Comparison Between Effectiveness of Ultrasound-Guided Corticosteroid Injection Above Versus Below the Median Nerve in Mild to Moderate Carpal Tunnel Syndrome

A Randomized Controlled Trial

Abstract

Objective 

In this study, the clinical effectiveness of ultrasound-guided corticosteroid injection “above” versus “below” the median nerve for treatment of patients with mild to moderate carpal tunnel syndrome was compared.

Design 

This prospective randomized double-blind clinical trial included 44 patients with mild to moderate carpal tunnel syndrome. The subjects were randomly assigned to two groups to receive ultrasound-guided injection of 40 mg of triamcinolone either “above” or “below” the involved median nerve. Outcome measures were the Boston Carpal Tunnel Questionnaire, visual analog scale, electrophysiological tests, and ultrasonographic measurement of the median nerve cross-sectional area at baseline, 6, and 12 wks after the injection.

Results 

All outcome measures improved significantly in both groups at 6 wks after intervention, and these improvements were persevered up to 12 wks of follow-up (all P values <0.05). However, there was no significant difference in measured outcomes between the two groups. No adverse effects were observed.

Conclusions 

Both above and under median nerve ultrasound-guided steroid injection techniques were effective in reducing the symptoms, improving the function, and improving electrodiagnostic and sonographic findings of carpal tunnel syndrome. However, the amount of improvement in the outcomes did not differ between groups, implying that none of technique has the superiority over another.

Carpal tunnel syndrome (CTS) is the most common peripheral nerve entrapment neuropathy of the upper limb. The prevalence of CTS in the general adult population is reported to be approximately 2.7% to 5.8% in different studies, and its lifetime incidence is estimated to be nearly 10%.^1,2 Carpal tunnel syndrome is more prevalent among women aged between 40 and 60 yrs. Symptoms are often gradual in onset and typically include numbness, tingling, and pain in the distribution of median nerve in the hand. Carpal tunnel syndrome is associated with considerable discomfort and disability for the patients, and it may progress to cause permanent sensory and/or motor deficits.^3,4

Local corticosteroids injection into the carpal tunnel has had long impact in the treatment of patients with mild to moderate CTS. The effectiveness of carpal tunnel injections has been established in several studies, and a recent meta-analysis demonstrated the efficacy of local corticosteroids injection in providing significant short-term improvement in symptom severity (SS) and functional status (FS) in patients with CTS.^5–8 In clinical practice, this intervention is generally performed using surface anatomic landmarks, which is typically referred to as “blind injection technique.”^9–11 In the recent years, ultrasound (US)-guided musculoskeletal injections have been increasingly used to improve accuracy of the neuromusculoskeletal interventions. Ultrasound is a relatively low cost, portable, readily available, and without radiation adverse effects, which can be easily implemented in clinical practice.¹² Because of these advantages, several US-guided carpal tunnel injection techniques, including in-plane, out-plane, and longitudinal approach for needle placement, have been introduced in the recent literature.^13–16 The primary goals of these techniques are to improve the accuracy of steroids injection in the carpal tunnel and to achieve maximum clinical effectiveness. In nearly all existing injection techniques, the authors have typically reported injecting steroids “around” or “close to” the median nerve, and no description is provided about the delivery of the injectate with respect to superficial or deep aspect of the median nerve. Indeed, it is questionable whether delivery of the steroids “above” or “below” the median nerve is associated with more clinical improvement. Theoretically, it may be argued that delivering the steroids above the median nerve (beneath the flexor retinaculum) might be associated with release of the nerve from the inflamed and adhesive layers of the retinaculum, which may potentially result in more favored clinical outcomes. However, to date, there is no well-established study to support the superiority of the above the median nerve injection in comparison with below the median nerve injection in the literature.

In this study for the first time, we compared the effectiveness of US-guided above versus below the median nerve corticosteroid injection in patients with mild to moderate CTS.

