The Influence of Glenohumeral Prosthetic Mismatch on Glenoid Radiolucent Lines: Results of a Multicenter Study

Walch, Gilles MD; Edwards, T. Bradley MD; Boulahia, Aziz MD; Boileau, Pascal MD; Mole, Daniel MD; Adeleine, Patrice PhD

Journal of Bone & Joint Surgery - American Volume:
Scientific Article
Supplementary Content
Abstract

Background: In shoulder arthroplasty, mismatch is defined as the difference in the radius or diameter of curvature between the humeral head and glenoid components. Recommendations for mismatch have not been substantiated scientifically. The purpose of this study was to evaluate the effect of mismatch on glenoid radiolucent lines.

Methods: The results of 319 total shoulder arthroplasties performed for the treatment of primary osteoarthritis were evaluated. All of the arthroplasties were performed with a single type of prosthesis (Aequalis; Tornier, Montbonnot, France) that included a cemented, all-polyethylene glenoid component. Three sizes of glenoid components and seven humeral head diameters were utilized. Radial mismatch was categorized as ≤4 mm, 4.5 to 5.5 mm, 6 to 7 mm, or >7 to 10 mm. Radiographs were evaluated at a mean of 53.5 months (range, twenty-four to 110 months) postoperatively. Glenoid radiolucent lines were scored with a scale ranging from 0 points for no radiolucency to 18 points for radiolucent lines exceeding 2 mm in six zones. Variance, linear contrasts polynomial, quadratic polynomial contrast statistical, and linear regression analyses were performed to evaluate the relationship between radial mismatch and glenoid radiolucent lines.

Results: A significant linear relationship was found between mismatch and the glenoid radiolucency score (p < 0.0001), with significantly lower (better) radiolucency scores associated with radial mismatches of >5.5 mm.

Conclusions: In this study of glenohumeral prosthetic mismatch ranging from 0 to 10 mm, the mismatch had a significant influence on the scores for the glenoid radiolucent lines, which were best when the radial mismatch was between 6 and 10 mm. The theoretical risk of prosthetic instability with larger mismatch values was not demonstrated within the range of mismatch values evaluated in this series.

Author Information

Gilles Walch, MD; Aziz Boulahia, MD; Department of Orthopaedic Surgery, Clinique Sainte Anne Lumière, 85 Cours Albert Thomas, 69003 Lyon, France. E-mail address for G. Walch: walch.gilles@wanadoo.fr

T. Bradley Edwards, MD; Minneapolis Sports Medicine Center, 7201 Washington Avenue South, Edina, MN 55439

Pascal Boileau, MD; Department of Orthopaedic Surgery, Hôpital de l'Archet, 151 Route de Saint Antoine de Ginestière BP 79, 06202 Nice CEDEX, France

Daniel Molé, MD; Clinique Traumatologique et Orthopédique, 49 Rue Hermine, SC 5211, F-54052 Nancy CEDEX, France

Patrice Adeleine, PhD; Department of Biostatistics, Hôpitaux de Lyon, 162 Avenue Lacassagne, 69424 Lyon CEDEX 03, France

Article Outline

Glenohumeral mismatch is the difference in the curvature between the glenoid and the humeral head and may be expressed in terms of the radius or diameter of curvature. In a congruent articulation, the radii of curvature of the glenoid and the humeral head are identical, whereas in a noncongruent articulation, they differ. These terms should not be confused with constrained and nonconstrained articulations, which are a function of glenoid depth. Anatomic glenohumeral mismatch varies, depending on whether the osseous or the cartilaginous anatomy is considered. Osseous glenohumeral radial mismatch ranges from 8 to 9 mm, whereas glenohumeral radial mismatch may be as little as 0.1 mm if the articular cartilage and the glenoid labrum are also considered 1-4.

