Thumb Radial Collateral Ligament Injuries Show Greater Rates of Metacarpophalangeal Subluxation Compared With Thumb Ulnar Collateral Ligament Injuries: A Comparative Imaging Analysis

Abstract

Background:

The ulnar and radial collateral ligaments (UCL and RCL) of the thumb are essential lateral stabilizers of the thumb metacarpophalangeal (MCP) joint. Despite anatomical and biomechanical findings suggesting a higher rate of MCP joint subluxation after RCL injury compared with that of UCL injury, this has not been clinically substantiated. This study examines and compares imaging of the thumb MCP joint in patients with operative RCL versus UCL injuries of the thumb.

Methods:

One hundred seventy-five operative thumb injuries were retrospectively reviewed, consisting of 131 UCL injuries, 40 RCL injuries, and 4 concomitant UCL and RCL injuries. Preoperative radiographs and magnetic resonance imaging (MRI) images were assessed for volar MCP joint subluxation, classified by severity based on percentage subluxation from grade I to IV.

Results:

On radiographs, the mean percent subluxation was 16.6 ± 10.2% for UCL injuries compared with 35.5 ± 16.2% for RCL injuries. On MRI, the mean percent subluxation was 30.4 ± 16.5% for UCL injuries compared with 43.8 ± 18.8% for RCL injuries. On radiographs, 0.8% of UCL injuries were classified as grade III or IV subluxation, while 22% of RCL injuries were classified as grade III or IV subluxation. On MRI, 11.6% of UCL injuries were classified as grade III or IV subluxation, while 23.5% of RCL injuries were classified as grade III or IV subluxation.

Conclusions:

Radial collateral ligament injuries show significantly higher volar subluxation of the MCP joint compared with UCL injuries on both plain radiographs and MRI. This has implications for operative indications of complete RCL injuries to help restore joint congruity.

Introduction

The ulnar collateral ligament (UCL) and radial collateral ligament (RCL) of the thumb are essential lateral stabilizers of the thumb metacarpophalangeal (MCP) joint. ¹ Injuries to these collateral ligaments can result in pain, instability, and diminished pinch and grip strength, comprising up to 86% of all injuries to the thumb.^{2 -5} Radial collateral ligament injuries are less common than UCL injuries, accounting for 10% to 42% of collateral ligament injuries, but are more likely to require surgical intervention.^6,7 While RCL injuries usually result from sudden thumb MCP joint adduction (ulnarly deviated force) or axial loading, UCL injuries typically result from sudden thumb MCP joint abduction and/or extension (radially deviated force).^5,8

Incomplete tears are generally treated conservatively using a cast or splint immobilization. ⁵ For complete tears, the treatment course typically involves surgical intervention; however, for complete RCL ruptures specifically, this has been debated.^5,6 In the past, conservative management was recommended, as, in contrast to UCL injuries, there is usually no interposed aponeurosis that would prevent RCL healing. ⁹ However, other anatomic differences exist between the radial and ulnar sides of the MCP joint that could explain why surgery so commonly becomes necessary. Disruption of the RCL leads to unopposed dynamic force of the adductor pollicis and flexor pollicis longus tendons. In this setting, the MCP joint is prone to a sequence of ligamentous laxity, joint instability, volar and ulnar subluxation, and degenerative joint disease. ¹ Because of this, many surgeons now recommend surgical intervention for grade III RCL tears, especially when there is radiographic evidence of MCP joint subluxation. ⁶ Even in chronic injuries, it has been shown that surgical intervention of the RCL can reduce pain, improve strength, and allow earlier return to activities. ¹⁰ It has also been shown that earlier treatment leads to better outcomes.^{10 -12}

Despite the anatomical and biomechanical findings that suggest a higher rate of MCP joint subluxation after RCL injury compared with that of UCL injury, no study has been conducted to substantiate that this occurs clinically. The purpose of this study is to examine and compare the thumb MCP joint subluxation in patients with RCL versus UCL injuries of the thumb who ultimately underwent operative management. We hypothesize that RCL injuries will have greater rates of MCP joint subluxation compared with UCL injuries.

