chronic use in the US is extralabel and must be a documented veterinarian decision with renal monitoring, lowest-effective-dose titration and owner counseling on gastrointestinal, renal and hepatic warning signs. CO-MEDICATION AND PERI-PROCEDURAL RULES, FROM THE LABELS. Do not administer an NSAID together with any other oral or injectable NSAID or with a corticosteroid. Cats at highest risk are those with existing renal, cardiovascular and/or hepatic dysfunction, those that are dehydrated, and those receiving diuretics or other nephroactive drugs; NSAID plus RAAS blockade (benazepril, telmisartan) plus dehydration is a recognized route to acute kidney injury, and the 2024 ISFM/AAFP consensus explicitly addresses interactions with other prescribed drugs. Withhold NSAIDs around dental and surgical procedures and during any period of fasting, vomiting or dehydration, and use perioperative fluid therapy where an NSAID is given with anesthesia, because anesthetic drugs may compromise renal perfusion - directly relevant here, since dental disease is both a named differential for this condition and the commonest recorded feline primary-care diagnosis. Robenacoxib may prolong the QT interval, with the associated arrhythmia risk unknown, so avoid co-administration with other QT-prolonging drugs.
Named products: Meloxicam (METACAM, Boehringer Ingelheim) - US FDA injectable label: single, one-time SC dose 0.14 mg/lb (0.3 mg/kg) for control of post-operative pain in cats, with the boxed warning 'Repeated use of meloxicam in cats has been associated with acute renal failure and death. Do not administer additional injectable or oral meloxicam to cats.' The label also states do not follow that single, one-time dose with any other NSAID, and do not administer METACAM Oral Suspension following it. Chronic oral meloxicam in cats in the US is extralabel. Post-approval reports include elevated liver enzymes. Robenacoxib (ONSIOR 6 mg tablets for cats, Elanco) - US FDA NADA 141-320, for the control of post-operative pain and inflammation associated with orthopedic surgery, ovariohysterectomy and castration, in cats 5.5 lb (2.5 kg) and over and 4 months of age and over, for up to a maximum of 3 days. Contraindicated in cats with hypersensitivity to robenacoxib or known NSAID intolerance. There is no US chronic-osteoarthritis label for robenacoxib in cats. Elevated liver enzymes (AST, ALT) were observed in safety studies. CARPROFEN - NOT A FELINE CHRONIC-PAIN OPTION. Carprofen appears in the 2024 ISFM/AAFP guideline's keyword set, but the US FDA-approved carprofen products (Rimadyl and generics) are labeled for DOGS and carry no feline indication, and we could not verify any label in any jurisdiction authorizing repeat or chronic carprofen dosing in cats. Where carprofen is licensed for cats at all it is as a single perioperative dose. Do not present carprofen as a feline chronic-pain option, and never give a cat a dog's carprofen tablet.
The governing consensus document covers mechanism of action, indications for use, screening prior to prescription, use in the presence of comorbidities, monitoring of efficacy, and avoidance and management of adverse effects; its indexed keyword set names carprofen, meloxicam and robenacoxib (guideline, cats). It PROVIDES INFORMATION on pre-prescription screening - it does not 'require' it, and it is scoped to long-term NSAID use, so it is not the authority that makes frunevetmab first-line. Efficacy and long-term tolerability evidence: a prospective study of 40 cats with OA on oral meloxicam 0.01-0.03 mg/kg once daily for a mean 5.8 months found gastrointestinal upset in 2/46 (4%) as the only adverse effect, no detected deleterious effect on renal function, and owner-assessed efficacy good or excellent in 34/40 (85%) - open-label, no placebo arm (cats). A retrospective case-control study in 38 cats OLDER THAN 7 YEARS (that is the cohort's age floor, not a treatment duration), identified from a 2005-2009 record period, with median treatment duration 467 days in the CKD subgroup and 327 days in the non-CKD subgroup and a median titrated maintenance dose of 0.02 mg/kg/day (range 0.015-0.033), found no difference in sequential serum creatinine or urine specific gravity versus untreated controls, and less progression of renal disease in the meloxicam-treated CKD group than in age- and IRIS-matched untreated CKD cats. Bound that inference: of the 38 cats, 22 had stable CKD at the start (IRIS stage 1 n=8, stage 2 n=13, stage 3 n=1) and none were stage 4, the design is retrospective and hypothesis-generating, and the authors state prospective studies are required. This must NOT be marketed as 'meloxicam protects kidneys'. Species: CAT.
