Transungual Surgical Excision of Subungual Glomus Tumour

INTRODUCTION

Glomus tumour is considered to be a hamartoma developing from the neuromyoarterial glomus body, which is a highly specialised arteriovenous anastomosis responsible for thermoregulation. The glomus body consists of an afferent arteriole, a tortuous arteriovenous anastomosis, a system of collecting veins and a neurovascular reticulum that regulates the flow of blood through the anastomosis.[1] Glomus bodies are present in the reticular dermis throughout the body, but are highly concentrated in the digits, palms and soles. Reflecting a similar distribution, glomus tumours may affect any area of the body, but up to 75% occur in the hand and approximately 65% of these are in the fingertips, particularly in the subungual space.[23] Diagnosis is mainly on clinical grounds although imaging has been reported to be useful in localising the lesion and determining its size. Magnetic resonance imaging (MRI) has been reported to be the most useful modality in this regard and can pick highly vascular lesions as small as 2 mm.[4] A complete surgical excision is the only effective treatment modality; however, the risk of post-operative nail deformity and recurrences are the major concerns in this regard. We present our experience with the management of subungual glomus tumours using the transungual excision approach.

PATIENTS AND METHODS

Patients

A total of 10 patients presenting with a clinically suspected subungual glomus tumour, over a period of 4 years (2009-2012) were included in this study. All patients consented to inclusion in the study and agreed to long-term follow-up. The principles of the 1975 Declaration of Helsinki with respect to the ethical treatment of study subjects were followed throughout the study. A record of clinical history and any apparent nail changes was kept [Figure 1a–d]. Plain radiographs were done for all cases. An MRI was done for five of these due to resource constraints.

Close

Figure 1

Varying clinical presentations of subungual glomus tumour. All the patients presented with complaints of pain in the involved digit with point tenderness (shown with an arrow) and cold sensitivity (a) Patient 2 presenting with severe pain with an apparently normal nail plate, (b) Patient 3 with a diffuse, erythematous to bluish discoloration visible over the nail plate surface, (c) Patient 4 with frank distal onycholysis, along with a larger erythematous, diffuse area visible over the nail plate. Slight swelling over the proximal nail fold can be appreciated, (d) Patient 6 presented with distal and lateral onycholysis, nail plate discoloration, subungual debris and partial destruction

Export to PPT

The lesion was localised on the basis of clinical signs that are the point of maximal tenderness (Love's test); the point of visible erythema of the nail plate; or the painful base of an erythronychial band. The surgical site was prepared antiseptically.

• The digit to be operated was anaesthetized with proximal nerve block using the 2% lignocaine hydrochloride with epinephrine (1:2,00,000) - 1% of lidocaine hydrochloride solution

• Digital exsanguination and tourniquet for relatively avascular operating field was done

• The nail plate overlying the tumour was carefully removed. For nail matrix glomus, the proximal nail fold (PNF) was also retracted (lateral release incisions and stay sutures) as shown in Figures 2–4

• As seen in Figure 5, blunt dissection with fine forceps enabled separation of the tumour from surrounding tissues in seven patients. Punch excision of a small lesion was performed in two cases [Figure 6], while in a single patient a longitudinal, fusiform excision was done [Figure 7]

• Subsequent defect was primarily repaired with 6-0 absorbable sutures

• The avulsed nail plate was trimmed and repositioned wherever possible [Figure 3]

• The lateral nail fold (LNF), if retracted was sutured back with 4-0 non absorbable sutures

• Dressing was done using antibiotic cream and gauze. The excised tissue was sent for histopathological confirmation.

