Slow Resolution of Clinically Active Trachoma Following Successful Mass Antibiotic Treatments

LETTERS

RESEARCH LETTERS Clinically active trachoma Chlamydial infection

100

Slow Resolution of Clinically Active 80 Trachoma Following Successful Mass 60 Antibiotic Treatments

40 rachoma, caused by infection with ocular strains Prevalence, % of chlamydia, is the leading infectious cause of 20 T blindness worldwide. The World Health Orga- nization recommends that in districts where the prevalence of clinically active trachoma exceeds 10% in chil- 0 6 12 18 24 Time Since First Azithromycin Distribution, mo dren aged 1 to 9 years, communities should receive 3 annual mass antibiotic distributions followed by clinical reassessment; any communities with persistent tra- Figure. Prevalence of ocular chlamydial infection and clinically active trachoma in 24 Ethiopian villages. At months 0, 6, 12, and 18, mass choma should continue receiving annual mass antibi- azithromycin treatments were distributed subsequent to monitoring. Each otic treatments until the prevalence of clinically active solid gray line represents an individual village; the solid black line represents trachoma in children aged 1 to 9 years falls below 5%.1 the summary of all 24 villages (assessed by generalized estimating equation, Although trachoma treatment decisions are based on assuming the mean baseline prevalence of clinically active trachoma). the prevalence of clinically active trachoma, it is un- clear how quickly the clinical signs of trachoma resolve monitoring visits, using time and the log baseline preva- once infection has been cleared, especially in areas with lence of clinically active trachoma as fixed effects, vil- severe trachoma. We recently performed a series of cluster- lage as a random effect, a first-order autoregressive cor- randomized clinical trials for trachoma in an area of Ethio- relation structure, and the Huber-White sandwich estimate pia with hyperendemic trachoma. In these trials, infec- of variance. Use of the log-transformed outcome was de- tion was brought to a low level in 24 villages randomized cided a priori and was supported by plots of the residual to receive mass azithromycin treatments every 6 months. vs fitted values. Analyses were performed with Stata ver- This provided an opportunity to determine the rate of sion 10.0 statistical software (StataCorp LP, College Sta- resolution of the clinical signs of trachoma given little tion, Texas). to no chlamydial reinfection. Results. The mean prevalence of ocular chlamydial in- Methods. Twenty-four randomly selected villages re- fection in children before the baseline mass azithromy- ceived mass azithromycin treatment every 6 months for cin treatment was 52.9% (95% confidence interval, 44.2%- 18 months.2,3 In these villages, we monitored the preva- 61.3%) (Figure).4 Mass azithromycin distribution lence of ocular chlamydia and clinically active tra- subsequently occurred at months 0, 6, 12, and 18. The choma in all children aged 1 to 5 years 2 months after mean prevalence of ocular chlamydia in children dur- baseline and then every 6 months after baseline for 24 ing posttreatment biannual monitoring visits was 5.1% months. Monitoring preceded mass antibiotic treat- (95% confidence interval, 3.6%-6.6%). During these 24 ment at each of the biannual visits. Ocular chlamydial months, the prevalence of clinically active trachoma de- prevalence was assessed using maximum likelihood es- creased by 33.3% (95% confidence interval, 28.6%- timations from pooled chlamydial polymerase chain re- 38.0%) per year as estimated from the generalized esti- action (Amplicor; Roche Diagnostics, Branchburg, New mating equation (Figure). Jersey) as described previously.2,3 Clinically active trachoma was defined as follicular trachoma and/or in- Comment. In a hyperendemic area, repeated biannual tense inflammatory trachoma according to the World mass antibiotic treatments reduced the level of ocular chla- Health Organization’s simplified grading system.1 Clini- mydial infection in children to a low level. The preva- cal examiners were trained and validated against gold- lence of the clinical signs of active trachoma declined at standard examiners as described previously.4 a much slower rate of 33.3% per year. This rate of reso- The primary outcome was the resolution rate of clini- lution is somewhat lower than that reported else- cally active trachoma at the village level. We used a gen- where.5,6 However, prior reports have been conducted in eralized estimating equation to predict the log preva- fewer villages, with less severe trachoma. In contrast, this lence of clinically active trachoma during biannual study was conducted in many villages in an area with se-