Omega-3 (EPA/DHA)-enriched therapeutic diet or supplementation - a low-risk ADJUNCT, never a substitute for analgesia
Controlled trial in this speciesThe best-supported non-drug intervention, easy for owners to comply with, and it can be run alongside any of the drugs above. This is the nutritional intervention to recommend - not glucosamine-chondroitin. POSITION IT HONESTLY AND NEVER AS A DRUG REPLACEMENT. Do not tell owners a diet works as well as the approved analgesics: no head-to-head trial exists, and on a public page that claim is what makes an owner decline analgesia for a painful cat. Describe it as LOW RISK rather than risk-free - absence of documented harm in a feline OA management literature the leading review group counts at only 29 publications is absence of evidence, not evidence of absence - and note the real practical considerations: gastrointestinal upset, altered platelet function relevant to any planned surgery, the caloric load in a cat on a weight-management plan, and palatability failure with oxidized oil.
A 2022 PROSPERO-registered systematic review and meta-analysis (CRD42021279368) of 72 trials across dogs and cats reported, verbatim, 'an evident clinical analgesic efficacy for omega-3-enriched diets, omega-3 supplements and cannabidiol (to a lesser degree)', a weak efficacy for collagen, and 'a very marked non-effect of chondroitin-glucosamine nutraceuticals'. Report that results list COMPLETE, including cannabidiol; PetSmartMeds separately does not recommend CBD for cats, for stated reasons and not by quietly editing a source's results: there is no FDA-approved cannabidiol product for cats, legal status varies by jurisdiction, feline dosing and long-term safety data are limited, and the meta-analysis itself graded its efficacy as the lesser of the three. NOTE also that this meta-analysis pooled dogs AND cats; the feline-only trial base is small. THE DEDICATED FELINE EVIDENCE IS ONE SMALL TRIAL, so give its actual numbers: a randomized, double-blinded, placebo-controlled crossover trial in 16 cats with radiographically documented naturally occurring OA (10 weeks fish oil delivering 1.53 g EPA and 0.31 g DHA per 1000 kcal ME vs 10 weeks corn oil) found 2 of 5 owner-rated outcomes reaching significance - less stiffness during gait (P=0.03) and higher jumps (P=0.03) - with activity level, stair use and owner interaction all at P=0.07, and the authors concluded only that supplementation 'changes the owner's perception of some aspects of behaviour and locomotion'. An earlier feline review stated plainly that 'unlike the dog, there is limited evidence for the effectiveness of omega-3 fatty acid-rich diets in managing feline OA and further work is required' (Bennett 2012, PMID 22247327). DO NOT REPUBLISH THE COMPARATIVE SENTENCE. The 2025 GREPAQ feline OA review writes that an omega-3 enriched diet 'has demonstrated therapeutic efficacy comparable to standard pharmacological treatments, without side effects and with high compliance'. That sentence is quoted accurately from PMID 40685570 but is not republishable here: it rests on no head-to-head trial, the review attributes it to 'Research from the GREPAQ' - the authors' own group, i.e. self-citation rather than independent replication - and 'without side effects' is an absolute zero-risk claim. A diet marketed as having efficacy comparable to approved analgesics for a named disease also reads as a drug claim. Species: CAT (plus pooled canine/feline meta-analysis).
Gabapentin as an adjunct for cats that remain painful, or that cannot take an NSAID - WITH MANDATORY DOSE REDUCTION IN RENAL IMPAIRMENT
Controlled trial in this speciesUseful where NSAIDs are contraindicated by kidney disease, or as add-on analgesia. Two trade-offs must be disclosed honestly. FIRST, THE RENAL DOSING TRAP, WHICH IS THE WHOLE POINT OF THIS STEP. Gabapentin is eliminated largely unchanged by renal excretion, and this page routes the very population that most often has CKD toward it. Cats with stable IRIS stage 2 and 3 CKD given 10 mg/kg PO had significantly HIGHER dose-normalized serum gabapentin concentrations than healthy cats given 20 mg/kg, with concentrations correlating with serum creatinine and SDMA, and the authors concluded this supports the need to dose-reduce in this patient population. So do NOT start a CKD cat at the feline OA trial dose. Reduce the dose and/or extend the interval in any cat with renal impairment - start at the low end (on the order of 5 mg/kg or less, or the same dose given less often) and titrate upward to effect. Be explicit that NO feline study has established a CKD-specific gabapentin dose for osteoarthritis pain: the dose-reduction evidence comes from single-dose serum concentrations, not from an OA efficacy trial, so this is titration under monitoring rather than a validated regimen. Warn the owner to stop the next dose and call for excessive sedation, wobbliness, weakness or inability to stand. SECOND, IT IS A COMFORT-AND-SEDATION DRUG IN THIS SETTING, NOT A MOBILITY DRUG. In the only feline OA trial, gabapentin improved the activities owners cared about but measurably reduced overall activity, and sedation was the commonest adverse effect. SPECIFY THE PRODUCT FORM. Prescribe a formulation intended for the cat. Human gabapentin oral liquids may contain excipients that have not been evaluated for cats - some human oral solutions contain xylitol - and compounded preparations vary in concentration, excipients and stability, so require the compounder to state the excipients and use a preparation you can verify.