Close

Figure 2

(a-c) Patient 4 presenting with a large subungual glomus tumour (11 mm) involving the nail matrix. The lesion is exposed with the help of nail plate avulsion and retraction of proximal nail fold. Post excision defect can be seen in Figure 2c

Export to PPT

Close

Figure 3

(a-f) Patient 10 presenting with subungual glomus tumour involving the thumb. Pre-operative localization with magnetic resonance imaging helped confirm the location of the tumour. Partial proximal nail avulsion is done. The proximal nail fold (PNF) is retracted and the tumour is removed. The resultant defect and the PNF are sutured. The avulsed nail plate is repositioned and secure dressing is applied

Export to PPT

Close

Figure 4

(a-d) A large subungual glomus tumour arising from the matrix is removed and the resultant defect is primarily repaired. However, on follow-up, a residula median nail plate defect is seen which persists in the distal part even on 1 year follow-up

Export to PPT

Close

Figure 5

(a-d) A relatively small nail bed glomus tumour, removed with the help of blunt dissection. The surgical defect is closed in a longitudinal fashion (Patient 2)

Export to PPT

Close

Figure 6

(a-d) A more proximal, smaller lesion is excised with the help of a punch. The residual defect is small and could be closed primarily (Patient 1)

Export to PPT

Close

Figure 7

(a-d) A larger nail bed glomus tumour is excised in a longitudinal fashion. The resultant longitudinal defect is also closed primarily (Patient 3)

Export to PPT

Patients returned for post-operative visits at 2 days and then on a weekly basis until the site healed. Any non-absorbable sutures used were removed at the end of the 1^st week. After complete healing, monthly review was done for assessing recurrence and post-operative nail plate deformity.

RESULTS

Of the 10 patients included in the study, 7 were females and 3 were males. Age at presentation ranged from 24 years to 47 years, with a mean age of 33.3 ± 7.55 years among females and 40 years for males. Most of the patients had glomus tumour involving the non-dominant hand (9 out of 10) with a single female patient having involvement of the dominant right hand. Among the digits involved, five patients had lesion on the fourth digit; three had involvement of the fifth digit, while two patients had thumb involvement. The duration of complaints at presentation ranged from 1 to 16 months (average duration 6.5 months). The details of the patients are summarised in Table 1.

Table 1 Clinical presentation and surgical outcome of patients with subungual glomus tumours involving the fingernails (n=10)

On examination, Love's sign was present in all patients (100%), while Hildreth's test was positive in 4 (40%). Nail plate showed variable manifestations, with a longitudinal red band (erythronychia) being the most common, seen in 50% of cases. Other nail plate signs included subtle tumefaction in three, distal onycholysis in one, distal and lateral onycholysis in two and nail plate discoloration in two, sub-ungual hyperkeratosis in one and thinning in one [Figure 1]. Importantly, the nail plate was normal in three of our patients [Figure 1a]. Secondary nail fold features in the form of erythema and tumefaction were noticed in five cases (on the PNF in three cases and LNF in two cases).

X-ray of the involved digit was done for all the cases and it showed only cortical thinning in three patients and was normal in the rest. MRI was done in five affording patients. Only in three of these, the tumour could be well visualized (Table 1). MRI in these three patients showed the classical findings of low signal intensity on T1-weighted images, marked hyper intensity on T2-weighted images and strong enhancement after the injection of gadolinium-based contrast material [Figures 8a–b and 9a–e].

Close

Figure 8

(a and b) Sagittal gadolinium-enhanced fat-saturated T1-weighted MR images, reveal a well-defined mass with homogenous enhancement beneath the nail plate (Patients 6 and 8)

Export to PPT

Close

Figure 9

(a-e) Glomus tumour in patient 10 presenting with a painful left thumb (a) Sagittal T1-weighted magnetic resonance (MR) image shows a focal lobulated hypointense lesion in the nail bed of the first distal phalanx (b and c) Sagittal T2-weighted MR (b) and short TI inversion recovery (c) Image shows the tumour with high signal intensity, (d and e) On gadolinium-enhanced fat-saturated T1-weighted MR image, the tumour shows prominent enhancement with a homogeneous pattern

Export to PPT

Surgical excision was performed by transungual route in all patients. The tumour was located over the nail bed in five and nail matrix in the other five patients. The excised tumours ranged in size from 3 mm to 11 mm (mean 6.1 mm). The post-operative period was uneventful, except for a mild continuous pain, persisting for about a week, which was reported by all the patients. The excised lesion was sent for histopathological evaluation which confirmed the clinical diagnosis in all 10 patients.

