Anterolateral rotatory instability of the elbow: a possible etiology of primary osteoarthritis
Article information
Abstract
Background
The purpose of this study is to describe anterolateral rotatory instability (ALRI) as a possible etiology of primary osteoarthritis (OA) of the elbow.
Methods
We examined 76 fresh frozen cadaveric elbows (male:female, 56:20; mean age, 81 years) for patterns of cartilage erosion that could be due to ALRI. These included erosions on the lateral trochlear ridge (LTR) lesion, crescent rim of the radial head (RC) lesion or the ventral capitellum (VC) lesion. The extent and location of the lesions were mapped by image processing of photographs of the humeral and radial articular surfaces, and the degeneration of the articular surface was graded.
Results
Ten of 76 specimens (13%) had one or more lesions consistent with ALRI. LTR lesions were most common and were seen in 10 of 10 specimens (100%), typically involving the distal 30% of the LTR. RC lesions were seen in 9 of 10 and were located on anteromedial crescent of the radial head ranging from 6 to 10 o’clock. VC lesions were seen in 8 of 10 specimens directed anteroinferiorly about 60° to the long axis of the humerus.
Conclusions
ALRI is a possible mechanism initiating primary OA of the elbow. It has a characteristic pattern of triple lesions involving the LTR, the RC, and the VC.
Level of evidence
IV.
INTRODUCTION
Primary osteoarthritis (OA) of the elbow is rare compared with that of other joints, affecting approximately 2% of the population, however it has been reported that the prevalence rate is increased by 10% to 30% in individuals who engage in heavy labor [1,2]. Although the understanding and treatment of the disease has steadily developed, our understanding of the initiation and natural history of the disease is limited. Previous researchers described consistent articular cartilage erosion patterns at specific locations of the humerus and radial head in the initiation of OA of the elbow, and explained that such lesions are due to the aging process or significant joint contact [3-5]. Yet there remains no definitive model that explains how this cluster of degenerative changes are initiated and how they develop to advanced OA of the elbow.
In early osteoarthritic cadaveric specimens, the typical combination of triple cartilage erosions that comprises (1) the ventral top of the capitellum, (2) distal end of the lateral trochlear ridge (LTR), and (3) the beveled rim of the radial head has been well described previously (Fig. 1) [3-6]. Based on the consistent combination and distribution of early cartilage lesions described in several previous studies, we explored the hypothesis that there is a comprehensive and consistent mechanism to initiate and develop lesions, and that they are not due to simple aging or normal significant contact.
We weighted our study towards the specific process of forming abnormal joint contact because we discovered exceptional lesion characteristics that were not explained by existing theories while observing lesions with reference to previous studies. For example, we observed a lesion with extended LTR cartilage damage toward the trochlear surface (Fig. 1B) covered by the coronoid and olecranon alternately throughout the entire flexion-extension arc, suggesting the possibility of a latent pathomechanism such as instability of the joint forming abnormal contact reaching to that area. Previous studies of early cartilage lesions formed in a certain combination at specific locations transitioning to OA did not take into account dynamic factors such as normal elbow motion or instability. Thus, we attempted to elucidate how these lesions develop by combining various motions and positions in cadaveric elbows and finally confirmed the existence of anterolateral rotatory instability (ALRI) of the elbow, which initiates and develops a cluster of typical cartilage erosions in early phase of primary OA of the elbow. ALRI is a pattern of instability in which the proximal radioulnar complex (radial head and ulna) internally rotates as a unit about the long axis of the ulna beyond the normal physiologic limit relative to the humeral articular surface with the elbow in mid-flexion, around 90°.
METHODS
The study was conducted using donated cadavers for research purposes, with approval from the institutional review board and an exemption from the requirement for informed consent.
