Fibrolipomatous hamartoma of the median nerve in a child with carpal tunnel syndrome: imaging findings and literature review

Abstract

Fibrolipomatous hamartoma of the median nerve is an uncommon benign tumour of the childhood, which usually manifests in adolescents or adulthood with signs of compressive neuropathy at wrist. Symptomatic tumour is unusual in children below 5 years age and can be underdiagnosed. Magnetic resonance imaging provides pathognomonic features for the diagnosis, obviating the need for biopsy. Although standard ultrasonography is frequently the first-line imaging approach in the evaluation of soft-tissue masses, sonographic findings of this lesion are less frequently reported and have to be kept in mind by radiologist. We report the unusual case of carpal tunnel syndrome secondary to fibrolipomatous hamartoma of the median nerve in a 4-year-old child successfully treated with surgical carpal tunnel release.

Introduction

Fibrolipomatous hamartoma (FLH) is a very rare benign tumour in childhood and consists of an overgrowth of fibrous and adipose tissue within the peripheral nerve sheath. The median nerve (MN) is often the most affected (80% of cases) [1, 2]. FLH of MN (FLHMN) can be present since the birth as an asymptomatic, slow growing mass leading to progressive compressive neuropathies such as carpal tunnel syndrome (CTS) usually in adolescence or adulthood [1–3]. A symptomatic FLHMN is rather rare in children below the age of 5 years and can be easily misdiagnosed [1]. Ultrasound (US) is usually the first imaging technique in the evaluation of this soft tissue mass based on its accessibility, zero radiation, high resolution on superficial soft tissues and the allowance of real-time imaging [4–6]. Although several imaging studies emphasize magnetic resonance imaging (MRI) features as pathognomonic of FLHMN, US findings are poorly discussed [3]. On US, the typical FLHMN at wrist appears as a hyperechoic fatty mass containing coaxial hypoechoic bands, which correspond to the neural bundles. Sonographic findings overlap with the MRI features, which consisting in hypointense longitudinal structures (the thickened nerve fibres) encased in a well-circumscribed fibro-adipose mass, hyper-intense on T1 and T2 scans and hypointense on fat-suppressed images [7, 8]. We highlight the role of standard US in the first diagnostic suspicion of FLHMN and its useful correlation with the MRI for the definitive diagnosis.

Case report

A 4-year-old boy with a faint swelling at the right palm hand was admitted to the vascular anomalies center of our hospital for pain and numbness at wrist occurred about 6 months before. Physical examination revealed soft elastic, poorly compressible, non-fluctuating and non-pulsating mass with normochromic skin located at level of the second and third metacarpal bone, medially to the thenar eminence. No enlargement was detected at wrist, which appeared painful to palpation. No anomalies were observed at the fingers (Fig. 1). Parents reported first appearance of the lesion about 2 years earlier as a little asymptomatic slow growing expansion. According to the “Face, Legs, Activity, Cry, Consolability” (FLACC) scale, the patient had a spontaneous pain score of 0, while during activity or digital pressure he had a score of 5. US examination was performed in the suspicion of a low-flow vascular malformation. US images revealed a solid fusiform well-defined hyperechoic formation with multiple internal hypoechoic bands organized as fascicle configuration. Neither lymphatic cysts nor venous lakes containing stagnant blood, clots or calcifications were observed within the lesion. No vascular signals on color-Doppler scans were identified. The lesion extended superficially within the subcutaneous fat from the second and third metacarpal head level to the distal forearm on the volar side, while appeared subfascial, narrowed and closely bordering to the flexor tendons within the carpal tunnel (Fig. 2). Moreover, the mass appeared to expand from the MN in the distal forearm, suggesting an origin from the peripheral nerve. FLHMN was hypothesized and MRI was performed. MRI confirmed the sonographic features and especially showed an enlarged hourglass-shaped MN mass consisted of tubular low-intensity structures (representing thickened nerve fascicles) surrounded by a hyperintense fibro-adipose tissue presenting both “coaxial cable” and “spaghetti string” appearance on axial and coronal images respectively. Slight and inhomogeneous contrast enhancement of the lesion was observed (Fig. 3). MRI findings were pathognomonic for FLHMN. Electroneurography displayed both absence of sensory nerve action potential (sNAP) and significantly lower value of the compound muscle action potentials (cMAPs) in the right MN (Table 1). Due to axonal sensory-motor neuropathy, child underwent surgical decompression of the MN through open carpal tunnel release (CTR) and transverse carpal ligament section (Fig. 4). Symptoms of CTS completely disappeared few weeks after surgical treatment. At 4-month follow-up, the child continued to be fully asymptomatic.