MATERIALS AND METHODS

Study Design and Setting

This prospective, double-blind (participants and outcome assessor) trial was conducted from April 2016 to September of 2016 on patients with mild to moderate CTS presenting to the physical medicine and rehabilitation outpatient clinics affiliated to Iran University of Medical Sciences, in Tehran, Iran. All participants gave informed written consent before enrollment in the study. This study conforms to all CONSORT guidelines and reports the required information accordingly (see Checklist, Supplemental Digital Content, https://links.lww.com/PHM/A538). The study protocol was approved by the ethics committee of Iran University of Medical Sciences, and the trial was registered on the Iranian Clinical trial Registry with identification number “IRCT2015101724572N1.”

Study Population

We consecutively recruited 44 patients (44 median nerves) with mild to moderate CTS for this study. In patients with bilateral CTS, the dominant hand was included in the study. The inclusion criteria for trial were: (1) age between 15 to 60 yrs; and (2) having signs/symptoms of CTS including pain, paresthesia, numbness, and/or tingling along the distribution area of median nerve in the hand (the first three radial digits and the radial side of the forth finger). The patients were also screened for positive Phalen, Tinel, and compression tests. The inclusion criteria required at least two symptoms or one symptom plus one positive test. The mild to moderate CTS was defined based on electrophysiological criteria set by Dumitru et al.¹⁷ as having at least one of the following findings: (1) median motor distal latency of 4.2 millisecs or greater or median sensory peak latency of 3.6 millisecs or greater; (2) a median sensory peak latency ratio of 50% or more between 7 and 14 cm; (3) a median to radial distal sensory latency difference of greater than 0.5 millisecs; and (4) a median to ulnar distal sensory latency difference of greater than 0.5 millisecs. The electrodiagnostic study was performed by a single expert physiatrist who was blinded to random assignments. The temperature of patients' hands was checked before electrodiagnostic study and was kept greater than 32°C.

Exclusion criteria were included patients with severe CTS (thenar atrophy), patients with contraindications for corticosteroid injection (including sensitivity to corticosteroids or presence of infection or skin lesion at the injection site), presence of conditions that mimic CTS symptoms (such as polyneuropathy, cervical radiculopathy, and thoracic outlet syndrome), history of steroid injection to the same wrist within the previous 6 mos, surgical history for CTS on the affected hand, neoplastic or traumatic source of pain, and history or presence of wrist bones fracture.

Sample Size and Assignment

To have a power of 0.80 with an α value of 0.05 and a β value of 0.20 and accounting for 10% dropout, a sample of 22 participants was calculated for each group. The patients were randomly assigned into two groups of US-guided injection “above the median nerve” group (n = 22 wrists) and the US-guided injection “below the median nerve” group (n = 22 wrists) using randomly generated treatment allocations within sealed envelopes generated by a statistician who was not involved in the recruitment. The flow chart of the included participants is presented in Figure 1. Both injection techniques were conducted using the ulnar side approach and with real-time in-plane sonographic guidance. A total of 40 mg of triamcinolone (1 ml of 40-mg/ml triamcinolone) was injected in both groups. All patients were educated regarding activity modifications and were asked to avoid repetitive flexion and extension of the wrist. A nightly wrist splint (Standard Cock-up) was prescribed for all participants in both groups and they were asked to wear it for 6 wks.

FIGURE 1: Flow chart of included and excluded participants.

Outcome Measures

To assess the effectiveness of the two injection techniques, subjective evaluation with visual analog scale (VAS) and Boston Carpal Tunnel Questionnaire (BCTQ) and objective comparison with electrodiagnostic and ultrasonographic criteria were conducted at baseline and at 6 and 12 wks after injection by a physician blinded to the allocation. In addition, the injection adverse effects were observed during the trial.

Visual Analog Scale

After an adequate description to the patient, the pain intensity measured by a 10-cm ruler where 0 indicated “no pain” and 10 indicated “unbearable pain.”