We studied 76 cadaveric fresh, frozen cadaveric specimens (male:female, 56:20; age, 76–93 years) without radiologic evidence of arthritis. The cadaveric upper arm specimens included the mid-humerus to hand, and were provided for biomechanical tests by Mayo Clinic Anatomy Laboratory. After disarticulation of the elbow joint, the distal forearm was fixed in a clamp with the forearm in full pronation. The elbow was positioned at approximately 90° of flexion by rearticulating the elbow joint. External rotation torque was manually applied to the humerus shaft, allowing medial gapping of the ulnotrochlear joint, and the elbow was examined for points of abnormal cartilage contact due to derailment of the ulnohumeral and radiocapitellar joints. The joint surfaces were examined to localize cartilage lesions and to determine whether or not they were correlated with a mechanism of ALRI (See Supplementary Video for ALRI simulations in cadaveric elbows to reproduce anterolateral rotatory subluxation and the resultant cartilage lesions). Among the 76 joint specimens, 27 specimens showing arthritic changes were initially selected. Of these, 12 specimens with advanced arthritis showing full-thickness cartilage loss dispersed over a wide joint surface for which the mechanism of cartilage loss could not be identified were excluded. Another five specimens with lesions confined to the cartilage layer that did not match the ALRI mechanism defined by the authors (Fig. 2) were also excluded from the study.
Among the 10 specimens finally selected by the ALRI criteria, articular cartilage status was categorized as normal (grade 0) if it was grossly intact, grade 1 if there was fibrillation and/or fissures without loss of cartilage, grade 2 if there was partial thickness cartilage loss without exposure of the subchondral bone, or grade 3 if subchondral bone was exposed due to cartilage loss. Cartilage erosions were localized to: (1) the inferior aspect of the LTR, (2) the ventral surface of the capitellum (VC), and (3) the anteromedial crescent rim of the radial head (RC) (Fig. 1).
For all included specimens with characteristic lesions limited to the cartilage, photos were taken at 0°, 45° and 90° angles to the axis extended from the humeral shaft and vertical to the articular surface of the radial head to document the articular cartilage lesions. All photos were edited semi-transparently and inverted to represent the right elbow with the borders of the lesions marked in Adobe Photoshop (Adobe Inc.). The photos were then superimposed into a single image by resizing them with the anatomical boundaries matched to each other so that the location, size, and incidence of each lesion were represented by color contour maps.
RESULTS
Cartilage erosions including LTR, RC, and VC lesions, consistent with an ALRI mechanism, were observed in 10 of 76 specimens (13%). LTR lesions, involving the inferior one of three of the LTR, were seen in all 10 specimens (100%). In four of 10 specimens, the erosions extended onto the anterior one of three of the LTR. Degenerative changes in this lesion were most severe compared to the other two lesions. This articular cartilage damage was concentrated at the most inferior portion of the LTR, which conforms to the curvature of the radial head at full extension of the elbow, which we termed the “buffer stop zone” (Fig. 3). Of the 10 specimens with LTR lesions, three were grade 2 and seven were grade 3 (Fig. 4).
On the radial heads, RC lesions were seen in nine of 10 specimens (90%). If we imagine the radial head as a clock face with the thumb pointing at the 12, all lesions were located on the anteromedial rim of the articular dish of the radial head surface (“the crescent”) from 6 to 10 o'clock with the forearm in neutral rotation. The lesions of nine specimens were mostly concentrated from 7 to 9 o'clock (Fig. 5). This area of the radial head is directly posterior with the forearm in full pronation, which means that this surface would experience elevated contact pressures and shearing stresses if the radial head subluxated anteriorly with the forearm pronated. Of nine specimens with RC lesions on the radial head, one was grade 1, four were grade 2, and four were grade 3 (Fig. 4).
VC lesions were observed in eight out of 10 specimens (80%). All lesions were located on the VC and varied in extent and shape (Fig. 3). The lesions were most frequently directed anteroinferiorly about 60° to the anterior humeral line parallel to the long axis of the humerus (Fig. 6). Of 8 specimens with VC lesions, one was grade 1, five were grade 2, and two were grade 3 (Fig. 4). Triple lesions were seen in seven of the 10 specimens (70%). Triple lesions refer to the presence of all three cartilage lesions: LTR, RC, and VC.
DISCUSSION
Numerous latent mechanisms might be associated with the occurrence of primary OA of the elbow, but there have been few explanations describing specific pathological mechanisms other than putative hypotheses such as overuse or age-related degeneration of articular cartilage. This study is a preliminary report about ALRI of the elbow based on the visual inspection of early cartilage erosions occurring as clusters at specific locations in the cadaveric elbow, so it could involve subjective assumptions and immature interpretations.