Fig. 1.

Right palm hand on coronal (A) and left-oblique (B) views. A faint central swelling (arrows) at level of the third metacarpal bone and medially to the thenar eminence is shown

Fig. 2.

Transverse (A and B) and sagittal (C) US images of the right carpal-metacarpal region showing markedly enlargement of the median nerve (dotted line) consisting of hypoechoic nerve fascicles surrounded by fibro-adipose hyperechoic tissue. The mass appears closely bordering to the underlying flexor tendons (arrows). c: cortex of capitate; r: cortex of distal radius

Fig. 3.

Axial T1-weighted DIXON in-phase (A) and fat only (B), coronal T1-weighted VIBE fat saturated post contrast (C) and sagittal T2-weighted DIXON fat only (D) MRI images. The tumour contains serpentine hypointense nerve bundles embedded within the fat and enhancing after gadolinium-based contrast administration. Transverse (A and B) images show the “coaxial cable-like” appearance of the lesion (blue arrows) contacting and displacing the underlying flexor tendons (red arrows). Coronal (C) and sagittal (D) views demonstrate hourglass and fusiform enlargement of the median nerve with “spaghetti-like” aspect (yellow arrows)

Table 1.

Electroneurography (ENoG) examination showing sensory and motor axonal mononeuropathy affecting the right median nerve

	Onset (ms)	Duration (ms)	Amplitude (µV)	Area (µVms)	Distance (cm)	Velocity (m/s)
Sensory nerve conduction velocity
Right: median
I digit	–	–	–	–	7.0	-
II digit	–	–	–	–	10.0	-
III digit	–	–	–	–	10.5	-
Left: median
I digit	2.0	1.3	48.4	29.6	10.0	50.0
Motor nerve conduction velocity
Right: median
Wrist	2.4	14.3	6.1	24.0	–	–
Elbow	5.3	17.6	5.7	27.5	19.0	65.5
Axilla	6.7	16.2	5.4	29.1	8.5	60.7
Left: median
Wrist	2.8	8.9	13.7	49.9	–	–

Values of the test demonstrate absence of SNAP (sensory nerve action potentials) from right sensory median nerve, and CMAP (Compound muscle action potentials) from right motor median nerve reduced in amplitude, without conduction blocks

Fig. 4.

Intra-operative view of the FLHMN at wrist after section of the transvers carpal ligament during surgical CTR

Discussion

In contrast to the adult patients, nontraumatic compressive neuropathies of the upper limb and particularly the CTS are rare conditions in children, [7, 9, 10]. Paediatric CTS is usually secondary to genetic underlying pathologies such as lysosomal storage disease, bony dysplasia or malformations of the anatomical structures of the carpal tunnel. Other less frequent causes are the enlargements of the MN by tumours, pseudotumours or vascular malformations [7, 9].

FLH represents an unusual cause of CTS in childhood [7]. FLH is a very rare benign tumour, especially in paediatric patients, characterized by an abnormal diffuse growth of fibro-adipose tissues within the nerve sheath [1, 2, 7, 11, 12]. This entity usually afflicts the upper extremity with particular predilection for the MN (60–80% of cases) [1, 2, 7]. Varying different terms have been employed to describe the lesion, including fibrolipoma, fibrofatty proliferation, intraneural lipoma, neurofibroma, lipomatous hamartoma or fatty infiltration of the median nerve [1, 2, 7, 11, 12]. Word Health Organization included FLH within the group of lipomatosis of the nerve in the 2013 [2]. FLH usually occur unilaterally, mainly in the right side, but few cases of bilateral localization were described [7, 12–14]. Females and males can be equally affected [12]. FLH can be also associated with other pathologies such as Klipplel–Tranaunay–Weber Syndrome, Proteus Syndrome, vascular malformations, bone and fat tumours and particularly the macrodactyly in the macrodystrofia lipomatosa [11, 12]. The most common sign of FLHMN is a slowly increasing soft tissue enlargement at wrist or forearm, often present since birth, which clinically reveals during the first two or three decades of life because of nerve compression symptoms, mostly CTS [1, 11, 15]. Biazzo et al. affirm that the prevalence of FLH is very low and most publications refer to isolated cases [12]. Symptomatic patient are commonly adolescents or adults [1]. Tahiri et al. provided the first comprehensive review in the literature on FLHMN reporting 180 cases from 1946 to November 2012. Of those, 88 had age at consultation between 0 and 19 years [11]. Ranyan et al. reviewed 16 cases between adolescents, young adults and elderly patients in a literature research from 2012 to 2019 [2, 11]. However, in children below the age of 10 years this rare entity is rather unknown to the clinician and therefore can be easily misdiagnosed. Moreover, symptomatic FLHMN is uncommon in children below the age of 5 years [1]. Senger et al. report a literature review from 1986 to 2012 of 15 only paediatric (< 18 years of age) cases affirming that CTS is a relatively rare condition in children and moreover FLHMN an unusual cause of CTS in paediatric patients [7]. Several differential diagnoses have been described, including ganglionic cysts, hereditary hypertrophic neuritis of Dejerin-Sottas syndrome, vascular malformations, traumatic neuromas, schwannomas, neurofibromas and intraneural lipomas [7, 11].

