Shoulder arthrodesis as a treatment for severe shoulder deformity secondary to septic arthritis in two cats

Abstract

Case series summary

Two rescued juvenile cats with unilateral severe shoulder deformities (SDs) suspected to be secondary to septic arthritis were referred for further management. Comprehensive diagnostic evaluations confirmed SD with suspected septic arthritis in both cases. Because of the poor prognosis for preserving shoulder joint function, shoulder arthrodesis was performed to relieve pain and restore forelimb weightbearing ability. At 6 and 8 years postoperatively, both cats exhibited normal motor function in the affected limbs, no signs of lameness and maintained a good quality of life. No evidence of osteomyelitis or implant failure was noted. These findings suggest that arthrodesis may be a viable and effective long-term treatment option for irreparable shoulder joint disorders in cats.

Relevance and novel information

Shoulder joint injuries and disorders are less common in cats than in dogs, and only a limited number of reports exist in the small animal orthopaedic literature. In clinical practice, SD in small animals is rarely documented, with few references regarding its aetiology and pathophysiology. This report presents two cases of SD in rescued cats, demonstrating that shoulder arthrodesis can be an effective treatment for irreparable feline shoulder joint disorders. To the best of our knowledge, this is the first report to describe long-term clinical and radiographic outcomes of shoulder arthrodesis in cats with SD.

Plain language summary

Surgical fusion of the shoulder joint to treat severe shoulder deformity caused by infection in two cats

Two young rescued cats were referred for treatment of severe deformities of one shoulder, likely caused by infection in the joint at an early age. Because the shoulder joints could not be repaired, a surgical procedure called shoulder arthrodesis (joint fusion) was performed to reduce pain and allow the cats to use their front legs normally. Both cats recovered well and were able to walk and use their affected limbs without lameness. Long-term follow-up at 6 and 8 years after surgery showed stable implants, successful bone fusion, and no signs of infection, with both cats maintaining a good quality of life. Shoulder joint problems are less common in cats than in dogs, and there are very few reports describing this condition or its treatment in cats. This case report shows that shoulder arthrodesis can be an effective long-term treatment option when the shoulder joint cannot be preserved. To our knowledge, this is the first report to describe long-term clinical and imaging outcomes of this procedure in cats with severe shoulder deformity.

Case series description

Case 1

A rescued kitten aged approximately 1–2 weeks was presented to a private animal hospital with suspected subcutaneous abscesses in the dorsal neck and right scapular regions.

No radiographic examination was performed at that time, and joint involvement was not assessed.

Both abscesses were incised, drained and irrigated with saline and neutral electrolysed water. Aspirated material was submitted for culture (Noba Medical Laboratory). Four bacterial species were identified, and ampicillin trihydrate (20 mg/kg q8h for 7 days) was prescribed based on susceptibility testing (Table 1).

Table 1.

Case 1: antimicrobial susceptibility profiles of bacterial isolates cultured from a subcutaneous abscess in the right scapular region

Antibiotic name	Pasteurella multocida	Bacteroides species	Peptostreptococcus species	Porphyromonas species
Ampicillin	S	S	S	S
Piperacillin	S	S	S	S
Clavulanic acid/amoxicillin	S	S	S	S
Cefazolin	S	S	S	S
Cephalexin	S	S	S	S
Cefaclor	S	S	S	S
Cefmetazole	S	S	S	S
Ceftazidime	S	S	S	S
Meropenem	S	S	S	S
Erythromycin	S	N	S	S
Clarithromycin	S	N	S	S
Azithromycin	S	N	S	S
Clindamycin	S	N	S	S
Ofloxacin	S	S	S	S
Levofloxacin	S	S	S	S
Enrofloxacin	S	S	S	S
Moxifloxacin	S	S	S	S
Doxycycline	S	S	S	S
Minocycline	S	S	S	S
Amikacin	S	S	S	S
Gentamicin	S	N	S	N
Vancomycin	S	N	S	N
Chloramphenicol	S	S	S	R
Sulfamethoxazole/trimethoprim	S	R	S	R
Fosfomycin	S	R	S	R

Antimicrobial susceptibility testing was performed using swab samples collected after incision and drainage of the abscess. Testing was conducted by Noba Medical Laboratory (Japan) using the disk diffusion method

N = not tested; R = resistant; S = susceptible

Both abscesses resolved within 1 week; however, the lesion in the right scapular region recurred three times at intervals of approximately 3 weeks.

