Evaluation of a Joint Distractor to Facilitate Arthroscopy of the

Canine Stifle

TOBY J. GEMMILL, BVSc MVM DSAS(Orth) Diplomate ECVS and MICHAEL FARRELL, BVetMed CertVA CertSAS Diplomate ECVS

Objectives—To assess a novel technique for arthroscopic evaluation of the canine stifle avoiding theneed for fat pad debridement.Study Design—Prospective study.Sample Population—Cadaveric canine stifles (n¼ 10); Client-owned dogs (15).Methods—In cadaveric stifles, the femoropatellar space was assessed before placement of a jointdistractor and examination of the femorotibial joint. Time to complete evaluation of the femoro-patellar space, to observe the cruciate ligaments, and to completion of examination of the femoro-tibial joint were recorded. Distractor-assisted arthroscopy was then performed in 15 consecutiveclinical cases. Clinical information, time to complete joint evaluation, and ability to perform anyadditional arthroscopic surgery were recorded.Results—Observation and palpation of intra-articular structures were possible in all cadavericstifles. Mean time to evaluate the femoropatellar space was 3.2 minutes (range, 2–4 minutes); toobserve the cruciate ligaments, 5.8 minutes (range, 3–8 minutes); and to completion of joint in-spection, 9.1 minutes (range 6–13 minutes). In the clinical cases, observation of intra-articularstructures was also possible in all stifles. Mean time to complete joint inspection was 21 minutes(range, 10–40 minutes). Partial meniscectomy was performed successfully in 5 stifles.Conclusions—Use of a joint distractor allowed arthroscopic observation of all relevant structures.Partial meniscectomy was readily performed.Clinical Relevance—Use of a joint distractor may simplify arthroscopic assessment of the caninestifle and avoid potential morbidity associated with fat pad debridement.r Copyright 2009 by The American College of Veterinary Surgeons

INTRODUCTION

ARTHROSCOPICALLY ASSISTED surgery of thecanine stifle joint is increasingly popular in small

animal orthopedic practice and has been used to treatcranial cruciate ligament deficiency,1 meniscal tears,2,3

and medial patellar luxation.4 The standard technique forarthroscopic evaluation of the stifle joint described byWhitney5 uses a craniolateral subpatellar arthroscopeportal and a craniomedial instrument portal. Observationof structures within the femorotibial space is achievedafter debridement of the infrapatellar fat pad using either

a motorized shaver5 or a radiofrequency probe3; how-ever, viewing after fat pad debridement can be challeng-ing, and conversion to arthrotomy can be necessary insome joints. Fat pad debridement can result in substan-tial hemorrhage5 that can further impair visibility. Re-sidual hematoma within a joint after surgery can lead tosubstantial inflammation and cartilage damage.6 Thelonger-term effects of fat pad debridement are unclear.Although more extensive than typically performed duringarthroscopy, fat pad ablation in humans undergoing totalknee replacement has been associated with increasedpostoperative anterior knee pain.7 In a rabbit model,8

Presented in part at the ESVOT Congress in Munich, Germany, September 2006.

Address reprint requests to Toby J. Gemmill, BVSc MVM DSAS (orth) Diplomate ECVS Willows Referral Service, 78 Tanworth

Lane, Solihull B90 4DF, UK. E-mail: toby.gemmill@btinternet.com.

Submitted July 2008; Accepted February 2009

From Willows Referral Service, Solihull UK and Division of Companion Animal Sciences, Institute of Comparative Medicine,

University of Glasgow Veterinary School, Glasgow, UK

r Copyright 2009 by The American College of Veterinary Surgeons

0161-3499/09

doi:10.1111/j.1532-950X.2009.00516.x

588

Veterinary Surgery

38:588–594, 2009

cranial stifle fibrosis after fat pad ablation led to a re-striction in patellar tendon growth and cartilage degen-eration. It is known that the fat pad is intricately involvedin the vascularity of the patellar tendon and cranialcruciate ligament.9 Further, the rich innervation of the fatpad suggests that it may be involved in proprioception.Fat pad fibrosis has been reported after debridement inboth humans10 and dogs11 and it has been suggested thatthis may be associated with accelerated cartilage degen-eration, especially of the femoropatellar joint.

