The Role of the Flexor Carpi Radialis Groove in Trapeziometacarpal Osteoarthritis

Abstract

Background:

Thumb carpometacarpal (CMC) osteoarthritis (OA) is a common condition. The contribution of surrounding ligaments and tendons to the stability of the CMC joint is likely altered in OA. The flexor carpi radialis (FCR) tendon runs in the trapezial FCR groove and is often noted to be frayed during CMC arthroplasty. We hypothesized that decreased integrity of the FCR tendon is related to FCR groove morphology and is associated with increased severity of CMC OA.

Methods:

We examined 3-dimensional surface models based on computed tomography (CT) scans of explanted trapezia from patients who underwent thumb CMC arthroplasty. Fraying of the FCR tendon was rated intraoperatively. Measurements were taken of the FCR groove to evaluate its morphology. Preoperative thumb CMC radiographs for each patient were scored using the modified Eaton classification system and the Thumb Osteoarthritis Index. Differences in the tendon groups were examined, and multivariable linear regression models were used to test the association between tendon group and FCR groove measurement.

Results:

There were 136 patients who were categorized into 4 tendon groups: intact, minor fraying, fraying, and ruptured. There were no differences between the tendon groups on any measures.

Conclusions:

Our findings do not demonstrate a significant influence of FCR groove morphology on FCR tendon fraying in CMC arthroplasty patients. We also did not find a significant association between the FCR tendon state and degree of radiographic CMC OA. Further studies should investigate the in vivo FCR tendon to evaluate its tearing and inflammation in relation to basilar thumb pain.

Introduction

Thumb carpometacarpal (CMC) osteoarthritis (OA) is a common condition that particularly affects postmenopausal women and can be highly disabling for patients.^{1 -3} The thumb CMC joint articular surfaces provide limited intrinsic stability; therefore, its ligaments as well as surrounding muscles play an important role in the joint’s stability.^4,5

The flexor carpi radialis (FCR) tendon runs near the thumb CMC joint, and travels volar to the trapezium to insert at the second metacarpal. The tendon runs through a canal in the trapezium called the FCR groove.^2,5,6 The FCR groove is a common site of FCR tendinitis, and the FCR tendon is often noted to be frayed when the trapezium is removed during thumb CMC arthroplasty. ⁷ The FCR tendinitis and rupture have also been associated with scaphoid-trapezium-trapezoid (STT) joint OA, which is typically associated with trapeziometacarpal OA.^{8 -11} Isolated STT arthritis is quite rare, and usually associated with calcium pyrophosphate dihydrate deposition (CPPD). ¹¹ A potential cause for the tendon fraying and local inflammation could be the shape or size of the FCR groove itself. To date, there is a lack of studies evaluating the morphology of the trapezial FCR groove. No studies have been done to study FCR groove anatomy with its relationship to FCR tendon pathology. As the FCR tendon runs in such close proximity to the trapezium, rupture or attenuation of the tendon could affect joint stability, by increasing joint laxity and promoting CMC OA development. ¹² There is also a lack of prior research evaluating the relationship of FCR tendon integrity with the thumb CMC arthritis.

The goals of our study were to determine whether the state of the FCR tendon was related to the morphology of the FCR groove and whether the FCR tendon integrity was associated with thumb CMC OA. Our hypothesis was that FCR tendon pathology would have a direct correlation with thumb CMC OA and that the FCR groove morphology is related to FCR tendon integrity.

Methods

This is a retrospective study performed at a single institution with a single surgeon (A.L.L.). As part of a larger ongoing study (institutional review board [IRB] approval no: 21458), explanted trapezia of patients who underwent trapeziometacarpal arthroplasty were collected after informed consent was obtained. All patients who underwent CMC arthroplasty were approached for the study. Patients were recruited between 2011 and 2020. Patients were excluded from the study if they had rheumatoid arthritis or an history of trauma at the CMC joint.

