JFP_03.06_PhotoRounds.Final 2/15/06 3:46 PM Page 225

PHOTO ROUNDS

Adam Leight, MD A non-healing ulcerated Department of Family Medicine, University of Kansas fingertip following Medical Center

man went to his primary care surgery clinic nearly 6 months after his physician 3 months after slamming original office visit. He was diagnosed A his right thumb in a car door. The clinically as having a giant pyogenic had turned black and sloughed off granuloma and was given antibiotics as several weeks later, leaving a red, draining well as silver nitrate sticks to cauterize on the tip of his thumb. The wound the wound daily. After missing several drained continuously for the next 2 months more follow-up appointments, the and showed little progress in healing. patient returned with a spongy, weeping His physician started him on anti- soft-tissue wound over the dorsum of his biotics, but the wound still showed no right thumb [that] doubled in size progress in healing over the next 6 weeks. over the past 3 months (FIGURE). Cultures were obtained that grew out Radiographs obtained at that time were Staphylococcus and Streptococcus spp. normal, but a bone scan revealed late Another course of antibiotics was given, uptake, cause for concern that this was but the patient’s condition failed to osteomyelitis. improve. At this point the patient was referred ■ What is the differential to a surgeon. He missed several appoint- diagnosis, and what tests ments before finally presenting to the are necessary?

FIGURE Ulcerated tip of the thumb

FEATURE EDITOR Richard P. Usatine, MD University of Texas Health Sciences Center at San Antonio

CORRESPONDENCE A spongy, weeping soft-tissue wound over the dorsum of the right Adam Leight, MD. E-mail: thumb. [email protected]

www.jfponline.com VOL 55, NO 3 / MARCH 2006 225 JFP_03.06_PhotoRounds.Final 2/15/06 3:46 PM Page 226

PHOTO ROUNDS

■ Diagnosis: lesions need to be biopsied. Subungual melanoma Subungual melanoma disproportion- Due to the aggressive nature of the wound ately affects nonwhites. While the total and the large soft-tissue defect, a plastic number of cases of subungual melanoma surgeon was consulted. Biopsy showed accounts for only 0.3% to 3% of all new malignant melanoma, and the thumb was cases of malignant melanoma, one study amputated at the base of the proximal found subungual melanoma accounted for phalanx. Pathology revealed a 3 x 3.5 cm up to 23% of all malignant melanomas in ulceration at the distal portion of the spec- Japanese persons, 17% of malignant imen with malignant melanoma involving melanomas in Hong Kong Chinese per- the skin, subcutaneous tissue, and bone sons, and 25 % of malignant melanomas marrow. Two of 3 axillary lymph nodes in African Americans.1 were also positive for metastatic malignant The thumb and big are the most melanoma. frequently involved regions, perhaps due to the larger proportion of nail matrix on History and epidemiology these digits, with 55% of lesions arising on While public awareness of cutaneous the hands, and more than half of those melanoma has been increasing, subungual involving the thumb.2 melanoma remains obscure. First described in 1834 by Alexis Boyer, surgeon to Cause is unclear Napoleon, it was later characterized in The cause of subungual melanoma is 1886 by Sir Jonathan Hutchinson, who thought to be different from that of cuta- reported 6 cases of “melanotic whitlow.”1 neous melanoma, but it remains unclear. Hutchinson reported that the lesion was Because the nail filters out UVB light and usually first attributed to an injury, and subungual melanoma most frequently aris- because of this the diagnosis was nearly es from a portion of the nail matrix that is always missed in the early stages. not sun-exposed, UV exposure is not Today, subungual melanoma is often thought to play the same role in pathogen- FAST TRACK neglected by patients and frequently mis- esis of subungual melanoma as it clearly More than half diagnosed by physicians. The estimated does in cutaneous melanoma.1 mean delay in diagnosis ranges from 3 to 24 Since the time of Hutchinson’s original of subungual months—nearly double the diagnostic delay report there have been numerous case melanomas are observed with cutaneous melanoma.1,2 reports and several series that describe mistaken for One study found 52% of subungual antecedent trauma in subungual melanomas were mistaken for benign or melanoma; however, there has never been benign/traumatic traumatic lesions of the nail bed such as conclusive evidence that trauma is a lesions; two thirds pyogenic granuloma, , ony- causative factor.3,4 of patients have chomycosis, chronic , subungual inappropriate , or pigmented nevus. This mis- take is not surprising as these lesions are all ■ Clinical presentation surgery in the differential diagnosis of a single pig- Nail pigmentation is the first clinical mented nail streak, and all are far more sign of subungual melanoma in more common than subungual melanoma. than 75% of cases, but few patients Another study found that two thirds present at this stage. Instead most of patients underwent some inappropriate patients delay presentation until changes surgical procedure before the correct diag- in the nail contour are evident, second- nosis was considered.1 Because of its poor ary infection supervenes, or ulceration of prognosis, often related to delay in diagno- the nail bed with granuloma formation sis, maintain a high index of suspicion for manifests.1 subungual melanoma in the proper setting Levit et al5 surveyed the world litera- and a sound understanding of which ture on “subungual melanoma” in 2000