Named products: Gabapentin (no FDA-approved veterinary product for feline analgesia; extralabel use). Renal impairment requires dose reduction and/or interval extension. Avoid human oral liquid formulations whose excipients have not been evaluated in cats.
Blinded, placebo-controlled, randomized crossover trial in 20 osteoarthritic cats 10 years and older (gabapentin 10 mg/kg PO q12h for 2 weeks vs placebo, no washout). Gabapentin was associated with approximately 3-fold greater odds of client-specific outcome measure ratings indicating improvement in impaired activities, but also with significantly LOWER mean daily accelerometer activity counts (39,038 vs 48,333 counts/day). Adverse events were recorded in 10 of 20 cats on gabapentin (sedation, ataxia, weakness, muscle tremors) vs 1 on placebo (lethargy) - a 50% adverse-event rate in a cohort that was NOT stratified or selected by renal function. RENAL DOSING EVIDENCE: in a pharmacokinetic and limited-sampling study, 5 healthy cats received a single 20 mg/kg oral dose and then 5 healthy cats plus 25 cats with stable IRIS stage 2 (n=14) and stage 3 (n=11) CKD received a single 10 mg/kg oral dose. Cats with CKD had significantly higher dose-normalized serum gabapentin concentrations than normal cats at 3 h and 8 h, concentrations correlated with serum creatinine and symmetric dimethylarginine, and the authors concluded this supports the need to dose-reduce in this patient population (Quimby 2022, PMID 35195476, DOI 10.1177/1098612X221077017). Species: CAT. Gabapentin appears on the FDA list of bulk drug substances for compounding office-stock drugs for use in non-food-producing animals (dogs, cats, horses); it is not an FDA-approved veterinary analgesic, so use in cats is extralabel.
Environmental modification and slow, supervised weight optimization
Early evidencePractical, free or cheap, and LOW RISK - not zero-risk: step platforms or ramps to favorite high spots, a large litter box with one low or cut-down side placed on every floor the cat uses, soft warm bedding away from drafts, food and water at floor level and not stacked, non-slip runners on smooth floors, and gentle grooming help for areas the cat can no longer reach. THE RESIDUAL RISKS ARE REAL AND SHOULD BE NAMED. Ramps and steps can cause falls in an ataxic, sedated or visually impaired senior cat, so introduce them low, wide and non-slip and supervise at first. Changing the litter box can itself precipitate inappropriate elimination, so add the new box alongside the old one rather than replacing it. WEIGHT: SLOW, SUPERVISED, AND ABANDONED IF THE CAT STOPS EATING. Weight optimization is worth doing for the cat's overall health and mobility, but rapid caloric restriction in a senior, CKD or partly inappetent cat risks loss of lean body mass and hepatic lipidosis. A weight-loss rate of 0.5% to 2% of body weight per week is what is recommended for cats, precisely to preserve lean mass and reduce hepatic-lipidosis risk. Any weight plan must be veterinary-supervised, and must be stopped and reviewed if the cat's food intake falls. Present the arthritis rationale truthfully as well: the feline data do not establish that excess weight causes or worsens arthritis the way it does in dogs and people, but they also do not establish the opposite (see signalment), and lameness and idiopathic hepatic lipidosis are both documented obesity-associated risks in cats.
Recommended by review and guideline consensus but NOT supported by feline controlled trials. Bennett et al. state plainly that research into environmental enrichment as a way of improving quality of life in cats with painful OA is lacking, but that it is an approach worth using where possible, and that modifications such as comfortable bedding and ramps are also important (review, cats). The 2025 GREPAQ review is explicit that proposed alternatives including food restriction, weight loss, physiotherapeutic modalities and lifestyle changes are primarily based on human clinical studies and veterinary research on other species, which introduces a high risk of bias, and counts only 29 publications on feline OA management (review, cats). On weight and OA severity specifically: in 64 cats that all had confirmed OA, severe knee OA was detected in 10-16% of affected joints with no difference between underweight, normal-weight and overweight groups - but the same authors add that 'the general prevalence of clinical symptoms and radiographic signs is different in overweight cats', and with no OA-free comparison group the study cannot test causation at all (cats). On the rate of weight loss: a feline obesity management review states that a weight-loss rate of 0.5-2% per week is recommended, to increase the chance of preserving lean body mass and decrease the risk of developing hepatic lipidosis (review, cats). Idiopathic hepatic lipidosis and lameness are listed among documented health risks associated with obesity in cats (review, cats).
Glucosamine-chondroitin joint supplements - recommend AGAINST, and say why in full
No evidence locatedThese are the most commonly bought over-the-counter cat joint products and the best available pooled evidence says they do not relieve pain. PetSmartMeds should say so. Money spent here is money not spent on the interventions that do have evidence. State the recommendation on the meta-analysis, and quote the dissenting source in full rather than truncating it.