Materials and Methods

Patient Selection

This project was performed under an exemption for informed consent from the institutional review board. Patients with collateral ligament injuries were identified via Current Procedural Terminology (CPT) procedure codes 26540 (Primary repair of collateral ligament, metacarpophalangeal joint), 26541 (Primary repair of collateral ligament, metacarpophalangeal joint; with tendon or fascial graft), and 26542 (Primary repair of collateral ligament, metacarpophalangeal joint; with local tissue [eg, adductor advancement]) from February 1, 2014, to February 1, 2024. All patients were seen at a single academic institution by 1 of 5 fellowship-trained hand surgeons. Intraoperative confirmation of collateral ligament injury was required for inclusion.

Charts were reviewed, and patients were included if they had an injury of the thumb RCL, UCL, or concomitant thumb RCL and UCL injuries. Exclusion criteria included the presence of other hand or wrist pathology that could alter the appearance of the first MCP joint on imaging. Only radiographs with proper positioning (minimal rotation and appropriate alignment) and sufficient image quality to visualize anatomical landmarks necessary for our measurements were included. A priori power analysis was not conducted as this cohort comprises the complete operative population from our institution over the study’s timeframe aside from those that did not meet the inclusion and exclusion criteria.

All data were collected from existing preoperative and postoperative notes found in the patients’ electronic health record and preoperative imaging. Preoperative radiographic and magnetic resonance imaging (MRI) parameters were assessed by 2 authors (J.F.H. and C.M.W.).

Measuring Thumb MCP Subluxation

Radiographs were assessed on posterior-anterior (PA) view (if hand radiograph) or oblique view (if thumb radiograph). Magnetic resonance imaging images were assessed on T1-weighted sagittal cuts. To assess subluxation of the thumb MCP joint, 2 lines were created along the center axes of the thumb metacarpal and proximal phalanx. The distance between these 2 axes at the center of rotation of the thumb MCP joint, where the UCL originates, was measured (Figure 1). In addition, the angle between these 2 axes was measured (Figure 2). Finally, the articular surface of the proximal phalanx of the thumb was measured, and the amount of dorsal “gap” secondary to proximal phalanx volar subluxation was measured (Figure 3). These measurements were divided, giving the percentage of subluxation of the thumb MCP joint.

Figure 1.

Illustration of the distance between the center axes of the thumb metacarpal and proximal phalanx measured at the center of rotation of the thumb metacarpophalangeal joint where the ulnar collateral ligament originates.

Note. Yellow—center axis of proximal phalanx, orange—center axis of metacarpal, and red—distance between central axes, measured at the center of rotation of the metacarpal head.

Figure 2.

Illustration of the angle measured between the center axes of the thumb metacarpal and proximal phalanx. Yellow—center axis of proximal phalanx, orange—center axis of metacarpal, and red—angle between central axes.

Figure 3.

Illustration of how the percentage of subluxation of the thumb metacarpophalangeal joint was measured.

Note. The dorsal “gap” secondary to proximal phalanx volar subluxation was measured and divided by the articular surface of the proximal phalanx of the thumb. Yellow—dorsal articular edge of metacarpal, orange—dorsal articular edge of phalanx, red—amount of subluxation, and green—proximal articular surface of P1.

Classification of Severity of Subluxation

As a classification system to grade subluxation of the thumb MCP joint has not been established, we modified the classification system used by Steinmetz et al ¹³ for the thumb carpometacarpal (CMC) joint to use to grade subluxation of the thumb MCP joint. We defined grade I subluxation as 0% to 25% subluxation, grade II subluxation as 25% to 50% subluxation, grade III subluxation as 0% to 75% subluxation, and grade IV subluxation as 75% to 100% subluxation.

Statistical Analysis

Descriptive statistics were performed to analyze patient demographics and follow-up period. One-way analysis of variance (ANOVA) with post hoc Tukey testing was performed to assess for differences in measurements of subluxation on MRI and radiographs between thumb RCL, UCL, and concomitant RCL and UCL injuries. A P value less than .05 was considered statistically significant.