During the third recurrence, radiographs revealed right shoulder dislocation, deformation and pseudoarthrosis, prompting referral to a university hospital. At the time of referral, no signs of pain were reported by the referring veterinarian. The cat, an intact male aged approximately 4 months and weighing 1.5 kg, was in good condition. Although no overt gait disturbance was noted, intermittent right forelimb lifting was observed. The cat was fed a commercial kitten diet. Orthopaedic examination and radiography revealed painful restriction of right shoulder extension and stressed radiography confirmed limited extension of the joint (Figure 1).

Figure 1.

Case 1: preoperative radiographs: mediolateral views showing dislocation, structural collapse, and scapular glenoid and humeral head deformity. (a) Neutral position; (b) during extension stress

After referral, discontinuation of antibiotics at the referring clinic led to recurrence of the right scapular abscess. Repeat culture using the same method and laboratory confirmed a mixed infection. Ampicillin was continued until surgery (Table 2), and no further recurrence was observed.

Table 2.

Case 1: antimicrobial susceptibility profiles of bacterial isolates cultured from a recurrent subcutaneous abscess in the right scapular region

Antibiotic name	Pasteurella multocida	Peptostreptococcus species	Bacteroides species
Ampicillin	S	S	S
Piperacillin	S	S	S
Clavulanic acid/amoxicillin	S	S	S
Cefazolin	S	S	S
Cephalexin	S	S	S
Cefaclor	S	S	S
Cefmetazole	S	S	S
Ceftazidime	S	S	S
Meropenem	S	S	S
Erythromycin	N	S	S
Clarithromycin	N	S	S
Azithromycin	N	S	S
Clindamycin	S	S	S
Ofloxacin	S	S	S
Levofloxacin	S	S	S
Enrofloxacin	S	S	S
Moxifloxacin	S	S	S
Doxycycline	S	S	S
Minocycline	S	S	S
Amikacin	S	N	S
Gentamicin	S	N	S
Vancomycin	S	N	S
Chloramphenicol	S	S	S
Sulfamethoxazole/trimethoprim	S	R	R
Fosfomycin	S	R	R

Bacterial culture was repeated after recurrence of the subcutaneous abscess after discontinuation of antimicrobial therapy. Antimicrobial susceptibility testing was conducted using swab samples collected after incision and drainage and was performed by the same laboratory (Noba Medical Laboratory, Japan) using the disk diffusion method

N = not tested; R = resistant; S = susceptible

As a result of growth-related changes, shoulder arthrodesis (SA) was performed at approximately 6 months of age (3.55 kg) using a standardised lateral approach to relieve pain and restore function (Figure 2; see Box 1 for details). At the referring hospital, the cat received only rest during the 2 months before surgery, without analgesics.

Figure 2.

Case 1: (a) mediolateral and (b) craniocaudal/anteroposterior postoperative radiographs and (c) postoperative external appearance

Box 1. Shoulder arthrodesis (SA) procedure ¹ .

In both cases, SA was performed using an identical surgical approach. Intraoperative photographs from case 2 are shown (Figures 7–10).

Figure 7.

To minimise heat generation during osteotomy, an electric sagittal saw was used with continuous saline irrigation. A 105–110° intersection was maintained between scapular and humeral axes

Figure 8.

A K-wire was implanted transarticularly, extending from the supracondylar space of the scapula to the proximo-medial area of the proximal humerus in an antegrade fashion

Figure 9.

(a) A K-wire was inserted from the supraspinous fossa of the scapula to the medial surface of the proximomedial area of the proximal humerus. Compression fixation was achieved between the resected surfaces following the tension band wiring technique. (b) The scapulohumeral angle at the fixation was maintained at approximately 110° even after compression fixation

Figure 10.

