Preliminary results of five feline total hip replacements



Five feline total hip replacements (including one case of staged bilateral total hip replacement) for management of proximal femoral physeal fractures are reported. The surgical procedure was a modified version of that widely used in dogs. Craniodorsal luxation was the only major complication encountered. Subjectively outcomes appeared to be good in all cases.


Total hip replacement (THR) has been available as a salvage procedure for the coxofemoral joint in medium and large breed dogs for over three decades (Hoefle 1974, Olmstead and others 1983). Before the introduction of smaller THR prostheses in Europe in 2006 (Liska 2008), the only salvage procedure available for the coxofemoral joint in cats was femoral head and neck ostectomy (FHNO), resulting in a pseudoarthrosis. Where salvage procedures of the coxofemoral joint are indicated, restoration of normal hip joint function is the preferred outcome.


Five THRs were performed in four cats. The mean age at surgery was 18·2 months (range 14 to 24 months) and mean weight was 5·0 kg (range 4·5 to 5·3 kg). All cats were neutered males. The breeds included were domestic shorthair, British blue, Siamese × Abyssinian and Bengal. Three cats (four hips) had spontaneous femoral capital physeal fracture (FCPF) and one had combined Salter-Harris type I and type III traumatic FCPF.


All cats were premedicated with 0·3 mg/kg im acepromazine (ACP injection, Novartis, Camberley, UK) and 0·3 mg/kg im papaveretum (Papaveretum injection BP, Martindale Pharmaceuticals, Romford, UK). Anaesthesia was induced with 4 mg/kg iv propofol (Propflo™, Abbott Laboratories Ltd, Maidenhead, UK) and maintained with isoflurane (Isoflo™, Abbott Laboratories Ltd). A fentanyl transdermal patch (Duragesic DTrans, Janssen-Cilag, Titusville, FL, USA) 25 μg/hour was applied 12 hours before surgery to provide analgesia. Amoxycillin/clavulanate (Noroclav, Norbrook Pharmaceuticals Worldwide, Newry, Northern Ireland 8·75 mg/kg im) was given before surgery and intraoperatively (Augmentin, GlaxoSmithKline, Brentford, UK 20 mg/kg iv initially, and thereafter 10 mg/kg iv at 2-hourly intervals). Meloxicam (Metacam, Boeringer Ingelheim, Bracknell, UK) was administered 0·3 mg/kg sc on the day of surgery followed by 0·05 mg/kg/day orally starting 48 hours after surgery for 14 days. Amoxycillin/clavulanate (Clavaseptin, Vetoquinol UK Ltd Buckingham, UK) 12·5 to 25 mg/kg orally twice daily in one case; cephalexin (Ceporex, Galen Craigavon, Northern Ireland) 10 to 25 mg/kg orally twice daily in three cases and cefovecin sodium (Convenia, Pfizer Animal Health, Kent, UK) 8 mg/kg sc in one case were given postoperatively.

The BioMedtrix CFX™ Micro Hip System (BioMedtrix, Boonton, NJ, USA) was used in all five cases. This is a cemented system utilising polymethylmethacrylate (PMMA) bone cement (Surgical Simplex P, Stryker Howmedica Osteonics, Mahwah, NJ, USA). Decisions regarding prosthesis sizes (femoral stems #2 or #3, acetabular cup 12, 14 or 16 mm) were made before surgery using acetate templates overlaid on standard radiographic views.

The surgical procedure was similar to that previously described for cemented THR in canine and feline patients (Olmstead and others 1981, Olmstead 1995, Liska and others 2009). Preparation of the acetabulum and femoral canal was performed using power tools. PMMA was vacuum mixed (Stryker Mixevac III, Mahwah, NJ, USA) and injected in liquid phase into the acetabulum and femur. Central placement of the femoral stem in the femoral canal was assessed subjectively. A femoral canal cement restrictor plug and femoral stem centralizer are not available for the Micro Hip System. Trial heads were used to determine the appropriate femoral neck length (+0/+2). Closure following reduction was performed routinely.

Immediate postoperative radiographs (ventrodorsal pelvis with hindlimbs extended, lateral pelvis, caudocranial femur and mediolateral femur) were used to assess the cement mantles and positioning of the prostheses. Three independent reviewers (an orthopaedic surgeon, HWS; an orthopaedic intern, PGW; and an imaging resident) used trigonometric techniques previously described for human patients (Ackland and others 1986, Pierchon and others 1994) and dogs (Dyce and others 2001) to measure the angle of lateral opening (ALO) of the acetabular cup on postoperative ventrodorsal radiographs.