Preoperative radiographs for each eligible patient were scored using the Thumb Osteoarthritis Index (ThOA) as well as the modified Eaton classification system. ¹³ The modified Eaton classification system scores radiographs on a scale of 0 to 4 using posteroanterior, lateral, Robert, and stress views of the thumb CMC joint. Stage 0 correlates to no evidence of any arthritis; Stage 1 correlates to minimal joint space narrowing with minimal contour changes and no debris; Stage 2 correlates with joint space narrowing, small debris, subchondral sclerosis, some contour changes, and some joint space narrowing; Stage 3 correlates with significant narrowing or no joint space, contour changes, large debris, and osteophytes and joint subluxation; Stage 4 correlates with Stage 3 and additional scaphotrapezoid and/or pan-trapezial arthritis. ¹³ The ThOA index scoring system uses a single Robert view of the thumb CMC joint and measures the width relative to the height of the trapezium using the Robert view. A higher ratio corresponds to trapezial width expansion with radial and ulnar osteophytes, compared with erosion of the first metacarpal articular surface, thus measuring severity of CMC arthritis. ¹³

During surgery, the trapeziometacarpal joint was approached using the Wagner approach. The trapezium was carefully extracted from each patient using a beaver blade as well as a 0.045 Kirschner-wire to manipulate the trapezium during extraction (Figure 1). Care is taken to not violate the structure of the trapezium as well as to ensure there is no iatrogenic damage to the FCR tendon.

Figure 1.

Intraoperative view of a trapeziectomy. Incision (a) and extraction were performed in such a way to not damage the trapezial articular facets and the flexor carpi radialis tendon. A Kirschner-wire was used to extract the trapezium intact (b).

After excision of the trapezia, the surgeon rated the state of the FCR tendon as 1 of 4 groups: intact, minor fraying fraying or ruptured. Tendons with fraying less than 30% were categorized as “intact,” those with 30% to 50% fraying as “minor fraying,” those with fraying more than 50% as “fraying,” and complete ruptures as “ruptured.” After surgical excision, the trapezia were stored in 4% formaldehyde solution. All specimens underwent computed tomography (CT) scanning. Medical image processing software (Mimics 21.0, Materialise) was used to create 3-dimensional (3D) surface models of the trapezia. The 3D surface models based on CT scans of these trapezia were analyzed. Three different measurements were used to quantify the FCR groove shape: (1) the angle at the deepest part of the groove between the proximal and distal borders; (2) the largest diameter circle that can fit within the groove at 3 different points; and (3) the distance over surface between the most proximal and distal border at the deepest part of the groove (Figure 2).

Figure 2.

(a) Volar view of left trapezium and flexor carpi radialis (FCR) groove. White arrow demarks location of the FCR tendon. (b) Proximal view, angle and diameter measurements. (c) Volar-proximal view depicts the 3 locations of groove diameter measurement.

Descriptive statistics of demographic characteristics (eg, age) and FCR groove measurement (angle, diameter, distance) were calculated for each tendon group. Differences in the tendon groups were examined using a Fisher Exact Test for sex and Eaton Stage and 1-way analysis of variance (ANOVA) tests for all other measurements. A simple linear regression was used to test whether there was a relationship between the state of the FCR tendon and the various shape parameters of the FCR groove. Multivariable linear regression models were used to test the association between tendon group and FCR groove measurement, adjusting for sex and age. All analyses were performed with a 2-sided level of significance of .05.

Results

There were 136 participants who were categorized into 4 tendon groups. There were 87 participants in group 1 (intact) with intact tendon, 16 participants in group 2 (minor fraying), 13 participants in group 3 (fraying), and 20 participants in group 4 (ruptured). Descriptive statistics for each group are reported in Table 1. There were no statistical differences in morphology, demographics, and radiographic OA stage between the tendon groups.

Table 1.

Descriptive Statistics of Demographic and Tendon Characteristics.

	Group 1 (n = 87)	Group 2 (n = 16)	Group 3 (n = 13)	Group 4 (n = 20)	p
Characteristic	n (%) or mean (SD)	n (%) or mean (SD)	n (%) or mean (SD)	n (%) or mean (SD)
Sex					.443
Female	70 (80%)	10 (63%)	10 (77%)	16 (80%)
Male	17 (20%)	6 (38%)	3 (23%)	4 (20%)
Age	63.55 (9.55)	61.31 (9.44)	67.38 (11.12)	65.10 (10.40)	.341
Eaton stage					.590
2	26 (30%)	3 (19%)	2 (15%)	5 (25%)
3	35 (40%)	6 (38%)	8 (62%)	11 (55%)
4	26 (30%)	7 (44%)	3 (23%)	4 (20%)
Thumb OA Index	2.61 (0.42)	2.13 (0.69)	2.43 (0.91)	2.26 (0.78)	.278
Angle deep	107.91 (17.57)	102.31 (17.87)	104.86 (19.71)	104.45 (17.82)	.330
Diameter 1	5.18 (1.74)	5.04 (1.50)	4.53 (1.09)	4.82 (1.09)	.187
Diameter 2	4.42 (1.17)	4.56 (1.39)	4.37 (1.43)	4.34 (1.17)	.807
Diameter 3	4.44 (1.24)	4.33 (0.91)	5.09 (1.85)	4.51 (1.41)	.434
Distance surface	7.51 (1.66)	7.80 (1.16)	7.56 (1.62)	7.54 (1.75)	.869
STL volume	2379.38 (659.08)	2632.00 (626.92)	2604.46 (522.18)	2495.75 (819.78)	.261
STL surface area	1144.18 (205.75)	1231.00 (185.81)	1225.92 (214.49)	1214.15 (310.80)	.100

