Evaluation of Gabapentin and Transforaminal Corticosteroid Injections for Brachioradial Pruritus

Abstract

This case series compares the reported outcomes of transforaminal injections and oral medications in patients with a brachioradial pruritus diagnosis.

Brachioradial pruritus (BRP) is a neuropathic condition affecting the skin of the neck, upper back, and arms. Diagnostic techniques are limited.¹ Effective treatment options, including gabapentinoids and transforaminal corticosteroid injections, remain few and understudied.^2,3 Herein, we outline key diagnostic modalities and therapeutic benefits of transforaminal injections and high-dose gabapentin for BRP.

Methods

The University of California, San Francisco (UCSF), Institutional Review Board approved this case series and waived the informed consent requirement because the large number of records used made contacting all participants impractical. We followed the AJO reporting guideline.

We identified the cohort from 539 patients referred for unspecified pruritus to the UCSF Itch Center from 2010 to 2020. One of us (T.G.B.) manually selected cases based on BRP diagnosis. Benefit was defined as reduction in numerical rating scale (NRS) score at follow-up. eMethods in the Supplement describes imaging modalities and analysis.

Statistical analysis was conducted with IBM SPSS, version 27 (IBM). Two-sided P < .05 indicated statistical significance.

Results

The cohort included 53 patients with BRP (mean [range] age, 67 [35-96] years; 36 women [68%], 17 men [32%]), consistent with previous studies.⁴ Seven of 53 patients (13%) had an accurate diagnosis despite mean (median) symptom duration of 6.29 (2.5) years (Table).

Table. Demographic Characteristics, Presentation, and Diagnosis.

Characteristic	No. (%) (n = 53)
Sex
Female	36 (68)
Male	17 (32)
History
Cervical spine disease	39 (74)
Trauma	20 (38)
Motor vehicle accident	9 (45)
Fall	4 (20)
Unspecified	7 (35)
Age, mean (range), y	67 (35-96)
Presentation
Previous diagnosis of BRP	7 (13)
Symmetrical	35 (66)
Location of symptoms
Forearms and other sites	28 (53)
Forearms only	17 (32)
Other sites onlya	8 (15)
Duration of symptoms, mean (median), y	6.29 (2.5)
Diagnosis
Imaging	31 (58)
Noncontrast MRI	30
Anteroposterior and lateral radiographs	1
Cervical spine disease	39 (74)
Foraminal stenosis	28
Spondylosis	2
Otherb	1

Abbreviations: BRP, brachioradial pruritus; MRI, magnetic resonance imaging.

^a Other sites included neck, shoulders, and upper arms.

^b Cervical spine fusion procedure.

Thirty-two patients (60%) received gabapentin. Nine of 12 patients (75%) who received a mean (range) dose of 1500 (1500-3900) mg/d or more benefited, compared with 11 of 20 (55%) who received a mean (range) dose less than 1500 (100-1200) mg/d. Responding patients received a mean (range) dose of 1455 (300-3900) mg/d, compared with a mean (range) dose of 900 (100-2100) mg/d in nonresponding patients.

Thirty of 31 patients with available cervical (C)-spine imaging underwent noncontrast magnetic resonance imaging; radiographs were obtained in 1 patient (Table). In all cases, imaging confirmed C-spine abnormalities involving 1 or more vertebral levels associated with the dermatomal distribution of their symptoms. Radiographic severity was not associated with symptoms. However, patients with severe symptoms (NRS ≥7 or sleep interruption) had significantly more C-spine levels affected than patients with NRS lower than 7 (mean difference 1.76; 95% CI, 0.80-2.72; P = .002).

Ten patients received computed tomography–guided C-spine transforaminal epidural corticosteroid injections. Six of 9 patients (67%) reported benefits at follow-up. In 5 patients, transforaminal injection was associated with a mean NRS reduction of 5.8 at 9- to 12-month follow-up, compared with a mean NRS reduction of 3.3 in 13 patients who received medical therapy alone. Nine of 10 patients (90%) who received transforaminal injection also received oral medical therapy (gabapentinoids, antidepressants, and/or cannabinoids), with 1 patient receiving topical calcineurin inhibitor monotherapy. All 10 patients received gabapentin, but concurrent gabapentin therapy (in 5 patients) did not anticipate treatment responses (3 patients benefited; 2 did not).

Vertebral levels injected included C2-3 to C6-7, with some patients receiving multilevel injections. Injections were most frequently placed at C5-6 (n = 6), followed by C4-5 (n = 3) and C6-7 (n = 2). In postprocedure imaging analysis, injectate was observed in the epidural space in all responding patients but did not reach the epidural space and was distributed foraminally or extraforaminally in nonresponding patients (Figure).

Figure. Transforaminal Corticosteroid Injection Computed Tomography (CT) Scan Images of Responding and Nonresponding Patients.

A and B, Needle tract (red arrowhead) terminating within the neural foramen (circled) delivers the corticosteroid injectate with contrast into the neural foramen and epidural space. The corticosteroid has not yet been injected to the contralateral side. C, Bilateral contrast (red arrowheads) is visible foraminally and extraforaminally, but not within the epidural space.

Discussion

Despite having classic BRP symptoms, few patients had an accurate diagnosis when referred to the center. Transforaminal corticosteroid injections were effective for BRP, including in cases lacking sufficient improvement from oral therapies. Whether injectate reached the epidural space of affected spinal nerves appeared to be the most important variable in good clinical response. Similarly, the findings support the benefits of gabapentin. We observed a pattern of better treatment response associated with higher daily doses. Thus, clinicians should consider uptitrating gabapentin to doses greater than that prescribed for other indications.⁵

Study limitations included a cohort drawn from a single urban tertiary care center; thus, findings may not be generalizable to other populations. Furthermore, follow-up NRS data were not available for all patients. These findings may guide dermatologists in all settings in treating patients with BRP.

Supplement.

eMethods.