Clinical Profile of Herpes Zoster Ophthalmicus in Ethiopians

Transcription

1 1256 INTERNATIONAL REPORT Clinical Profile of Herpes Zoster Ophthalmicus in Ethiopians Samson Bayu and Wondu Alemayehu From the Department of Ophthalmology, Faculty of Medicine, Addis Ababa University, Ethiopia, East Africa We conducted a prospective study of 100 consecutive Ethiopian patients with herpes zoster ophthalmicus (HZO); this study revealed a high incidence of HZO among the young (mean age, 35 years). Eighty-one (95%) of 85 patients who underwent serological testing were seropositive for antibodies to human immunodeficiency virus (HIV). Unlike previous investigators, we found a marked increase in the incidence and severity of eyelid (25%) and ocular (78%) complications as well as postherpetic neuralgia (55%). Visual loss occurred in 56% of the cases. Lack of medication, delay in presentation, severity of HIV-related HZO, and application of herbal medications adversely affected the outcomes for these patients. We conclude that all patients with HZO, especially those younger than 45 years of age, should be screened for HIV infection. Because HZO is a visionthreatening problem, all health care workers should become aware of its management. pian patients; to determine the demographics of patients with HZO (with special reference to age, sex, marital status, and occupation); and to determine the prevalence of HIV infection in Ethiopian patients with HZO. Patients and Methods Herpes zoster ophthalmicus (HZO) is a maculopapular rash that becomes a vesicular rash and later leaves a scar on the dermatomal distribution of the ophthalmic division of the trigeminal nerve. Few cases of HZO that did not involve a cutaneous eruption have been reported in the literature [1]. HZO is most often due to reactivation of latent varicella-zoster virus from the gasserian ganglion, although exogenous exposure to the virus may occasionally be responsible for the rash [2]. HZO occurs in 10% 17.5% of patients with herpes zoster [3, 4]. Aging, malignancies, immunosuppressive therapy, chemother- apy, tuberculosis, malaria, syphilis, and trauma are predisposing factors [1, 5, 6]. Since the advent of the HIV pandemic, HZO has been recognized as an early clinical marker of HIV infection and as an ocular manifestation of AIDS [7 10]. Ophthalmic findings vary considerably. Ocular complications occur in 50% 89% of patients with HZO; these compli- cations lead to substantial visual disability, severe postherpetic neuralgia, and, rarely, fatal cerebral complications [9, 11 13]. Systemic zoster encephalitis and pneumonitis may develop in immunocompromised patients [3, 5, 7, 11, 12]. In one Ethiopian study, HZO was found to be the second leading reason for seeking care at the neuro-ophthalmic clinic of a tertiary eye-care referral center [14]. Health care workers at two hospitals in Addis Ababa have reported cases of HZO in association with AIDS and HIV infection [15, 16]. Our objectives in conducting the present study were to de- scribe the clinical features and complications of HZO in Ethio- From 1 October 1993 to 31 May 1995, we prospectively studied the cases of 100 consecutive Ethiopian patients with clinically diagnosed HZO who were seen at the Menelik II Hospital in Addis Ababa. Facilities for viral culture are not available in Ethiopia. Patients with presumptive diagnoses of zoster sine eruptione were excluded because of the rarity of this form of the disease and because an assay of serum CF antibody to varicella-zoster virus, needed for definitive diagnosis, could not be done in Ethiopia. All patients were examined within 30 days of onset of the rash. An open-ended questionnaire was used to collect demographic data including age, sex, address, occupation, and mari- tal status, as well as the following medical data: the date of onset of prodromal symptoms and rash; a history of weight loss or known systemic illnesses (i.e., tuberculosis, sexually transmitted diseases [STDs], and diabetes mellitus); a history of immunosuppressive drug intake, irradiation, blood transfu- sions, or repeated injections before the onset of HZO; a history of