Results

In total, 332 CPT codes were generated and reviewed, which revealed 211 injuries to the thumb UCL, RCL, or both. The remaining cases were injuries to digits other than the thumb. Thirty-six patients did not have imaging available for review or were not of sufficient quality to be included. and were excluded. Of the 175 included, 131 were UCL injuries, 40 were RCL injuries, and 4 were concomitant RCL and UCL injuries.

Demographics

One hundred twenty-one injuries were in men and 54 were in women. Average overall age at the time of surgery was 36.05 ± 14.63 years. Eighty-eight cases involved the right hand, and 87 cases involved the left hand. One hundred fifty-six patients were right-hand dominant, 17 were left-hand dominant, and 2 were ambidextrous. Fifteen patients were current smokers, 12 were former smokers, and 148 were never-smokers. Between groups, there was a significant difference in age at the time of surgery and sex. There were no significant differences in hand dominance, smoking status, and workers’ compensation status (Table 1).

Table 1.

Demographics of Included Patients.

Demographics	UCL (n = 131)	RCL (n = 40)	Both (n = 4)	P
Mean age at surgery, y	36.73 (14.81)	32.43 (13.63)	50.00 (6.78)	.041*
Gender
Male	85	34	2	.039*
Female	46	6	2
Hand dominance
Right	120	32	4	.140
Left	9	8	0
Ambidextrous	2	0	0
Smoking status
Never	111	34	2	.172
Current or former	20	6	2
Workers’ compensation status
Yes	20	30	3	.343
No	111	10	1

Note. This table shows demographic information of included patients. UCL, isolated UCL injury; RCL, isolated RCL injury; both, concomitant RCL and UCL injury. UCL = ulnar collateral ligament; RCL = radial collateral ligament.

^*Indicates a significant difference between groups on 1-way ANOVA or χ² test for quantitative and qualitative variables, respectively.

Radiographic Subluxation Measures

Of 175 cases, 163 had radiographs available for review. Percent MCP subluxation, center axis distance, and angle for each group were measured (Table 2). The ANOVA showed significant differences between groups for all 3 measurements (P < .001). For percent MCP subluxation, post hoc testing revealed significant differences between percent subluxation in UCL and RCL (P < .001) as well as between UCL and concomitant UCL and RCL injury (P = .002). No difference in percent subluxation was found in the RCL-only cohort compared with the concomitant RCL and UCL injury cohort (P = .973). Percent MCP subluxation measurements were further stratified by severity using a 4-tiered grading system as described in the methods section (Table 3).

Table 2.

Measures of Thumb MCP Joint Subluxation by Ligaments Injured.

	UCL	RCL	UCL + RCL	P
Radiographs
Percent subluxation	16.6 (10.2)	35.5 (14.7)	36.8 (16.2)	<.001*
Distance between center axes at center of rotation, mm	1.32 (1.05)	2.42 (1.39)	2.13 (1.09)	<.001*
Angle between center axes, °	9.04 (6.59)	14.54 (9.29)	15.73 (3.74)	<.001*
MRI
Percent subluxation	30.4 (16.5)	43.8 (18.8)	44.3 (14.3)	<.001*
Distance between center axes at center of rotation, mm	1.23 (0.80)	1.89 (1.11)	1.73 (0.87)	<.001*
Angle between center axes, °	14.78 (8.96)	16.37 (8.03)	16.25 (4.88)	.63

Note. Values are presented as means with standard deviations in parentheses. MCP = metacarpophalangeal; UCL, isolated thumb ulnar collateral ligament injury; RCL, isolated thumb radial collateral ligament injury; UCL + RCL, concomitant thumb ulnar and radial collateral ligament injury; MRI = magnetic resonance imaging.

^*Indicates statistical significance on 1-way analysis of variance.

Table 3.

Classification of Subluxation Severity.

