Introduction

The clinical significance related to stability of the ulnar nerve continues to be a source of controversy. Ulnar nerve instability (UNI) at the cubital tunnel has been reported in healthy individuals [1,2,3], and Zaltz et al. reported UNI in 17.7% of asymptomatic children [4]. However, UNI may predispose to cubital tunnel ulnar neuropathy [5,6,7]. The position of the elbow joint may be a relevant biomechanical variable in UNI [5]. Elbow flexion is associated with tightening of the retinaculum which decreases volume in the cubital tunnel resulting in increased pressure on the ulnar nerve which may alter intraneural blood flow triggering edema and enlargement [3]. Chronic nerve compression with intraneural edema results in inflammation, fibrosis, demyelination and eventual axonal loss [10]. In addition, during elbow flexion, an unstable ulnar nerve displaces medially, impacts and is transiently deformed by the humeral medial epicondyle [8]. Abnormal repetitive ulnar nerve dynamics associated with flexion may produce shear stress when the nerve impacts the edge of the humeral retroepicondylar groove [1, 9]. Nerve enlargement arising from repetitive mechanical stressors results in increased cross-sectional area (CSA), a metric that defines peripheral neuropathy. However, even though CSA changes are reported on ultrasonography (US), altered morphology of the nerve may be accompanied by normal electrodiagnostic studies [6, 10, 11].

US is emerging as an important imaging resource for evaluation of the musculoskeletal and peripheral nervous systems [2, 12, 13]. US may detect ulnar neuropathy before other imaging modalities, such as MRI, in asymptomatic patients with UNI [11]. In healthy individuals, the ulnar nerve is located posterolateral to the medial epicondyle and resides within the cubital tunnel. The normal ulnar nerve located within the cubital tunnel is hypoechoic and unifascicular [12]. However, some nerves are hyperechoic or multifascicular. In addition, US has also been shown to reliably detect dynamic movement of the ulnar nerve and accurately evaluate peripheral nerve abnormalities [14, 15]. Rutter et al. found preoperative US agreed with the intraoperative findings of UNI in 88% of patients [16]. From a therapeutic standpoint, US is currently used as the standard for guidance in regional anesthesia procedures [17].

UNI has been defined as either subluxation or dislocation of the ulnar nerve from the cubital tunnel as described by Kawahara et al. [1]. Swelling with increased CSA, hypofascicularity [18], increased elastography-derived metrics [19], and nerve hyperemia [20] are the US findings of cubital ulnar nerve neuropathy [14]. US studies have defined upper limits of the ulnar nerve with CSA values of 0.075–0.09 cm2 [12]. The presence of UNI increases the risk for ulnar neuropathy [5,6,7, 10, 21,22,23]. However, neuropathy of the ulnar nerve may occur in the stable or unstable (UNI) ulnar nerve, as underlying pathoetiologies differ [9, 22].

Elastography is an emerging technique that involves an US image-based quantification of tissue strain in response to an applied force [24]. There are several techniques in use. Shear-wave elastography (SWE) has emerged as a valuable technique for assessing peripheral neuropathy [25,26,27]. If the CSA is increased, the elastographic measurements typically increase as strain results from reduced fluid diffusion across the cell membrane [28]. Recent studies evaluating the median nerve in non-diabetic patients with carpal tunnel syndrome and the tibial nerve in diabetic patients without neuropathy have reported an increase in nerve stiffness with SWE [25, 26, 29]. Although SWE has been widely employed for the evaluation of tendons and muscles, it is likely that it will increasingly contribute to the diagnosis of peripheral neuropathy [19, 27, 30].

Detection of the pressure pain threshold (PPT) has proven reliable and valid in quantitatively assessing sensitization in muscle pain syndromes [31]. PPT occurs at the transition point when applied pressure is sensed as pain (pain threshold). Algometry devices identify the applied force which elicits the sensation of pain [32]. The use of algometry in assessing peripheral nerves is novel, and PPT evaluation of the ulnar nerve in UNI may prove valuable as it has not reported. Algometry in peripheral nerve assessment has been reported in controls and peripheral neuropathy [33, 34].

The aim of our study was to define and characterize the sonographic dynamics of UNI, ulnar nerve CSA, and SWE in asymptomatic volunteers. We hypothesized that UNI would be associated with increased CSA and SWE values. We also utilized algometry to evaluate the ulnar nerve in the cubital tunnel. We hypothesized that UNI would display lower PPT since recurrent subluxation and dislocation of the ulnar nerve have been correlated with friction neuralgia, a source of peripheral sensitization [35].