Note. OA = osteoarthritis; STL = stereolithography.

There were no statistically significant differences in Eaton stage between each group. Group 1 had 26 (30%) Eaton stage 2 patients, 35 (40%) Eaton stage 3 patients, and 26 (30%) Eaton stage 4 patients. Group 2 had 3 (19%) Eaton stage 2 patients, 6 (38%) Eaton stage 3 patients, and 7 (44%) Eaton stage 4 patients. Group 3 had 2 (15%) Eaton stage 2 patients, 8 (62%) Eaton stage 3 patients, and 3 (23%) Eaton stage 4 patients. Group 4 had 5 (25%) Eaton stage 2 patients, 11 (55%) Eaton stage 3 patients, and 4 (20%) Eaton stage 4 patients. There were no statistically significant differences in Thumb OA Index between the groups. Group 1 mean Thumb OA Index was 2.61 ± 0.42, group 2 was 2.13 ± 0.69, group 3 was 2.43 ± 0.91, and group 4 was 2.26 ± 0.78 (P = .278).

Multivariable linear regression results are reported in Table 2. There were no statistically significant differences between groups in their FCR groove measurements.

Table 2.

Linear Regression Models Examining Association Between Tendon Group and FCR Groove Measurements.

FCR groove measures	Group 2		Group 3		Group 4
	Est. (SE)	P value	Est. (SE)	P value	Est. (SE)	P value
Angle deep	–4.21 (4.86)	.388	–2.58 (5.27)	.625	–3.35 (4.37)	.446
Diameter 1	–0.18 (0.44)	.686	–0.62 (0.48)	.197	–0.35 (0.39)	.383
Diameter 2	0.05 (0.34)	.874	–0.03 (0.37)	.933	–0.07 (0.30)	.827
Diameter 3	–0.15 (0.36)	.670	0.61 (0.39)	.120	0.06 (0.32)	.853
Distance surface	0.04 (0.42)	.930	0.05 (0.46)	.919	0.04 (0.38)	.918
STL volume	147.39 (163.03)	.368	164.59 (176.71)	.353	98.74 (146.65)	.502
STL surface area	60.94 (53.48)	.256	53.48 (57.92)	.358	60.60 (48.07)	.210

Note. Reference group = group 1. All models adjust for sex and age of bone. FCR = flexor carpi radialis; STL = stereolithography.

Two-group analysis was also performed to test for a relationship between intact versus non-intact tendon. There were 87 patients with an intact tendon (group 1) and 49 patients with a non-intact tendon (groups 2, 3, and 4). There were no statistical differences in morphology, demographics, and radiographic OA stage between these 2 groups. There was also no statistical difference between intact versus non-intact tendon in any FCR groove measurements.

Discussion

Our findings did not demonstrate a significant relationship between FCR groove morphology on FCR tendon fraying in CMC arthroplasty patients. We also did not find a relationship with tendon integrity to either modified Eaton or ThOA index severity.

Studies have shown that arthritis at the thumb CMC joint and STT joint can lead to FCR tendon attenuation.^8,14 However, to date, there is no literature linking specific patient factors to FCR tendon integrity. Studies have hypothesized that the FCR becomes frayed from nearby osteophytes and joint degeneration, yet this has not been proven.^8,9 Although these studies typically mention the narrow FCR groove being associated with tendon vulnerability to tearing, this did not correlate with our study as the groove morphology was not associated with tendon integrity. Other factors likely contribute to the degree of FCR tendon pathology in patients with thumb CMC OA. It is possible that symptom duration and other medical comorbidities may be associated with tendon integrity; however, this was not evaluated in our study. It is also possible that the tendon can be damaged iatrogenically during a CMC arthroplasty. To minimize this risk in our study, the attending surgeon performed all trapezial excisions with trainees acting only as assistants.