A prolonged post-operative follow-up was done for all cases. The period of follow-up ranged from 5 months to 14 months (mean follow-up duration of 9.9 months). Only two of our patients went on to develop a persistent longitudinal ridging of the nail plate as a residual defect post-surgery [Figure 4]. Rest of the eight patients did not develop any permanent, visible nail abnormality [Figure 10]. A recurrence of pain was reported in the same digit in two cases after a period of 5 and 12 months respectively; however, in both cases, it was attributed to an isolated lesion arising in a different quadrant as compared to the one previously operated. Both the recurrences were again dealt with surgically.

Close

Figure 10

(a-d) Post-operative growth of a normal nail plate. The residual defect is seen to grow out distally and an essentially normal nail plate can be seen at 8 months of follow-up

Export to PPT

DISCUSSION

Glomus tumour was first described by Wood in 1812 and was named by Masson in 1924.[5] Although, the exact aetiology of glomus tumour is unknown, it has been proposed that a structural weakness in the glomus body could lead to reactive hypertrophy following trauma; or alternatively, they may represent hamartomas in which individual components are hypertrophied.[5] The glomus cells are specialized smooth muscle cells derived from Zimmerman pericytes, particularly concentrated around the dilated vascular spaces. In addition, nerve fibres and mast cells may be present in increased numbers. The tumour is exquisitely sensitive to changes in temperature which leads to contraction of myofilaments. This results in an increase in intracapsular pressure, transmitted by unmyelinated nerve fibres, leading to the perception of pain.[6]

Glomus tumours most commonly affect patients in third to fifth decades of life, as was also seen in our study, although cases have been described in all age groups.[7] A female predominance has been reported previously,[7] as was seen in our series as well. Patients with subungual glomus tumours often present to multiple specialties and quite often the diagnosis is delayed; hence, the duration of symptoms at presentation may be long. In our patients, however, the mean duration of symptoms was relatively less at 6.5 months (range 1-16 months). This may be because of the structure of our health system where patients can directly reach specialists, without having to be referred step by step.

Diagnosis of glomus tumours can largely be made on the basis of clinical signs and symptoms. In a series of 51 patients with glomus tumours of the hand, Van Geertruyden et al. noted spontaneous pain in 80% of patients, sensitivity to touch in 100% of patients, cold sensitivity in 63% and blue-reddish discoloration in 29% and nail deformity in 33%.[8] Pain may be described by the patient as severe and paroxysmal, often radiating to shoulder. Severe pain was the presenting feature in all our patients and all of them also had significant cold sensitivity. However, three of our patients had no signs suggestive of a subungual mass and only presented with pain localised to subungual region. This is the perfect setting for misdiagnosis.

Love's sign and Hildreth's test are often helpful in achieving the diagnosis. Love's sign refers to severe localized tenderness elicited on applying pressure with a pinhead, along with relief of pain when the pressure is removed. Hildreth's test is performed by elevating the patients’ arm to exsanguinate it. A tourniquet is placed over arm and inflated to 250 mm Hg. This causes reduction or abolition of lesional pain and tenderness. A test is positive when releasing the cuff causes a sudden return of pain. According to Bhaskaranand and Navadgi's series of 18 patients, Love's test was 100% sensitive and 78% accurate, while Hildreth's test was 71.4% sensitive, 100% specific and 78% accurate.[9] All ten of our patients (100%) also tested positive for Love's sign and Hildreth's test was positive in 40% of cases.

Radiology has a small but definite role in the management of glomus tumours. X-ray may reveal thinning or erosions of the underlying cortical bone, especially with larger lesions; however, the frequency of this observation ranges from less than 30% to up to 60% of cases in various series.[5] It may be useful for differentiating glomus tumours from subungual exostoses.[1011] We found cortical thinning in three of our patients (30%), two of whom had lesions of size 6 mm each and the third had an 11 mm tumour. Ultrasonography has also been used to localise and assesses the size of glomus tumours pre-operatively. These are visible as well-circumscribed hypoechoic masses. The technique has limited utility for small, flattened lesions and in the artifacts that may be generated by the nail preventing glomus tumours being distinguished from other hypoechoic masses.[5] Doppler ultrasonography can identify low signals within the hypoechoic mass and is preferred over standard ultrasonography. Chen et al. have reported 100% detection rates and no false negative results with Doppler imaging in one series.[12]