Methods

Participants

The Institution Review Board of the university approved this study. Participants were recruited internally through the university by class announcements. All participants provided a written informed consent. Criteria for inclusion were either gender, age 18-65, and provision of informed consent. Exclusion criteria included a history of shoulder, elbow, or wrist pain; peripheral neuropathy; systemic disease, such as diabetes mellitus or rheumatoid arthritis; and upper extremity fracture or surgery. Body mass index (BMI) was calculated for all participants. Additionally, demographic information regarding gender, height, weight, age, and handedness was recorded. Both elbows were examined in each participant.

Sonography

All US examinations were performed in the sonography suite of the radiology department. US examination of the ulnar nerve in the cubital tunnel was performed with participants in a supine position with the arm abducted (75°), elbow extended, and wrist supinated. The sonographic parameters of the ulnar nerve were measured with the elbow in three different positions: extension (as previously described), 45° flexion, and full flexion. The sonographic probe position was standardized for each exam with the footprint on an imaginary line between the medial epicondyle and the olecranon process. All images were acquired by a chiropractic radiologist with 4 years of experience in musculoskeletal ultrasound. A Logiq E9 (GE Healthcare, Wauwatosa, WI) ultrasound system operating at 15 MHz with high-frequency linear array transducer and coupling agent was employed.

Ulnar nerve CSA (cm2) and SWE (kPa) were obtained within the cubital tunnel in all three elbow positions. The ulnar nerve was located in short axis within the cubital tunnel. CSA of the ulnar nerve was then measured at its maximum diameter using the digital trace function tool. The region of interest (ROI) for elastography was set to the CSA of the ulnar nerve.

The normal position of the ulnar nerve is typically central within the cubital tunnel, with minimal nerve movement during elbow positional changes [36]. When the ulnar nerve was confined to the cubital tunnel, this finding was classified as a control. Instability of the ulnar nerve was classified into two types: subluxation (partial movement of the ulnar nerve to the 12 o’clock position on the medial epicondyle), or dislocation (movement of the ulnar nerve beyond the 12 o’clock position on the medial epicondyle) [1] (Fig. 1).

Fig. 1
figure 1

Created in Adobe Photoshop (color figure online)

UNI dynamics of the cubital tunnel. Diagram and correlative US images demonstrating the normal (a, d), subluxation (b, e), dislocation (c, f) of ulnar nerve (yellow oval) adapted from Kawahara et al. [1]. The top images (a, d) show the elbow in extension, while the subsequent images (b, c, e, f) are in full flexion. ME medial epicondyle, MT medial triceps muscle, OP olecranon process, U ulnar nerve.

Full size image

All US images were assessed for technical quality prior to analysis. Images were analyzed and stored in the Logic E9 hard drive.

Pressure pain threshold

Pressure pain threshold (PPT) was obtained utilizing an algometer. This measure was obtained bilaterally overlying the ulnar nerve in the cubital tunnel with all three positions of the elbow. The algometer measurements were expressed in pounds of pressure. The algometer, (CommanderTM Algometer; JTECH Medical Industries, Midvale, UT) utilized a 1.0-cm circular probe with a rubber tip connected to the pressure transducer. Algometer pressure was gradually increased over the cubital tunnel until the participant indicated pain (PPT) by verbal response. The algometer measurement was recorded in each of the three elbow positions. The readings were obtained with a 1-min rest period between each position. All PPT measurements were independently performed by a single examiner who was blinded to the sonographic metrics.

Statistical analysis

Descriptive statistics were presented as mean, standard deviation (SD), and ranges for continuous data or numbers (n) and percentages (%).

A two-way analysis of variance (ANOVA) was conducted in Matlab (The MathWorks, Natick, MA) to evaluate the influence of the three independent variables (arm position and condition) on each measurement (CSA, PPT, and SWE). Arm position included three levels (extension, 45° flexion, full flexion), and condition also consisted of three levels (control, subluxation, dislocation). Both main effects and interactions between factors were evaluated, and statistical significance was set at p ≤ 0.05. Post hoc comparisons were performed to investigate the significant effects revealed by the ANOVA, and statistical significance was set at p ≤ 0.05, Bonferroni-corrected for multiple comparisons.

Both intra- and inter-rater reliability of the ulnar nerve CSA within the cubital tunnel was performed using Cohen’s Kappa and calculated in Bland–Altman plots [37]. Images of the ulnar nerve within the cubital tunnel were randomly selected for rater evaluation and performed independently. Ulnar nerve CSA measures for reliability were obtained in elbow extension. For intra-rater reliability, rater #1 re-measured 84 ulnar nerve CSAs 3 weeks after the original data collection. For the inter-rater reliability, a second rater randomly selected and analyzed 42 ulnar nerve CSAs 1 week later after rater #1.

