Article Text

Download PDFPDF

Ocular findings as a presenting sign of hydroa vacciniforme
  1. B H Jeng1,
  2. T P Margolis1,
  3. N S Chandra2,
  4. T H McCalmont3
  1. 1Francis I Proctor Foundation and the Department of Ophthalmology, University of California San Francisco, San Francisco, CA, USA
  2. 2Department of Ophthalmology, Kaiser Permanente Medical Center, Martinez, CA, USA
  3. 3The UCSF Dermatopathology Service, Departments of Dermatology and Pathology, University of California San Francisco, San Francisco, CA, USA
  1. Correspondence to: Todd P Margolis MD PhD Francis I Proctor Foundation, University of California, San Francisco, 95 Kirkham Street, San Francisco, CA 94143, USA;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Hydroa vacciniforme (HV) is an unusual photosensitivity disease of unknown aetiology which starts in childhood and is characterised by vesiculopapular eruptions on the exposed area of the body.1,2 Ocular manifestations of this disease are uncommon but can include conjunctivitis, vesicular eruptions of the conjunctiva and cornea that resemble phylctenular keratoconjunctivitis,3 corneal infiltration with vascularisation,4 and keratouveitis.5,6 However, no cases of HV have been reported in the ophthalmic literature in over 40 years, and in contrast with previously reported cases, to our knowledge, this is the first case of a patient with HV who presented initially only with ocular findings.

Case report

A 6 year old Indian boy was referred to the Proctor Foundation with a 3 month history of recurrent, self resolving episodes of redness, pain, and photophobia in his right eye associated with a “bump” on his temporal conjunctiva. Aside from an episode of “cold sores” on his lips 2 weeks before his presentation, he had no cutaneous complaints. On examination, visual acuity was 20/20 right eye and 20/25 left eye. The conjunctiva of the right eye was injected temporally, and the inferotemporal cornea had a faint anterior and diffuse deep stromal haze with vessels, consistent with an interstitial keratitis. The patient was given a presumed diagnosis of phlyctenular keratoconjunctivitis and treated with topical corticosteroid drops. Three months later, his symptoms returned in association with a vesicular crusting reaction of his lower lip and upper ears. A conjunctival scraping was obtained for polymerase chain reaction testing, and a purified protein derivative (PPD) and chest x ray were ordered to rule out herpes simplex virus and Mycobacterium tuberculosis, respectively, as causative aetiologies of the patient’s findings. The tests yielded negative results.

Twenty three months after his initial presentation, the patient returned with an acute, unilateral, granulomatous, anterior uveitis in the left eye and vesicular crusting of his lips and upper ears. The uveitis responded to treatment with topical corticosteroid drops. Serological testing for varicella zoster virus IgG, toxoplasma IgG, erythrocyte sedimentation rate (ESR), angiotensin converting enzyme (ACE), lysozyme, rapid plasma reagin (RPR), and fluorescent treponema antibody absorption (FTA-abs) test were all negative.

Six months later, the patient returned with red eyes and vesicular skin lesions of his ears (fig 1A), lips (fig 1B), arms, and fingers. Slit lamp examination revealed a sclerokeratitis temporally in both eyes (fig 1C and D). Both corneas were slightly oedematous temporally with deep stromal vessels. There was also a mild iritis in the left eye. Oral prednisone was started to treat a possible systemic inflammatory cause for his sclerokeratitis, and a papulovescicular lesion of the ear was biopsied. Serological testing for ANA, ANCA, anti-SNA, anti-SS-A, anti-RNP, and anti-Smith was performed; all results were negative.

Figure 1

 Hydroa vacciniforme lesions on the left ear (A) and lower lip (B). Sclerokeratitis temporally in right (C) and left (D) eyes.