a similar previous attack of zoster in any other part of the body; recent exposure to persons with chickenpox; and use of herbal medication for treatment of the illness. The presence or absence of postherpetic neuralgia was recorded. Postherpetic Received 23 April 1996; revised 27 November neuralgia was defined as the presence of continuous or frequent Reprints or correspondence: Dr. Samson Bayu, P.O. Box 34263, Addis pain for 6 months after the onset of disease [13]. Ababa, Ethiopia, Africa. The dermatomal distribution of the rash (75% of patients) Clinical Infectious Diseases 1997;24: by The University of Chicago. All rights reserved. or scars (25%) was evaluated, and an eyelid examination was /97/ $02.00 performed, after which a thorough ocular examination that

2 CID 1997;24 (June) Herpes Zoster Ophthalmicus in Ethiopia 1257 Table 1. Distribution of 100 patients with herpes zoster ophthal- [{SD], { 10.2 years). The male-to-female ratio was micus in Ethiopia by age and sex. 2:1. Eighty percent of the patients were aged 45 years. Age range (y) No. of males No. of females Eighty-four percent of the patients were from Addis Ababa or resided within a 100-km radius of the city, while the remaining patients (16%) came from different regions of the country Forty-three of the patients were married, which was not a statis tically significant factor (P Å.9). Fifty percent of the patients were government employees; taxi, bus, and truck drivers; or housewives Fourteen patients were found to have tuberculosis, for which they received treatment. Sixteen patients had histories of STDs One patient developed HZO 1 month after undergoing surgical intervention (orbitotomy). He was seropositive for antibodies Total to HIV before the operation. None of the patients was diabetic or iatrogenically immunosuppressed (either as a result of chemotherapy or radiation therapy); none had a history of recent included determinations of visual acuity and intraocular presinvolvement; exposure to persons with chickenpox; none had bilateral sure, and biomicroscopic examination of the anterior and posterior and none had had a previous episode of HZO. segments (i.e., the conjunctiva, cornea, anterior chamber, Because of the unavailability of acyclovir and its costliness pupil, lens, and vitreous) was done. The patients pupils were when available, only 22 patients received the drug (18 received dilated, and funduscopy was done. The functioning of cranial topical therapy for established ocular complications and four nerves II, III, IV, V, VI, and VII was assessed. received oral therapy soon after the development of the dis- Baseline laboratory data, including the hemoglobin concencase ease). Fifteen patients fulfilled the criteria of the WHO clinical tration, WBC count and differential, erythrocyte sedimentation definition of AIDS [18]. One patient with AIDS died of rate, and fasting blood sugar level were obtained, and a urinalytion diarrheal disease. Thirty-four patients applied herbal medica- sis was done. Serological examinations, including VDRL (Venereal to the rash soon after the onset of the disease. Disease Research Laboratory) and fluorescein treponeophthalmic The dermatomal distribution of HZO was localized in the mal antibody absorption (FTA-ABS) tests for syphilis and an division of the trigeminal nerve in 95% of cases. ELISA for antibodies to HIV were done after verbal informed Three patients had mixed ophthalmic- and maxillary-division consent was obtained from 85 of the patients. involvement, and two patients had involvement of all divisions The ELISA for detection of antibodies to HIV was done by of the trigeminal nerve. The right eye was affected in 49 pausing first-generation Wellcozyme kits, and positive results tients, and the left was affected in 51. were confirmed by using second-generation kits (Wellcome Ocular complications were observed in 78% of the patients. Diagnostics, Dartford, UK). Patients with double positive We observed zoster corneal involvement in its protean manifes- ELISAs were considered to have HIV infection. This determiobserved tations in 65% of our patients. Anterior uveitis, which was