MRI is a non-invasive technique that is most useful in evaluating glomus tumours pre-operatively. It can detect lesions as small as 2 mm. The lesions appear as hypointense on T1 and hyperintense on T2 images, with marked enhancement on injection of gadolinium contrast. A nidus appearance with a high signal central dot surrounded by a zone of less signal intensity is characteristic of glomus tumours.[13] Al-Qattan et al. reported a sensitivity of 90% and positive predictive value of 97% in their series of 42 patients. However, specificity was only 50% and negative predictive value 20%.[13] MR angiography gives a typical appearance of strong enhancement in the arterial phase and tumour blush that increases in size in the delayed phase.[114] MRI can also be useful for assessment of patients with recurrence or incomplete resolution of symptoms following surgery.[15] However the high cost precludes its routine use. In our study, MRI was not helpful in detecting a glomus tumour, later confirmed pathologically, in three of the five patients in whom it was performed.

Histologically, glomus tumours are made up of three components: Glomus cells, vascular structures and smooth muscle cells; therefore, according to the predominant component, they can be categorised into three types. Solid glomus tumours (75% of cases) comprise of a few vascular structures and very few, if any, smooth muscle cells. Glomangiomas (20% of cases) have a prominent vascular component, whereas glomangiomyomas (5% of cases) show a predominance of vascular structures and smooth muscle cells.[516] Originally, glomus cells were believed to arise from endothelial cells; however, recent histologic evidence points towards their being a type(s) of smooth muscle cell.[171819] Histochemical staining reveals glomus cells to be positive for vimentin and alpha-smooth muscle actin, but negative for desmin.[2021] All our patients showed the typical findings associated with solid glomus tumours, but immunochemistry could not be performed because of resource constraints.

Treatment of glomus tumours is purely surgical. However, the major concerns are the risk of nail deformity and recurrences and rates of recurrence reported by various authors range from 4% to 15%.[8222324] Incomplete surgical removal may lead to persistent/recurring pain. However, it is now believed that most recurrences result from pre-existing synchronous satellite lesions not seen at the time of initial surgery and which become larger and symptomatic with time.[25] This was demonstrated well by Gandhi et al. in five patients of recurring glomus tumours, all of whom developed lesions in different locations of the same digit.[25] MRI may be useful in localising the recurrent lesions. Recurrence developed in two of our cases (20%). In both cases, the recurrent lesion was located in the same nail, however, in a different quadrant when compared to the previously excised one.

Two main approaches for the surgical excision of glomus tumours are - transungual and lateral (lateroungual or laterodigital Keyser-Littler approach).[26] We used the transungual approach in all our patients, as it provides for a better view of the whole subungual region and makes precise and complete excision possible. It is also useful when the exact location is not known beforehand. If the lesion is believed to be involving the nail matrix, an eponychial flap has to be raised allowing for a clearer view. It is prudent to inspect the whole subungual region for reasons mentioned above. Previous reports have indicated a higher rate of the nail plate deformity with transungual approach when compared to lateral approaches.[26] This may be because of the nail bed incongruities created during the excision or formation of adhesions between matrix and PNF.[26] Careful handling of tissues during surgery and meticulous repair of the defect created by excision can largely prevent this outcome. This fact is substantiated in our series of cases, where only two patients developed a post-operative longitudinal ridging of the nail plate, owing largely to the initial large size of the lesions. Nail plate repositioning has also been used as a method to prevent adhesions of the eponychium to the matrix. It also favours repair of the nail bed and reduces the pain occurring when dressings are changed, as the wound is kept covered.[26] This proved to be helpful in our patients. Li et al. have reported the use of autologous fat grafting to repair the defect in nail bed with good results.[27] We simply used 6-0 absorbable sutures to close the defect and found encouraging outcomes as mentioned previously.

CONCLUSION

The diagnosis of subungual glomus tumour is largely clinical. Radiological examination, including MRI may not offer much help in the majority of cases. Complete surgical excision offers the only prospect of complete cure. We found transungual approach to be a simple procedure, which can reduce the chances of recurrence by allowing examination of the entire subungual region. Nail deformities are minimal if adequate care is taken during surgery.