Use of a more proximal lateral subpatellar portal hasbeen proposed to improve observation within the caninestifle joint.12,13 Rather than locating the portal immedi-ately proximal to the tubercle of Gurdy as described byWhitney,5 it is established lateral to the patellar tendonapproximately half way between the distal pole of thepatellar and the tibial tuberosity. This allows insertion ofthe arthroscope proximal to the fat pad, which reducesthe extent of fat pad debridement required to view thefemorotibial space. Observation of structures withinthe femorotibial space can also be aided by levering theproximal tibia cranially using a small Hohmann retractorinserted through a second craniomedial subpatellar por-tal12 although this can be challenging, especially for theunassisted surgeon. Furthermore, use of a Hohmann re-tractor can distort the caudal horns of the menisci, whichmay complicate assessment of meniscal integrity.14

A joint distractor (Veterinary Instrumentation, Shef-field, UK) has been developed for improving exposure ofthe femorotibial space during stifle arthrotomy (Fig 1A).With the stifle flexed, the jaws of the distractor are placed

between the intercondylar notch and the fat pad or thecranial intermeniscal ligament. Opening the jaws thenopens the femorotibial space. Because the distractor isdesigned with arms of slightly different lengths the tipscan be crossed over (Fig 1B) which facilitates insertioninto the joint. The instrument is self retaining and can bemaintained in a distracted position using a speed lock.We hypothesized it may be possible to use this instrumentto facilitate timely arthroscopic evaluation of the caninestifle joint, allowing observation of all relevant intra-ar-ticular structures without the need for fat pad debride-ment. To test this, we conducted a pilot study in 10cadaveric canine stifle joints and then used this techniquein 15 dogs for diagnostic or therapeutic stifle arthroscopy.

MATERIALS AND METHODS

Cadaver Study

Stifles (n¼ 10) from 6 canine cadavers were collected fromdogs euthanatized for reasons unrelated to this study. Pelviclimbs were clipped and cadavers stored at �201C until study.Cadavers were thawed for 24 hours at 181C, weighed, and aclinical examination of the stifles performed to excludepre-existing disease. Each cadaver was positioned in dorsalrecumbency with the pelvic limb to be evaluated extendedcaudally and secured in a brace as described by Schulz et al.15

A lateral subpatellar stab incision was made 1–3mm lateral tothe patellar tendon at the mid point between the distal pole ofthe patellar and the tibial tuberosity. A medial suprapatellaregress portal was established as described by Whitney.5 Fluidflow was maintained with a pressurized fluid pump (A106;

Fig 1. (A) Stifle Joint Distractor. The arms are of slightly different lengths. This allows the tips to be crossed over (B) to facilitate

placement into the joint.

589GEMMILL AND FARRELL

World of Medicine, Berlin, Germany) set at 50mm Hg. Flowwas fine controlled manually using the tap on the arthroscopecannula. A 2.7mm 301 fore-oblique arthroscope (Olympus;MGB, Berlin, Germany) was inserted through the lateral sub-patellar portal for joint evaluation.

Initially, the supratrochlear region of the joint, the medialand lateral recesses, and the femoropatellar joint were in-spected. A joint distractor was then positioned via a medialsubpatellar stab incision located 1–3mmmedial to the patellartendon at the mid point between the distal pole of the patellarand the tibial tuberosity (Fig 2). The proximal jaw was placedinto the intercondylar notch, and the distal jaw into the fatpad. The femorotibial space was then examined. The cranialand caudal cruciate ligaments, tibial eminence, and medial andlateral menisci were observed and palpated with a blunt probeplaced between the arms of the distractor. The ability to ob-serve and palpate intra-articular structures was recorded. Instifles where evaluation of all structures was not possible viathe lateral subpatellar portal, the arthroscope was withdrawnfrom the joint and re-introduced through the medial subpa-tellar portal to allow completion of the examination. The timeto perform evaluation of the femoropatellar space, time toobservation of the cranial cruciate ligament—which wouldcorrespond to the time taken to complete fat pad debridementusing the technique described by Whitney5—and time tocompletion of examination of each joint was recorded. After

completion of the procedure, stifles were dissected to excludegross pre-existing disease and assess trauma caused by theprocedure.