A 2.0 mm String of Pearls (SOP) plate was placed from the proximal scapula to the proximal third of the humerus. Positioned for screw fixation at the scapular spine–supraglenoid cavity intersection, it was secured using self-tapping 2.0 mm cortical screws (DePuySynthes) for neutralisation fixation

The patient was placed in lateral recumbency under general anaesthesia. The affected limb was clipped, and the surgical field extended from the scapula to the distal forearm and from the lateral neck to the caudal forelimb. After aseptic preparation and draping, a skin incision was made from the dorsal scapula along the acromion to the mid-humerus. The cephalic vein was ligated (4-0 PDS).

The trapezius and omotransversarius muscles were separated and retracted cranially. The acromial part of the deltoid muscle was detached with bipolar cautery and reflected ventrally. The supraspinatus muscle was separated from the scapular spine. The supraspinatus, infraspinatus and teres minor tendons were transected at their humeral insertions.

After the scapular spine was grasped and stabilised with a Kern clamp, the lateral joint capsule was incised. The glenoid cavity and humeral head were exposed, and the biceps brachii tendon was transected at the supraglenoid tubercle. Articular surfaces of the glenoid and humeral head were excised using an electric sagittal saw (Figure 7) and submitted for bacterial culture.

After aligning the resected surfaces, a K-wire was placed (Figure 8).

For internal fixation, a figure-of-eight 22 G orthopaedic wire was applied as tension band wiring (Figure 9a,b), and 1–2 transarticular K-wires were added to prevent rotation. A 2.0 mm String of Pearls (SOP) plate (Orthomed) was contoured to the scapula and proximal humerus and secured with self-tapping cortical screws (Figure 10).

The deltoid muscle was reattached to the supraspinatus and infraspinatus muscle using 3-0 PDS. Closure was performed in layers, and the skin was closed with 4-0 polypropylene in an interrupted pattern. Postoperatively, both cats received analgesics and antibiotics. Monitoring included checks for implant displacement, infection and lameness. Culture results from surgical specimens were negative. This technique achieved stable fixation, proper alignment and favourable long-term function.

Preoperative blood work, including haematology and biochemistry, showed all values within reference intervals (RIs).

Anaesthesia was induced with midazolam (0.3 mg/kg IV) and propofol, and maintained with isoflurane in oxygen with positive-pressure ventilation. Intraoperative fentanyl infusion (3–10 µg/kg/h) was continued for 24 h postoperatively, followed by a transdermal fentanyl patch until day 6. No additional analgesics were given, and daily behavioural assessment showed no signs of discomfort.

Antibiotics included intraoperative ampicillin (25 mg/kg IV every 90 mins, three doses) and a single dose of gentamicin (5 mg/kg IV), followed by ampicillin (25 mg/kg IV q12h for 12 days) and gentamicin (6 mg/kg IV q24h for 7 days). Gentamicin was discontinued after negative intraoperative cultures.

Although occult infection could not be excluded, clinical stability and lack of radiographic or pain signs suggested no active infection at arthrodesis.

Clinical parameters were monitored daily, but renal function was not evaluated. Intraoperative fluids consisted of lactated Ringer’s (3 ml/kg/h). Postoperatively, intravenous (IV) fluids were discontinued owing to catheter issues and replaced with subcutaneous fluids (250 ml/day).

The cat was discharged after suture removal on day 12 with unrestricted indoor activity. Follow-ups at 3 weeks, 1.5, 3 and 6 months showed no lameness or pain, and radiographs confirmed implant stability and bone fusion by 3 months. Subsequent follow-up every 6 months for 3 years and then annually showed intact function without signs of osteomyelitis or implant failure at 8 years postoperatively (5.0 kg) (Figure 3).

Figure 3.

Case 1: (a) mediolateral and (b) craniocaudal radiographs and (c) external appearance at 89 months postoperatively

Case 2

A rescued kitten aged approximately 1–2 weeks was initially presented to a private hospital with suspected subcutaneous abscesses in the right scapular and left hip regions. Swab cultures identified group G Streptococcus species (Table 3). Both abscesses resolved within 1 week after incision and irrigation without antibiotics and did not recur.