nation is in accord with the 64% and 94.2% specificities of the in 50% of our patients, occurred with or without first- and second-generation Wellcozyme Kits, respectively, for corneal involvement. (It is usually chronic, tends to recur, and the population of Ethiopia with low risk (i.e., without specific requires topical corticosteroid therapy for a long period). Sec- risk factors) [17]. Counseling of the patients was done on the ondary glaucoma, which developed in 50% of the patients with basis of the serological test result. anterior uveitis, was difficult to control. We detected cataracts, Patients were treated with acyclovir (topical and systemic), which occur secondary to the inflammatory process (uveitis) topical steroids, cycloplegics, antibiotics, topical beta-blockers, or long-term use of corticosteroids, in 14 of the patients. We oral carbonic anhydrase inhibitors, and analgesics or with surartery also observed cases of scleritis, optic neuritis, hypopyon, retinal gery, as indicated and when available. Patients were followed occlusion, and phthisis bulbi. The rates of the different up for a minimum of 6 months. ophthalmic complications in the present study, as compared The provisional case definition of the World Health Organiand/or with those in other studies, are shown in table 2. Frontal zation (WHO) and a positive serology for antibodies to HIV corneal anesthesia occurred in 47% of the patients. were the criteria for the diagnosis of AIDS [18]. For the statistiand Surgical correction of the eyelid, temporary tarsorrhaphy, cal analysis of the demographic, clinical, and laboratory data, evisceration were performed for six patients, four patients, we used the x 2 test. and three patients, respectively. The follow-up period ranged from 6 months to 18 months, with an attrition rate of 15%. Results The final visual acuity was õ20/200 for 40% of the patients, The distributions of the patients by age and sex are shown and another 36.47% (31 of 85) went blind (visual acuity, in table 1. Ages ranged from 18 years to 70 years (mean age õ20/400) in the involved eye (on the basis of the WHO defini-

3 1258 Bayu and Alemayehu CID 1997;24 (June) Table 2. Comparison of ophthalmic complications in patients with herpes zoster ophthalmicus who were from Ethiopia, the United States, the United Kingdom, and Rwanda. Country [reference] Ethiopia United States [12] United Kingdom [11] Rwanda [9] Complication (n Å 100) (n Å 86) (n Å 1,356) (n Å 19) Eyelid involvement Canalicular scarring Ocular complications Corneal involvement Anterior uveitis Secondary glaucoma Cataracts Scleral involvement Optic neuritis Hypopyon Phthisis bulbi Central renal artery occlusion Extraocular muscle palsy Facial palsy Postherpetic neuralgia 55* NOTE. The study periods were as follows: Ethiopia, ; United States, ; United Kingdom, ; and Rwanda, Data are percentages of patients; some patients had more than one complication. * Of 85 patients who were followed for 6 months in our study. tion of blindness). There was no statistically significant differ- studies haven t shown a sexual predilection [1, 3, 4, 12]. The ence in the final visual acuity between patients who did or did higher rate of HZO among our male patients is not due merely not receive topical acyclovir after the development of ocular to a higher prevalence of this disease in males; as has been complications. shown in other hospital-based studies in Ethiopia, our findings The results of baseline laboratory investigations were all reflect the fact that males have better access to health care normal. Nineteen (22.35%) of the 85 patients who volunteered institutions [14]. to undergo VDRL and FTA-ABS testing had reactive tests. Marital status didn t appear to play a role in the development The double ELISA technique revealed HIV infection in 81 of HZO in our series of patients. In terms of occupation, most (95.3%) of the 85 patients. All of the patients aged 45 years of the patients were government employees, drivers and their were infected with HIV. Four patients ú45 years of age were assistants, house wives, and soldiers, which parallels the find- found to be