INTRODUCTION

Haemangiomas are common tumours of infancy and their presence in the anogenital region is well known. Their presentation as an ulcer however is extremely rare and so far only three such cases have been reported in the literature.^1,2 In this regard, we present two further cases.

Case 1

A general practitioner referred a 4-week-old boy with an ulcer in the perianal region. The child's mother noticed the ulcer that had been thought to develop from a nappy rash, during the second week after birth. At admission the boy was apyrexial, healthy and feeding very well on breast milk. On examination there was an 8 mm, round, shallow ulcer on the left buttock adjacent to the natal cleft. There was no discharge or bleeding but the surrounding skin was mildly erythematous. A culture swab was taken and the patient was empirically commenced on topical 2% mupirocin ointment. As the presentation was very unusual a Medline search was conducted by using the keywords "perianal" and "ulcer", which suggested that haemangioma could be a likely diagnosis. At 2 weeks, the ulcer remained stable while the surrounding skin showed features of a capillary haemangioma confirming our early suspicion (Figure 1). At 11 weeks, the ulcer had improved considerably and the strawberry naevus had become more evident.

Figure 1.

Case 1.

Full figure and legend (126K)

Case 2

A 2-week-old girl, with a history of a perianal ulcer for 2 days, was referred by her general practitioner. The ulcer had developed following a nappy rash. The patient was apyrexial and feeding well on bottle. On examination the ulcer was 6 mm in size and appeared similar to the previous case. The patient was discharged on topical 2% mupirocin ointment. At 6 weeks, the ulcer had almost healed while the lesion developed into a haemangioma.

Discussion

While perianal haemangiomas are not uncommon their presentation as an ulcer in the neonate is extremely rare and consequently poses a diagnostic dilemma. Stanley and colleagues reported 2 cases, both these were boys who presented with an ulcer on the left buttock 2 weeks after birth.¹ Following treatment with topical antibiotics for a month, the ulcers healed completely with scarring.

Knispel and Shaw² reported an African American boy who, during the third week after birth, presented with an ulcer on the right buttock. The lesion had been noted on the third day of life and was treated with low-dose systemic steroids. The ulcer healed with significant re-epithelialization but the report did not mention the time it took for the healing to occur.

In a series of 10 female infants with ulcerated anogenital haemangiomas, presented by Achauer and Vander Kam,³ only three patients had lesions in the buttock region. Of these, only one case presented as a neonate but the ulcer did not develop until the child was 2 months old. All their patients were treated with argon laser. Healing of the ulcers occurred between 1 and 5 weeks following treatment while the ulcer in the patient mentioned above healed in 4 weeks.

Both of our patients were treated conservatively as mentioned above together with instructions to maintain local hygiene. It yielded satisfactory results with spontaneous healing of the ulcer within 10 weeks, hence obviating the need for steroids. The use of argon laser has gained popularity in the treatment of haemangiomas and it has shown that it may considerably reduce the time for the involution of these lesions.³ The aesthetic end result, however, is considered to be indifferent from that as could be expected from natural involution of haemangiomas that occurs usually by the 10th birthday of the child. Owing to the inevitable scarring following laser therapy it remains controversial whether all haemangiomas should be treated by this modality.⁴

Lastly, an important thing to note regarding haemangiomas around the lumbosacral spine is their association with occult spinal dysraphism. In these cases, however, there are other cutaneous markers as well, which should be looked for and their presence should prompt further investigation with either an ultrasound or magnetic resonance imaging.⁵ This is an exception to the rule that haemangiomas, unlike vascular malformations, are not usually associated with extracutaneous findings.

In conclusion, haemangioma should be considered in the differential diagnosis in perianal ulceration in neonates. In this regard, the presence of a surrounding red patch, which in both of our cases was thought to be nappy rash, perhaps is a vital clue which should alert the clinician about the possible diagnosis. It is indeed possible that irritation from the nappy may actually be responsible for the development of the ulcer. Nevertheless, conservative treatment by topical antibiotics remains a suitable treatment until the ulcer epithelialises. Later, the management can follow the same lines as that of haemangiomas elsewhere.

References

1. Rekant SI, Katz IL. Perianal hemangioma appearing as an ulcer. Arch Dermatol 1972;106:382–383. | Article | PubMed |
2. Knispel J, Shaw JC. Nonhealing perianal ulcer. Arch Dermatol 2001;137:365–370. | PubMed |
3. Achauer BM, Vander Kam VM. Ulcerated anogenital hemangioma of infancy. Plast Reconstr Surg 1991;87:861–868. | PubMed |
4. Waner M. Recent developments in lasers and the treatment of birthmarks. Arch Dis Child 2003;88(5):372–374. | Article | PubMed |
5. Allen RM, Sandquist MA, Piatt Jr. JH, Selden NR. Ultrasonographic screening in infants with isolated spinal strawberry nevi. J Neurosurg 2003;98:247–250. | PubMed |

Acknowledgments

The authors wish to thank Drs. P MacMahon and J Cosgrove for allowing the use of their cases for this report.