The 2022 systematic review and meta-analysis of 72 canine and feline trials reported 'a very marked non-effect of chondroitin-glucosamine nutraceuticals, which leads us to recommend that the latter products should no longer be recommended for pain management in canine and feline osteoarthritis.' That sentence carries this recommendation. An earlier systematic review of nutraceuticals in horses, dogs and cats found only 1 of 22 included controlled trials was performed in cats, and rated the strength of evidence low for all nutraceuticals except omega-3 fatty acids in dogs. QUOTE BENNETT IN FULL, INCLUDING THE PART THAT CUTS AGAINST US. Bennett et al. write: 'There is no published data as yet for the usefulness or otherwise of nutraceuticals (glucosamine and chondroitin) in managing feline OA; studies in the authors' clinic suggest some pain-relieving effect.' Stopping that sentence at the semicolon, as an earlier draft of this page did, misrepresents a cited author and is not an acceptable way to build a recommendation. The honest position: one review group reports an unpublished clinical impression of benefit, while the only pooled quantitative synthesis found a marked non-effect - and a published meta-analysis outweighs an unpublished in-house impression. Species: pooled canine/feline; feline-specific controlled data essentially absent.
Physical rehabilitation, therapeutic laser, acupuncture, pulsed electromagnetic field therapy
No evidence locatedThese are widely offered for cat arthritis and are low-risk, and some cats clearly enjoy the handling and warmth. But we could locate no controlled trial of any of them in cats with naturally occurring OA. If offered, they should be presented as unproven adjuncts that a willing cat may tolerate, not as evidence-based therapy, and they must not displace the approved drug, the diet or the kidney monitoring.
NO EVIDENCE LOCATED in cats with naturally occurring OA. The 2025 GREPAQ feline OA review states there is limited knowledge regarding the management of feline OA, with only 29 publications available, many of which lack rigorous methodology, and that proposed alternatives including physiotherapeutic modalities are primarily based on human clinical studies and veterinary research on other species, which introduces a high risk of bias. The positive photobiomodulation/PEMF combination trial that does exist was performed in 30 DOGS with hip OA, on top of bedinvetmab, not in cats, and must not be cited as feline evidence (Cidral LO, et al. Vet Res Commun. 2026;50(3). PMID 41824140. DOI 10.1007/s11259-026-11133-3).
2024 ISFM and AAFP consensus guidelines on the long-term use of NSAIDs in cats — International Society of Feline Medicine (ISFM, International Cat Care) and American Association of Feline Practitioners (AAFP). This is the governing document for chronic NSAID pharmacotherapy in cats, and note its scope precisely: it is scoped to the long-term use of NSAIDs, and it PROVIDES INFORMATION on mechanism of action, indications for use, screening prior to prescription, use in the presence of comorbidities, monitoring of efficacy, and avoidance and management of adverse effects - it does not 'require' screening, and it is not the authority that establishes frunevetmab as first-line therapy. Its indexed keyword set names carprofen, meloxicam and robenacoxib; the guideline also notes that the cat's unique metabolism should be considered when prescribing any medication and that management of chronic pain may be complicated by prescription of other drugs with the potential for interactions with NSAIDs. There is NO feline-osteoarthritis-specific ISFM/AAFP consensus guideline - we searched and it does not exist, so do not cite one. The complementary documents are the 2022 AAHA Pain Management Guidelines for Dogs and Cats (Gruen ME, Lascelles BDX, Colleran E, et al. J Am Anim Hosp Assoc. 2022;58(2):55-76. PMID 35195712. DOI 10.5326/JAAHA-MS-7292), which include chronic-pain assessment algorithms, owner pain-assessment instruments and a tiered therapeutic decision tree; and the 2022 ISFM Consensus Guidelines on the Management of Acute Pain in Cats (Steagall PV, et al. J Feline Med Surg. 2022;24(1):4-30. PMID 34937455. DOI 10.1177/1098612X211066268).
How it is diagnosed
- History-led mobility and behavior interview with the owner - jumping height, stair use, grooming, litter-box use, sleeping locations, sociability. Owner-reported behavioral change, not lameness, is the primary clinical signal in this species (Bennett 2012, PMID 22247326; Lascelles 2010, PMID 20210938).
- Full orthopedic and neurological examination including joint palpation and range of motion. The examination IS informative - Slingerland concluded that 'clinical examination of the larger peripheral joints had the highest sensitivity and specificity for radiographic OA' (PMID 20083417) - and lumbosacral DJD was associated with palpable back pain in 7 cats in the Kimura series. What must NOT be inferred is that a normal appendicular palpation excludes OA: in that same 101-cat orthogonal-radiograph series, no obvious palpable pain was detected in any cat with appendicular OA (PMID 32296718). Net message: a negative appendicular palpation does not rule out OA; it does not mean palpation is useless.
- Orthogonal (two-plane) radiographs of the appendicular joints, and of the lumbosacral spine where indicated. Single-plane films under-diagnose hip and stifle DJD (PMID 32296718).