Clinical Study

After completion of the cadaveric study, the distractor wasused in 15 consecutive dogs (16 stifles) that had diagnostic ortherapeutic stifle arthroscopy. Briefly, dogs were positioned indorsal recumbency with the pelvic limb to be examined ex-tended caudally and secured in a brace. Positioning was fa-cilitated by use of a tilting surgical table. Each stifle wasexamined arthroscopically as described in the cadaveric studyand additionally, an attempt was made to observe structuresin the femorotibial space before distractor insertion. After ex-amination, therapeutic procedures were undertaken as deter-mined by the surgeon. Clinical information, time to completediagnostic evaluation of the joint, and the time to completeany therapeutic procedures were recorded in each case.

RESULTS

Cadaver Study

Mean bodyweight of the cadavers was 22.2kg (range,19–28 kg; Table 1). There were 4 left and 6 right stiflejoints. Observation and palpation of intra-articular struc-tures was possible in all stifles; however, some difficultywas encountered completely observing the lateral menis-cus in 2 joints that necessitated insertion of the arthro-scope through the medial subpatellar portal to completeassessment. Mean time to complete evaluation of the fe-moropatellar space was 3.2 minutes (range, 2–4 minutes),until observation of the cranial cruciate ligament, 5.8minutes (range, 3–8 minutes), and to completion of theexamination, 9.1 minutes (range, 6–13 minutes).

Fig 2. Instrumentation of cadaver stifle demonstrating portal

positions. The egress cannula (E) is placed through a medial

suprapatellar portal, the arthroscope (A) through a lateral

subpatellar portal, and the distractor (D) through a medial

subpatellar portal. The probe is easily introduced between the

arms of the distractor. If desired, the arthroscope can be placed

through the arms of the distractor via the medial subpatellar

portal.

Table 1. Population Data and Results of Arthroscopy in Cadaveric

Joints

Specimen

Cadaver

Weight (kg)

Time from Start of Arthroscopy (Minutes)

Examination of

Supratrochlear

Space

Observation

of Cranial

Cruciate

Ligament

Completion of

Arthroscopy

1 22 4 8 10

2 19 3 5 9

3 19 4 6 7

4 21 3 7 9

5 21 3 8 12

6 21 3 5 13

7 21 2 3 6

8 22 4 6 8

9 28 3 5 8

10 28 3 5 9

Mean time (minutes) 3.2 5.8 9.1

590 DISTRACTOR-ASSISTED CANINE STIFLE ARTHROSCOPY

In 8 stifles, no gross evidence of degenerative jointdisease was seen after dissection. Very mild chondrom-alacia of the proximal trochlear groove that had not beenobviously apparent on arthroscopic examination wasnoted in 2 joints. Examination after dissection did notreveal any evidence of osteophyte formation or synovialhyperplasia. Superficial cartilage abrasion in the distalaspect of the trochlear groove, attributed to instrumen-tation, was noted in 6 stifles. In 3 stifles, this traumaconsisted of punctate lesionso1mm in diameter whereasin 3 stifles, the lesions consisted of linear partial thicknessabrasions ranging from 3 to 4mm in length.

