**1. Introduction**

576 Recent Advances in Arthroplasty

[27] Tugwell P, Boers M. OMERACT conference on outcome measures in rheumatoid

Despite improvements in glenoid prosthesis design, materials and surgical techniques, complications related to the glenoid component continue to be a leading cause of failure after total shoulder arthroplasty. Although previously felt to be of little clinical significance, radiolucent lines around the glenoid prosthesis are now recognized as a sign of impending mid and long term fixation problems. While much attention in shoulder prosthesis design has focused on anatomical reconstruction of the humerus through increasing modularity, comparatively little progress has been made in solving the problems of glenoid wear and fixation failure. Resolving these issues on the socket side of the equation remains a challenge for the shoulder arthroplasty surgeon as the population ages, as young patients present with terminal shoulder arthritis and as patients demand higher performance from their implant.

Fig. 1. In the normal glenohumeral joint, the compliance of the articular cartilage and the labrum provide for conforming surfaces which optimize stability and load distribution.

Humeral Hemiarthroplasty with Spherical Glenoid

survivorship data for hip and knee arthroplasty.

Reaming: Theory and Technique of The Ream and Run Procedure 579

Several studies have looked at glenoid prosthesis design parameters and at cementation and bone preparation techniques seeking to optimize fixation to the host bone and resist tensile stresses imparted by eccentric loading conditions.8-10 There is general consensus that pegged glenoid components tend to outperform keeled components and that third generation cementation techniques have reduced the incidence of early radiolucent lines. While most outcome studies show survivorship of the prosthesis at midterm, clinical outcome studies demonstrate deterioration in glenoid component longevity as the time interval from surgery increases. Walch and colleagues in a multicenter study of 333 shoulder arthroplasties performed with the same cemented convex-backed glenoid component showed that nearly 50% were radiographically loose at 10 years.11 The need to eccentrically ream the glenoid to correct posterior erosion was associated with a higher rate of loosening. Young et al performed a similar multicenter study looking at long-term survivorship of a cemented flatbacked, keeled all-polyethylene glenoids and found radiographic loosening of 20% at 10 years and 67% at 15 years.12 Kasten et al similarly noted at 9% rate of loosening at 5 years, which increased to 33% at 9 years. Radiolucent lines were noted to progress over time.13 As with arthroplasty of other joints that rely on cement fixation, progressive radiolucency is an "at risk" sign for eventual loosening and though mid-term data show little correlation between lucency and revision, it is expected that longer follow-up will bear out eventual clinical failure. Bartelt et al followed a cohort of patients younger than age 55 years who underwent total shoulder arthroplasty and found a 30% rate of radiographic loosening at a

mean of 6.6 years.14 The implant survival rate in this group was only 92% at 10 years.

In light of these sobering data, there have been historical and resurging attempts to improve fixation durability by capitalizing on the proven success of bone ingrowth into porous metal surfaces. However, experience with early metal-backed designs demonstrated failure rates much higher than for cemented polyethylene components.15 Taunton et al reported on revision or radiographic failure for a metal-backed, bone-growth glenoid of 20% at 5 years and 48% at 10 years.16 Unacceptably high rates of radiographic failure have also been reported for cemented metal-backed glenoid components.17 Although some cases of fibrous ingrowth have been documented, many of these failures were not related to fixation problems between the metal and the bone, rather to excessive polyethylene wear, locking mechanism problems and severe osteolysis.18 Tensile stress seems to be transferred to the fixation between the polyethylene and metal baseplates leading to the potential for excessive wear, fracture and dissociation of the components. While much of this experience is based on implants used in the 1980's to 1990's, there is renewed interest in metal-backed designs given the success of metal base plates in reverse shoulder arthroplasty and the desire to have modularity on the glenoid side facilitating conversion between primary and reverse shoulder replacement. Clinical outcome studies on newer designs have yet to prove their merit versus their historical counterparts and the current standard set by cemented all polyethylene designs, albeit a fair standard compared to

These data are alarming considering the growing epidemic of degenerative conditions in younger patients, the higher expectations of patients who wish to remain physically active into their older years and the general aging of the population. The conclusion one can draw from the current clinical and basic science literature is that modern prosthetic glenoid components are destined to fail by wear and loosening with repeated eccentric loading. Thus, for younger patients or those whose life-expectancy is greater than 15 years there is a

In the normal shoulder, the compliance of the articular cartilage and labrum allow the mating surfaces to conform under applied loads throughout a wide range of motion (Figure 1). Thus, normal, physiological glenohumeral translations can occur between congruent surfaces without introducing a kinematic conflict caused by excessive constraint. Because polyethylene glenoids are not compliant, they cannot instantly deform to remain congruent as normal translations occur. To resolve the potential conflict between conformity and constraint, total shoulder systems have introduced mismatch between the radii of curvature of the head and glenoid to permit small translations that increase range of motion and resolve some of the stresses transmitted to the fixation interface.