ALRI is a rotatory instability pattern that belongs to the same category of rotatory instability of the elbow as posterolateral rotatory instability (PLRI) [7,8], but shows the opposite case in the direction of rotation of the forearm (pronation vs. supination) and the angle of flexion of the elbow (flexion vs. extension). In contrast to PLRI, which is essentially supination instability of the forearm, ALRI can be defined as pronation instability of the forearm at the elbow joint. Any rotatory instability of the elbow, including ALRI and PLRI, causes the forearm to over-rotate in either direction, which results in derailment of the ulnotrochlear joint accompanied by anterior or posterior subluxation of the radial head on the capitellum depending on the direction of instability. In ALRI, unrestrained over-pronation of the forearm results in internally rotated derailment of the ulnotrochlear articulation combined with anterosuperior subluxation of the radial head producing pathologic contacts between specific sites in the joint leading to initiation of primary OA of the elbow.
In cadaveric elbow specimens with initial cartilage erosion considered as a precursor lesion of primary OA that is not radiologically detected, we noticed cartilage damage in the joint typically showed a triple combination in a cluster: erosion of distal portion of the LTR, ventral surface of capitellum (VC lesion), and anteromedial beveled rim of the radial head crescent (RC lesion). These lesions have been described by previous authors [3-6]. To explain the formation of such early cartilage lesions that precede advanced OA, previous studies suggested hypotheses as individual lesions of early primary arthritis of the elbow without consistent mechanism and interrelation of these lesions. In 1967, Goodfellow et al. [4] studied the pattern of articular cartilage degeneration limited to particular areas in the elbow joint and concluded that the characteristics of movement that occur at the joint surface is the most significant factor in the mechanical environment of its articular cartilage. They believed that the almost inevitable degeneration of the radiohumeral joint in old age was related to the combination of rotation and hinge movements occurring at that joint in contrast with the relative immunity of the ulnohumeral articulation, which has the hinge movement only. They hypothesized that any irregularity of either surface under such complexity of the motion of the radial head (“the movement of a pot scourer”) in radiohumeral articulation would produce a devastating effect upon the other side. On the other hand, they believed that impingement of the posteromedial segment of the radial head rim (identical area to the crescent of radial head in present study) to the posterior part of the crest (identical area to the “buffer stop zone” of the LTR in present study) (Fig. 3) would produce cartilage injuries upon the other side.
Murata et al. [6] also reported the frequency, location, and aspect of the same three lesions in good agreement with our study. They concluded that degenerative changes in the radiohumeral joint were always more advanced than those in the humeroulnar joint and the mode of radial head cartilage degeneration was well correlated with cartilage degeneration in the radiohumeral articulation and OA of the elbow joint. Ahrens et al. [3] also described three distinct lesions with the most frequent degenerative changes in the radiohumeral compartment. But unlike their observation that lesions were concentrated on the posteromedial quadrant of the radial head, our study revealed the preponderance of lesions on the anteromedial rim of the radial head (Fig. 5). Interestingly, they also referred to degenerative changes on the lateral facet of the olecranon on occasion.
In the most recent study related to this issue, Jeon et al. [5] reintroduced the term “zona conoidea,” which refers to the lateral slope of the LTR that forms significant contact with the posteromedial rim of the radial head in supination as a factor initiating elbow OA. We agree in part with their hypotheses and findings, but with more detailed observation, we had noticed that pronation results in tighter fit contact between the zona conoidea and the medial rim of the radial head. As such, previous studies have described the combination of lesions characteristic of early degenerative arthritis in a consistent manner, but our more detailed observations have revealed that quite exceptional lesions exist. For example, the variant form of the LTR lesion (Fig. 1B) showing that cartilage erosion extended to the trochlear joint surface, which is always alternately covered by the coronoid and olecranon in the entire flexion-extension arc, could not be explained by the normal joint contact mechanisms on which the previous studies are based. In addition to this paradoxical lesion characteristic, the combination of site-specific lesions observed in our sample strongly suggests that the contacts in the joints come into abnormal loading during some phase of motion under specific circumstances rather than the aging process or the significant contact described in previous studies [3-6]. Through a series of simulations that reproduced normal and abnormal contact by combining various elbow flexion angles and forearm rotation angles on the cadaveric elbow specimens, we discovered a unique instability that we defined as ALRI, which consistently integrated multiple cartilage lesions into a single mechanism (Fig. 2).