Table 3.

Case 2: antimicrobial susceptibility profile of group G Streptococcus species cultured from a subcutaneous abscess

Antibiotic name	Group G Streptococcus species
Ampicillin	S
Clavulanic acid/amoxicillin	S
Cefazolin	S
Cefalexin	S
Cefdinir	S
Cefpodoxime	S
Cefmetazole	S
Azithromycin	S
Clindamycin	S
Ofloxacin	S
Levofloxacin	S
Enrofloxacin	S
Doxycycline	R
Minocycline	R
Gentamicin	N
Chloramphenicol	S
Fosfomycin	S

Antimicrobial susceptibility testing was performed using a swab sample obtained during incision and drainage of the abscess and processed by the Institute of Health Sciences (Japan) using the disk diffusion method

N = not tested; R = resistant; S = susceptible

Radiographs were not taken at the initial presentation, so joint involvement and concurrent septic arthritis could not be confirmed or excluded. Although the clinical course suggested a localised soft tissue abscess, early joint involvement cannot be ruled out retrospectively.

By approximately 3 months of age, the kitten developed hindlimb gait abnormalities and radiographs showed osteolysis of the left femoral head. At approximately 5 months of age, similar lesions were noted in the right humeral head, prompting referral to a university hospital.

At referral, the cat, an intact male aged approximately 5 months and weighing 2.5 kg, was in good condition and eating a commercial kitten diet. Orthopaedic examination revealed pain and crepitus in the left hip and restricted, painful right shoulder movement. Radiographs showed severe deformities of the scapular glenoid and humeral head with no swelling, making joint preservation unfeasible (Figure 4).

Figure 4.

Case 2: preoperative radiographs: (a) mediolateral and (b) craniocaudal views of the shoulder; (c) ventrodorsal view of the hip joint in extension

At approximately 6 months of age, 2 weeks after referral to the university hospital, a CT scan was performed to further evaluate the joint abnormalities.

SA was performed the following day using a standardised lateral approach to relieve pain and restore function (Figure 5; see Box 1 for details).

Figure 5.

Case 2: (a) mediolateral and (b) craniocaudal postoperative radiographs and (c) postoperative external appearance

There were no records indicating that analgesics were used at the referring hospital during the 2 weeks before surgery.

Preoperative haematology and serum biochemistry were within RIs. Anaesthesia was induced with IV propofol and maintained with isoflurane in oxygen. Intraoperative analgesia included a paravertebral block with 0.25% bupivacaine, remifentanil infusion (20–40 µg/kg/h) and fentanyl boluses (2–3 µg/kg). Postoperative fentanyl (2 µg/kg/h) was administered for 24 h, followed by a transdermal patch (2.5 µg/kg/h) until day 6. No additional analgesics were given and pain was monitored behaviourally, as in case 1.

Cefmetazole (25 mg/kg IV) was administered intraoperatively (two doses 90 mins apart) and then q12h for 9 days. Gentamicin (6 mg/kg IV q24h) was given for 5 days and discontinued after negative intraoperative cultures.

Clinical parameters were monitored daily, but renal function was not evaluated.

No clinical signs of pain or lameness were observed, and a preoperative CT scan showed no evidence of osteomyelitis or active infection in the hip joints; therefore, left hip arthrocentesis was not performed. Although persistent infection could not be excluded, no active infection was considered present at arthrodesis.

Sutures were removed on day 15, and unrestricted indoor activity resumed. Lameness resolved within 1 month, with improved weightbearing at 1, 3 and 6 months. Radiographs confirmed implant stability and bone fusion by 3 months. Follow-up every 6 months for 1 year and annually thereafter showed intact motor function, with no osteomyelitis or implant failure at 6 years (4.25 kg) (Figure 6). As no hip-related clinical signs were observed, follow-up radiographs were not performed.

Figure 6.

Case 2: (a) mediolateral and (b) craniocaudal radiographs and (c) external appearance 72 months postoperatively

Discussion

Shoulder deformity (SD) is rarely reported in small animal veterinary medicine. ² Compared with dogs,^3–5 feline shoulder injuries are uncommon and published cases are limited. ⁶

In human medicine, SD is often linked to incomplete ossification of the scapular neck and glenoid cavity, typically unilateral.^7,8 In our two juvenile rescue cats, marked deformities of the scapular glenoid and humeral head were observed, consistent with SD.