While mismatch has proven to reduce earlier fixation failure, it also facilitates eccentric loading by allowing the center of rotation of the humerus to translate away form the center of the glenoid concavity. As the contact point on the prosthetic glenoid surface changes, there are corresponding marked changes in the cement mantle stress than ultimately result in micromotion at the bone cement interface.1 Oosterom and colleagues performed biomechanical studies looking at the effect of varying degrees of conformity on rim loading and found that mismatch increases rim displacement.2 Furthermore, mismatch results in abrasive wear at the prosthetic surface and a higher likelihood of material failure under eccentric loading conditions.3 Retrieval studies of failed glenoids that have radial mismatch, have repeatedly shown characteristic changes in the prosthetic surface including edge deformation, rim fracture, broad surface irregularity and wear to conformity (Figure 2).4,5

Fig. 2. Wear to conformity (left), broad surface abrasion and rim wear (right).

Modes of damage have been found most commonly in the inferior quadrant suggesting calcar impingement with the humerus.6 Braman et al further showed that alterations in the surface geometry of the damaged prosthesis compromised its intrinsic stability, thus exacerbating the pathomechanics that result in surface damage to begin with.4

In the normal shoulder, the compliance of the articular cartilage and labrum allow the mating surfaces to conform under applied loads throughout a wide range of motion (Figure 1). Thus, normal, physiological glenohumeral translations can occur between congruent surfaces without introducing a kinematic conflict caused by excessive constraint. Because polyethylene glenoids are not compliant, they cannot instantly deform to remain congruent as normal translations occur. To resolve the potential conflict between conformity and constraint, total shoulder systems have introduced mismatch between the radii of curvature of the head and glenoid to permit small translations that increase range of motion and

While mismatch has proven to reduce earlier fixation failure, it also facilitates eccentric loading by allowing the center of rotation of the humerus to translate away form the center of the glenoid concavity. As the contact point on the prosthetic glenoid surface changes, there are corresponding marked changes in the cement mantle stress than ultimately result in micromotion at the bone cement interface.1 Oosterom and colleagues performed biomechanical studies looking at the effect of varying degrees of conformity on rim loading and found that mismatch increases rim displacement.2 Furthermore, mismatch results in abrasive wear at the prosthetic surface and a higher likelihood of material failure under eccentric loading conditions.3 Retrieval studies of failed glenoids that have radial mismatch, have repeatedly shown characteristic changes in the prosthetic surface including edge deformation, rim fracture, broad surface irregularity and wear to conformity (Figure 2).4,5

resolve some of the stresses transmitted to the fixation interface.

Fig. 2. Wear to conformity (left), broad surface abrasion and rim wear (right).

exacerbating the pathomechanics that result in surface damage to begin with.4

Modes of damage have been found most commonly in the inferior quadrant suggesting calcar impingement with the humerus.6 Braman et al further showed that alterations in the surface geometry of the damaged prosthesis compromised its intrinsic stability, thus Several studies have looked at glenoid prosthesis design parameters and at cementation and bone preparation techniques seeking to optimize fixation to the host bone and resist tensile stresses imparted by eccentric loading conditions.8-10 There is general consensus that pegged glenoid components tend to outperform keeled components and that third generation cementation techniques have reduced the incidence of early radiolucent lines. While most outcome studies show survivorship of the prosthesis at midterm, clinical outcome studies demonstrate deterioration in glenoid component longevity as the time interval from surgery increases. Walch and colleagues in a multicenter study of 333 shoulder arthroplasties performed with the same cemented convex-backed glenoid component showed that nearly 50% were radiographically loose at 10 years.11 The need to eccentrically ream the glenoid to correct posterior erosion was associated with a higher rate of loosening. Young et al performed a similar multicenter study looking at long-term survivorship of a cemented flatbacked, keeled all-polyethylene glenoids and found radiographic loosening of 20% at 10 years and 67% at 15 years.12 Kasten et al similarly noted at 9% rate of loosening at 5 years, which increased to 33% at 9 years. Radiolucent lines were noted to progress over time.13 As with arthroplasty of other joints that rely on cement fixation, progressive radiolucency is an "at risk" sign for eventual loosening and though mid-term data show little correlation between lucency and revision, it is expected that longer follow-up will bear out eventual clinical failure. Bartelt et al followed a cohort of patients younger than age 55 years who underwent total shoulder arthroplasty and found a 30% rate of radiographic loosening at a mean of 6.6 years.14 The implant survival rate in this group was only 92% at 10 years.