Sutures were removed 10 to 14 days after surgery. Cage confinement for four weeks followed by a gradual increase in activity over the following eight weeks was recommended. Re-examinations were scheduled at four weeks for orthopaedic examination (Table 1) and 12 weeks postoperatively (orthopaedic examination in addition to radiography) with annual re-examinations thereafter. Measurements of left and right hindlimb girth at the level of the proximal third of the femur were made for case 2 at a re-examination 16 months after surgery.

Table 1. Indices for assessment of outcome following total hip replacement in feline patients
Lameness of operated limb0 – no overt
 1 – mild
 2 – mild to moderate
 3 – moderate
 4 – moderate to severe
 5 – severe
Comfort on palpation/manipulation of operated hip0 – comfortable
 1 – mild resentment
 2 – mild to moderate resentment
 3 – moderate resentment
 4 – moderate to severe resentment
 5 – severe resentment
Passive range of motion of operated hip0 – within normal limits
 1 – mildly reduced
 2 – moderately reduced
 3 – severely reduced
Muscle atrophy on operated limb0 – none appreciated
 1 – mild
 2 – mild to moderate
 3 – moderate
 4 – moderate to severe
 5 – severe


Table 2 lists the sizes of prostheses used. Case 1 received a #3 femoral stem and 14 mm acetabular cup, whereas the other cases all received #2 femoral stems and 12 mm acetabular cups. Femoral necks were +0 in all cases at initial surgery. Mean ALO measurements by three observers (Table 3) ranged from 38·3° to 61·4° (mean 51·5°). A minor complication was encountered in case 2 when sutures were removed by the cat two days after surgery necessitating repeat suturing.

Table 2. Sizes of acetabular and femoral prostheses utilised in five feline total hip replacements
CaseAcetabular cup (mm)Femoral head (mm)Femoral stem (mm)
1148 mm + 03
2128 mm + 02
3128 mm + 02
4128 mm + 02
5128 mm + 0/+22
Table 3. Mean angles of lateral opening for five feline total hip replacements measured by three separate observers from ventrodorsal radiographs
CaseMean ALO (degrees)

Lameness, muscle atrophy and resentment of coxofemoral joint manipulation ranged from mild to moderate at the initial examinations (Table 4), with passive range of movement recorded as within normal limits in all cases. At the second re-examination (Table 5), mild muscle atrophy was present in cases 3 and 4 (the left and right hips, respectively, of one cat with staged bilateral procedures). Lameness and resentment of hip manipulation scored 0 for cases 1 to 4 at this stage. Case 5 scored higher for lameness and resentment to hip manipulation (Table 5) and radiography confirmed craniodorsal luxation of the prosthetic joint (Fig 1). Open reduction was performed. Prosthesis positioning was not revised since ALO was satisfactory (Table 3) as measured from radiographs. The +0 femoral head was replaced with a +2 femoral head to improve stability by increasing the tension in the soft tissues. Radiography confirmed hip reduction at 4, 9 and 16 weeks after surgery.

Table 4. Assessment of five feline cases of total hip replacement (first re-examination)
CaseTime postop (weeks)LamenessComfort on palpation/manipulationPROMMuscle atrophy
  1. PROM, passive range of motion in hip joint.

Table 5. Assessment of five feline cases of total hip replacement (second re-examination)
CaseTime postop (weeks)LamenessComfort on palpation/manipulationPROMMuscle atrophy
  1. PROM, passive range of motion in hip joint.

Figure 1.

Craniodorsal luxation of a right total hip replacement in a cat (case 5) diagnosed radiographically 15 weeks following surgery despite satisfactory clinical assessment. Note also suboptimal cement fill of the femoral medullary canal and contralateral hip dysplasia.

Follow-up examinations ranged from 7 to 27 months (mean 15 months) after initial surgery (Table 6). Case 1 disappeared 27 months after THR surgery, so information was derived from owner questioning regarding hindlimb function before his disappearance. The remaining three cats (four hips) were re-examined, and all owners reported excellent outcomes with unlimited exercise and no visible gait abnormalities. Examinations of the three cats were graded 0 for all indices (Table 6). Pelvic limb girth at the proximal third of the femur measured 28 cm on the right hindlimb (THR) and 26 cm on the left hindlimb (FHNO) for case 2.