Although the number of specimens involved in the present study is small, the frequency and severity of lesions (Fig. 6) can be used to determine the order of lesion formation. Of the 10 specimens, we observed degenerative changes of grade 1 or more among 8 VC lesions, 9 RC lesions, and 10 LTR lesions. This order of frequency and severity is consistent with the findings of Murata et al. [6], indicating the order of development of these lesions. Because we included only specimens with early cartilage erosions transitioning to clinical OA, we did not observe any significant cartilage damage of the olecranon of coronoid. However, when the ulnotrochlear derailment was intensified and the subchondral bone in the LTR was widely exposed, we observed that another lesion could develop on the lateral facet of the olecranon, and such lesions were previously reported by Ahrens et al. [3] in addition to cartilage lesions identical to our triple lesions.
Along with this discovery, we investigated potential contributing factors to the initiation and development of ALRI. As biomechanical or anatomical key factors that are believed to contribute to the onset and development of ALRI, we concluded that, at present, four major elements interact. First, in order to understand the pathology of ALRI, specific spatial configuration of the pronation in the forearm should be considered. The pronation of the forearm occurs when the radius overrides the ulna [9], forming a lever system in which the radial shaft is the lever on the ulnar fulcrum. Assuming the position at 90° of elbow flexion with fully pronated forearm which is feasible for ALRI forming the lever system in the forearm, when the forearm is forced to turn into pronation further or weight in the hand is increased, the lever system will tend to dislocate the radial head upward.
Second, we hypothesize that the biomechanics of the biceps brachii are a dynamic factor that initiates and rapidly aggravates ALRI. The biceps muscle is a strong supinator of the forearm and a flexor of the elbow, but the effect on the joints, especially on the radiocapitellar joint, varies significantly depending on the degree of elbow flexion. When the elbow is extended, it compresses the radial head to the capitellum stabilizing the articulation, but when it reaches 90° flexion or more, it loses compressive effect gradually and acts only as a force to pull the radial head upward [10]. When the elbow is at 90° of flexion, the moment arm of the biceps muscle reaches its maximum value [11,12]. What should be considered more important in terms of joint stability in this situation is that there is no dynamic element that antagonizes the action of the strong biceps destabilizing the radiocapitellar articulation except a weak supinator. In addition, when the biceps brachii lever system is considered, it is evident that unopposed biceps pull on the radial head without a dynamic antagonist will increase when there are loads such as racquet or dumbbell in hand [10,12]. Further study will be needed to biomechanically verify this model.
Third, medial gapping of the ulnotrochlear joint (Fig. 2E) is a prominent phenomenon in ALRI and is an essential factor in ulnotrochlear joint derailment caused by over-pronation of the forearm complex. In order for medial gapping of the ulnotrochlear joint to occur in ALRI, we hypothesized that the restrained function of the posterior bundle of the medial collateral ligament (PMCL) might be completely lost. This hypothesis could be supported by the normal biomechanical observation that the PMCL is the most tightly strained at around 90º flexion where ALRI happens. The maximum tension of the PMCL can be achieved by end-pronation at the elbow with flexion over 90° [13,14], so that the inherent function of the PMCL could be declared to be a major restraint resisting against pronation at the elbow when flexed to 90° or more. Therefore, loss of this function of the PMCL can predict the increase in the internal rotation (pronation) of the radioulnar complex with medial gapping of ulnotrochlear joint in elbow flexion above 90°, which is the same as the ALRI phenomenon. Numerous researchers have repeatedly emphasized the anatomical and biomechanical importance of the PMCL, which is consistent with our hypothesis. Bellato et al. [15] reported that medial gapping caused PMCL injury in a posteromedial rotatory instability (PMRI) injury model accompanied with coronoid process fracture. Pollock et al. [16] stated that isolated sectioning of the PMCL may not be completely benign and may contribute to varus and rotation instability of the elbow. Golan et al. [17] also observed that isolated transection of the PMCL resulted in PMRI represented by medial gapping of the ulnotrochlear joint, and Gluck et al. [18] experimentally showed that reconstruction of the PMCL may confer some stability to recover the joint gapping.