Boston Carpal Tunnel Questionnaire

Boston Carpal Tunnel Questionnaire is the most commonly used outcome measure, which assesses improvements in the clinical symptoms and functional recovery of CTS patients. It includes two subscales: SS and FS. The SS scale (11 items) measures severity of hand symptoms and the FS scale (8 items) measures the FS of the hand. Five answers to each question (scores 1–5) according to the severity of the symptoms or difficulty in performing a defined activity are possible. The higher scores indicate more severe symptoms or poor function.^18,19

Electrodiagnostic Evaluation

In electrodiagnostic evaluation of subjects, we measured distal latency and amplitude of the “compound motor action potential” (CMAP) and “sensory nerve action potential” (SNAP) of the median nerve. The measurements were carried out based on descriptions provided by Dumitru et al.¹⁷ For obtaining median SNAP, the recording electrodes were placed on the third digit and antidromic sensory recordings were collected, while stimulating the median nerve at 7 and 14 cm proximal to the recording electrodes. Antidromic median and ulnar SNAP were also recorded from fourth digit, while stimulating median and ulnar nerves 14 cm proximal to the ring electrode. Antidromic median and radial SNAP were recorded from first digit, while stimulating median and radial nerves 10 cm proximal to the recording electrodes. The CMAP was obtained by placing the recording electrode on the motor point of Abductor Pollicis Brevis muscle and stimulating the median nerve 8 cm proximal to the recording electrode. Electromyography of the Abductor Pollicis Brevis muscle was also conducted. All measurements were conducted using a Natus Synergy Ultrapro S100 instrument (USA) at the baseline, 6, and 12 wks after treatment by the same physiatrist who was blinded to the groups.

Ultrasonographic Evaluation

The patients were positioned in the upright sitting with their hands on a pillow, the forearm supinated, and the wrist was in slight dorsiflexion using a rolled-up towel. A high-frequency linear transducer (5–12 MHz) was then placed transversely along the distal crease of the wrist at the inlet of the carpal tunnel. The median nerve was identified, and its cross-sectional area was determined. The median nerve cross-section area (CSA) was measured in transverse section by tracing the internal border of the perineural echogenic rim over the hypoechoic median nerve (Fig. 2). It is agreed upon by several authors that a median nerve cross-sectional area of 10 mm² or greater at distal wrist crease is the most sensitive parameter for diagnosis of CTS on ultrasonography.^20,21 A Medison Samsung Ultrasound with a 5–12-MHz linear array probe was used for this study.

FIGURE 2: Ultrasound image transverse view of the carpal tunnel at the distal wrist crease. The cross-sectional area of the median nerve measures 13 mm². A, ulnar artery; B, carpal bones; M, median nerve; T, flexor tendons.

Injection Technique

After preparing the skin with antiseptic and under sterile conditions while US probe was covered with a sterile barrier and using a sterile gel, the transducer was positioned transversely along the distal crease of the wrist, perpendicular to the median nerve at carpal tunnel inlet (short-axis view). A 23-gauge needle was inserted on the ulnar side of the proximal carpal tunnel at the level of the distal wrist crease. Under real-time US guidance and using in-plane technique, the needle was passed superficial to the ulnar nerve and artery toward the flexor retinaculum. After penetrating the flexor retinaculum, the needle was directed toward median nerve. In the “above the median nerve” injection group, the needle was directed between the median nerve and the flexor retinaculum (toward the superficial side of the median nerve) and then injection was performed using 1 ml of triamcinolone 40 mg/ml (Fig. 3). In the “below the median nerve” injection group, the needle was directed under the median nerve (between the median nerve and flexor tendons) and then the injection was performed (Fig. 4). All injections were performed by a same physiatrist with 8 yrs of experience in neuromusculoskeletal ultrasonography and US-guided interventions.

FIGURE 3: Ultrasound-guided CTS injection above the median nerve (beneath the retinaculum) in transverse (short-axis) view. A, ulnar artery; B, carpal bones; M, median nerve; T, flexor tendons; white arrows, needle.

FIGURE 4: Ultrasound-guided CTS injection below the median nerve in transverse (short-axis) view. A, ulnar artery; B, carpal bones; M, median nerve; T, flexor tendons; white arrows, needle.