(REPRINTED) ARCH OPHTHALMOL / VOL 129 (NO. 4), APR 2011 WWW.ARCHOPHTHALMOL.COM 512

©2011 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 vere trachoma and specifically included the variation among villages in the analysis. Furthermore, our analy- Intraocular Invasion by Microsporidial sis assessed the prevalence of clinically active trachoma Spores in a Case of Stromal Keratitis in a village as opposed to its presence in an individual. A village-level analysis is most useful for trachoma pro- icrobial keratitis is a visually disabling con- gram managers, who make treatment decisions based on dition caused by a variety of infectious organ- the prevalence of trachoma in a village. Our findings sug- isms. Although uncommon, microsporidial gest that in areas with hyperendemic trachoma treated M stromal keratitis has recently been reported as an emerg- through mass distribution of azithromycin, the signs of 1 active trachoma can be expected to resolve slowly, even ing cause of stromal keratitis. Intraocular microsporidi- if antibiotic distributions are quite successful in reduc- osis causing endophthalmitis and sclerouveitis with clear 2,3 ing the prevalence of ocular chlamydia. cornea has been reported, although there was no evidence of contiguous anterior chamber involvement. We Jeremy D. Keenan, MD, MPH report a rare case of microsporidial stromal keratitis in Takele Lakew, MD, MPH which microsporidial spores were detected from cor- Wondu Alemayehu, MD, MPH neal scraping, corneal tissue, and endothelial exudates. Muluken Melese, MD, MPH Jenafir I. House, MPH Report of a Case. A 40-year-old woman had intermit- Nisha R. Acharya, MD, MS tent pain, redness, photophobia, and decreased vision in Travis C. Porco, PhD, MPH her right eye for 1 year. She had trauma with a leaf 1 year Bruce D. Gaynor, MD before her initial visit to us. Since then, she had been Thomas M. Lietman, MD treated irregularly with topical antiviral medication (acyclovir, 3%) by various health care providers. Author Affiliations: F. I. Proctor Foundation (Drs Keenan, On examination, her best-corrected visual acuity was Acharya, Porco, Gaynor, and Lietman and Ms House), Departments of Ophthalmology (Drs Keenan, Acharya, 20/70 OD. Slitlamp biomicroscopic examination of the Porco, Gaynor, and Lietman) and Epidemiology and Bio- right eye revealed conjunctival congestion, central stro- statistics (Drs Porco and Lietman), and Institute for Global mal edema with Descemet folds, and keratic precipi- Health (Dr Lietman), University of California, San Fran- tates (Figure, A). Her intraocular pressure was normal. cisco; and ORBIS International, Addis Ababa, Ethiopia Based on results of the clinical evaluation, a diagnosis of (Drs Lakew, Alemayehu, and Melese). herpetic viral endotheliitis was made. She began treat- Correspondence: Dr Keenan, F. I. Proctor Foundation ment with systemic acyclovir, 400 mg 5 times daily, and and Department of Ophthalmology, University of Cali- topical corticosteroid (prednisolone acetate, 1%). Al- fornia, San Francisco, 95 Kirkham St, San Francisco, CA though her visual acuity and clinical picture improved, 94143 ([email protected]). she revisited us 4.5 months after the initial visit with a Financial Disclosure: None reported. decrease in vision, tearing, and pain that had lasted 15 Funding/Support: This work was supported by grants K23 days. She was using topical steroid once every other day. EY019071 and U10 EY016214 from the National Insti- Her visual acuity was light perception with projection OD. tutes of Health, the Heed Ophthalmic Foundation, the In- There was a 3.8ϫ2.7-mm central full-thickness stro- ternational Trachoma Initiative, the Bernard Osher Foun- mal infiltrate. The endothelium showed exudates ar- dation, That Man May See, the Peierls Foundation, the Bodri ranged as a sheet. The corneal scrapings examined in po- Foundation, the Harper Inglis Trust, the South Asia Re- tassium hydroxide and calcofluor white mount (Figure, search Fund, and Research to Prevent Blindness. B) showed multiple oval microsporidial spores. The spores Additional Contributions: The Ethiopian Ministry of were also seen in Gram staining of the corneal scraping Health and many health care professionals, including (Figure, C). She began intensive treatment with topical Tadesse Kebede, Berhanu Fikre, Mifta Shifa, MPH, and polyhexamethylene biguanide, 0.02%, eyedrops. Ow- Tadesse Birru, helped us organize and implement our fieldwork in Ethiopia. ing to progressive thinning and descemetocele forma- tion (Figure, D), she underwent tissue adhesive appli- 1. Solomon AW, Zondervan M, Kuper H, Buchan JC, Mabey DCW, Foster A. cation 4 days later. After a week with no response to Trachoma Control: A Guide for Programme Managers. Geneva, Switzerland: medical therapy, she underwent therapeutic penetrat- World Health Organization; 2006. 2. Melese M, Alemayehu W, Lakew T, et al. Comparison of annual and bian- ing keratoplasty. Histopathologic examination of the cor- nual mass antibiotic administration for elimination of infectious trachoma. neal tissue showed multiple microsporidial spores within JAMA. 2008;299(7):778-784. 3. Lakew T, House J, Hong KC, et al. Reduction and return of infectious tra- the stroma as well as in the subendothelial exudates with choma in severely affected communities in Ethiopia. PLoS Negl Trop Dis. 2009; intact Descemet membrane (Figure, E and F). The post- 3(2):e376. operative period was uneventful. At the last follow-up, 4. Keenan JD, Lakew T, Alemayehu W, et al. Clinical activity and polymerase chain reaction evidence of chlamydial infection after repeated mass antibi- 6 months postoperatively, the graft was clear with vi- otic treatments for trachoma. Am J Trop Med Hyg. 2010;82(3):482-487. sual acuity of 20/200 (Figure, G). 5. Bailey RL, Arullendran P, Whittle HC, Mabey DC. Randomised controlled trial of single-dose azithromycin in treatment of trachoma. Lancet. 1993;342(8869): 453-456. Comment. Stromal keratitis due to microsporidium is less 6. Fraser-Hurt N, Bailey RL, Cousens S, Mabey D, Faal H, Mabey DC. Efficacy of oral azithromycin vs topical tetracycline in mass treatment of endemic common than superficial keratoconjunctivitis. It is fre- trachoma. Bull World Health Organ. 2001;79(7):632-640. quently misdiagnosed as herpes simplex keratitis. Pre-

(REPRINTED) ARCH OPHTHALMOL / VOL 129 (NO. 4), APR 2011 WWW.ARCHOPHTHALMOL.COM 513