It is recognized that a congruent articulation allows optimal surface contact, minimizes the risk of surface wear of the glenoid component, and contributes to joint stability. However, with these advantages comes a lack of obligate translation 5 —i.e., translation between the articular surfaces that occurs with active and passive shoulder mobility and is absorbed by elastic deformation of the articular cartilage and the glenoid labrum in the normal shoulder 5-8. Obligate translation also occurs following shoulder arthroplasty 9-14, and a lack of this translation may lead to loosening of the glenoid component because of increased stresses at the implant fixation site 15-17. Alternatively, noncongruent articulations in which the radius of curvature of the glenoid component is larger than that of the humeral component allow obligate translation between the humeral head and the glenoid; however, surface wear of the glenoid component and joint instability are concerns.

No in vivo investigation has conclusively demonstrated superiority of congruent or noncongruent glenohumeral articulations. Karduna et al. and Iannotti and Williams performed a number of studies to obtain information regarding ideal glenohumeral prosthetic mismatch 14-16,18. They found that normal glenohumeral joint translation is best reproduced by a glenohumeral radial mismatch of approximately 4 mm, anterior-posterior translation is greater than superior-inferior translation (1.5 mm compared with 1.1 mm), and variations of 0 to 5 mm of radial mismatch do not alter prosthetic joint stability 18. Friedman et al. reported that when glenohumeral radial mismatch exceeds 10 mm, the risk of polyethylene fracture increases 17.

A review of recommendations regarding mismatch for various prosthetic designs reveals a lack of agreement regarding ideal mismatch. Neer employed a congruent prosthetic system in which the glenoid and the humeral head had identical radii of curvature 19. Between 3 and 5 mm of radial mismatch is recommended for the Global shoulder system (DePuy, Warsaw, Indiana) designed by Rockwood and Matsen 10. Bigliani and Flatow employed a unique system that uses two different radial mismatches; the central portion of the glenoid component is congruent (no mismatch), while the peripheral portion is noncongruent 20. The Aequalis prosthesis (Tornier, Montbonnot, France) allows variable mismatch between 0 and 12 mm, depending on the sizes of the humeral head and glenoid components chosen by the surgeon; however, a mismatch between 5 and 7.5 mm is recommended 21,22. Despite the diverse nature of recommendations regarding mismatch, we are not aware of any data regarding the influence of prosthetic mismatch on clinical or radiographic results. Therefore, the purpose of this study was to examine the effects of glenohumeral radial mismatch on clinical and radiographic results in a series of patients treated with total shoulder arthroplasty with a single type of prosthesis.

Back to Top | Article Outline

Materials and Methods

This investigation was based on a series of 1542 primary shoulder arthroplasties reported as part of a multicenter study in Nice, France, in September 2001 23. Forty-two surgeons at thirty-eight centers located in eight countries contributed cases to this series. Only shoulders with primary glenohumeral osteoarthritis as described by Neer 24 and a single type of glenoid implant (all-polyethylene, cemented, flat-backed) were chosen for the present investigation. Shoulders affected by an inflammatory arthropathy (rheumatoid arthritis, systemic lupus erythematosis, or ankylosing spondylitis), osteochondromatosis, acromegaly, Paget disease, postinfectious arthropathy, skeletal dysplasia, neurological problems (Charcot arthropathy or Parkinson disease), or osteonecrosis (posttraumatic, radiation-induced, or idiopathic) were excluded. Additional exclusion criteria included a history of shoulder trauma (fracture or soft-tissue injury), instability (surgically or nonsurgically treated), or a prior shoulder operation. Finally, shoulders with a marked pathological condition of the rotator cuff (acromiohumeral arthritis, a massive rotator cuff tear, or a rotator cuff tear involving the infraspinatus) as seen on imaging or at the time of surgery were excluded. Three hundred and forty-eight shoulders met the aforementioned criteria. Twenty-nine of these shoulders had missing information or were lost to follow-up before a minimum of two years postoperatively, leaving 319 shoulders in 300 patients available for evaluation.