Table 6. Assessment of five feline cases of total hip replacement (third re-examination)
CaseTime postop (months)LamenessComfort on palpation/manipulationPROMMuscle atrophy
  1. PROM, passive range of motion in hip joint

  2. *By owner questioning

  3. Twelve months following contralateral femoral head and neck ostectomy


Follow-up radiography revealed substantial femoral cortical bone hypertrophy (up to a 20% increase in total femoral diameter in mediolateral radiographs and up to 10% increase in caudocranial radiographs) at the level of the femoral stem in three of five cases (Fig 4). Radiography revealed inconsistent cement fill of the femoral medullary canals in all cases and femoral stem positioning eccentric to the femoral long axis in cases 2 and 5 (Figs 3 and 4), with radiolucency noted at the stem-cement interface in case 5 (Fig 1). Follow-up radiography of case 5 (with a different imaging system) at seven months after THR confirmed a satisfactory stem-cement interface (Fig 4), albeit with incomplete femoral canal cement fill and a lucency lateral to the distal stem tip (probably a gas bubble which formed in the liquid phase PMMA).

Figure 4.

Cranio-caudal view of the same right femur as figure 1, seven months following total hip replacement (case 5). Linear lucency surrounding the femoral stem has apparently not progressed, although radiographic positioning is not identical.

Figure 3.

Ventrodorsal (hips extended) radiograph of the pelvis of a cat at 16 months post right total hip replacement and 12 months post left femoral head and neck ostectomy. (Case 2). Note proximal location of left proximal femur compared with the right in addition to the appearance of femoral foreshortening owing to reduced left hip extension during imaging. The femoral stem is positioned lateral to the midline. Note also substantial femoral cortical bone hypertrophy (especially laterally at the level of the femoral prosthesis) and intra-medullary cement extending to the level of the distal metaphysis.

Spontaneous FCPF occurred in the contralateral hindlimbs of cases 2 and 3 (16 and 4 weeks after THR, respectively). A second THR was performed in one cat (cases 3 and 4 are the left and right THRs of a single cat; Fig 2) and FHNO was performed in case 2 owing to financial constraints (Fig 3).

Figure 2.

Ventrodorsal (hips extended) radiograph of the pelvis of a cat with staged bilateral total hip replacements, 14 months following left THR (case 3) and 13 months following right THR (case 4).


THR surgery produced joints which appeared pain free and functionally normal. FHNO is an alternative to THR (Berzon and others 1980). However, high levels of postoperative rehabilitation are required for optimal long-term function (Levine and others 2005, Piermattei and others 2006) and in dogs the procedure results in long-term limb shortening, unpredictable pain control and muscle atrophy (Duff and Campbell 1977). Furthermore, force plate analysis shows that functional deficits persist after FHNO, even in small dogs (Off and Matis 1997). Miniature THR prostheses have been used effectively in small dogs (Matis and Holz 2008). Although no objective data relating to functional outcome in cats after FHNO have been reported, it seems likely that cats will benefit from joint replacement in preference to FHNO. Circumstantial evidence to support this suggestion is provided by Jeffery (1989) who reported that cats with FCPFs functioned better after fracture repair than after FHNO, and similarly Liska and others (2009) reported better functional outcomes in three cats after THR than five cats after FHNO. Hindlimb muscle mass disparity in the single case reported here of unilateral THR and contralateral FHNO (case 2) would appear to support this.

Indications for THR include severe osteoarthritis (OA) refractory to medical management, chronic coxofemoral luxation, irreparable femoral head and neck fractures and avascular necrosis of the femoral head (ANFH) (Olmstead and others 1983). Hip OA is relatively common in the domestic cat (Clark and others 2005), but its recognition is low, either because cat owners do not see the hindlimb lameness or because cats are able to compensate for functional impairment. Feline THR is likely to be performed infrequently as long as feline hip OA remains under-diagnosed. Chronic coxofemoral luxation where reduction is unsuccessful is an accepted indication for THR, although it is less frequently reported in cats than in dogs (Hammer 1980, Basher and others 1986, Sissener and others 2009). ANFH, an indication for THR in the dog, has not been reported in the cat, possibly owing to the relatively greater blood supply to the femoral head and neck via the round ligament (Scott and McLaughlin 2007).