226 VOL 55, NO 3 / MARCH 2006 THE JOURNAL OF FAMILY PRACTICE JFP_03.06_PhotoRounds.Final 2/15/06 3:46 PM Page 227

Non-healing ulcerated fingertip following injury ▲

and, based on their findings, described the TABLE 1 ABCDEF mnemonic (TABLE 1) to describe Salient features of subungual melanoma the salient features. When considering a nail bed lesion, the presence of any one of A = Age (5th to 7th decades), African-American, Asian, American these features should raise the clinician’s Indian index of suspicion for subungual mela- noma, while the presence of multiple B = Brown/Black pigment, Breadth (>3mm), Border variegation features should raise a significantly higher C = Change in nail band or lack of change in nail morphology concern. despite, presumably, adequate treatment

Nail pigmentation D = Digits most commonly involved (thumb, hallux) The 2 most important signs of subungual E = Extension of pigment onto the proximal/lateral nail fold melanoma are melanonychia striata (longi- (Hutchinson’s sign) tudinal brown to black pigmented streaks in the nail) and Hutchinson’s sign, which is F = Family or personal history of melanoma the spread of brown or black pigment from the nail bed, nail matrix, or nail plate onto the adjacent cuticle or onto the prox- Amelanotic subungual melanoma imal or the lateral nail fold. Fewer than 7% of cutaneous melanomas Many patients with subungual mela- lack pigment. In contrast, 20% to 33% of noma have a history of a thin pigmented subungual melanomas are amelanotic. streak that had remained unchanged for This makes the diagnosis of amelanotic years and then suddenly began to subungual melanoma at best difficult, and enlarge—eventually involving the entire often times impossible without a biopsy.1 nail bed with subsequent penetration to the eponychium or paronychium, ulcera- tion, or granuloma formation. ■ Management of Melanonychia striata. The differential subungual melanoma diagnosis of melanonychia striata is quite Nail biopsy FAST TRACK long, and most of these streaks are benign Untreated melanoma is potentially fatal, The early clinical (TABLE 2). Dark brown or black lines in while nail biopsy is technically difficult, the nails are common in Asians, African potentially disfiguring, and can be compli- signs of subungual Americans, and in dark-skinned individu- cated by scarring or pterygium formation. melanoma are nail als, and may simply represent ethnic varia- The physician must thus carefully deter- pigmentation, tion in pigment. Multiple streaks and mine which lesions require biopsy. Except streaks that do not extend distally from the in the special case when the clinician deems changes in nail proximal nail fold are nearly always the probability of subungual hematoma to contour, secondary benign. be very high (discussed below), it may be infection, and Single streaks greater than 6 mm wide, best to refer these patients to a specialist ulceration of those appearing in the sixth and seventh when available. decade of life, streaks with a variegated As a general rule any nail lesion, the nail bed color, or those that exhibit a broader prox- whether pigmented or not, that does not imal base or undergo any morphological heal with 6 to 12 weeks of conservative change (indicating an active process) are treatment should be biopsied. Lesions suspicious for subungual melanoma. causing nail dystrophy, ulcerating lesions, Hutchinson’s sign. Hutchinson’s sign is and those presenting with Hutchinson’s pathognomonic for subungual melanoma sign should all be biopsied. only when accompanied by ulceration of Caucasians. Managed conservatively, the nail bed or obliteration of the nail plate all single pigmented nail streaks in adult by granuloma. When present, a tissue Caucasians unresolved within 6 to 12 diagnosis must always be sought. weeks need to be biopsied. The clinician