The mean age was 66.4 years (range, fifty-four to ninety years) at the time of surgery. Two hundred and thirty-two of the shoulders were in women, and eighty-seven were in men. The forty-two surgeons each contributed an average of nine cases (range, one to eighty-three cases) to this series. All patients underwent primary total shoulder arthroplasty with use of an Aequalis prosthesis. Seven different humeral head diameters (ranging from 39 to 50 mm) and three different glenoid radii of curvature (27.5, 30, and 32.5 mm) were used. The mismatch values for the different head-glenoid combinations are shown in Table I . According to the manufacturer, with this prosthetic system, head sizes of 39/14 and 41/15 are best coupled with a small glenoid, head sizes of 43/16 and 46/17 are best coupled with a medium glenoid, and head sizes of 48/18, 50/16, and 50/19 are best coupled with a large glenoid. However, the system permits use of any head size with any glenoid size 21,22. The number of shoulders treated with each combination in our study is shown in Table II . At the time of surgery, various factors that could possibly influence the results of the arthroplasty, such as the condition of the supraspinatus tendon and the morphology of the glenoid (concentric or eccentric), were recorded.

For purposes of data analysis, the shoulders were divided into four groups based on glenohumeral prosthetic mismatch. Group 1 had a radial mismatch of ≥4 mm; group 2, a mismatch of 4.5 to 5.5 mm; group 3, a mismatch of 6 to 7 mm; and group 4, a mismatch of >7 to 10 mm. Demographic data were compared statistically among the groups.

Preoperative and postoperative information on each shoulder was collected with use of a standardized form. Clinical evaluation was performed with use of the absolute Constant score, which consists of four individual scores for pain (15 points), activity (20 points), active mobility (40 points), and strength (the number of kilograms of downwardly directed force that the patient is able to resist with the shoulder in 90&deg; elevation in the plane of the scapula as measured with a dynamometer and multiplied by two) 25. Additionally, active forward elevation in the plane of the scapula and active external rotation with the arm at the side were recorded.

Anteroposterior radiographs with the humerus in neutral, internal, and external rotation and an axillary radiograph made at the time of the latest follow-up were evaluated for humeral and/or glenoid periprosthetic radiolucent lines and/or component loosening and proximal migration of the humeral component. The operating surgeon or a member of the surgical team classified any glenoid radiolucent lines according to the previously validated system of Molé et al., which assigns scores ranging from 0 points for no radiolucency to 18 points for radiolucent lines exceeding 2 mm in six zones ( Fig. 1 ) 23,26. A numeric value determined by the thickness of the radiolucent line is assigned to each zone: 0 points is given for no radiolucent line; 1 point, for a radiolucent line of <1 mm in thickness; 2 points, for a radiolucent line of 1 to 2 mm in thickness; and 3 points, for a radiolucent line of >2 mm in thickness. The scores for all of the zones are then added to yield the radiolucency score. A score of 0 to 6 points is considered to represent no loosening, 7 to 12 points represents possible loosening, and 13 to 18 points represents definite loosening. Complications or subsequent reoperations were also noted.

Statistically, the data were analyzed with use of the Fisher exact test and a Pearson chi-square test for qualitative variables. When quantitative values were evaluated, analysis of variance was performed to determine significance. Additionally, because of the possibility of the concomitant presence of multiple factors, an analysis was performed with use of a linear regression for quantitative dependant variables. A multivariate analysis and a Student t test were utilized to evaluate any relationship between quantitative and qualitative variables. To analyze the relationship between mismatch and radiolucency score, we performed a linear regression analysis and a variance analysis with one factor with linear and quadratic trend test. Significance was set at p < 0.05.

Back to Top | Article Outline

Results

The duration of follow-up averaged 53.5 months (range, twenty-four to 110 months). Statistical analysis of the demographic data demonstrated no significant differences between the groups regarding gender, hand dominance, status of the rotator cuff at the time of surgery, age at the time of surgery, or duration of follow-up ( Table III ). Additionally, the experience of the surgeon did not influence the degree of mismatch that was selected.

Table IV shows the results of the clinical evaluation for each of the mismatch groups. With the numbers available, no significant difference was found between groups with regard to the Constant score, any component of the Constant score (pain, activity, mobility, or strength), or active forward elevation. Mismatch between 4.5 and 7 mm was associated with better active external rotation with the arm at the side (p = 0.001). With the numbers available, mismatch was not found to influence the rate or type of postoperative complications (anterior, posterior, or superior instability).