- Rule out the differentials that mimic reduced mobility: hyperthyroidism, cardiac disease and aortic thromboembolism, spinal disease, trauma, neoplasia, and pain of non-joint origin including severe dental disease. Auscultate and assess respiratory effort at every visit - in cats with hypertrophic cardiomyopathy, congestive heart failure and aortic thromboembolism are leading modes of cardiac death and both can be mistaken for 'slowing down with age' (Payne 2015, PMID 26776589, DOI 10.1016/j.jvc.2015.09.008).
- Stage the kidneys and measure blood pressure BEFORE selecting analgesia - serum creatinine, SDMA where used, urine specific gravity, urine protein:creatinine ratio, systolic blood pressure. CKD co-prevalence was 68.8% in DJD-recruited cats and 50% in a random age-stratified cohort, and the 2024 ISFM/AAFP consensus provides information on screening prior to prescription and on use in the presence of comorbidities (PMID 24217707; PMID 38587872).
- Baseline CBC/biochemistry and, in senior cats, total T4 - both to screen comorbidity and to give a reference point for repeat monitoring on treatment. Record the full current medication list, including cardiac and renal drugs, before prescribing an NSAID.
- Record a baseline owner-completed mobility score (client-specific outcome measures; FMPI short form only if licensed - see monitoring.licensing) so that response to treatment can be measured rather than guessed (Enomoto 2022, PMID 34002643, DOI 10.1177/1098612X211011984). RESEARCH TOOL, NOT A CLINICAL INSTRUMENT: a collar-accelerometer support-vector-machine classifier across 51 of 56 cats over 14 days achieved AUC 78% (confidence interval 0.65-0.88, with 68% sensitivity at 75% specificity) - but its ground truth was OWNER-REPORTED mobility change (24 cats with no reported change vs 27 with reported change 'suggestive of early DJD', validated through clinical orthopedic examination), i.e. it classifies the same subjective signal this page says must be supplemented, and the published confidence interval is not stated as 95% in the source record. An AUC of 78% with a lower bound of 0.65 is not a diagnostic test and there is no deployable clinical product to recommend (Montout 2025, PMID 40204089, DOI 10.1016/j.tvjl.2025.106352).
Where a biologic fits
Our own product, graded by the same rule
Short answer: no. We looked, and there is no good evidence a biologic helps here.
For osteoarthritis in cats there is no controlled trial evidence that mesenchymal stromal cell, exosome or peptide products relieve pain or slow joint disease: the only published feline OA cell-therapy report is an uncontrolled, unblinded, open-label pilot in six cats conducted largely by the product's own manufacturer (Mitani et al., Res Vet Sci 2024;182:105470), no in-vivo exosome study in feline joints has been published at all, no animal cell or exosome product is FDA-approved for any species, and the only treatment for feline osteoarthritis pain with placebo-controlled evidence and regulatory approval is the anti-nerve-growth-factor antibody frunevetmab (Solensia, US FDA NADA 141-546) - whose own benefit is real but modest, with a number needed to treat of 9.
What exists
MESENCHYMAL STROMAL CELLS (CATS): exactly one published in-vivo report in cats with naturally occurring OA. Mitani et al. gave a single intravenous injection of allogeneic feline adipose-derived MSCs to SIX cats with OA and reported that the Feline Musculoskeletal Pain Index improved significantly one month later, with no adverse effects observed (Res Vet Sci. 2024;182:105470. PMID 39612738. DOI 10.1016/j.rvsc.2024.105470). This design cannot support an efficacy claim: n=6, single-arm, open-label, NO placebo or control group, one-month follow-up, and three of the four authors were employees of the manufacturer (J-ARM Co., Ltd.); the authors' own conclusion is that ADSC therapy 'could have beneficial effects' and that 'further research is necessary'. The problem is compounded by the instrument used - the original 17-item FMPI was tested for responsiveness in a randomized, double-blinded, placebo-controlled crossover study of 25 cats with DJD and detected a significant positive response to PLACEBO (P=0.0001) while failing to distinguish meloxicam from placebo at all; 'controlling for significant covariates, most positive effects were seen for placebo treatment' (Benito 2013, PMID 23551140, DOI 10.1111/jvim.12077). An uncontrolled single-arm improvement on an instrument documented to register large placebo responses is not evidence of drug effect. EXOSOMES / EXTRACELLULAR VESICLES (CATS): zero. A PubMed search of extracellular vesicles or exosomes combined with cats and joint, cartilage or arthritis returned 0 records (run 2026-09-12). The feline EV literature that exists is in-vitro characterization or other organ systems; the single best-designed in-vivo feline EV trial located anywhere is in limbal stem cell deficiency of the eye, not joints. There is also no published in-vivo study of umbilical-cord MSC-derived exosomes in a dog, cat or horse with osteoarthritis in any species. PEPTIDES (CATS): zero. A PubMed search for 'BPC 157' combined with cats returned 0 