seronegative for HIV, a finding that decreases the ings of the recent report on Ethiopian patients with AIDS [20]. seropositivity rate to 76.47%. The fact that none of the patients had a history of recent expo- sure to chickenpox favors viral latency as a cause of herpes Discussion zoster. HZO has been found to be one of the ocular manifestations HZO was previously believed to be a disease of the elderly of AIDS [10, 16] and an early clinical marker of HIV infection [3, 4, 11 13]: in a study from the United States that was pub- [8, 9]. Fifteen of the patients in this study were found to have lished in 1983, the mean age of the patients was 65 years [12]. AIDS; 14 of these patients had concomitant tuberculosis. The In a prospective study from the United Kingdom in 1977 that prevalence of HIV infection among our patients (95.3% of the included 77 patients, the youngest patient was 46 years old [19]. total population and 100% of patients aged 45 years) is in Since the advent of the HIV pandemic, this pattern has changed. agreement with that reported from Rwanda in 1987 [9] but In our study, 83% of the patients were 45 years of age (mean differs from that reported in the United States in 1993, where age, 35 years). This finding is in agreement with that of Kestelyn only 21% of the total number of patients and 56% of the et al. [9], who in 1977 described 19 patients from Rwanda for patients aged 45 years were HIV infected [10]. whom the mean age was 28 years. Although children are said Nineteen (22.35%) of our 85 patients had positive serologies to constitute 7% of patients with HZO [1], the youngest patient for syphilis. Recent studies among donors of blood to the blood in our series was 18 years old. bank and of pregnant women receiving antenatal care showed Harding et al. [13] found a significant predominance of that 8% 13.1% had reactive VDRL tests and that the rate of males among patients õ60 years of age; this predominance positivity for antibodies to Treponema pallidum is higher was more pronounced among patients õ40 years of age. Other among HIV-infected patients [21, 22]. Syphilis has been

4 CID 1997;24 (June) Herpes Zoster Ophthalmicus in Ethiopia 1259 incriminated as a predisposing factor for the development of Oral acyclovir, if given in the first 72 hours after the onset HZO [6], and the strong association between syphilis and HIV of HZO, protects against ocular complications [24] and reduces infection is also a risk factor for acquiring HZO. the severity and incidence of postherpetic pain [25]. Because The dermatomal distribution of HZO is observed with equal an insufficient number of patients (eight) received this drug in frequency on the right and left sides [12]. Involvement of two our study, a valid comparison of outcomes with and without or more branches of the trigeminal nerve, which is seen in 5% treatment cannot be made. For established ocular complica- of cases, is said to be a sign of disseminated zoster in the tions, a controlled coded trial [26] showed that topical acyclovir immunocompromised patient [2]. The high degree of eyelid was significantly superior to topical steroids with respect to involvement that we observed among our patients (nearly twice treatment duration, and there were no recurrences of HZO after that seen among patients in the United States study [12]) might the patients stopped receiving treatment with acyclovir. The have resulted from the common practice of using herbal medi- reductions in treatment duration and recurrence rate would be cation and from the severity of HZO in the HIV-infected patients. expected to result in a reduced incidence of ocular damage The use of herbal medication was associated with both and visual loss among acyclovir-treated patients [26]. Topical bacterial superinfection and chemical toxicity, resulting in steroid therapy should be withheld in all but the most severe more-severe eyelid disease. cases and should be supplemented with topical acyclovir therapy The incidence of corneal involvement among our patients [26]. (65%) is higher than that observed in studies from the United It has been claimed that systemic steroids prevent