www.jfponline.com VOL 55, NO 3 / MARCH 2006 227 JFP_03.06_PhotoRounds.Final 2/15/06 3:46 PM Page 228

PHOTO ROUNDS

TABLE 2 nicity, and these warrant clinical observa- tion and a diligent search for another Common differential diagnoses 1 of multiple pigmented nail streaks cause. The single exception to this is or periungual pigmentation metastatic melanoma to the nail bed, which may present as multiple pigmented DRUGS lesions on the same or different nails. Children. Subungual melanoma is Antimalarials Hydroxyurea Sulfonamide quite rare in children of any race, though Bleomycin Ketoconazole Tetracycline the incidence is not zero. As such, any new Cyclophosphamide Methotrexate Zidovudine Doxorubicin Minocycline pigmented streak warrants observation by 5-FU Phenytoin the physician, while pre-existing streaks unchanged for several years may by moni- SYSTEMIC CONDITIONS tored by the parent and child. Any mor- Addison’s disease Peutz-Jeghers syndrome phological change in a pre-existing streak Hemosiderosis Vitamin B12 deficiency or change in a new streak requires biopsy Porphyria Pregnancy in all adults and children. Radiation therapy Laugler-Hunziker syndrome Subungual melanoma Conducting a nail biopsy When the clinician believes a diagnosis of MICROBIAL subungual hematoma is likely, a simple AIDS Proteus mirabilis office procedure and test will often con- Blastomycetes Secondary syphilis firm the diagnosis.6 Any family physician Candida Trichophyton can perform this procedure and test, saving the patient both time and the expense of obtaining a consultation, and sparing both should thoroughly investigate non-melan- patient and physician the anxiety of wait- otic causes of single pigmented streaks ing days or even weeks to rule out a poten- with a careful history and a thorough tially life-threatening diagnosis. FAST TRACK physical examination. Seek out clues, espe- Procedure. First, soak the hand (or Any nail lesion— cially from the patient’s drug and treat- ) in warm water for 10 to 15 minutes ment history. Unchanged, lightly pigment- to soften the nail plate. Next, drive a 3- or pigmented or not— ed streaks already present for many years 4-mm punch biopsy through the nail plate, that does not heal on adult Caucasians at the time of initial being careful not to injure the underlying in 6 to 12 weeks of presentation warrant close follow-up, and nail bed, as any bleeding will invalidate the those with such lesions that remain subsequent test. To minimize the risk of treatment should unchanged after 6 to 12 months of follow- traumatizing the nail bed, avoid local anes- be biopsied up may be safely discharged with instruc- thesia. Once you remove the specimen, tions to seek medical evaluation if the grasp it with a pair of forceps and inspect lesion shows any subsequent morphologi- it closely. cal change. If the pigment does not adhere to the Asians and dark-skinned persons. In undersurface of the nail plate, but instead contrast to pigmented nail streaks in the biopsy specimen is clean, with the pig- Caucasians, dark-complected persons mented part being the nail bed itself, the commonly exhibit melanonychia striata, diagnosis of subungual hematoma is virtu- and these streaks should not be routinely ally excluded, mandating a biopsy of the biopsied unless there has been some mor- nail bed. If the underside of the nail speci- phological change. Multiple streaks, men is pigmented, use a #15 blade to care- streaks present on multiple nails, and those fully scrape the pigment onto the sample streaks that do not originate from the surface of a standard hemoccult card. Add proximal nail fold are also unlikely to be a drop of water to the pigment scrapings; melanoma in adults or children of any eth- then the card is developed in the usual