records; there is no peptide in the MicroAmino catalogue with controlled efficacy evidence in dogs, cats or horses. PLATELET-RICH PLASMA (CATS, adjacent but not an MSC/exosome/peptide product): one exploratory prospective case series of SIX cats given intra-articular autologous PRP reported improvement in FMPI and clinician visual analogue scale at day 42-45 - again uncontrolled, unblinded, n=6, with two authors employed by a commercial device company (Huntingford 2024, PMID 38863451, DOI 10.3389/fvets.2024.1394055). OTHER SPECIES, LABELLED AS SUCH, AND WITH THE REGULATOR NAMED: MSC evidence for OA in DOGS consists of real but small randomized controlled trials. DogStem is a biological veterinary medicinal product containing EQUINE umbilical-cord-derived MSCs, authorised by the EUROPEAN MEDICINES AGENCY for use in DOGS (EMEA/V/C/005829, EU/2/22/285, EquiCord S.L.) - a xenogeneic product. It is NOT FDA-approved, it is not authorised for cats anywhere, and it must never be described as allogeneic, as a canine-donor product or as a US predicate. Its regulator-assessed pivotal field trial figures are 51% of DogStem-treated dogs versus 5% of placebo-treated dogs meeting the primary force-plate endpoint at 8 weeks, and 39% versus 11% at 12 weeks; the separately published single-arm rates from the sponsor's own JAVMA paper (63% gait analysis, 77% orthopedic examination, 65% owner-assessed quality of life at 8 weeks in 80 dogs with elbow or hip OA - Punzon 2022, PMID 36198051, DOI 10.2460/javma.22.06.0237) are NOT treatment-versus-placebo deltas and must not be quoted as though they were. None of that is feline data and none of it may be presented as feline data. MECHANISTIC WORK IN CATS EXISTS BUT IS NOT THERAPEUTIC: calcium imaging confirmed functional TRPV1, TRPV2, TRPA1 and TRPM8 in healthy cat dorsal root ganglion neurons and GFRA3 expression in cat DRG; in DJD cats the TRP transcript changes did NOT reach statistical significance, and serum artemin correlated with radiographic DJD score but NOT with veterinarian pain score (Wheeler 2026, PMID 41923874, DOI 10.3389/fpain.2026.1716651).
What has not been shown
There is no randomized, controlled, blinded trial of any mesenchymal stromal cell, exosome/extracellular-vesicle, conditioned-medium, secretome or peptide product in cats with osteoarthritis - not one, for any route, at any dose. There is no feline dose-ranging or pharmacokinetic study for any of these products in joint disease. There is no published in-vivo exosome or EV study in feline joints at all. There is no evidence that any of these products slows radiographic progression, preserves cartilage, or reduces the need for approved analgesia in cats. There is no measurement of anti-drug antibody development in cats after repeat xenogeneic (for example human- or equine-derived) cell or EV dosing. There is no established post-injection flare rate for EV products in cats. There is no FDA-approved animal cell, exosome or conditioned-medium product for any species or indication in the United States: such products are unapproved new animal drugs, and FDA CVM has issued warning letters on exactly this basis, including one naming canine and feline conditioned media (Safari Stem Cell LLC, MARCS-CMS 661023, 5 April 2024). TWO CLAIMS MUST NEVER BE MADE. First, these products are not immune-invisible and 'immune-privileged' is banned, together with any 'zero-flare' claim. Punzon et al. randomized 24 healthy police working dogs to intra-articular xenogeneic equine umbilical-cord MSCs, allogeneic canine adipose MSCs, or placebo and found that no adverse events were detected following single and repeated administration 'despite both equine and canine MSC generate antibody titres in the dogs' (Front Vet Sci. 2023;10:1098029. PMID 37266387, DOI 10.3389/fvets.2023.1098029). Note that BOTH the xenogeneic and the species-matched product raised titres. DogStem's own EMA Summary of Product Characteristics records that 50% of dogs in a laboratory study developed antibodies to the xenogeneic cells and that the effect of those antibodies on efficacy has not been evaluated. Immune-TOLERANT is the defensible word; immune-PRIVILEGED is not. Second, 'exosomes lack cell surface antigens' is factually wrong - EVs carry a surface proteome that includes MHC molecules. A THIRD CAUTION, FROM THE ONLY FELINE INTRAVENOUS CELL-SAFETY DATASET OF ANY SIZE: in feline chronic kidney disease pilot studies, intravenous cryopreserved allogeneic adipose MSCs caused vomiting during infusion in 2/5 cats and increased respiratory rate and effort in 4/5 cats, and the authors concluded that 'administration of cryopreserved aMSCs was associated with significant adverse effects and no discernible clinically relevant improvement in renal functional parameters' (Quimby 2013, PMID 23632128, DOI 10.1186/scrt198). Different indication, but it is not reassuring. Finally, the leading feline OA research group has said this in print: proposed alternatives including adjunctive musculoskeletal treatments with biologics are primarily based on human clinical studies and veterinary research on other species, which introduces a high risk of bias (PMID 40685570).