Results

Forty-two subjects (25 males, 17 females) aged 22–40 years (mean age 26.7 years) were enrolled. The average BMI of our participants was 27.4 lbs/in2 (range 18.6–40.2 lbs/in2). Thirty-eight of the participants were right-hand dominant (90.5%). Each subject contributed two sets of measurements (right and left elbows) that were considered as independent data points, since the same subject could exhibit subluxation or dislocation unilaterally. The final dataset included n = 84 elbows in three different positions, for a total of 252 data points for each measurement (111 in the control group, 63 in the subluxation group, 78 in the dislocation group). The SWE measurements for one subject were excluded as an outlier.

Mean and standard deviation values for the three measures (CSA, PPT, and SWE) across groups are reported in Table 1. The two-way ANOVA performed on CSA values revealed a significant main effect for the condition factor, yielding an F ratio of (F(2, 243) = 6.53, p = 0.002. Significance was not met for the main effect of position (F(2, 243) = 0.59, p = 0.557), or for factor interaction (F(4, 243) = 0.05, p = 0.995). Post hoc T tests showed that the control group had significantly smaller CSA values (0.057 ± 0.017 cm2, mean ± SD) compared to subluxation (0.066 ± 0.024 cm2, p = 0.027) and dislocation (0.067 ± 0.024 cm2, p = 0.003) groups (Fig. 2). No significant main effects or interactions were found on either algometry or SWE measures. However, subjects with dislocation demonstrated an increased mean SWE measurement and decreased PPT. Inter- and intra-rater reliability was calculated using intraclass correlation coefficient (ICC). The inter- and intra-rater agreements were moderate (0.54) and excellent (0.94), respectively (Fig. 3).

Table 1 CSA, PPT, and SWE values (mean ± SD) across groups (control, subluxation, dislocation)
Full size table
Fig. 2
figure 2

Post hoc comparisons revealed that both subluxation and dislocation (UNI) groups had significantly higher values of CSA of the ulnar nerve compared to the control group. Error bars indicate standard deviation

Full size image
Fig. 3
figure 3

Bland–Altman plots of the intraclass correlation coefficient for CSA of the ulnar nerve revealed moderate and excellent reliability for the inter- and intra-rater reliability, respectively

Full size image

Discussion

We hypothesized that UNI pathomechanics would increase the ulnar nerve CSA and SWE with a reduction of PPT. We observed significantly increased CSA measurements in the UNI group. There was no change in the PPT or SWE between the control and UNI groups. UNI was defined as either subluxation or dislocation of the ulnar nerve out of the cubital tunnel [1]. The pathomechanics of UNI are influenced by the elbow joint position. During elbow flexion, the unstable ulnar nerve can displace and undergo deformation by the medial epicondyle of the humerus [8]. As the elbow flexes, there is tightening of the retinaculum resulting in reduction of the cubital tunnel volume [22]. This elbow posture may prompt medial ulnar nerve movement with resulting dynamic compression [38]. The ulnar nerve normally undergoes some degree of movement within the tunnel during elbow flexion. A nerve without excursion from the cubital was identified as a control [1, 36, 38].

Flexion of the elbow commonly results in entry of the medial head of the triceps into the proximal aspect of the cubital tunnel [39]. The triceps tendon may contribute to ulnar movement and even instability in the snapping triceps syndrome. Interestingly, on this question, Michael et al. found that participants with hypertrophic triceps muscles had an increased frequency of ulnar nerve subluxation or dislocation compared to controls [39]. There are several etiologic sources of UNI. They include absent retinaculum (Osborne ligament) [3, 21], anconeus epitrochlearis muscle [40], tight posterior medial collateral ligament bundle (pMCL) [22], increased cubital tunnel pressure [9, 41, 42], and congenital anomalies such as a shallow groove or dysplasia [1, 9, 12]. Hypertrophy and snapping of the medial head of the triceps brachii have also been associated with UNI [8, 12, 35, 39, 40]. We assessed ulnar nerve stability with high-resolution US and categorized the findings as control, subluxation, or dislocation. Fifty-six percent of the asymptomatic volunteers demonstrated UNI (25% subluxation and 31% dislocation). Multiple previous studies report differing percentages, from 15.4 to 30% for subluxation and 5.8 to 19% for dislocation of the ulnar nerve at the cubital tunnel [1, 2, 8, 12, 22]. Not all of these studies utilized US when assessing UNI. Kawahara et al. evaluated UNI with MRI [1]. In concordance with our study, Kang et al. [8] and Michelin et al. [22] evaluated similar elbow positions and age groups in their sample. However, in our study, a much higher percentage of participants demonstrated UNI. Our convenience sample included a number of athletes, which may have influenced our prevalence of UNI. In contrast to our study, Kang et al. [8] and Michelin et al. [22] did not assess CSA of the ulnar nerve. The operator dependence of extremity nerve US may also underlie varying reports of reliability with the upper greater than the lower [43]. Michelin et al. reported difficulty in reliably differentiating subluxation and dislocation utilizing US [22], while Fink et al. demonstrated excellent reliability when sonographically assessing UNI [23].