Statistical Analysis

Statistical analyses were performed using SPSS V22 (SPSS Inc, Chicago, IL). Kolmogorov-Smirnov test showed normal distribution of the data. Independent t test and χ² test were used to compare parametric and nonparametric data at baseline between groups. Mixed analysis of variance and post-hoc tests were used to explore the interaction effects of time and group on outcomes. Greenhouse-Geisser estimates of sphericity were used to correct degrees of freedom wherever Mauchly's test was significant. Contrasts were performed to compare each level of interaction during different time points. Results are reported at 95% confidence interval and a P value of 0.05 or less was considered significant. Partial η² is reported as effect size.

RESULTS

A total of 44 patients (22 patients in each group) completed the study. The patients included 40 females (90.9%) and 4 males (9.1%). The mean (SD) age was 56.1 (6.6) yrs. The baseline demographic and clinical characteristics of the patients are presented in Table 1. There was no significant difference in age, sex, and body mass index between the two groups. No statistically significant difference was observed in the baseline values of study outcomes, except for the SNAP latency. To explore the possible effects of the baseline SNAP latency difference between the two groups, we added this variable to the model using analysis of covariance. No change was seen in the statistical significance of the results.

TABLE 1: Demographics and electrophysiologic findings of participants at the baseline

All patients tolerated the injection and diagnostic procedures well and no serious adverse events (vessel insult or median nerve injury) was observed in any of the cases.

Outcome Measures

Table 2 demonstrates the effects of the corticosteroid injection after 6 and 12 wks of treatment in both study groups. The VAS showed a significant improvement of pain in both study groups 6 wks as well as 12 wks after the injection. A similar observation was noted for SS and FS scales of the BCTQ. However, there was no significant difference between 6 wks versus 12 wks of injection in the VAS, SS, and FS scales in neither of the groups.

TABLE 2: Outcome measures in the study groups at baseline, 6, and 12 wks after treatment

With respect to electrophysiological outcomes, corticosteroid injection was associated with a significant improvement in distal motor latency 6 and 12 wks after the injection in both groups. Distal sensory latency was significantly improved only among the above the median nerve injection group. It did not show any significant improvement among the participants receiving below the median nerve injection during the same intervals. There was no significant change from the baseline in CMAP and SNAP amplitudes 6 or 12 wks after the injection in neither group. Likewise, there was no significant difference between the electrodiagnostic values collected 6 wks versus 12 wks after injection in both groups (Table 2).

A significant reduction in the cross-section area of the median nerve was observed 6 wks after the baseline injection in both groups, which was maintained 12 wks after the injection. Similar to other outcomes, there was no significant difference between the CSA measured 6 wks versus 12 wks after the injection in neither group.

Time by Group Interaction

The interaction effect of time and group on all outcome measures (including VAS score, BCTQ scores, electrophysiological and ultrasonographic findings) was not significant. This implies that the behaviors of both groups were similar regarding the changes in these outcomes (VAS effect size = 0.02; BCTQ-SS effect size = 0.01; BCTQ-FS effect size = 0.00). The details of time by group analyses for each study variable are characterized in Table 3.

TABLE 3: Interaction effects of time and group on outcome measures

All patients tolerated the injection procedures well, and no serious adverse events (vessel insult or median nerve injury) were observed in any of the cases.

DISCUSSION

The US-guided carpal tunnel corticosteroid injection is increasingly being used for the treatment of patients with mild to moderate CTS. Although several injection techniques have been introduced in the past decade, there are no data comparing the clinical outcomes of delivering steroids superficial or deep to the median nerve. This is the first study that compared the clinical effectiveness of US-guided above versus below the median nerve corticosteroid injection in patients with mild to moderate CTS. Our observations revealed that although both techniques were associated with significant clinical, ultrasonographic (CSA), and electrophysiological improvements, there was no difference between the two methods at 6 and 12 wks of follow-up. The findings of this study carry important clinical implications for physicians practicing neuromusculoskeletal interventions, because “below” the median nerve injection is less technically demanding and easier to perform in daily clinical practice.