Sources
- The only feline OA cell-therapy report: allogeneic feline adipose MSC, n=6, uncontrolled, manufacturer-run — Mitani K, Ito Y, Takene Y, Inaba T. Evaluation of the quality of life-enhancing effect of allogeneic feline adipose mesenchymal stem cells in cats with osteoarthritis: A pilot study. Res Vet Sci. 2024;182:105470. PMID 39612738. DOI 10.1016/j.rvsc.2024.105470 https://doi.org/10.1016/j.rvsc.2024.105470
- Why an uncontrolled FMPI improvement proves nothing: FMPI responded significantly to PLACEBO and could not separate meloxicam from placebo — Benito J, Hansen B, DePuy V, Davidson GS, Thomson A, Simpson W, Roe S, Hardie E, Lascelles BDX. Feline musculoskeletal pain index: responsiveness and testing of criterion validity. J Vet Intern Med. 2013;27(3):474-82. PMID 23551140. DOI 10.1111/jvim.12077 https://doi.org/10.1111/jvim.12077
- Leading feline OA group, in print: biologics for feline OA rest on human and other-species extrapolation, high risk of bias — Lefort-Holguin M, Delsart A, Frezier M, Martin L, Otis C, Moreau M, Castel A, Lussier B, Martel-Pelletier J, Pelletier J-P, Troncy E. Osteoarthritis in cats: what we know, and mostly, what we don't know. . . yet. J Feline Med Surg. 2025;27(7). PMID 40685570. DOI 10.1177/1098612X251347999 https://doi.org/10.1177/1098612X251347999
- The approved, placebo-controlled feline alternative - and its honest effect size (NNT 9, ES 0.3) — Gruen ME, Myers JAE, Tena J-KS, Becskei C, Cleaver DM, Lascelles BDX. Frunevetmab, a felinized anti-nerve growth factor monoclonal antibody, for the treatment of pain from osteoarthritis in cats. J Vet Intern Med. 2021;35(6):2752-2762. PMID 34724255. DOI 10.1111/jvim.16291 https://doi.org/10.1111/jvim.16291
- MSCs are NOT immune-privileged: BOTH equine and canine MSCs raised antibody titres in dogs, in the sponsor's own randomized safety trial — Punzon E, Salgueiro L, Ramirez L, et al. Local, systemic and immunologic safety comparison between xenogeneic equine umbilical cord mesenchymal stem cells, allogeneic canine adipose mesenchymal stem cells and placebo: a randomized controlled trial. Front Vet Sci. 2023;10:1098029. PMID 37266387. DOI 10.3389/fvets.2023.1098029 https://doi.org/10.3389/fvets.2023.1098029
- The n=80 canine RCT of equine UC-MSC in dogs - dog data, labelled as dog data — Punzon E, Salgueiro L, Beltran V, et al. Equine umbilical cord mesenchymal stem cells demonstrate safety and efficacy in the treatment of canine osteoarthritis: a randomized placebo-controlled trial. J Am Vet Med Assoc. 2022;260(15):1947-1955. PMID 36198051. DOI 10.2460/javma.22.06.0237 https://pubmed.ncbi.nlm.nih.gov/36198051/
- The only sizeable feline intravenous MSC safety dataset - cryopreserved cells caused significant adverse effects with no renal benefit — Quimby JM, Webb TL, Habenicht LM, Dow SW. Safety and efficacy of intravenous infusion of allogeneic cryopreserved mesenchymal stem cells for treatment of chronic kidney disease in cats: results of three sequential pilot studies. Stem Cell Res Ther. 2013;4(2):48. PMID 23632128. DOI 10.1186/scrt198 https://doi.org/10.1186/scrt198
- US regulatory status: FDA CVM warning letter naming canine and feline conditioned media and stem cells as unapproved new animal drugs — FDA Warning Letter, Safari Stem Cell LLC, MARCS-CMS 661023, 5 April 2024 https://www.fda.gov/inspections-compliance-enforcement-and-crimi
- Adjacent autologous biologic, also uncontrolled n=6: intra-articular PRP in feline DJD — Huntingford J, Looney A, Johnson J, Miller L. The use of platelet rich plasma in the treatment of degenerative joint disease in cats: an exploratory case series. Front Vet Sci. 2024;11:1394055. PMID 38863451. DOI 10.3389/fvets.2024.1394055 https://doi.org/10.3389/fvets.2024.1394055
- EMA authorisation, named regulator: DogStem is EQUINE umbilical-cord MSCs authorised by the EMA for DOGS - not FDA-approved, never for cats — European Medicines Agency, DogStem (equine umbilical cord-derived mesenchymal stem cells), EMEA/V/C/005829, EU/2/22/285, EquiCord S.L., status Authorised; regulator-assessed pivotal field trial 51% DogStem-treated vs 5% placebo-treated dogs on the primary force-plate endpoint at 8 weeks and 39% vs 11% at 12 weeks; SmPC records 50% antibody formation in a laboratory study with the effect on efficacy unevaluated https://www.ema.europa.eu/en/medicines/veterinary/EPAR/dogstem
Tracking response
How you know whether it is working
Client-specific outcome measures (CSOM) as the primary owner-completed instrument, in which the owner nominates three specific activities their own cat currently struggles with and scores those over time. CSOM is the instrument on which frunevetmab's pivotal approval data were generated. The Feline Musculoskeletal Pain Index short form (FMPI-sf, 9 items) is the better-validated questionnaire but is NOT freely licensable - see monitoring.licensing - so it is added only once written permission is on file. Optional objective adjunct, research-grade only: collar-mounted accelerometry.