Clinical Study

Mean age of dogs was 57 months (range, 8–100months), and mean bodyweight was 33kg (range, 16–53kg; Table 2). Complete observation of structures in thefemorotibial space before use of the distractor was chal-lenging. After placement of the distractor observation ofrelevant structures was straightforward in all stifles (Figs3–5), although use of both lateral and medial scope por-tals was necessary in 2 joints. Mean time to complete

diagnostic evaluation of the joint was 20.7 minutes(range, 10–40 minutes). After a diagnosis of a meniscaltear partial meniscectomy was successfully performed in 5stifles (Table 3). Mean additional time to perform partialmeniscectomy was 19.8 minutes (range, 12–28 minutes).The first dog in the series (stifles 1, 2) was diagnosed withbilateral osteochondritis dissecans. After arthroscopicevaluation of the stifles, it was felt that with our level ofexperience at that stage timely removal of the osteochon-dral flaps was best achieved via limited subpatellar ar-throtomies.

Manipulation of the stifles after arthroscopy did notreveal any evidence of ligamentous damage as a result ofthe procedure. Minor postoperative seromas around themedial subpatellar portal occurred in 2 stifles; these re-solved without further treatment. No other complicationswere encountered.

DISCUSSION

Given the technical difficulty, hemorrhage, and po-tential morbidity that can be associated with arthroscopicexamination of the canine stifle joint after fat pad debri-

Table 2. Clinical Study Arthroscopy Results

Dog Breed

Age

(months)

Weight

(kg)

Affected

Limb Diagnosis

All Structures

Observed

Arthroscope

Portals

Joint Evaluation

(minutes) Complications

1 Labrador

Retriever

8 28 Left Stifle OCD Yes Lateral 22 None

2 Labrador

Retriever

8 28 Right Stifle OCD Yes Lateral 25 None

3 Boxer 35 27 Left Primary gonarthritis Yes Lateral and

medial

26 None

4 Collie cross 86 16 Left Partial cranial cruciate

ligament deficiency

Yes Lateral 16 None

5 Boxer 67 36 Left Septic arthritis Yes Lateral 17 None

6 Boxer 23 27 Right Primary gonarthritis Yes Lateral and

medial

24 Small self-resolving

seroma

7 Labrador

Retriever

84 27 Left Isolated medial meniscal tear Yes Lateral 17 None

8 German

Shepherd dog

96 44 Left Partial cranial cruciate

ligament deficiency

Yes Lateral 22 None

9 Golden

Retriever

36 28 Right Partial cranial cruciate

ligament deficiency

Yes Lateral 21 None

10 Boxer 84 33 Left Late meniscal injury Yes Lateral 22 Small self-resolving

seroma

11 Labrador

retriever

36 33 Right Late meniscal injury Yes Lateral 15 None

12 Labrador

retriever

88 28 Left Late meniscal injury Yes Lateral 15 None

13 Dogue de

Bordeaux

18 53 Left Partial cranial cruciate

ligament deficiency

Yes Lateral 22 None

14 Labrador

Retriever

96 33 Left Partial cranial cruciate

ligament deficiency

Yes Lateral 17 None

15 Labrador

Retriever

100 31 Left Late meniscal injury Yes Lateral 10 None

16 Labrador

Retriever

48 31 Right Partial cranial cruciate

ligament deficiency

Yes Lateral 40 None

591GEMMILL AND FARRELL

dement, techniques that aim to simplify the procedureand reduce the need for fat pad debridement are of in-terest to the surgeon. We have shown that arthroscopicexamination of the stifle and basic intra-articular maneu-vers are possible using a distractor-assisted technique.

We positioned dogs in dorsal recumbancy with theindex limb extended caudally as described by Whitney.5

The limb was secured in a brace15 that prevents undesir-able limb movement during arthroscopy and, if desired,facilitates stressing of the joint in different directionswithout the requirement for a surgical assistant. Whenusing the joint distractor in cadaveric stifles, furtherstressing of the joint into varus or valgus positions wasunnecessary; however, clinically additional stressing ofthe stifles was helpful to facilitate observation of struc-tures in the femorotibial space and in performing partialmeniscectomy.