In light of these sobering data, there have been historical and resurging attempts to improve fixation durability by capitalizing on the proven success of bone ingrowth into porous metal surfaces. However, experience with early metal-backed designs demonstrated failure rates much higher than for cemented polyethylene components.15 Taunton et al reported on revision or radiographic failure for a metal-backed, bone-growth glenoid of 20% at 5 years and 48% at 10 years.16 Unacceptably high rates of radiographic failure have also been reported for cemented metal-backed glenoid components.17 Although some cases of fibrous ingrowth have been documented, many of these failures were not related to fixation problems between the metal and the bone, rather to excessive polyethylene wear, locking mechanism problems and severe osteolysis.18 Tensile stress seems to be transferred to the fixation between the polyethylene and metal baseplates leading to the potential for excessive wear, fracture and dissociation of the components. While much of this experience is based on implants used in the 1980's to 1990's, there is renewed interest in metal-backed designs given the success of metal base plates in reverse shoulder arthroplasty and the desire to have modularity on the glenoid side facilitating conversion between primary and reverse shoulder replacement. Clinical outcome studies on newer designs have yet to prove their merit versus their historical counterparts and the current standard set by cemented all polyethylene designs, albeit a fair standard compared to survivorship data for hip and knee arthroplasty.

These data are alarming considering the growing epidemic of degenerative conditions in younger patients, the higher expectations of patients who wish to remain physically active into their older years and the general aging of the population. The conclusion one can draw from the current clinical and basic science literature is that modern prosthetic glenoid components are destined to fail by wear and loosening with repeated eccentric loading. Thus, for younger patients or those whose life-expectancy is greater than 15 years there is a

Humeral Hemiarthroplasty with Spherical Glenoid

**2. Basic science and clinical support** 

technique, recovery and results.

with a convex metallic prosthesis.

articulating humeral hemiarthroplasty.

properties these relationships engender.

its ability to withstand its mechanical environment.

Reaming: Theory and Technique of The Ream and Run Procedure 581

rigors in which many of these patients wish to engage. In 1992, Frederick A. Matsen, III, MD began investigating the technique of humeral hemiarthroplasty with spherical glenoid reaming to restore a concavity to the glenoid and to reorient the worn glenoid perpendicular to the centerline of the scapular body. This technique has hence become known as the Ream and Run procedure. As follows is an in-depth description of basic science and clinical support for this technique, the principles of its application, patient selection, surgical

According to Matsen, "glenoid components fail as a result of their inability to replicate essential properties of the normal glenoid articular surface to achieve durable fixation to the underlying bone, to withstand repeated eccentric loads and glenohumeral translation, and to resist wear and deformation."37 The Ream and Run seeks to address these deficiencies by stimulating a biological response at the glenoid surface that can adapt to the applied stress through the process of healing and remodeling. Interest in this potential came from observations on retrieval studies of mold arthroplasty of the hip. Observations on this historical technique showed that the reamed acetabular bone was often covered with a smooth fibrous tissue layer that amounted to a biological resurfacing.38 In addition, histologic studies demonstrated a relatively normal subchondral bony architecture that had remodeled according to the loads born on the surface.39, 40 Failure of this technique was often due to loosening and bone resorption on the femoral side.41 These results suggest that reamed bone has a regenerative potential to yield a durable joint surface when articulating

Laboratory studies were then undertaken to determine if the reamed glenoid concavity was comparatively stable to either the native or a prosthetic glenoid. Weldon et al, using a cadaveric model, demonstrated that the intrinsic glenoid stability was compromised by loss of articular cartilage and that this stability could be restored to levels comparable to a prosthetic glenoid through spherical reaming.42 In other words, the surface geometry of the bone predicts its influence on glenohumeral kinematics. To further characterize the healing process that occurs at the reamed glenoid surface, Matsen and colleagues performed histologic analysis of retrieved glenoids at serial follow-up intervals in a canine model of the Ream and Run.43 At 24 weeks post-surgery specimens consistently showed growth of a thick fibrocartilaginous tissue covering and firmly attached to the glenoid surface.(Figure 3) This progressive maturation between 3 and 6 months and remained congruent with the