However, imaging is the most quick and reliable way to establish the correct diagnosis of FLHMN [1]. MRI is the gold standard diagnostic tool and shows pathognomonic features. On MRI, FLHMN appears as thickened longitudinal structures corresponding to the nerve fibres, hypointense in all sequences, encased in a well-circumscribed fibro-adipose mass, hyper-intense on T1 and T2 scans and hypointense on fat-suppressed images. Particularly, on axial images lesion presents the “coaxial cable-like” aspect, while coronal and sagittal scans show the “spaghetti string” appearance. No or mild enhancement on post-contrast images is usually detected within the lesion [1–3, 7, 11, 12, 16]. MRI can definitively rule out alternative diagnosis and avoid the biopsy, which represents an unnecessary risk and may cause functional sequelae [1, 7, 11, 12]. However, the first choice imaging approach to the soft tissue lesions is usually through standard US [4]. Although high-resolution nerve US is an emerging technique for the study of peripheral neuropathies, cannot be widely available in most imaging centers and most conventional US proves may not able to detect the fascicles of the nerve [1, 17]. Sonographic features of FLHMN have been scarcely defined and its diagnostic efficacy has been previously questioned [3]. Toms et al. affirm that for their knowledge firstly illustrated consisted echo patterns of FLHMN [8]. On US, FLH appears as coaxial hypoechoic bands, referring to the neural bundles, within a hyperechoic mass representing fibro-adipose tissue. No or poor vascularization on color-Doppler examination is usually demonstrated [3, 7, 8, 11]. Sonographic patterns are highly similar to the MRI findings and may be characteristic of the lesion [3, 8]. Some authors consider US the most convenient technique so that may obviate the need for MRI and biopsy [8, 18]. However, we agree that in suspicion of FLHMN US features have to be verified by MRI [3].

There is not definitive therapy for FLHMN and treatment options are determined case by case [7, 11]. FLHMN cannot be completely surgically excised because of the risk of significant neurological deficit. Conservative treatment is the primary gold standard whenever possible [7]. In patients with compressive neuropathy CTR is the recommended treatment of choice and full resolution of symptoms of CTS in paediatric patients is reported [7, 11, 12, 19]. Section of the flexor retinaculum helps to decrease motor and sensory impairments and provides symptomatic relief from compressive median neuropathy [11]. However, therapeutic management of FLHMN is controversial and has to be determined case-by-case [7, 11, 12]. In this view, Tahiri et al. propose useful diagnostic and therapeutic algorithms for FLHMN, indicating CTR for both symptomatic and asymptomatic patients [11].

In conclusion, it is rather rare that FLHMN becomes symptomatic in children below the 5 years. Although less frequently described, standard sonographic findings overlap with the MRI features and can be equally considered representative of the lesion. As for each soft tissue lesion usually analysed in first instance trough US, every paediatric radiologist has to be familiar with the sonographic pattern of FLHMN. Recognizing typical imaging findings of FLHMN may obviate histological confirmation for the definitive diagnosis.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Guglielmo Paolantonio, Marco Cirillo and Francesca Grussu. The first draft of the manuscript was written by Guglielmo Paolantonio and Marco Cirillo and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Declarations

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose.

Consent to participate and to publish

Written informed consent was not obtained because the submission does not include images that may identify the person.