Instrumentation of the joint was straightforward usingthe distractor technique. A medial suprapatellar egressportal was used as described by Whitney.5 It has beensuggested that a lateral portal may also be used.13 Thislateral suprapatellar portal could easily be used in con-junction with the distractor technique if desired. Whenplacing instrumentation it became apparent during eval-uation of cadaveric stifles that insertion of the proximalarm of the distractor well into the intercondylar notchwas important to prevent inadvertent displacement of theinstrument after distraction. Displacement of the dis-tractor after application was not encountered in any jointin the clinical study. Placement of the distal arm of thedistractor is less critical as this can engage either the fatpad or the cranial intermeniscal ligament.

Use of a joint distractor allowed complete observationand palpation of all relevant structures. Ease of palpationin the cadaver series and the partial meniscectomies per-formed in the clinical series suggest that arthroscopicsurgical procedures on structures in the femorotibialspace can be performed without undue difficulty. Fat paddebridement was not required in any stifle. Minor hem-orrhage from stab incisions was encountered in some sti-fles in the clinical study but this did not interfere withobservation and resolved without the need for specifichemostatic procedures. This suggests that when using thedistractor, fat pad debridement is unnecessary, leading to

Fig 3. Intraoperative view of the cruciate ligaments (Dog 3).

A hook probe is being used to palpate the cranial cruciate

ligament.

Fig 4. Intraoperative view of the lateral meniscus (Dog 3). F,

femur; T, tibia; M, meniscus.

Fig 5. Intraoperative view of the medial meniscus (Dog 3). F,

femur; T, tibia; M, meniscus.

592 DISTRACTOR-ASSISTED CANINE STIFLE ARTHROSCOPY

simplified surgery and potentially decreased morbidity.The distractor technique was rapid and simple to per-form. In addition, the distractor is self retaining and mi-nor manipulations of the joint to aid evaluation of thefemorotibial space and perform partial meniscectomy canbe easily performed by a single surgeon. Therefore, asurgical assistant is unnecessary and technical problemsarising from assistant fatigue are avoided.

In 2 cadaveric stifles and 2 stifles in the clinical series,complete examination of the lateral femorotibial spacewas not possible using the lateral portal alone. However,the space was easily investigated after insertion of thearthroscope through the medial portal between the jawsof the distractor. Instrumentation to probe intra-articularstructures or to perform partial meniscectomy was easilyintroduced through either portal.

In the cadaveric stifles dissection of the joints at thecompletion of arthroscopy allowed assessment for ia-trogenic trauma. Although abrasions were noted in 6 of10 joints, these were considered minor and of no greaterseverity than might be observed after arthrocentesis.Histological assessment of the traumatized regions wasnot performed, but they were considered unlikely to beassociated with clinical consequences. Abrasions were notapparent on arthroscopic evaluation of the trochleargroove before positioning the distractor. Therefore, it islikely that they were caused by improper placement of theproximal distractor arm or disengagement of the dis-tractor arm from the intercondylar notch. These abra-sions can be avoided by cautious placement of theproximal distractor arm well into the intercondylarnotch, as was performed in the clinical study. Indeed,re-evaluation of the trochlear groove just before jointclosure in the clinical cases failed to identify any evidenceof cartilage trauma.

Although our findings are encouraging, some limita-tions must be acknowledged. The number of stifles as-sessed was relatively small, and it is possible some intra-articular maneuvers may be challenging without fat paddebridement. Indeed, removal of OCD fragments fromthe first 2 stifles in the clinical study was performed viamini-arthrotomies; however, these were the first 2 stifles

in which the distractor technique had been attempted andit is possible that with further experience conversion toarthrotomy may have been unnecessary. Another studylimitation is that direct comparison with the traditionalfat pad debridement technique was not performed and assuch it is not possible to fully advocate 1 technique overthe other. Although we have demonstrated that stifle ar-throscopy is possible using the distractor technique, com-parative intraoperative and longer-term follow-up studieswould be useful.

