Unicompartmental arthroplasty of the knee with the cemented MOD3TM prosthesis

Unfallchirurgie �9 Urban & Vogel 1999

Originalarbeit

Unicompartmental Arthroplasty of the Knee with the Cemented MOD3 TM Prosthesis A Prospective Study

Ilan Cohen 1, Zeev Feldbrin 2, David HendeP, Alexander Blankstein 1, Aharon Chechick 1, Valentin Rzetelny 2

Abstract: A consecutive prospective series of 19 knees had unicompartmental arthroplasty between 1991 and 1992 for gonarthrosis and AVN. The MOD3 TM prosthesis was used in all cases. A total follow-up evaluation was undertaken after 6 to 7.5 years (mean 6.5 years) in 18 patients (1 patient has died during the observation period). Complications included 1 technical failure, 2 cases of thromboembolic phenomena and 1 case of a late deep in- fection. Knee score averaged 84.3 (pre-operative 54.7) and functional score averaged 68.9 (we-operative 34.6). There were no cases with aseptic loosening of the prosthesis. It is concluded that with the appropriate indications, the unicompartmental arthroplasty of the knee is an effec- tive and predictable procedure to achieve pain reduction and improved quality of life in patients with monocom- partmental arthritis. Furthermore, the cemented MOD3 TM prosthesis yields results that are comparable to other unicompartmental prostheses in current use.

Key Words: Unicondylar arthroplasty �9 UnicompartmentaI arthropIasty - Hemi-arthroplasty of the knee �9 Mono- compartmental gonarthrosis

Unikompartimenteller Kniegelenkersatz mit zementierter MOD3TM-Prothese

Zusammenfassung: In einer prospektiven Serie in den Jahren 1991 und 1992 wurden 19 unikompartimentelle Kniegelenkprothesen bei Gonarthrose und avaskul~irer Knochennekrose durchgefiihrt. In allen Ffillen wurde die MOD3TM-Prothese eingesetzt. Die Nachuntersuchung bei 18 Patienten (ein Patient verstarb bis zur Nachun- tersuchung) erfolgte nach durchschnittlich 6,5 Jahren (sechs bis 7,5 Jahre). Als Komplikationen wurden ein tech- nischer Fehler, zwei Thromboembolien und eine sprite tiefe Infektion beobachtet. Der Knie-Score hatte im Mit- tel 84,3 Punkte (prrioperativ 54,7 Punkte), und der Funktionsscore betrug im Mittel 68,9 Punkte (prrioperativ 34,6 Pnnkte). Aseptische Prothesenlockerungen wurden nicht beobachtet. Die Untersuchungsergebnisse legen nahe, dab bei regelrechter Indikation der unikompartimentelle Kniegelenk- ersatz ein effektives und vorhersehbares Verfahren zur Schmerzlinderung und Verbesserung der Lebensqualitfit darstellt. Dariiber hinaus sind die Ergebnisse mit der zementierten MOD3TM-Prothese vergleichbar mit anderen derzeit verwendeten unikompartimentellen Prothesen.

Schliisselw6rter: Unikondyl~irer Gelenkersatz . Unikompartimenteller Gelenkersatz �9 Monokompartimentelle Arthrose - Gonarthrose �9 Endoprothese

~Orthopedic Division, Sheba Medical Center, Tel-Hashomer, Israel, 2Orthopedic Department, Wolfson Medical Center, Holon, Israel, 3Orthopedic Department, Hasharon Medical Center, Petach-Tikva, Israel.

Accepted: 29 Nov 1999.

Unfallchirurgie 1999;25:287-93 (Nr. 6) 287

Cohen I, et al. M O D T M Unicondylar Herni-Arthroplasty

S urgical intervention in the early stages of osteo- arthritis of the knee, when the disease process is still

confined to 1 compartment only (medial or lateral), consists of i of 2 possible procedures: If the gonarthros- is is associated with a significant angulatory joint deform- ity (usually of the varus type), a realignment proced- ure such as high tibial osteotomy may be considered in an attempt to alter the biomechanics of the knee joint by lowering the loads in the affected compartment. The second option is the unicompartmental arthro- plasty which consists of replacing the cartilage and subchondral bone of both femoral condyle and tibial plateau in 1 compartment only by a small unconstrai- ned prosthesis that has 2 components: a metallic femoral part shaped in the contour of the condyle and a polyethylene tibial tray with or without metal backing [16, 20].

Indications for unicompartmental arthroplasty include osteoarthritis (either degenerative or posttraumatic) or avascular necrosis of the knee that cause significant pain and functional disability and are confined to 1 compartment only [1, 13, 14, 18, 20]. The ideal patient is non-obese, over 60 years of age and leads a seden- tary way of live [12, 14]. Accepted absolute contraindi- cations include infection, systemic diseases such as rheumatoid arthritis or hemophilia, an angulatory deformity (varus or valgus) of more than 20 ~ and an unstable knee (resulting from a significant ligamen- teous injury and not only to the sinking of the tibial plateau in the affected side due to the arthritic pro- cess) [1, 5, 12-14, 20]. Obesity and high functional demands are considered to be relative contraindica- tions since both factors might expose the prosthesis to unusually high loads that might bring about early loos- ening [13, 14, 18].

This technique which has gained much popularity in the early 70s has many advantages over a total knee replacement: It allows preservation of important ana- tomical structures such as the cartilage in the unaf- fected compartments, the cruciate ligaments which not only provide antero-posterior and rotational sta- bility to the knee but are also believed to play a role in the sense of proprioception within the joint [12, 20]. Another benefit is supposed to be the conserva- tion of the subchondral bone stock whose integrity is a major factor in the successful outcome of revision surgery [8, 12, 20]. Moreover, propagation of the disease to the contralateral compartment of the same knee at a later stage does not necessarily have to be

treated by conversion to a formal total joint arthro- plasty since it is instead possible to perform a second unicompartmental procedure, provided of course that the patello-femoral articulation is not signifi- cantly involved [4, 11, 16]. Two additional positive observations were reported: The amount of peri-ope- rative bleeding (and therefore the need for blood replacement) is less than in total knee replacement, and the final range of motion is achieved more rapidly thereby shortening both hospital stay and rehabilitation periods [1, 4, 8, 12, 16].

Kieser et al. [8] have compared infection rates of total knee replacement versus unicompartmental in 514 arthroplasties of the knee of which 2/3 were of the uni- condylar type. The results were striking: zero infec- tions in the unicondylar as opposed to 2.4% in the total knee replacement group.

Despite these many advantages, unicompartmental arthroplasty is still controversial since reports concern- ing success rates and complications have not been uniform. Padgett et al. [15] performed 21 cases of revi- sion surgery after a failed unicondylar prosthesis and were faced with a major osseous defect in 16 of them and many technical difficulties. Gill et al. [6] looked into 60 cases of total knee replacement, half of which followed failed unicondylar prostheses and the other half following a previous high tibial osteotomy. They pointed out that revision after unicondylar arthroplas- ty is a technically demanding procedure with consid- erable bone loss to deal with. Finally, the knee scores achieved in total knee replacement after high tibial osteotomy were better than those with total knee replacement after a unicondylar replacement. Jackson et al. [7] in a similar study support the findings about the significant bone loss found at the medial tibial pla- teau but in their hands the revisions after high tibial osteotomy were more difficult to perform and yielded a very high rate of serious complications including 20% (!) of deep infections. Swank et al. [19] reported a relatively high failure rate (12%) of unicompartmen- tal prostheses at 8 years (on the average) of follow-up. Reasons of failure included progression to pan-arthro- sis, component failure, component loosening, poly- ethylene wear and technical errors in more or less equal rates. Similar numbers were reported by Scott et al. [17]: 10% failure at 9 years and climbing drastically to 18% at 11 years. Better results were reported by Capra et al. [2] with a failure rate of 6.25% only at 10 years.

288 Unfallchirurgie 1999;25:287-93 (Nr. 6)

Cohen Let al. MOD T M Unicondylar Hemi-Arthroplasty

We have been using unicompartmental arthroplasty of the knee for many years and with a variety of prosthe- ses without in fact knowing how it affected our pa- tients in the long-run. We therefore decided to restrict ourselves to a single type of prosthesis and conducted the present prospective study. We hereby report our medium-term results and experience with this type of implant.

Patients and Methods

Nineteen patients aged 57 to 81 years (average age 68 years) were operated upon during the years 1991 and 1992 in our ward, 1 knee in each patient to a total of 19 knees. Sex distribution was 16 women and 4 men. The pre-operative diagnoses were osteoarthritis in 17 cases (16 degenerative and 1 posttraumatic) and avas- cular necrosis in 2 cases, both in the medial compart- ment but 1 in the femoral condyle and the other in the tibial plateau. There was involvement of the left side in 11 cases and of the right side in 9. Disease was con- fined to the medial compartment in the majority of the cases (n : 18) while the lateral compartment was involved in a single case only. Three patients had had a previous surgical intervention in the contralateral knee in the form of a total knee replacement. The indication for operation was functionally impairing pain in the knee associated with significant unicom- partmental damage to the hyaline cartilage in the weight bearing area.

in all cases (Figure 1). A suction drain was left for 48 hours at all times. Continuous passive motion (CPM) was started at day 2. Patients started to walk with a walker and began active and passive physiotherapy including muscle re-enforcement exercises by days 2 to 4 according to individual abilities.

Average hospitalization time was 1 week and stitches were taken out at 10 days in the out-patient clinic. Fol- low-up again included patient history, a physical examination and roentgenological findings. Patients were assessed pre-operatively and at follow-up accord- ing to the recommendations and scoring system of the Hospital for Special Surgery with separate scores for the knee (including parameters of pain, stability, and range of motion) and for function (including walk- ing, climbing of stairs and the need for using walking aids).

Results

One patient out of this series of 19 had unfortunately died of a cerebro-vascular accident 2 years after sur- gery, and we therefore remained with 18 knees (1 in

Pre-operative assessment included patient history, physical examination and plain standing and stress X- rays of both knees. Equivocal cases or those showing a discrepancy between complaints and radiological find- ings were further assessed by either diagnostic arthros- copy or an "under-vision" inspection of the cartilagi- nous surfaces through the arthrotomy at the beginning of the procedure. All patients received peri-operative antibiotic prophylaxis by way of a single dose of cefonicid (Monoceff M) 1 g given i.v. 30 minutes prior to surgery. Anticoagulation prophylaxis was not given routinely and it was prescribed to only 2 patients considered to be at "high risk" for developing throm- boembolic phenomena due to prior illnesses. These 2 patients received 40 mg/d s.c. of low molecular weight heparin (ClexanerM).

The operations were performed in an "open" fashion through a formal mid-line arthrotomy of the knee. We used the cemented MOD3 TM prosthesis (by Richard)

Unfallchirurgie 1999;25:287-93 (Nr. 6)

Figure 1. The cemented MOD3 T M unicompartmental prosthesis.

289

Cohen I, et al. MOD3 T M Unicondylar Hemi-Arthroplasty

Knee score Number of knees Percentage Functional score Number of patients Percentage

41-50 0 0 31-40 1 7.2 51-60 1 5.8 41-50 3 21.4 61-70 4 23.3 51-60 2 14.3 71-80 2 11.7 61-70 2 14.3 81-90 3 17.5 71-80 3 22.4 91-100 7 41.7 81-90 0 0

91-100 3 21.4 Total 17 100

Table 1. Distribution of the knees according to score at follow-up examination.

Degree of severity Pre-operative At follow-up

None 0 5 Mild 2 7 Moderate 9 5 Severe 6 0

Table 3. Pain pre-operative and at the time of follow-up examina- tion.

Number of knees Number of knees Pre-operative At follow-up

80 ~ 3 0 90 ~ 5 0

100 ~ 2 5 110 ~ 1 1 120 ~ 3 4 125 ~ 0 1 130 ~ 3 5 135 ~ 0 1

Total 17 17 Average flexion 103 ~ 117.6 ~

Table 5. Range of movement (flexion).

each pa t ient ) to evaluate at follow-up. Survival

prosthet ic rate was 94.5% overall (17 out of 18 knees).

Early Complications

There were 3 cases with early pos topera t ive problems:

We had a technical failure with 1 of our first pat ients

that requi red an early revis ion surgery due to malposi-

t ioning of the tibial c o m p o n e n t due to a great bu lk of

cemen t in the poster ior aspect which resul ted in a non-

hor izonta l posi t ion of the prosthesis. Two other pa-

t ients suffered f rom t h r o m b o e m b o l i c sequelae, that is

1 case of p u l m o n a r y embol i sm diagnosed 3 days after

surgery by a ven t i l a t ion-per fus ion lung scan and an-

o ther with deep vein thrombosis that occurred 1

m o n t h postoperat ively. Bo th cases were successfully

t rea ted by full dose ant icoagulat ion.

Total 14 100

Table 2. Distribution of the patients according to the functional score at follow-up examination.

Pre-operative At follow-up

Extension lag None 13 11 Up to 5 ~ i 4 6 to 10 ~ 1 2 More than 10 ~ 2 0

Total 17 17

Instability Up to 5 ~ 15 15 6 to 10 ~ 2 1 More than 10 ~ 0 1

Total 17 i7

Table 4. Extension lag and instability.

Late Complications

There was 1 pat ient who developed a late deep infec-

t ion 1 year after surgery and therefore had to be re-

opera ted and a 2-stage convers ion to total knee re-

p lacement was pe r fo rmed with a favorable end result.

O ne other pa t ien t compla ined of pa te l lo- femoral pa in

after 2 years and was therefore t rea ted by arthroscopic

abras ion chondroplas ty and lateral re t inacular release

with d i sappearance of her pain.

The average follow-up per iod was 6 years and 6

mon ths (with a range of 6 to 7.5 years). Seven teen

knees were eva lua ted by Hospi ta l for Special Surgery

scoring (after excluding the pa t ien t with the septic

knee) . We should emphasize that since 3 pat ients as

has a l ready b e e n m e n t i o n e d had had a previous total

knee r ep lacement in the cont ra la tera l knee, the func-

t ional score was found to be i r re levant since most of

their walking and cl imbing disabili ty were felt to be

caused by the knee with the total knee rep lacement

ra ther than by the knee with un i compar tmen ta l

arthroplasty. Tha t left us with 14 pat ients for calcula-

t ion of the funct ional score as compared to the 17 pa-

t ients avai lable for the assessment of the knee score.

Results are shown in Tables 1 through 5.

290 Unfallchirurgie 1999;25:287-93 (Nr. 6)

Cohen I, et aL MOD3 TM Unicondylar Hemi-Arthroplasty

90

80

7O

60

40

2 0 -

t 0 -

0 Pre-op

Figure 2. The average knee score.

_ I

[ ] Average knee score (points)

Follow-up

The average knee score was found to have improved from a pre-operat ive value of 54.7 to 84.3 (Figure 2). There were 13 knees falling into the Hospital for Spe- cial Surgery category of a good or an excellent result (70 to 100 points) and 4 knees rated as a fair result (60 to 69 points). The average functional score also im: proved, that is f rom 34.6 to 68.9 (Figure 3). Eight pa- tients achieved a good or excellent functional result, 2 were in the fair range and 4 remained in the poor result zone (less than 60 points).

Roentgenological Findings

Radiolucent lines, as can be seen in Table 6, were of no clinical significance since they occurred in 2 cases only and with a thickness of 1 m m or less. Moreover, these osteopenic lines that were both present in the tibial aspect, did not encircle the tibial component en- tirely but were rather intermittent and measured in length about 1/3 of the total length of the bone- prosthesis interface (Figures 4 and 5).

Discussion

The unicompartmental arthroplasty is a controversial procedure since early enthusiastic reports were follow- ed by later studies that showed relatively high failure rates due either to component loosening or to conti- nuing pain and functional impairment despite the ope- ration [1, 9, 13, 16, 20J. This, however, does not seem to be accurate anymore as we have learned from review- ing the literature of the last 20 years that shows long- term results with rates of success and complications that are comparable to both high tibial os teotomy and total knee replacement for up to 10 years.

70

60

5O

4O

30

20

10

0 Pre-op

Figure 3. The average functional score. Follow-up

Christiensen [5] reported a series of 575 knees with gonarthrosis in 415 patients. The indications for sur- gery were very liberal so that the series included pa- tients suffering f rom obesity or chondrocalcinosis and even f rom instability of the knee due to A C L tear or insufficiency. Patients were followed for an average period of 9 years with 96% of successful results. There were 7 knees who required revision surgery due to aseptic loosening and 14 knees which were re-opera- ted due to continuing pain after the operation.

Rougraff et al. [16] compared 120 cases of unicompart- mental arthroplasty to 81 cases of total knee replace- ment followed for 10 years on the average. They found a 92% success rate with the unicompartmental arthro- plasty with a bet ter final range of motion and a great- er functional improvement after unicompartmental arthroplasty than after total knee replacement.

Stockelman et al. [18] have followed a series of 42 knees after unicompartmental arthroplasty for 5 to 12 years (average 7.4 years) with a similar success rate of 92%, 4 cases that required revision surgery due to

aseptic loosening and 2 more knees in which conver- sion to total knee replacement had to be done due to

Number of arthroplasties

No evidence of resorption 15 Up to t mm 2 1 to2mm 0 More than 2 mm 0

Total 17

Table 6. Roentgenologica[ assessment of bone-cement interphase at follow-up examination.

Unfallchirurgie 1999;25:287-93 (Nr. 6) 291

Cohen I, et aL M O D TM Unicondylar Herni-Arthroplasty

Figure 4. Anterior-posterior view roentgenograms of 4 patients at follow-up evaluation.

early propagation of the arthritic disease to the other compartments.

Lindstrand et al. [12] have compared 2 groups of pa- tients that had had unicompartmental arthroplasty with a porous coated type of prosthesis. Fixation to bone was cemented in 1 group and cementless in the other. Results were found to be statistically identical with only 1 failure in each group.

Results of an 8-year follow-up of 102 knees with uni- compartmental arthroplasty in 90 patients were report- ed by Larsson et al. [10] with again a success rate of 92% while about half of the failures were due to asep- tic loosening and the other half to advancement of the degenerative process to a tr icompartmental involve- ment.

This difference between earlier reports and those published in the 90s may be due to 1 or more of the

Figure 5. Lateral view roentgenograms of the same 4 patients as in Figure 4 at follow-up evaluation.

following factors: Improvement of the prostheses both in design and in chemical structure of the components, a greater experience with the surgical technique and re-assessment of the primary indications and contrain- dications by analysis of the accumulating data.

MOD3 TM is a prosthesis developed by Cartier et al. [3] and is a newer generation based on the former MOD2 which provided promising results with 95% survival rate and 10% of complications overall at 2 to 10 years (average 4.5 years) of follow-up and which has been further improved by its developer.

Our findings in the present study with a survival prosthesis rate of 94.5% (17 out of 18 knees) at more than 6 years stand in accordance with Cartier's [3] experience with the former MOD2 prosthesis and with all the other above-mentioned reports that have found unicompartmental arthroplasty to be an effective and reliable way of surgically treating early gonarthrosis

292 Unfallchirurgie 1999;25:287-93 (Nr. 6)

Cohen I, et al. M O D TM Unicondylar Hemi-Arthroplasty

with good results, provided that the arthritic process is indeed confined to a single compartment. We there- fore advocate its use whenever it is indicated, although we have not found a particular advantage of this speci- fic prosthesis over other types of commercially avail- able prosthetic devices. We also recommend not to rely solely upon X-rays in cases that are in doubt, but rather to postpone the final decision whether or not to actually perform a unicompartmental arthroplasty until after the cartilaginous surfaces in the other compartments have been inspected by either arthros- copy or direct vision by the time of arthrotomy.

References

1. Bert JM. Universal intramedullary instrumentation for unicom- partmental total knee arthroplasty. Clin Orthop 1991;271:79-87.

2. Capra SW Jr, Fehring TK. Unicondylar arthroplasty. A survi- vorship analysis. J Arthroplasty 1992;7:247-51.

3. Cartier P, Cheaib S. Uuicondylar knee arthroplasty. 2-10 years of follow-up evaluation. J Arthroplasty 1987;2:157-62.

4. Chesnut WJ. Preoperative diagnostic protocol to predict candi- dates for unicompartmental arthroplasty. Clin Orthop 1991;273: 146-50.

5. Christiensen NO. Unicompartmental prosthesis for gonarthro- sis. Clin Orthop 1991;273:165-9.

6. Gill T, Schemitsch EH, Brick GW, et al. Revision TKR after failed unicompartmental knee arthroplasty or high tibial osteo- tomy. Clin Orthop 1995;321:10-18.

7. Jackson M, Sarangi PP, Newman JH. Comparison of outcome following primary proximal tibial osteotomy or unicompart- mental arthroplasty. J Arthroplasty 1994;9:539-42.

8. Kieser C, Raber D. Expenses and risk of artificial knee joint. Schweiz Med Wochenschr 1996;126:1047-53.

9. Koshino T, Morii T, Wada J, et al. Unicompartmental replace- ment with the Marmor modular knee - operative procedures and results. Bull Hosp Joint Dis 1991;51:119-31.

10. Larsson SE, Larsson S, Lundkvist S. Unicompartmental knee arthroplasty. Clin Orthop 1988;232:174-81.

11. Laurencin CT, Zelicof SB, Scott RD, et al. Unicompartmental versus total knee arthroplasty in the same patient. Clin Orthop 1991;273:151-6.

12. Lindstrand A, Sterstrom A, Egund N. The PCA unicompart- mental knee. Acta Orthop Scand 1988;59:695-700.

13. Marmor L. Unicompartmental arthroplasty of the knee with a minimum of 10 years follow-up period. Clin Orthop i988;228: 171-7.

14. Marmor L. Unicompartmental knee arthroplasty. Clin Orthop 1988;226:14-20.

15. Padgett DE, Stern SH, Insall JN. Revision TKR for failed uni- compartmental replacement. J Bone Joint Surg 1991;73:186-90.

16. Rougraff BT, Heck DA, Gibson AE. A comparision of tricom- partmental and unicompartmental arthroplasty for the treat- ment of gonarthrosis. Clin Orthop 1991;273:157-64.

17. Scott RD, Cobb AG, McQueary FG, et al. Unicompartmental knee arthroplasty. 8 to 12 year follow-up evaluation with survi- vorship analysis. Clin Orthop 1991;271:96-100.

18. Stockelman RE, PohI KP. The long term efficacy of unicom- partmental arthroplasty of the knee. Clin Orthop 1991;271: 88-95.

19. Swank M, Stulberg SD, Jiganti J, et al. The natural history of unicompartmental arthroplasty. An 8 year follow-up study with survivorship analysis. Clin Orthop 1993;286:130-42.

20. Thornhill TS, Scott RD. Unicompartmental total knee arthro- plasty. Orthop Clin North Am 1989;20:245-56.

Address for Correspondence: Dr. Ilan Cohen, 34 Kehilat-Zion Street, Hertzlia 46382, Israel, Fort (+972/52) 436003, Fax (+972/3) 6436447, e-mail: ilanco l @ hotmail.com

Unfallchirurgie 1999;25:287-93 (Nr. 6) 293