In the recent years, multiple studies described novel US-guided injection techniques with the purpose of delivering adequate amount of steroids close to the median nerve without causing iatrogenic injury to the structures in close proximity. The first report in the medical literature describing US-guided carpal tunnel injection was presented by Grassi et al.²² They described successful management of CTS in a 50-yr-old man with CTS secondary to rheumatoid arthritis. The authors placed the needle radial to the median nerve at the distal crease (i.e., between the median nerve and the flexor carpi radialis tendon) and injected the steroids “around” the median nerve under visualization of a high-frequency linear transducer. Smith et al.²³ introduced the ulnar approach method in which the needle was placed at the ulnar side of the proximal carpal tunnel and the transducer was placed transversely (short-axis view) at the level of pisiform bone. In this approach, the needle is directed superficial to the ulnar nerve and artery toward the ulnar side of the median nerve. Since then, the ulnar approach has been used by several authors including Eslamian et al.,¹³ Ustun et al.,¹⁴ Lee et al.,¹⁵ and Makhlouf et al.²⁴ who reported significant improvements in VAS, BCTQ, and electrophysiological parameters using this approach. Technically, the US-guided ulnar approach can be performed in either “in-plane” or “out-plane” methods, depending on the orientation of the needle with respect to the plane of the US transducer (parallel or perpendicular, respectively). In a meta-analysis by Chen et al.,²⁵ the authors compared these methods and showed that the pool estimate of BCTQ functional and SS scales improvements favored “ulnar side in-plane approach” in comparison with “ulnar side out-plane approach.” There has been no report of serious adverse events using either of these techniques. In a recent study, Babaei-Ghazani et al.²⁶ performed a meta-analysis of randomized trials comparing the effectiveness of US-guided (ulnar side approach) versus landmark-guided injections on SS, FS, and electrophysiological outcomes of patients with CTS. Their findings demonstrated that US-guided injection was more effective in improving SS; however, no significant differences were found with respect to the FS or electrophysiological improvements between the two methods.

The US-guided CTS injection is most commonly performed using a transverse-view US image of the carpal tunnel.^13–15 However, in one study, Hong et al.¹⁶ introduced CTS injection using longitudinal view of carpal tunnel where the injection was performed superficial to the long-axis view of the median nerve and beneath the flexor retinaculum. In a comment letter, Ata et al.²⁷ challenged this technique and noted several drawbacks of CTS injection in the long-axis view. These included uncertainty regarding distribution of steroids beneath the flexor retinaculum, difficulties in distinguishing flexor retinaculum from the swollen superficial edge of the median nerve, and higher risk of median nerve insult when using long-axis view for injection. To date, no study has compared the efficacy and adverse events of US-guided CTS injection in transverse view versus longitudinal view. Further studies are warranted to compare treatment outcomes of these two approaches.

In our study, we used in-plane technique for both “above” and “under” the median approaches. The present study is probably the first report evaluating these two techniques, and therefore, we cannot compare our findings with previous ones. The strength of this study was the use of objective (i.e., electrophysiological and ultrasonographic parameters) along with subjective (VAS and BCTQ) outcome measures for comparing the effectiveness of two injection techniques. However, the findings of this study must be interpreted in view of its limitations. The main limitation was the relatively small sample size and short duration of follow-up period. Another limitation was the fact that we were not able to completely blind the injecting physician because of the nature of the interventions. Finally, we did not include a control group (receiving no injection but only cock-up splint) in this study, and therefore, we cannot evaluate the extent of splint use or natural recovery in the improvement of clinical findings. This study was the first on this topic, and further studies, with larger sample size and longer follow-up, which also include US diagnostic parameters (including median nerve vascularity, mobility, and wrist to forearm cross-sectional ratio), are warranted to compare these methods.

CONCLUSIONS

In conclusion, our study demonstrated that both US-guided above and below median nerve injection techniques were effective in reducing the symptoms and improving the FS, cross-section area of the median nerve, and electrodiagnostic findings of patients with mild to moderate CTS. However, no difference was observed between the two techniques at 6 and 12 wks of follow-up evaluation. No serious adverse effects were observed in either of the methods.