Ulnar neuropathy at the cubital tunnel is the second most common entrapment neuropathy after median nerve in the carpal tunnel [41]. A few studies have demonstrated an increased risk of developing ulnar neuropathy with subluxation or dislocation of the ulnar nerve [5, 6]. Schertz et al. sonographically examined subjects with positive electrodiagnostic studies. Their study assessed the ulnar nerve for morphologic and dynamic differences between patients with and without neuropathy. They observed 49% of their subjects with ulnar neuropathy dislocated from the cubital tunnel compared to 23% of controls [5]. While we did not employ electrodiagnostic testing in our asymptomatic subjects, we found that our sample had significantly increased CSA of the ulnar nerve when associated with UNI. Omejec et al. found patients with ulnar nerve US morphologic abnormalities but normal electrodiagnostic studies. Those patients demonstrated dislocation more commonly when compared with controls [6].

Intriguingly, Leis et al. demonstrated that dislocation of the ulnar nerve seemed to have a protective effect when compared with subluxation. Ulnar nerve subluxation was shown to produce abnormal sonographic and electrodiagnostic findings compared to dislocation, suggesting subluxation may invoke greater risk for ulnar neuropathy [7]. Van den Berg et al. demonstrated that subluxation occurred more frequently, although not significantly, in patients with ulnar neuropathy compared to healthy controls. In contrast, he found no significant differences in electrodiagnostic or sonographic findings between the two groups [10]. However, Pelosi et al. demonstrated US had higher sensitivity than electrodiagnostic testing in detecting clinically mild ulnar neuropathy [44]. In our study, 56% of the asymptomatic subjects displayed UNI with abnormally increased CSA. The question of whether UNI predisposes to the development of ulnar neuropathy is debated. Our data derived by high-resolution US suggest UNI is associated with abnormally increased ulnar nerve CSA and may predispose to ulnar neuropathy.

Elastography has become a useful imaging technique for evaluating the mechanical property of tissue [45]. SWE is one of three types of elasticity imaging. SWE tracks propagation of sound waves through tissue estimating the tissue stiffness, and has been widely applied to the musculoskeletal system. SWE tracks healing progress, and predicts injury risk and/or prognosis of healing [24]. However, very few studies have been performed with SWE on peripheral nerves [25, 27, 29]. Paluch et al. demonstrated subjects with ulnar neuropathy had three times greater SWE values compared to controls [19]. We evaluated SWE of the ulnar nerve within the cubital tunnel to detect changes in tissue stiffness in different biomechanical positions [46]. Previous studies have reported that SWE can significantly improve the US evaluation of the median nerve within the carpal tunnel [29]. Our study demonstrated no elastographic changes of the ulnar nerve between the control and UNI groups in all three elbow positions. Although we observed increased CSA, the lack of accompanying elastographic changes is difficult to explain. One possibility is the asymptomatic group did not have sufficient levels of abnormal fluid diffusion to provoke abnormal mechanical strain detectable by SWE.

PPT is measured when a pressure stimulus transitions into a painful one [32]. Algometry provides a quantitative measure of the pain threshold [32]. Graven-Nielsen et al. revealed that cuff and manual pressure algometry are comparable in the assessment of PPT [47]. Our study employed manual pressure algometry over the cubital tunnel to measure PPT. Although between groups (controls versus UNI) PPT measurements were not significantly different, subjects with dislocation of the ulnar nerve had a decreased mean PPT compared to controls. The absence of peripheral sensitization was likely responsible for the negative group differences.

We are reporting the largest sample size of UNI investigated by US. The finding of increased CSA in UNI is novel. A few limitations have been identified that require discussion. First, this was a cross-sectional study of asymptomatic participants who were aged 22–40 years, limiting its generalizability. Second, our subjects were evaluated by self-report without a clinical evaluation. Third, our study did not include electrodiagnostic measurements as the participants were asymptomatic, which may have inadvertently included borderline, but asymptomatic ulnar neuropathy. Last, increased BMI has been shown to elevate the CSA of peripheral nerves, a potential confound for our CSA measures [48].

In conclusion, UNI is prevalent in asymptomatic elbows as measured by high-resolution US, and our study demonstrated a higher prevalence of UNI than previously reported. UNI has been implicated as a risk factor in ulnar neuropathy. In our study of asymptomatic volunteers, UNI was associated with an increased CSA. Although not significant, when compared to controls, subjects with dislocation demonstrated increased mean elastographic measurements and decreased values of PPT. A longitudinal study will inform the risk of progression from UNI to ulnar neuropathy. US is an available, noninvasive, and reliable means of assessing UNI.