Use of a joint distractor allowed assessment and pal-pation of all relevant intra-articular structures withoutthe need for fat pad debridement. Small cartilage abra-sions caused by instrumentation of the stifles were con-sidered unlikely to be of clinical consequence and can beavoided by correct placement of the proximal distractorarm. Continued use of this technique in clinical patientscan be recommended.

REFERENCES

1. Hulse DA, Beale BS: Arthroscopically assisted ‘under and

over’ reconstruction of the cranial cruciate ligament in the

dog. Proceedings of the 27th Annual Conference of the Vet-

erinary Orthopaedic Society, March 2000, Val d’Isere,

France, 22pp

2. Ralphs SC, Whitney WO: Arthroscopic evaluation of menisci

in dogs with cranial cruciate ligament injuries: 100 cases. J

Am Vet Med Assoc 221:1601–1604, 2002

3. Ridge PA: Isolated medial meniscal tear in a Border Collie.

Vet Comp Orthop Traumatol 19:110–112, 2006

4. Bevan JM, Taylor RA: Arthroscopic release of the medial

femoropatellar ligament for canine medial patellar luxat-

ion. J Am An Hosp Assoc 40:321–330, 2004

5. Whitney WO: Arthroscopically assisted surgery of the stifle

joint, in Beale BS, Hulse DA, Schulz KS, Whitney WO

(eds): Small Animal Arthroscopy. Philadelphia, PA, Saun-

ders, 2003, pp 117–157

6. Roosendaal G, Tekoppele JM, Vianen ME, et al: Blood in-

duced joint damage: a canine in vivo study. Arth Rheum-

atol 42:1033–1039, 1999

7. Tanaka N, Sakahashi H, Sato E, et al: Influence of the in-

frapatellar fat pad resection in a synovectomy during total

Table 3. Details of Partial Meniscectomies Performed

Dog Diagnosis Treatment Arthroscope Portals

Additional Time for

Meniscectomy (minutes)

7 Isolated medial meniscal tear Partial medial meniscectomy Lateral and medial 14

10 Late meniscal injury Partial medial meniscectomy Lateral and medial 28

11 Late meniscal injury Partial medial and lateral meniscectomy Lateral and medial 25

12 Late meniscal injury Partial medial meniscectomy Lateral 12

15 Late meniscal injury Partial medial meniscectomy Lateral 20

593GEMMILL AND FARRELL

knee arthroplasty in patients with rheumatoid arthritis. J

Arthroplasty 18:897–902, 2003

8. Takatoku K, Sekiya H, Hayashi M, et al: Influence of fat pad

removal on patellar tendon length during growth. Knee

Surg Sports Traumatol Arthroscopy 13:706–713, 2005

9. Arnoczky SP, Rubin RM, Marshall JL: Microvasculature of

the cruciate ligaments and its response to injury. An ex-

perimental study in dogs. J Bone Joint Surg (Am) 61:1221–

1229, 1979

10. Paulos LE, Wnorowski DC, Greenwald AE: Infrapatellar

contracture syndrome. Diagnosis, treatment and long-term

follow up. Am J Sports Med 22:440–449, 1994

11. Hulse DA: Second look following TPLO. Proceedings of the

14th European Society of Veterinary Orthopaedics and

Traumatology Congress, Munich, Germany. September

10–14, 2008. pp 99–100

12. Mahn MM, Cook JL, Cook CR, et al: Arthroscopic verifi-

cation of ultrasonographic diagnosis of meniscal pathology

in dogs. Vet Surg 34:318–323, 2005

13. Innes J: Advances in canine stifle arthroscopy. Arthroscopy

Working Group Seminar, September 2006, Munich, Ger-

many

14. Moses PA: A technique for the surgical repair of caudal pe-

ripheral detachment and longitudinal peripheral tears of

the medial meniscus in dogs. Vet Comp Orthop Traumatol

15:92–96, 2002

15. Schulz KS, Holsworth IG, Hornoff WJ: Self retaining braces

for canine arthroscopy. Vet Surg 33:77–82, 2004

594 DISTRACTOR-ASSISTED CANINE STIFLE ARTHROSCOPY