Radiographically, 243 of the 319 shoulders had a well-fixed glenoid component (a radiolucency score of <7 points), forty-nine shoulders had a glenoid component that was possibly loose (a radiolucency score of 7 to 12 points), and twenty-seven shoulders demonstrated loosening of the glenoid component (a radiolucency score of >12 points). Overall, the radiolucency score averaged 4.7 points (range, 0 to 18 &plusmn; 4.8 points).

The effect of glenohumeral mismatch on the radiolucency score was found to be highly significant (linear trend; p < 0.0001) as shown in Figure 2 and the scatterplot in Figure 3 . Table V shows the radiolucency scores for each category of mismatch. Group 1 was significantly different from group 3 (p < 0.002) and group 4 (p < 0.008). Group 2 was not found to be significantly different from the other groups with the numbers available.

Statistical analysis was performed to evaluate the potential influence of other factors (experience of the surgeon, presence of a partial or full-thickness tear of the rotator cuff, eccentric glenoid morphology, and occurrence of a postoperative complication) on the relationship between mismatch and the radiolucency score. The experience of the surgeon—that is, whether the surgeon had contributed ten cases or more to this series or had contributed less than ten cases—did not influence this relationship. Similarly, preoperative glenoid morphology, the occurrence of a postoperative complication, and the presence of a supraspinatus tear did not affect the relationship between mismatch and the radiolucency score.

Back to Top | Article Outline

Discussion

This multicenter investigation demonstrated the relationship between glenoid component radiolucency and glenohumeral prosthetic mismatch. Despite the aforementioned in vitro investigations of glenohumeral mismatch 14-16, to our knowledge this in vivo relationship has not been described previously. The results of our investigation indicate that prosthetic glenohumeral radial mismatch in excess of 5.5 mm but no more than 10 mm most favorably influences the glenoid radiolucency score for the Aequalis prosthesis. Although a linear trend was discovered, the desired upper limit of radial mismatch has not been conclusively determined at this time since data obtained in this investigation are not amenable to extrapolation.

The importance of radiolucent lines adjacent to a glenoid component is unclear. Although some authors have reported that such lines have no influence on results or the revision rate, other investigators have found an association between the lines and compromised outcomes 27,28. Interestingly, in the present series, despite the relationship between glenohumeral mismatch and the formation of radiolucent lines, the mismatch had minimal effect on clinical results or complication rates. The only clinical parameter that it affected was postoperative external rotation, and the reason for this remains unclear. Although our findings appear to add more confusion to the issue of the importance of glenoid radiolucent lines, the lack of clinical influence may be explained by the relatively short-term follow-up of the patients in this study. In a long-term follow-up investigation, Torchia et al. found a relationship between compromised clinical results and glenoid radiolucent lines 28.

The current investigation is not without limitations. Multicenter study designs have inherent flaws. Different surgeons have different levels of experience, which can influence outcome. Additionally, many conditions (diagnosis, status of the rotator cuff, and glenoid morphology) can introduce confounding variables into an analysis of the outcomes of shoulder arthroplasty. To minimize the influence of these factors, we limited this series to patients with primary glenohumeral osteoarthritis, using strict exclusion criteria, even though that limited the extrapolation of our data to other diagnoses. Additionally, the influence of other confounding variables (surgeon experience, status of the rotator cuff, and glenoid morphology) was analyzed statistically to further validate the findings of this investigation.

Other factors limiting application of the data obtained in this study are the influence of prosthetic mismatch on glenohumeral stability and our inability to evaluate in vivo polyethylene wear. Joint conformity has been proposed as a factor affecting stability of the glenohumeral joint 29,30, and several investigators have studied this effect in a cadaveric model 9,11,13,31, with measurement of the minimum force necessary for joint dislocation. Severt et al. 11 and Fukuda et al. 31 concluded that greater joint conformity leads to higher subluxation forces. Karduna et al. showed that variation of radial mismatch between 0 and 5 mm changed dislocation forces by an average of only 3% 18. They concluded that these small differences were not clinically relevant. In our series of 319 total shoulder arthroplasties, only two patients had anterior instability; both were in the 4.5 to 5.5-mm-mismatch group. One patient, with &le;4 mm of mismatch, had posterior instability. Similarly, we evaluated the influence of mismatch and superior instability of the humeral head as characterized by a decrease in the acromiohumeral distance. Although eight shoulders had a decreased acromiohumeral distance at the time of follow-up, this factor had no significant relationship with prosthetic mismatch. In summary, after an average of fifty-four months of follow-up, we did not observe an influence of mismatch on anterior, posterior, or superior prosthetic stability.