Histopathological examination of the biopsied lesion revealed an epidermotropic lymphoid infiltrate with focal epithelial necrosis consistent with the diagnosis of hydroa vacciniforme (fig 2A). The infiltrate was of T cell lineage, confirmed by immunoperoxidase staining, with diffuse expression of CD3 (fig 2B); scattered CD8 positive lymphocytes were present (fig 2C). In situ hybridisation identified the presence of Epstein-Barr virus encoded small nuclear RNA in a minority (∼10%) of the lymphocytes present, suggesting the diagnosis of a hydroa vacciniforme-like lymphoproliferative disorder (fig 2D). Genotypic analysis via a polymerase chain reaction method revealed no evidence of a clonal rearrangement of T cell receptor genes.

Figure 2

 (A) Dense infiltration of lymphoid cells with epidermal necrosis consistent with hydroa vacciniforme (haematoxylin and eosin, original magnification ×40). (B) Immunoperoxidase staining demonstrating diffuse expression of CD3 (original magnification ×100). (C) Immunoperoxidase staining demonstrating scattered staining for CD8 (original magnification ×100). The immunoperoxidase chromagen in (B) and (C) is diaminobenzidine (DAB) with a haematoxylin counterstain. Appropriate positive and negative control staining was also observed (data not shown). (D) In situ hybridisation demonstrating the presence of Epstein-Barr virus encoded small nuclear RNA transcripts in a minority of the lymphocytes (original magnification ×100). The antisense probe in this preparation shows dark black nuclear staining in positive cells, while the sense (control) probe demonstrated no nuclear labelling (data not shown). The counterstain is nuclear fast red.

The patient was treated for hydroa vacciniforme with oral prednisone, ganciclovir, and precautions about ultraviolet light exposure. Since starting therapy for hydroa vacciniforme, he has not had any flares of ocular inflammation, and he has only suffered one episode of dermal inflammation. At the last follow up visit, 40 months after his initial presentation, the patient demonstrated no ocular inflammation, and his best spectacle corrected visual acuity was 20/20 both eyes. He is currently being maintained on ganciclovir therapy. The patient’s parents deny recalling a history of Epstein-Barr virus (EBV) infection in the past.


Hydroa vacciniforme was initially described by Bazin in 1862,7 and the estimated prevalence of this disease is at least 0.34 cases per 100 000.8 Patients typically present with vesicles or bullae on an erythematous base that occur primarily on light exposed body areas and develop within several days of sun exposure. With time, these lesions become progressively necrotic and ultimately heal with varioliform scars.9

Although laboratory testing has revealed no haematological, biochemical, or immunological abnormalities in affected patients, recent investigations have found that the cutaneous lesions of hydroa vacciniforme are associated with latent infection with the EBV,1,2,10 and in situ hybridisation confirmed the presence of EBV RNA synthesis in our patient’s skin lesions.

Although many dermatologists recognise that EBV associated hydroa vacciniforme-like skin lesions may have malignant potential, it is not yet clear whether this disease is inflammatory or neoplastic.10 Iwatsuki and associates2 have recently reported that three of their six patients with atypical hydroa vacciniforme progressed to overt haematological neoplasms 2–14 years after onset of their cutaneous findings. Chen and associates11 subsequently reported a patient with a CD8+ cutaneous T cell lymphoma that presented with, rather than progressed from, hydroa vacciniforme-like skin lesions. Together, these reports suggest that hydroa vacciniforme may not only progress from a smouldering stage to a lymphoid malignancy, but that it could itself be a lymphoid neoplasm. Therefore, our patient may be at increased risk for the development of an EBV related lymphoma, for which he will be monitored closely.

Ocular involvement secondary to hydroa vacciniforme is uncommon and typically occurs coincidentally with an outbreak on the face.3 Occasionally, ocular findings occur at a later time than cutaneous findings, which may be the result of the protection afforded by the eyelids.4 Although his later ocular findings occurred simultaneously with facial lesions of hydroa vacciniforme, in the very beginning, our patient manifested only ocular symptoms and findings, including an interstitial keratitis. To our knowledge, this is the first case of hydroa vacciniforme in which ocular findings preceded the onset of cutaneous lesions. Therefore, based on our report, the differential diagnosis for interstitial keratitis in young children should include hydroa vacciniforme.



  • Supported by the Heed Ophthalmic Foundation Fellowship, Cleveland, Ohio (BHJ); Research to Prevent Blindness, New York, New York (TPM).