In humans, SD can result from obstetric brachial plexus palsy or growth disturbances due to epiphyseal osteomyelitis or septic arthritis in infancy.^9–13 Although the exact cause in our cases remains unclear, four bacterial species, including Pasteurella multocida in case 1 and group G Streptococcus in case 2, were isolated, suggesting infection as a possible contributor.^14–16 Trauma or bite wounds may have served as entry points, and given the young age of the kittens, haematogenous spread from an umbilical source should also be considered. Although synovial fluid analysis and histopathology were not performed, clinical and imaging findings supported infection-related joint destruction.

No analgesics were prescribed at the referring clinic, and a more proactive approach after referral might have improved pain management. Given the severity of the radiographic findings, intermittent right forelimb lifting in case 1 likely indicated chronic pain; therefore, pre-emptive analgesia should have been initiated despite the absence of overt clinical signs.

Given the invasiveness of the procedure, postoperative pain management based solely on behavioural observations at that time may not have been sufficient, and according to current guidelines, a longer duration of analgesic treatment would have been preferable.^17,18

Excision arthroplasty would have been a logical approach for these two cats given the history of septic arthritis in both cases. However, in addition to poor function reported after shoulder excision in clinical cases, there is also a very low risk that the instability created after this procedure could result in further bacterial proliferation. The risks of performing SA in the face of persistent joint infection would be the development of a surgical site infection and formation of a glycocalyx or biofilm with the necessity for implant removal in the future.

To better guide surgical planning, synovial fluid should have been collected under anaesthesia immediately before surgery in both cases to confirm infection status, including of the hip joint in case 2.

In retrospect, gentamicin may not have been the most appropriate choice of antimicrobial, as sensitivity testing did not support its use against all isolated organisms and it poses a known nephrotoxicity risk in young cats. ¹⁹ Given the availability of safer, effective alternatives based on susceptibility results, different agents should have been considered.

Cefmetazole, used in case 2, is a second-generation cephalosporin; however, other options may have better aligned with current guidelines. ¹⁹

These points emphasise the need to align antimicrobial therapy with current clinical standards and culture/sensitivity data.^20,21 In addition, despite the known nephrotoxicity associated with gentamicin, renal monitoring was not performed in these cases; this should be considered in future case management. ²²

After evaluating the risks and benefits, SA was selected in both cases to provide stability and function as a salvage option when joint preservation was not feasible. Follow-up over 72–89 months confirmed bone fusion, implant stability and sustained forelimb function without infection recurrence, suggesting the infection was inactive at surgery.

In case 2, despite the absence of clinical signs and the lack of CT findings suggestive of osteomyelitis or active infection, prereferral radiographs had demonstrated osteolysis of the femoral heads. Therefore, further radiographic follow-up or additional diagnostic evaluation of the hips would have been warranted to ensure comprehensive assessment, and might have allowed consideration of additional therapeutic interventions, potentially improving the cat’s welfare.

String of Pearl (SOP) plates provide precise alignment through multidirectional contouring and mechanical strength comparable to or exceeding locking compression plates. ²³ Our findings align with previous canine studies and support their use in cats. Although SA for feline SD has been reported, long-term outcomes have not been described; this study is the first to report long-term clinical and radiographic results. In cats, the scapulothoracic joint provides high mobility and compensates for shoulder fixation.^24–26

Fixation was set at 105–110°, approximating the natural standing angle in cats, with a postoperative scapulohumeral angle of approximately 25° likely supporting joint biomechanics. Although the optimal angle remains undetermined, a 25–35° range appears appropriate; however, further research is needed to optimise surgical parameters for feline function.

Conclusions

SA was performed in two juvenile cats with SD secondary to septic arthritis. At 6 and 8 years postoperatively, both cats maintained intact limb function without implant failure, supporting arthrodesis as an effective treatment for irreparable shoulder joint disorders in cats.