Femoral head fracture is an indication for THR in dogs. Case 1 was a FCPF with a bone fragment too small to fix, necessitating either joint replacement or FHNO. The indication for THR in the four remaining cases was spontaneous FCPF. While repair of acute/traumatic FCPF using Kirschner wires can result in good functional outcome with short recovery time and at minimal cost (Fischer and others 2004), there are reports of continuing lameness in 7 to 14% of cases. (Jeffery 1989, Culvenor and others 1996). FCPF is associated with sclerosis and osteolysis of the femoral neck (Queen and others 1998, McNicholas and others 2002), complicating repair. THR should be considered to be the primary recommendation for treatment of cases of FCPF with sclerosis and osteolysis where remodelling of the femoral neck is not yet so severe as to preclude appropriate placement of the femoral stem.

In case 5, craniodorsal luxation was revealed radiographically after 15 weeks. Mild lameness and mild resentment to hip manipulation were noted on orthopaedic examination although the client gave a subjectively satisfactory assessment. A similar report of acceptable comfort and function despite long-term craniodorsal luxation of a canine THR case was given by Warnock and others (2003). Luxation after THR in the dog most commonly occurs within six weeks of surgery (Dyce and others 2000), and it is possible that the luxation in case 5 was missed at the four-week re-examination owing to apparently satisfactory comfort owing gait. Radiography is now routinely performed as part of the first and second re-examinations. Inappropriate acetabular cup positioning is the primary cause of postoperative luxation in human THR (Fackler and Poss 1980, Ali Khan and others 1981), and acetabular cup positioning, particularly the ALO, has been shown to be important in canine THR, with the recommended range between 35° and 55° (Dyce and others 2000). At 61·4°, the ALO for case 4 was outside the optimum range, although luxation, commonly associated with an ‘open’ cup (or high ALO), did not occur. In contrast, the ALO for case 5 (which luxated) was 46·3°, close to the ideal of 45°. Revision of case 5 involved open reduction and replacement of the +0 femoral head with a +2 femoral head to increase the soft-tissue tension. Open reduction was successful, with reluxation not occurring during the subsequent four-month follow-up period. Factors such as gluteal muscle atrophy and excessive activity in the postoperative period may have been contributory to luxation.

Without a femoral canal cement restrictor plug, injected PMMA fills the medullary canal inconsistently (Figs 1 and 3). Ota and others (2005) found significantly less femoral implant loosening in dogs when a cement restrictor was used, and consideration should be given to adding a suitably sized cement restrictor to the Micro Hip kit.

Femoral cortical remodelling at the distal end of the femoral stem has been recognised in association with aseptic loosening of femoral prostheses following canine (Bergh and others 2004) and human THR (Oh and Harris 1978, Lewis and others 1984). The femoral cortical thickening seen in three cases reported here may be related to thermal necrosis associated with bone cement (Little and others 2008), reaming and filing of the femur during surgery or stress concentration at the distal stem tip which, in turn, might be due to excessive patient activity, stem instability, stem malpositioning or stem undersizing). Decisions on prosthesis sizes were made preoperatively using acetate templates over radiographs. The micro THR system includes two femoral stem sizes: numbers 2 and 3. Although there are currently no recommendations regarding stem size and cement mantle thickness in cats, review of postoperative radiographs suggests that the femoral stem was too small (Fig 3) in some cases. Follow-up radiography of future feline THR patients with number 3 femoral stems may reveal whether femoral stem undersize may be a factor in femoral cortical remodelling.

Radiographs of case 5 three months after THR (Fig 1) show lucency at the stem-cement interface, consistent with debonding. Stem-cement debonding in the early postoperative period has been implicated in stem instability in human THR (Jasty and others 1991), although debonding may be non-progressive and insignificant in the long-term stability of femoral prostheses (Berry and others 1998). Seven months after THR, radiographs revealed no progression of the lucency at the stem-cement interface (Fig 4). Future annual re-examinations of the cats will reveal whether femoral cortical remodelling and/or stem-cement interface deterioration continue in these cases and whether they affect long-term durability of the prostheses.


Indications for THR in cats include FCPF. Feline THR produced subjectively good short-term results in five cases. Long-term success of the procedure is currently unknown.

Conflict of interest

None of the authors of this article has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper.