San Diego Shoulder Course San Diego, California, June 14-17, 2000

Hemiarthroplasty vs Total Shoulder Arthroplasty Sandy Kirkley, MD, FRCSC Primary osteoarthritis of the shoulder •

Signs and symptoms: Diffuse, achy, progressive pain Stiffness progressing to permanent loss of motion Sleeping difficulties and limitation of function

•

Radiographic findings: Joint space narrowing Humeral head flattening Osteophyte formation at the inferior margin of the humeral head Subarticular cyst formation in the superior aspect of the head Head may be enlarged to twice normal size Glenoid flattening with sclerosis Posterior erosion of the glenoid in severe cases

•

Rotator cuff is usually intact, as forces necessary to produce deformities would require significant muscle forces.

•

It is estimated that over 10,000 shoulder arthroplasties are performed annually in the USA currently8

•

Total shoulder arthroplasty has been shown to be as cost effective as total hip or knee arthroplasty16

Surgical Treatment Options • • • • •

Joint Debridement – limited indications in minor to moderate disease Joint Resection – salvage procedure in face of chronic infection, infected arthroplasty or non-reconstructable bone loss following arthroplasty Shoulder Arthrodesis – indicated in failed arthroplasty not amenable to revision and chronic infection Hemiarthroplasty Total Shoulder Arthroplasty

Theoretical advantages and disadvantages of hemi and TSA •

Hemiarthroplasty Easy procedure Short operating time Less risk of instability Can be revised to total shoulder arthroplasty later if necessary Less consistent pain relief Progressive erosion of the glenoid may cause deterioration of the results with time

•

Total Shoulder Arthroplasty More consistent pain relief Better fulcrum for active motion More difficult procedure Longer operating time Subscapularis rupture more common High-density polyethylene wear debris may cause loosening of both components Glenoid loosening with associated bone loss

Recommendations based on experience •

Neer9 1998 -- “When the articular surface of the glenoid is good, the results of hemiarthroplasty are similar to those of TSA. Wear on the glenoid has not been a problem if the articular surface was good at the time of surgery and glenohumeral motion was re-established.”

Comparative non randomized trials •

Boyd et al1 1990 -- 24-124 months follow up of 210 shoulders (64 Neer hemis in 59 patients and 146 Neer TSA in 134 patients) with RA, OA, AVN, and 4 part fractures of proximal humerus. No significant difference in results for hemi and TSA in terms of functional improvement Pain relief, ROM, satisfaction better with TSA in the RA population No significant difference in the OA population

•

Cofield 13 1998 -- 5 year minimum follow up of 114 shoulders (108 shoulders reviewed) – 74 hemi in 64 patients and 34 total shoulder arthroplasty (TSA) in 31 patients Mixed bag of patients (OA, RA, cuff disease) all less than 50 years of age Only one half had a successful result at 5 years in both groups Hemi – 83% survival @ 10 years and 73% @ 15 years TSA – 97% survival @ 10 years and 84% @ 15 years

•

Rockwood11 1998 -- 88patients (117 shoulders) 42 hemi, 75 TSA Average 4-year follow up No significant difference in results for hemi and TSA in terms of pain relief, ROM, ASES 15 activities and patient satisfaction. Concluded that hemi is successful in patients with OA when the glenoid is concentric and not eroded anteriorly or posteriorly but not successful when there is a flat eroded glenoid with posterior subluxation of the head of the humerus.

Randomized Clinical Trials •

Jonsson5 1998 -- RCT of hemi vs TSA with randomization stratified for RA and OA 56 randomized but 7 excluded – 5 unable to do glenoid after being randomized to the TSA group, 1 infection, 1 nerve injury Results showed no significant difference in night pain, pain with activities, ROM but trend in favour of TSA at 2 years ****May have created bias in favour of TSA because most severe cases (considerable glenoid wear which did not allow for glenoid replacement) were excluded form the TSA group but not from the hemi group

•

Sandow12 1998 -- RCT of hemi vs TSA (Global Shoulder System) in patients with OA 30 patients randomized after soft tissue release sufficient to allow for glenoid component insertion. Results on 27 patients at 1 year using UCLA, Constant, ROM showed no significant difference at 6 months, one year but less pain in the TSA group at 1 year 2 patients in hemi group crossed over to TSA at 2 years 2 patients in TSA group dislocated

•

Gartsman4 2000 -- 47 patients (51 shoulders) degenerative OA randomly assigned to hemi or TSA Followed for mean of 35 months No significant differences with ASES, UCLA. Significantly greater pain relief and internal rotation with TSA 3 patients in hemi group have crossed over to TSA

•

Kirkley et al6 2000 -- 42 patients with primary OA failed conservative treatment randomized to hemi or TSA Randomization stratified for surgeon’s (3) No significant differences with WOSI, ASES, Constant, ROM, DASH at 1 year follow-up 2 patients crossed over to TSA after one-year follow-up Complications in TSA group included 1 intra-op glenoid fracture,1 late subscap rupture and 1 deep infection Complications in hemi group included 1 fracture Trend for better pain relief in TSA group

Does glenoid condition predict results of hemi? •

Levine et al7 1997 -- No statistically significant difference between patients with concentric and nonconcentric glenoid wear at time of hemi for ASES, Constant, ROM, and pain relief but trend for better results in concentric wear group.

•

Rockwood11 1998 – anecdotally yes

•

Warner15 1999 -- Although progressive glenoid wear occurred in all patients undergoing hemi there was no statistically significant correlation between glenoid erosion and poor functional outcome or need for subsequent TSA

•

Kirkley et al6 2000 -- Concentric glenoid wear in both patients crossed over to TSA

Is glenoid loosening a problem in TSA? •

Cofield 3 1984 – at 2-6 years follow-up, significant glenoid radiolucency in 60 of 73 glenoid components. Eight of these cases were considered grossly loose and 3 were revised for pain.

•

Brostrom2 1992 -- 96% of 26 unconstrained total shoulder replacements with a mean follow-up of 47 months showed radiolucent zones around the glenoid. This development was associated with a decrease in function and increased pain

. •

O’Driscol et al10 1998 – 91% of 81 shoulders with a mean follow-up of 4 years had radiolucent lines at the glenoid and 25/81 components were judged to be radiographically “loose”. These adverse radiographic features were associated with a higher level of pain.

•

Rockwood11 1998 – not really

•

Walch et al14 2000 -- Aequalis shoulder arthroplasties (259 TSA and 17 hemis) for primary OA reviewed at mean of 30 months 61% of glenoids cemented and 39 % not cemented. Glenoid radiolucent lines were present in 58% of the cases (23% progressive) and were associated with a less satisfactory result

Conclusions It is currently unknown whether the operation of choice for patients with primary OA of the shoulder is a hemi arthroplasty or total shoulder arthroplasty. A meta-analysis of the 3 recently presented or published randomized clinical trials will shed considerable light on this important question. It may be that with current surgical techniques there is a trade off between better early results with TSA and later failures due to glenoid loosening. At this point in time the consensus of most experienced shoulder surgeons is that:

Hemiarthroplasty should be considered in patients with concentric wear of the glenoid, in the younger active patient and the patient where it is not possible to insert a glenoid component. Although progressive radiolucent lines surrounding the glenoid component are a common finding and are associated with poorer results, the fact remains that the revision rate of total shoulder arthroplasties is still quite low.

References 1. Boyd, AD Jr, Thomas WH, Scott RD, Sledge CB, Thornhill TS: Total shoulder arthroplasty versus hemiarthroplasty: Indications for glenoid resurfacing. J Arthroplasty, 5(4):329-36, 1990. 2. Brostrom LA, Kronberg M, Wallensten R: Should the glenoid be replaced in shoulder arthroplasty with an unconstrained Dana or St. Georg prosthesis? Ann Chir Gynaecol, 81(1):54-7, 1992. 3. Cofield, RH: Total shoulder arthroplasty with the Neer prosthesis. JBJS, 66A(6):899-907, 1984. 4. Gartsman GM, Roddey TS, Hammerman SM: Shoulder arthroplasty with or without resurfacing of the glenoid in patients who have osteoarthritis. JBJS, 82A(1):26-34, 2000. 5. Jonsson U, Abbaszadegan H, Revay S, Salomonsson B: Better function with a glenoid component 2 years after shoulder arthroplasty. International Congress on Surgery of the Shoulder, Sydney, Australia, 1998, pp98. 6. Kirkley A, Lo, IKY, Griffin S, Faber K, Patterson SD, Litchfield RB: Hemiarthroplasty versus total shoulder arthroplasty in the treatment of osteoarthritis of the shoulder: A prospective randomized trial. American Shoulder and Elbow Meeting, Miami Beach, Florida, 2000, pp115. 7. Levine WN, Djurasovic M, Glasson JM, Pollock RG, Flatlow, EL, Bigliani LU: Hemiarthroplasty for glenohumeral osteoarthritis: Results correlated to degree of glenoid wear. J Shoulder Elbow Surg, 6(5):449-454, 1997. 8. Matsen FA, Rockwood CA, Wirth MA, Lippitt SB: Glenohumeral arthritis and its management. In Rockwood CA, Matsen FA (eds): The Shoulder, 2nd Edition. Philadelphia, WB, Saunders CO. 1998, pp840-964. 9. Neer II, CS: Shoulder Prosthetics (Indications, hemi vs TSA, outcome, factors affecting outcome). American Academy of Orthopaedic Surgeons, Kiawah Island, SC, 1998, pp3 10. O’Driscol SW, Wright T, Cofield RH: The glenoid problem: Radiographic assessment of the glenoid component in total shoulder arthroplasty. International Congress on Surgery of the Shoulder, Sydney, Australia, 1998, pp53. 11. Rockwood Jr CA, Jensen KL, Wirth MA: Hemiarthroplasty vs total shoulder arthroplasty in patients with osteoarthritis. International Congress on Surgery of the Shoulder, Sydney, Australia, 1998, pp99. 12. Sandow MJ, David HG, Bentall SJ: Hemi-arthroplasty or total replacement for shoulder osteoarthritis preliminary results of a PRCT with intra-operative randomization. International Congress on Surgery of the Shoulder, Sydney, Australia, 1998, pp100. 13. Sperling JW, Cofield RH, Rowland CM: Neer hemiarthroplasty and Neer total shoulder arthroplasty in patients fifty years old or less. Long-term results. JBJS, 80A(4):464-73, 1998. 14. Walch G, Godeneche A, Boileau P, Nove-Josserand L: Clinical and radiographic results of 268 shoulder arthroplasties in primary osteoarthritis. American Shoulder and Elbow Surgeons 16h Open Meeting, Orlando, Florida, 2000, pp21. 15. Warner JP, Parsons IM: Quantification of radiographic glenoid wear after shoulder hemiarthroplasty. American Shoulder and Elbow Surgeons 15th Open Meeting, Anaheim, California, 1999, pp29.

16. Wirth MA, Rockwood CA: Complications of shoulder arthroplasty. Clin Orthop Rel Res, 307:47-69, 1994.