Enomoto M, Lascelles BDX, Robertson JB, Gruen ME. Refinement of the Feline Musculoskeletal Pain Index (FMPI) and development of the short-form FMPI. J Feline Med Surg. 2022;24(2):142-151. PMID 34002643. DOI 10.1177/1098612X211011984 - pooled data from 180 cats across four studies; the original 17-item FMPI had reasonable reliability but low or no responsiveness, and stepwise item elimination produced a 9-item short form whose between-group treatment effect was always greater than the original. CRITICAL CAVEAT to quote alongside any FMPI result: Benito J, et al. J Vet Intern Med. 2013;27(3):474-82. PMID 23551140. DOI 10.1111/jvim.12077 found the original FMPI detected a significant positive response to PLACEBO (P=0.0001) and could not distinguish meloxicam from placebo - so an uncontrolled before-and-after FMPI improvement is not evidence that a treatment worked. CSOM basis: Gruen ME, et al. J Vet Intern Med. 2021;35(6):2752-2762. PMID 34724255. DOI 10.1111/jvim.16291. Accelerometry classifier, RESEARCH TOOL ONLY: Montout AX, et al. Vet J. 2025;311:106352. PMID 40204089. DOI 10.1016/j.tvjl.2025.106352 - AUC 78% with a confidence interval of 0.65-0.88 (the source record does not state the interval is 95%) and 68% sensitivity at 75% specificity, in 51 of 56 cats over 14 days, against a ground truth of OWNER-REPORTED mobility change validated by clinical orthopaedic examination, not against radiographic or independently diagnosed DJD. There is no deployable clinical product to recommend.
Suggested cadence: Baseline owner score plus baseline renal panel, urine specific gravity, urine protein:creatinine ratio and systolic blood pressure BEFORE the first treatment, together with a full current medication list. Re-score the owner instrument at 4 weeks and again at 8 weeks after starting or changing any treatment - the frunevetmab pivotal trial read out at days 28, 56 and 84, so 8 to 12 weeks is the honest window in which to judge whether it is working, and a cat with no improvement on the owner instrument after 2 to 3 monthly injections should be recorded as a non-responder and have its plan changed rather than injected indefinitely. Once stable: owner score every 4 weeks (that is, at each monthly injection or refill), renal panel and blood pressure at 3 months then every 6 months, and sooner in any cat with IRIS-staged CKD or on chronic NSAID therapy, in any cat on gabapentin whose renal function changes, or after any episode of vomiting, anorexia or dehydration. Photograph or video the cat's jump onto one nominated surface at baseline and at each recheck - it is the single most useful owner-collected objective record. At every check-in, re-ask the two safety questions: has anyone given this cat a human medicine or another animal's medicine, and has anything about the dose or the routine changed without a call?
Questions you can answer at home
- How high can your cat jump right now, in one go - and is that higher, the same, or lower than last time we asked?
- Does your cat go up and down stairs normally, one step at a time, or avoid the stairs altogether?
- Is your cat's coat clean and smooth everywhere, or are there matted or greasy patches over the back, hips or tail base?
- Is your cat using the litter box every time, or have there been accidents outside it?
- How much of the day does your cat spend playing, exploring or interacting with you, compared with sleeping or hiding?
- Since the last check-in, is your cat clearly better, about the same, or worse - and what one thing tells you that?
- Has your cat been given ANY medicine we did not prescribe - anything from your own medicine cabinet, or anything prescribed for a dog or another pet? Answer honestly; we are not judging, we need to know.
- Has anything about the dose, timing or routine changed since we set the plan - a missed dose, a doubled dose, a new supplement or joint chew, a restarted old prescription?
- Is your cat eating normally? If your cat is on a weight-loss plan and has gone off its food, tell us today.
- Since the last injection or dose change, have you seen any itching, scabs or hair loss around the head and neck, any vomiting or diarrhoea, any yellowing of the gums or eyes, or any unusual sleepiness or wobbliness?
Last reviewed: 2026-09-20 · Every claim on this page carries a citation you can check. If one does not hold up, tell us and we will correct it. science@petsmartmeds.com
Build a profile for your pet
Tell us what you are dealing with and the site will show you only what applies.