Lastly, although not to the point of high-degree rupture causing PLRI [7], some degree of relaxation due to degenerative effacement or stretch on the lateral collateral ligament (LCL) complex of the elbow, including the lateral ulnar collateral ligament, may also be necessary to allow abnormal anterosuperior displacement of the radial head. The degree of ligament involvement was not determined in the present study because of the paucity of lesions involved. Some of the specimens with established OA revealed varying degrees of degenerative changes of the LCL complex, which suggests it is a component of recalcitrant lateral elbow pain such as tennis elbow or any latent instability [19,20]. When ligaments reach states of irreversible insufficiency, overt ALRI forming pathologic contact in cluster might be induced, and eventually develop into advanced degenerative arthritis of the elbow over time (Figs. 7 and 8). This hypothesis is a rational speculation based on visual inspection of the ligaments and spatial configuration in the case of subluxation of joints due to ALRI, but it needs to be explored through biomechanical studies in the future.
It is difficult to generalize the prevalence of ALRI over the entire age range based on this cadaveric study. Nevertheless, the prevalence rate of 13% (10/76) according to strict criteria is high, and might be even higher given that specimens were excluded from the study due to advanced arthritis. Although known other joint instabilities are associated with arthritis in different ways [8,21], a number of unknown factors still remain latent. We anticipate that ALRI is prevalent in sports players or manual workers [22] using repetitive strong and fast pronation with a racquet, ball, or hammer in the hand. In addition, the mechanism underlying the VC lesion of ALRI shows the existence of shearing force sweeping on the capitellum, and may explain osteochondritis dissecans lesions in the elbows, which occur in the same location as VC lesions.
Through this observational study, we elucidated the role of the PMCL in terms of elbow stability, and hypothesized that osteophytes will form more prominently around the ulnotrochlear joints [23] to compensate for instability in the process of extending OA from the radial side to the ulnar side [3,4,24]. Based on this study, in arthroscopic osteocapsular arthroplasty in OA patients [1,2], osteophytes that restrict the range of flexion-extension should be excised, but in areas resisting pronation, such as the medial lip of anteromedial coronoid or the lateral lip of the anterolateral olecranon, osteophytes should be preserved for compensation against any kind of rotatory instability that could derail the ulnotrochlear joint. However, we believe that this issue needs to be addressed in future clinical trials and more data should be obtained.
Since this study is based on simple observation of the cadaveric specimens of elderly persons, considerable selection bias is possible, and potential interpretations and clinical application should remain cautious. This study revealed that separate articular cartilage erosions previously reported by several researchers consistently occur as one mechanism. However, at present, the study comprises immature interpretations at the level of assumption or hypothesis, so it is meant to provide clues for subsequent research, and active refutations and corrections through various studies are awaited until clinical applications.
CONCLUSIONS
ALRI of the elbow is a pathomechanism that comprises rotatory instability, in which the radioulnar joint surface is subluxated in a unit as the forearm complex pronates excessively with respect to the joint surface of the humerus. This phenomenon explains why several early cartilage erosions reported in elbow joints are not independent lesions but are formed in a cluster with particular wear patterns at specific locations by one consistent pathomechanism, suggesting that it explains the etiology of morbidity of primary arthritis of the elbow.
Notes
Author contributions
Conceptualization: YBK, JSF, EB, SWO, HSJ, DWK. Data curation: YBK, JSF, EB, HSJ, DWK. Formal analysis: YBK, JSF, HSJ, DWK. Methodology: YBK, EB, SWO, HSJ, DWK. Project administration: YBK. Software: YBK, DWK. Supervision: YBK, JSF, SWO, DWK. Validation: YBK, SWO, DWK. Visualization: YBK, DWK. Writing – original draft: YBK, EB. Writing – review & editing: YBK, JSF, EB, SWO, HSJ, DWK.
Conflict of interest
None.
Funding
This work was supported by a grant from the research year program of Inje University in 2015.
Data availability
Contact the corresponding author for data availability.
Acknowledgments
We would like to express our gratitude to Dentium (Seoul, Korea) for their assistance with the reconstruction of the finite element models and virtual surgery using MIMICS Research 22.0, 3-matic 14, NRecon, and Solidworks 2019. Additionally, we would like to thank Tae Sung S&E (Seoul, Korea) for their technical support with the operation of Ansys software, and TDM (Seoul, Korea) for providing the three-dimensional computer-aided design (CAD) file of a cannulated screw with a 6.5 thread diameter.
This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (No. HI22C0494), and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. 2022R1G1A1003299). No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found via https://doi.org/10.5397/cise.2024.00416.