Subluxation grade	UCL	RCL	UCL + RCL
Radiographs
Grade I	84.5 (104)	30.6 (11)	0 (0)
Grade II	14.6 (18)	47.2 (17)	75 (3)
Grade III	0.8 (1)	22.2 (8)	25 (1)
Grade IV	0 (0)	0 (0)	0 (0)
MRI
Grade I	42.7 (44)	14.7 (5)	0 (0)
Grade II	45.6 (47)	61.8 (21)	50 (2)
Grade III	9.7 (10)	14.7 (5)	50 (2)
Grade IV	1.9 (2)	8.8 (3)	50 (2)

Note. Values are presented as percentages with counts in parentheses. Grade I, 0% to 25% thumb metacarpophalangeal subluxation; grade II, 25% to 50% thumb metacarpophalangeal subluxation; grade III, 50% to 75% thumb metacarpophalangeal subluxation; grade IV, 75% to 100% thumb metacarpophalangeal subluxation. UCL, isolated thumb ulnar collateral ligament injury; RCL, isolated thumb radial collateral ligament injury; UCL + RCL, concomitant thumb ulnar and radial collateral ligament injury; MRI = magnetic resonance imaging.

For center axis distance, post hoc testing revealed significant differences between the UCL and RCL cohort (P < .001). There was no significant difference between the UCL and concomitant UCL and RCL cohorts (P = .34) or between the RCL and concomitant UCL and RCL cohorts (P = .879). For center axis angle, post hoc testing revealed significant differences between the UCL and RCL groups (P < .001). No significant difference was found between RCL and concomitant UCL and RCL cohorts (P = .166) or between RCL and concomitant UCL and RCL cohorts (P = .948).

Magnetic Resonance Imaging Subluxation Measures

Of 175 cases, 141 had MRI images available for review. Percent MCP subluxation, center axis distance, and angle for each group were measured (Table 2). The ANOVA showed significant differences between groups for both percent MCP subluxation and center axis distance (P < .001). For percent MCP subluxation, post hoc testing revealed significant differences between percent subluxation in UCL and RCL (P < .001). No significant differences were found between the UCL and concomitant UCL and RCL cohorts (P = .246) or between RCL and concomitant UCL and RCL cohorts (P = .999). Percent MCP subluxation measurements were further stratified by severity using a 4-tiered grading system as described in the methods section (Table 3).

For center axis distance, post hoc testing revealed significant differences between the UCL and RCL cohort (P < .001). There was no significant difference between the UCL and concomitant UCL and RCL cohorts (P = .50) or between the RCL and concomitant UCL and RCL cohorts (P = .973). The ANOVA showed no significant differences between groups for the center axis angle measurement (P = .63).

Discussion

Our study showed a significant, measurable difference in volar subluxation, measured in 3 separate ways, between patients with isolated operative RCL versus isolated operative UCL injury. Differences in RCL versus UCL injuries have been previously documented in anatomical and biomechanical studies, showing that RCL injury has a greater impact on volar, ulnar, and pronation instability of the MCP joint. ¹⁴ However, this has not yet been borne out in the clinical setting. Prior studies have commented on subtle findings of volar subluxation but have not attempted to quantify it. ¹⁵

Historically, the treatment of complete RCL injuries has been debated because the anatomy on the radial aspect of the thumb should, in theory, lend itself to healing under conservative treatment in the absence of an interposed aponeurosis as opposed to the ulnar side of the thumb. ⁹ However, there remain clinical findings of significant volar subluxation of the MCP joint in the setting of RCL injuries compared with UCL injuries. There are factors that either compound or protect against the volar laxity prompted by an RCL injury. Most notably, there is an anatomic difference between the musculotendinous insertions on the radial and ulnar aspects of the MCP joint. On the radial side, the aponeurosis of the abductor pollicis brevis and flexor pollicis brevis runs longitudinally and dorsally, while on the ulnar side the intact adductor runs transversely and is volar to the axis of the MCP joint. Thus, with radial instability of the thumb MCP joint, there is a net volar deforming force. ⁴ The integrity of the dorsal capsule and dorsal fibrocartilage, often found to be concurrently injured with an RCL injury, can further exacerbate volar subluxation. ¹⁴ In contrast, a flatter metacarpal head provides biomechanical stability in the setting of an incompetent RCL, providing protection against volar subluxation. ¹⁶ While these additional factors are important to note, an RCL injury causing a clinically significant volar subluxation of the joint leads to joint incongruity and places the articular cartilage under abnormal stress, which can lead to degeneration of the joint if not addressed.