The implication of these findings are: 1) healing and remodeling of the reamed bone is a progressive process demonstrating continued biological activity in response to the mechanical environment; and 2) progressive maturation of the regenerative surface suggests

Based on this background, the Ream and Run procedure been in clinical application now for well over a decade, and its indications and techniques have been refined with increasing experience. As the length of follow-up continues for these patients, the foundations for sustained positive outcomes and the modes of failure have become clearer. These foundations are anchored in the principles behind this technique all of which relate to replicating the anatomical relationships of the glenohumeral joint and the biomechanical

high likelihood that revision surgery will be necessary to address glenoid component failure. In many of these cases, reimplantation of a glenoid component is not possible due to structural deficiency of the remaining glenoid vault, and functional outcomes are often uncertain.19,20 Studies on structural grafting of bone deficits have shown a high rate of shortterm reconstitution but long-term subsidence indicating that durable and reliable solutions for the failed glenoid are not yet available for this growing cohort of patients.21,22

To address the need for surgical options to treat shoulder arthritis in younger patients, biological resurfacing procedures using a number of different interposition materials have been evaluated with largely varying short and mid-term success. Autogenous fascia lata, Achilles tendon23-24 and glenohumeral capsule,25,26 allograft lateral meniscus27,28 and dermal scaffolds,29 and xenographic tissues patches30 have all been used to resurface the worn glenoid. Both arthroscopic and open techniques have been reported and techniques have included humeral chondroplasty, humeral prosthetic surface replacement versus stemmed humeral hemiarthroplasty.23 While some of these reports demonstrate initial improvement in pain and function, progressive joint space narrowing and glenoid erosion are common and consequent eventual revision to definitive arthroplasty. Gerber has nicely summarized the literature on this field, stating "Biologic resurfacing of the glenoid has hitherto failed to adequately restore the geometry and biology of the glenoid."31 It is fair to say that lesser invasive approaches such as the arthroscopic techniques can be used as an interval step to delay arthroplasty but durable long-term results from these procedures seem to be the exception rather than the rule.

Hemiarthroplasty without glenoid resurfacing or reshaping is yet another alternative to total shoulder replacement which avoids the risk of glenoid failure. There has been extensive comparison between hemiarthroplasty and total shoulder arthroplasty in the literature looking at comparative outcomes. It is fairly well-established based on these studies including a meta-analysis of the existing literature, that total shoulder arthroplasty provides superior pain relief and range of motion over time.32-34 Nevertheless, some series do show comparable outcomes recognizing that some patients with progressive glenoid erosion do require conversion to total shoulder arthroplasty.35 Levine et al have shown that results of hemiarthroplasty are inferior if preoperative glenoid erosion or posterior wear exists.36 This highlights the importance of recentering the humeral head and restoring proper load-bearing mechanics at the glenohumeral joint after prosthetic reconstruction. Sperling and colleagues studied long-term results of hemiarthroplasty versus total shoulder in patients 50 years or younger.33 While glenoid wear after hemiarthroplasty was present in 72% of cases, radiolucencies around the glenoid prosthesis were present in 76% of patients. The risk of painful glenoid erosion necessitating revision glenoid replacement lead to the conclusion that patients with total shoulder replacement have superior clinical outcomes. The authors also concluded however based on survivorship of total shoulders in this cohort that "great care must be exercised, and alternative methods of treatment considered before either hemiarthroplasty or total shoulder arthroplasty is offered to patients aged 50 years or younger."

Based on this background, a definitive treatment option for young and physically demanding patients with end-stage shoulder arthritis remains both a need and a challenge. Experience with the failure modes of both hemiarthroplasty and total shoulder arthroplasty, along with a better understanding of glenohumeral biomechanics have laid a foundation on which such a treatment must be based to provided a lasting solution that withstands the rigors in which many of these patients wish to engage. In 1992, Frederick A. Matsen, III, MD began investigating the technique of humeral hemiarthroplasty with spherical glenoid reaming to restore a concavity to the glenoid and to reorient the worn glenoid perpendicular to the centerline of the scapular body. This technique has hence become known as the Ream and Run procedure. As follows is an in-depth description of basic science and clinical support for this technique, the principles of its application, patient selection, surgical technique, recovery and results.