A greater degree of glenohumeral mismatch allows more translation and a higher potential for polyethylene wear. To our knowledge, there is no reliable technique to evaluate wear of the polyethylene component quantitatively in total shoulder arthroplasty. Until it is possible to establish the influence of increased glenohumeral mismatch on polyethylene wear and its importance, the findings of this study must be applied with a degree of caution.

Despite these acknowledged imperfections, we believe that our study is the first in vivo investigation of the effects of glenohumeral prosthetic mismatch on clinical and radiographic parameters. On the basis of the results of this investigation, it appears that prosthetic radial mismatch in excess of 5.5 mm but no more than 10 mm most favorably influences the glenoid radiolucency score. Although a linear trend was discovered, the desired upper limit of radial mismatch was not conclusively determined since the data obtained in this investigation are not amenable to extrapolation. It is our hope that this investigation will be the first step in establishing a scientific basis for recommendations regarding glenohumeral prosthetic mismatch.

Investigation performed at the Department of Orthopaedic Surgery, Clinique Sainte Anne Lumière, Lyon, France

In support of their research or preparation of this manuscript, one or more of the authors received grants or outside funding from Association pour le Développement de la Pathologie de l'Epaule. In addition, one or more of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commer- cial entity (Tornier Company). Also, a commercial entity (Tornier Company) paid or directed, or agreed to pay or direct, benefits to a research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.