postherpetic States (54.6%) [12] and the United Kingdom (49%) [11] and neuralgia, but the studies from which this claim originates close to that observed in a recent study of HIV-infected patients had many pitfalls in terms of meeting the necessary standard from Rwanda [9]. The combination of eyelid abnormalities and for drug trials [2]. Investigators who conducted a randomized neurotropic keratopathy is likely to lead to permanent corneal placebo-controlled study of steroid therapy for HZO concluded damage [1]. Scleritis, optic neuritis, retinal artery occlusion, that prednisolone does not prevent postherpetic neuralgia [27]; and phthisis bulbi have also been described in association with therefore, clinicians should be cautious in prescribing systemic HZO [1, 2, 11, 12]. steroids, as there is increased potential for the development of A prospective study of 61 patients in the United Kingdom systemic zoster in immunocompromised patients [1, 5]. showed that 31% had extraocular muscle palsy [19]; most of the Late presentation, application of herbal medication, and the patients developed this complication during the first 2 weeks of unavailability and cost of acyclovir were the possible causes of the rash. In our study, the rate of this complication was 12%, the increased incidence of ocular complications in our patients. which is much higher than the rate of 3% in a study from the Thus, the poor visual outcomes and high incidence of postherpetic United States [12]. All of our patients with extraocular muscle neuralgia among these patients is much like the sce- palsy, except two with external ophthalmoplegia, recovered nario that prevails in Malawi [23]. completely. Seroprevalence studies in Ethiopia have shown that Ç20% Postherpetic neuralgia, which is usually severe and intracta- of urban antenatal-care recipients, 9% of blood donors, ble, occurred at an alarmingly high frequency in our study: in 32% 42% of patients, and 0 6% of the general rural popula- other reports, the rates ranged from 9% to 42% [3, 4, 9, 12, tion had become infected with HIV by the end of 1994 [20]. 13], whereas 55% of our patients developed this complication. The prevalence of HIV infection in our series of patients with Unlike Harding et al. [13], we did not find a statistically sig- HZO was 95.3%, a finding indicative of the strong association nificant association between postherpetic neuralgia and increasing with HIV infection. Thus, a randomized controlled study is age in our patients. Although we didn t encounter any needed. All patients with HZO, especially those aged 45 patient with systemic complications secondary to HZO, cases years, should be screened for HIV infection; if such patients of hemiplegia, cerebral angiitis [12], and other cerebral compli- are found to be seropositive, they should receive counseling to cations leading to death [7, 11] have been reported. prevent the spread of this fatal disease. There is a significant rate of visual loss following the devel- With the presently increasing magnitude of HIV infection, opment of HZO. Our finding is similar to that of a study from there will be increasing numbers of patients with HZO, espe- Malawi, in which 66% of patients had a final visual acuity of cially in the young, economically productive age group. Thus, õ20/60, and 40% had light perception only or no light perception in a country like Ethiopia where the number of trained ophthal- [23]. In contrast, a study from the United States that was mologists is limited, primary care physicians should be well conducted before the AIDS era showed a visual acuity of versed in the management of HZO, which affects all structures õ20/60 for only 24 (28%) of 86 patients [12]. of the eye and may lead to blindness. In general, the incidence and severity of ocular complications and postherpetic neuralgia were higher for our patients than for patients described in studies from the developed world [3, 4, References 10, 12, 13]; however, our findings are in agreement with those 1. Karabassi M, Raizman MB, Schuman JS. Herpes zoster ophthalmicus. of other recent African studies on HIV-related HZO [9, 23]. Surv Ophthalmol 1992;36:

5 1260 Bayu and Alemayehu CID 1997;24 (June) 2. De-Luise VP, Wilson FM II. Varicella and herpes zoster ophthalmicus. Ethiopian Medical Association (Addis Ababa). Addis Ababa, Ethiopia: In: Duane TD, Jaeger EA, eds. Clinical ophthalmology. Philadelphia: Ethiopian Medical Association, J.B. Lippincott, 1988 (vol 4, chap 20): Daniel A. Ocular manifestation of patients with AIDS [abstract no 36]. 3. Ragozzino MW, Melton LJ III, Kurland LT, Chu CP, Perry HO. Popula- In: Program and abstracts of the 28th Annual Medical Conference of tion-based study of herpes zoster and its sequelae. Medicine (Baltimore) Ethiopian Medical Association (Addis Ababa). Addis Ababa, Ethiopia: 1982;61: Ethiopian Medical Association, Burgoon CF, Burgoon JS, Baldrige GD. The natural history of herpes 17. Messele T, Zewdie D. Comparison of two generations of Wellcozyme zoster ophthalmicus. JAMA 1957;164: kits. Ethiopian Journal of Health Development 1990;4: Mazur MH, Dolin R. Herpes zoster at the NIH: a 20 year experience. Am 18. Acquired immunodeficiency syndrome (AIDS): provisional WHO clinical J Med 1978;65: case definition for AIDS. Wkly Epidemiol Rec 1986;61: Weller TH. Varicella and herpes zoster. Changing concepts of the natural 19. Marsh RJ, Dudley B, Kelly V. External ocular motor palsies in ophthalmic history, control, and importance of a not-so-benign virus. N Engl J Med zoster: a review. Br J Ophthalmol 1977;61: ;309: National AIDS control programme, review report of the Ministry of Health, 7. Cole EL, Miesler DM, Calabrese LH, Holland GN, Mondino BJ, Conant Ethiopia. Addis Ababa: National AIDS Control Programme of the Min- MA. Herpes zoster ophthalmicus and acquired immune deficiency synistry of Health, 1994: December drome. Arch Ophthalmol 1984;102: Yohannes G, Molla K, Rahlenbeck S. Infection with hepatitis-b virus, HIV and syphilis in north-west Ethiopia: prevalence of serologic markers in 8. Sandor EV, Millman A, Croxson TS, Mildvan D. Herpes zoster ophthalblood donors [abstract no 25]. In: Program and abstracts of the 31st micus in patients at risk for the acquired immune deficiency syndrome Annual Medical Conference of Ethiopian Medical Association (Addis (AIDS). Am J Ophthalmol 1986;101: Ababa). Addis Ababa, Ethiopia: Ethiopian Medical Association, Kestelyn P, Stevens AM, Bakkers E, Rouvroy D, Van de Perre P. Severe 22. Gebrekidar K, Fantahun M, Azeze B. Seroprevalence of HIV and its herpes zoster ophthalmicus in young African adults: a marker for association with syphilis seropositivity among antenatal care clinic at- HTLV-III seropositivity. Br J Ophthalmol 1987;71: tenders at Debretabor rural hospital, north-west Ethiopia [abstract no 10. Selliti TP, Huang AJ, Schiffmann J, Davis JL. Association of herpes zoster 24]. In: Program and abstracts of the 31st Annual Medical Conference of ophthalmicus with AIDS and acute retinal necrosis. Am J Ophthalmol Ethiopian Medical Association (Addis Ababa). Addis Ababa, Ethiopia: 1993;116: Ethiopian Medical Association, Marsh RJ, Copper M. Ophthalmic herpes zoster. Eye 1993;7: Lewallen S. Herpes zoster ophthalmicus in Malawi. Ophthalmology 1994; 12. Womack LW, Liesegang TJ. Complications of herpes zoster ophthalmicus. 101: Arch Ophthalmol 1983;101: Cobo LM, Foulks GN, Liesegang T, et al. Oral acyclovir in the treatment 13. Harding SP, Lipton JR, Wells JC. Natural history of herpes zoster ophthal- of acute herpes zoster ophthalmicus. Ophthalmology 1986;93: micus: predictors of postherpetic neuralgia and ocular involvement. Br 25. Harding SP, Porter SM. Oral acyclovir in herpes zoster ophthalmicus. Curr J Ophthalmol 1987;71: Eye Res 1991;10(suppl): Bayu S, Alemayehu W. Pattern of neuro-ophthalmic disorders in a tertiary 26. McGill J, Chapman C. A comparison of topical acyclovir with steroids in eye care center in Addis Ababa. Ethiop Med J 1997;35: the treatment of keratouveitis. Br J Ophthalmol 1983;67: Quana a P, Zewde M. AIDS amongst ophthalmic population [abstract no 27. Esmann V, Geil JP, Kroon S, et al. Prednisolone does not prevent postherpetic 35]. In: Program and abstracts of the 28th Annual Medical Conference of neuralgia. Lancet 1987;2:126 9.