228 VOL 55, NO 3 / MARCH 2006 THE JOURNAL OF FAMILY PRACTICE JFP_03.06_PhotoRounds.Final 2/15/06 3:46 PM Page 229

Non-healing ulcerated fingertip following injury ▲

fashion. A positive test indicates the pres- typically has a worse prognosis than cuta- ence of old blood and confirms the diagno- neous melanoma.1 Racial differences in sis of subungual hematoma. You should outcomes may also exist, as one study make sure to follow-up periodically to found that African Americans with subun- assure that the pigment recedes as the nail gual melanoma have a 3.5 times greater grows, as well as to confirm the absence of rate of death than Caucasians (95% con- any further lesion is still required—keeping fidence interval, 1.4–8.6). When rates in mind that nail trauma precedes a large were controlled for Clark and Wihm’s percentage of subungual hematoma. level, they continued to have a 2.6 times greater rate of death.2 This indicates that Staging the greater mortality rate is not due entire- Once the diagnosis of subungual mela- ly to either later stage of disease at presen- noma is confirmed the depth of invasion tation or to later diagnosis. ■ must be determined. Clark’s level of inva- sion may be poorly defined in subungual REFERENCES melanoma, due to unique histopathologi- 1. Thai K, Young R, Sinclair R. Nail apparatus melanoma. cal characteristics of the nail bed. Breslow’s Australian J Dermatol 2001; 42:71–83. absolute depth correlates less closely with 2. O’Leary J, Berend K, Johnson J, Levin S, Seigler H. Subungual melanoma: A review of 93 cases with iden- prognosis of subungual melanoma than tification of prognostic variables. Clinical Orthopedica with cutaneous melanoma, but tumor and Related Research 2000; 378:206–212. thickness has been found to be a good 3. Blessing K, Kernohan N, Park K. Subungual malignant melanoma: clinicopathological features of 100 cases. prognostic indicator of subungual Histopathology 1991; 19:425–429. 1 melanoma. 4. O’Toole E, Stephens R, Young M, Tanner A, Barnes L. Subungual melanoma: a relation to direct injury? J Am Acad Dermatol 1995; 33:525–528. ■ 5. Levit E, Kagen M, Scher R, Grossman M, Altman E. The Treatment: Excision ABC rule for clinical detection of subungual and melanoma. J Am Acad Dermatol 2000; 42:269–274. Amputation or wide excision are the only 6. Elpern D. Subungual melanoma vs hematoma. Virtual Grand Rounds in Dermatology. Presented in accepted treatments for subungual mela- Williamstown, MA, 16 June 2001. Available at: FAST TRACK noma. However, there is no clear consen- www.vgrd.org. The only accepted sus regarding the most appropriate type of 7. Harrington P, Kelly A, Trail I, Freemont A. Amelanotic subungual melanoma mimicking pyogenic granuloma 5 excision, biopsy, or level of amputation. of the hand. J R Coll Surg Edin 2002; 47:638–640. treatments for Limb perfusion chemotherapy is thought subungual to increase survival in selected cases. melanoma are Outcomes amputation and Subungual melanoma typically has a wide excision worse outcome than cutaneous mela- noma, but the causes remain unclear. Anatomic location has often been cited as a prognostic marker, and many studies attribute the unusual location of subun- gual melanomas to their poorer progno- sis.7 Some recent studies have found that worse outcomes are tied to the fact that subungual melanoma often presents at a much more advanced stage than cuta- neous melanoma, while other studies have shown that even when controlled for dis- eased and stage, subungual melanoma

www.jfponline.com VOL 55, NO 3 / MARCH 2006 229