1. Boileau P, Walch G. Normal and pathological anatomy of the glenoid: effects on the design, preparation and fixation of the glenoid component. In: Walch G, Boileau P, editors. Shoulder arthroplasty. New York: Springer; 1999. p 127-40.
2. Bonnel F, Dusserre F, Canovas F, Assi C. Morphologie et biométrie de l'articulation scapulo-humérale. In: Mansat M, editor. Prothèses d'épaule. Paris: Expansion Scientifique; 1999. p 18-26.
3. Iannotti JP, Gabriel JP, Schneck SL, Evans BG, Misra S. The normal glenohumeral relationships. An anatomical study of one hundred and forty shoulders. J Bone Joint Surg Am, 1992;74: 491-500.
4. Soslowsky LJ, Flatow EL, Bigliani LU, Mow VC. Articular geometry of the glenohumeral joint. Clin Orthop, 1992;285: 181-90.
5. Harryman DT 2nd, Sidles JA, Harris SL, Matsen FA 3rd. The role of the rotator interval capsule in passive motion and stability of the shoulder. J Bone Joint Surg Am, 1992;74: 53-66.
6. Poppen NK, Walker PS. Normal and abnormal motion of the shoulder. J Bone Joint Surg Am, 1976;58: 195-201.
7. Howell SM, Galinat BJ, Renzi AJ, Marone PJ. Normal and abnormal mechanics of the glenohumeral joint in the horizontal plane. J Bone Joint Surg Am, 1988;70: 227-32.
8. Harryman DT 2nd, Sidles JA, Matsen FA 3rd. Range of motion and obligate translation in the shoulder: the role of the coracohumeral ligament. Trans Orthop Res Soc, 1990;15: 273.
9. Kelkar R, Flatow EL, Bigliani LU, Mow VC. The effects of articular congruence and humeral head rotation on glenohumeral kinematics. ASME Adv Bioeng, 1994;BED28: 19-20.
10. Collins D, Tencer A, Sidles J, Matsen F 3rd. Edge displacement and deformation of glenoid components in response to eccentric loading. The effect of preparation of the glenoid bone. J Bone Joint Surg Am, 1992;74: 501-7.
11. Severt R, Thomas BJ, Tsenter MJ, Amstutz HC, Kabo JM. The influence of conformity and constraint on translational forces and frictional torque in total shoulder arthroplasty. Clin Orthop, 1993;292: 151-8.
12. Friedman RJ. Glenohumeral translation after total shoulder arthroplasty. J Shoulder Elbow Surg, 1992;1: 312-6.
13. Harryman DT, Sidles JA, Harris SL, Lippitt SB, Matsen FA. The effect of articular conformity and the size of the humeral head component on laxity and motion after glenohumeral arthroplasty. A study in cadavera. J Bone Joint Surg Am, 1995;77: 555-63.
14. Karduna AR, Williams GR, Williams JL, Iannotti JP. Glenohumeral joint translations before and after total joint arthroplasty. A study in cadavera. J Bone Joint Surg Am, 1997;79: 1166-74.
15. Karduna AR, Williams GR, Iannotti JP, Williams JL. Total shoulder arthroplasty biomechanics: a study of the forces and strains at the glenoid component. J Biomech Eng, 1998;120: 92-9.
16. Iannotti JP, Williams GR. Facteurs anatomiques et biomécaniques influençant le design des prothèses. In: Mansat M, editor. Prothèses d'épaule. Paris: Expansion Scientifique; 1999. p 68-80.
17. Friedman RJ, An YH, Draughn RA. Glenohumeral congruence in total shoulder arthroplasty. Orthop Trans, 1997;21: 17.
18. Karduna AR, Williams GR, Williams JL, Iannotti JP. Joint stability after total shoulder arthroplasty in a cadaver model. J Shoulder Elbow Surg, 1997;6: 506-11.
19. Neer CS 2nd. Shoulder arthroplasty: history, new designs and new complications. In: Mansat M, editor. Prothèses d'épaule. Paris: Expansion Scientifique; 1999. p 1-11.
20. Bigliani LU, Flatow EL. Bigliani/Flatow shoulder prosthesis manufacturer's brochure, 2001. Zimmer, Warsaw, Indiana.
21. Walch G, Boileau P. The Aequalis shoulder system. In: Walch G, Boileau P, editors. Shoulder arthroplasty. New York: Springer; 1999. p 185-9.
22. Walch G, Boileau P. Prosthetic adaptability: a new concept for shoulder arthroplasty. J Shoulder Elbow Surg, 1999;8: 443-51.
23. Walch G, Boileau P, Molé D, editors. 2000 shoulder prostheses . . . two to ten year follow-up. Montpellier: Sauramps Medical; 2001.
24. Neer CS 2nd. Replacement arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am, 1974;56: 1-13.
25. Constant CR, Murley AH. A clinical method of functional assessment of the shoulder. Clin Orthop, 1987;214: 160-4.
26. Molé D, Roche O, Riand N, Levigne C, Walch G. Cemented glenoid component: results in osteoarthritis and rheumatoid arthritis. In: Walch G, Boileau P, editors. Shoulder arthroplasty. New York: Springer; 1999. p 163-71.
27. Boyd AD Jr, Thomas WH, Scott RD, Sledge CB, Thornhill TS. Total shoulder versus hemiarthroplasty. Indications for glenoid resurfacing. J Arthroplasty, 1990;5: 329-36.
28. Torchia ME, Cofield RH, Settergren CR. Total shoulder arthroplasty with the Neer prosthesis: long-term results. J Shoulder Elbow Surg, 1997;6: 495-505.
29. Lettin AW, Scales JT. Total replacement of the shoulder joint (two cases). Proc R Soc Med, 1972;65: 373-6.
30. Reeves B, Jobbins B, Dowson D, Wright V. A total shoulder endo-prosthesis. Eng Med, 1972;1: 64-8.
31. Fukuda RJ, Chen CM, Cofield RH, Chao EY. Biomechanical analysis of stability and fixation strength of total shoulder prostheses. Orthopedics, 1988;11: 141-9.
Copyright 2002 by The Journal of Bone and Joint Surgery, Incorporated