Our study provides 3 methods to quantify the amount of subluxation present at the MCP joint. It is important to note that there can be patients with normal physiological volar subluxation of the MCP joint, documented to be as high as 35% with a mean of 18%. ¹⁷ Based on our grading system, most of these ligamentously lax patients would fall into the grade I subluxation category. Our results show that fewer than 15% of patients with an operative RCL injury would be described as grade I subluxation and the vast majority are at least a grade II subluxation or higher on MRI. This would mean that only the top percent of ligamentously lax patients (found to be as high as 35% volar subluxation, putting them in a grade II subluxation category) could be mistaken for having a potential RCL injury. This is important for clinicians to keep in mind and to use our measurements and grading system within a full evaluation of the clinical picture.

One of the additional benefits on these findings is the ability to evaluate that patient on static imaging rather than dynamic stress imaging under fluoroscopy in clinic, which not only induces pain during the examination but has also been shown to have complications such as a “locked” MCP joint after stress testing. ¹⁸ All 3 measurements were found to be statistically significant on plain radiographs. Only 2 measurements were found to be statistically significant on MRI. The measurement found not to be statistically significant on MRI was the angle between the center axis of the proximal phalanx and the metacarpal. This could be due to a wide variety of thumb positioning during the MRI with some patients having relatively extended versus flexed thumb joints at rest. As MRI images undergo re-formatting to obtain true sagittal and coronal cuts of the thumb, positioning at the time of MRI may not be as standardized as hand or thumb radiographs. Magnetic resonance imaging has been shown in other studies to be a reliable way of assessing the presence of injury to the RCL or UCL. Shin et al ¹⁹ found MRI to have a sensitivity, specificity, and accuracy of 85.7%, 91.4%, and 85.7%, respectively, compared with thumb MCP arthroscopy. Given this, MRI is likely still the superior imaging modality to assess for RCL injury both in the quality of the soft tissue and the 2 sets of significant measurements as demonstrated by our study, specifically the amount of dorsal “gap” measured due to volar MCP subluxation.

A strength of our study is that 175 thumbs were reviewed with 2 imaging modalities, one of which is readily available to all providers in the community in the form of plain radiographs. One limitation of our study is the lack of consistency in our plain radiograph cohort, as the preoperative radiographs available included both hand and thumb series radiographs. Because the majority were hand radiographs, we made the choice to use the PA view for these, and then choose the thumb series image that most closely resembled a hand PA view, which is the oblique thumb radiograph. In addition, while images were screened for quality and proper positioning, there is always some degree of variability in the positioning of the hand on radiographs that could affect the measurements taken. Another limitation of our study is that all included cases were operative and, therefore, are more likely to be severe. It is unclear whether the differential rates of subluxation would be apparent in nonoperative collateral ligament injuries, and this should be an area of future research using the preoperative image-based grading system introduced.

An additional weakness is the presence of significant differences in age and sex between the 3 cohorts. While the difference in age between the RCL only and UCL only cohorts is negligible, the age of the RCL and UCL cohort is approximately 15 years higher. As for the difference in sex, a much larger proportion of RCL only injuries were in men. These differences have been seen repeatedly in the clinical RCL literature. Catalano et al ⁹ had 65% men included in their study and Huag et al ¹¹ had 77%. Other studies have looked exclusively at male patients such as Werner et al ¹⁵ examining RCL injuries in the National Football League.

Future studies should examine whether there is clinical evidence of statistically significant ulnar subluxation of the joint in addition to volar subluxation, as it has been described in biomechanical studies. ¹⁴ This would require dedicated thumb radiographs with a true anterior-posterior of the thumb or coronal slices on MRI. Correlating preoperative clinical imaging with surgical findings of structures such as the dorsal capsule or dorsal fibrocartilage could help further inform our preoperative evaluation of what structures are involved and how much volar subluxation would confer concern about the RCL as opposed to both the RCL and dorsal structures. Finally, stratifying acute and chronic RCL injuries and combining it with either long-term outcomes of radiographic degenerative changes or an intraoperative examination of the articular cartilage of chronic injuries could inform our level of concern regarding the effect of long-term volar subluxation on the health of the joint. Evaluating the correlation between the subluxation measured and long-term patient outcomes could help determine the clinical significance of our findings.