Overlying integrity of supporting tendons such as the FCR has not yet been studied in their association with CMC OA. The relationship of joint stability to the surrounding ligaments, however, has been extensively reported. Jónsson et al ¹⁵ found the female patients with thumb base OA had a higher prevalence of hypermobility features compared with age-matched controls. Numerous biomechanical studies have evaluated the importance of the volar anterior oblique ligament, the dorsal deltoid ligament, and other surrounding ligaments in stability, and therefore as a purported contributor to CMC OA.^16,17 The thumb CMC joint is also surrounded by muscles that help provide dynamic stabilization.^12,18,19 As the FCR tendon is both close to and in contact with the trapezium through the FCR groove, it likely plays some role in joint stability. Our evaluation of clinically relevant arthritic specimens, rather than cadaveric biomechanical evaluation, did not find a relationship with FCR fraying and arthritis severity. Future biomechanical studies to evaluate whether and to what degree the FCR tendon contributes to CMC joint stability may support and promote intervention for patients with early CMC OA, if indeed stability reflects joint preservation or delays arthritis progression.

The integrity of the FCR tendon is also relevant to preoperative planning in thumb CMC arthroplasty, yet to our knowledge, no studies have been done regarding the examination of the FCR tendon before surgery. The FCR tendon commonly has tendinitis at the level of the trapezial FCR groove. ⁷ Many surgical options use the FCR tendon for ligament reconstruction and tendon interposition (LRTI). ¹⁴ ,^{20 -22} In our study, 49 of 136 patients had FCR tendons that were not fully intact, 20 of which were completely ruptured. Such foreknowledge may be helpful in choosing arthroplasty options to maximize surgical success. Jones et al ¹⁴ presented several techniques for salvage procedures for the intraoperative discovery of a ruptured or insufficient FCR tendon. As studies have shown STT arthritis to be associated with FCR tendinitis and rupture,^{8 -10} evaluating the STT joint on preoperative radiographs is an important preoperative step for surgeons planning to use the FCR tendon for LRTI. Ultrasound evaluation has been shown to be able to identify FCR tendinopathy as well as partial and complete tears of the tendon. ²³ Future directions may include predicting or diagnosing FCR tendon disruption prior to a CMC arthroplasty using ultrasound or magnetic resonance imaging (MRI). Future studies should also investigate the in vivo FCR tendon to evaluate its tearing and inflammation in relation to basal thumb pain. Tearing or degradation of the FCR tendon could be associated with levels of pain at the thumb CMC joint. It should also be noted that our study had a high rate of FCR rupture, about 15%, which has not been appreciated in the literature. We believe that FCR rupture is underreported. This may be due to surgical techniques that do not use the FCR tendon, such as tightrope suspensionplasty, or techniques that can use the FCR stump even if it has ruptured, such as the suture suspensionplasty.

Our preliminary study has several limitations. One limitation is that the FCR tendon state was performed by a single experienced surgeon using a non-validated estimate of tendon fraying. We had anticipated that differences between “minor fraying” and “fraying” might be relevant, and collectively these disruptions would be distinct from “ruptured.” If any misclassifying occurred within this subjective group allocation, it is likely not relevant given that no difference was identified among the groups.

A further limitation of our study is that the degree of arthritis of the specimens was graded using radiographic measures, the modified Eaton classification system and the ThOA Index, rather than direct visual findings. Evaluating the severity of arthritis of the thumb CMC joint has classically been done with standard radiographs. The authors believed that using the radiographic classification systems was a better measure for characterization of the degree of OA for each specimen, given that they are reliable and reproducible, and to correlate with articular surface wear identified at surgery. ¹³ Advanced imaging such as MRI and CT scans are not currently recommended for evaluation of thumb basal joint arthritis, although a consideration for future investigation to address the role of soft tissue envelope stabilizers. ²⁴

Preoperative evaluation for FCR tendon integrity in patients undergoing thumb CMC arthroplasty may enhance surgical planning, especially when a surgeon’s procedure of choice relies on the FCR for anchor stabilization, source of interposition, or both. The presence of 15% complete FCR rupture in our preliminary study underscores the consideration of preoperative planning. Future areas of investigation include examining the biomechanical role the FCR tendon contributes to CMC joint stability, and the role fraying or rupture may contribute to pain and disability in CMC OA.