Trachoma
BMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7306.218 (Published 28 July 2001) Cite this as: BMJ 2001;323:218
- Denise Mabey, consultant ophthalmologist (Denise.Mabey{at}mabeys.demon.co.uk)a,
- Nicole Fraser-Hurt, epidemiologistb
- Correspondence to: D Mabey
Background
Definition Active trachoma is chronic inflammation of the conjunctiva caused by infection with Chlamydia trachomatis. The World Health Organization's classification for acute trachoma defines mild trachoma (grade TF) as the presence of five or more follicles in the upper tarsal conjunctiva of at least 0.5 mm diameter. Severe trachoma (grade TI) is defined as pronounced inflammatory thickening of the upper tarsal conjunctiva that obscures more than half of the normal deep vessels. Repeated infection causes scarring trachoma, in which the upper eyelid is shortened and distorted (entropion), and the lashes abrade the eye (trichiasis). Blindness results from corneal opacification, which is related to the degree of entropion or trichiasis.
Interventions to prevent scarring trachoma by reducing active trachoma
Likely to be beneficial:
Promotion of face washing
Children having a sustained clean face
Fly control using insecticide
Antibiotics (versus placebo or no treatment)
Oral azithromycin (versus topical tetracycline)
Treatment of scarring trachoma
Likely to be beneficial:
Bilamellar tarsal rotation (versus other eyelid surgery), when performed by an experienced operator
Incidence/prevalence Trachoma is the world's leading cause of preventable blindness and is second only to cataract as an overall cause of blindness.1 Globally, active trachoma affects an estimated 150 million people, most of them children. About 5.5 million people are blind or at risk of blindness as a consequence of trachoma. Trachoma is a disease of poverty regardless of geographical regions. Scarring trachoma is prevalent in large regions of Africa, the Middle East, south west Asia, the Indian subcontinent, and Aboriginal communities in Australia, and there are small focuses in Central and South America.1 In areas where trachoma is constantly present at high prevalence, active disease is found in more than 50% of preschool children and may have a prevalence of 60-90%.2 As many as 75% of women and 50% of men over the age of 45 may show signs of scarring disease.3 The prevalence of active trachoma decreases with increasing age, with less than 5% of adults showing signs of active disease.2 Although similar rates of active disease are observed in male and female children, the later sequelae of trichiasis, entropion, and corneal opacification are more common in women than men.2
Aetiology/risk factors Active trachoma is associated with young age and where there is close contact between people. Discharge from the eyes and nose may be a source of further reinfection.4 Sharing a bedroom with someone who has active trachoma is a risk factor for infection.5 Facial contact with flies is held to be associated with active trachoma, but studies reporting this relation had weak methods.6
Prognosis Corneal damage from trachoma is caused by multiple processes. Scarring may cause an inadequate tear film, and a dry eye may be more susceptible to damage from inturned lashes, leading to corneal opacification. The prevalence of scarring and consequent blindness increases with age and blindness is most commonly seen in older adults.7
Aims To prevent active trachoma; to reduce the rate of progression to scarring; and to relieve entropion and trichiasis in people with scarring trachoma, with minimal side effects of treatment.
Outcomes Rates of active trachoma; clinical signs of active trachoma using the WHO grading scale; laboratory evidence of Chlamydia trachomatis infection; eyelid position; and degree of entropion or trichiasis.
Methods
Clinical Evidence search and appraisal March 2000.
Question What are the effects of interventions to prevent scarring trachoma by reducing active trachoma?
Option Public health interventions
Summary One randomised controlled trial (RCT) found that promotion of face washing, compared with no advice, significantly reduced the rate of severe trachoma. It found no significant reduction of the overall rate of trachoma, but the study was not powered to exclude a clinically important effect. One small RCT found that the number of new cases of trachoma was reduced by fly control using insecticide compared with no intervention.
Benefits
Promotion of face washing: We found no systematic review but found one RCT that randomised six villages to either promotion of face washing plus 30 days of daily topical tetracycline ointment or 30 days of daily topical tetracycline ointment only.8 It found that in 1417 children aged 1-7 years the promotion of face washing increased the likelihood of children having a clean face on at least two of three follow up visits, but the result was not significant (odds ratio (OR) for having a clean face with intervention v control 1.6, 95% confidence interval 0.94 to 2.74). After one year, promotion of face washing plus topical tetracycline compared with topical tetracycline reduced the risk of severe trachoma (OR for severe trachoma 0.62, 0.40 to 0.97), but the reduction was not significant for all grades of trachoma (OR for mild and severe trachoma 0.81, 0.42 to 1.59). When all participants from intervention and control villages were pooled, children who had a sustained clean face were less likely to have active trachoma than those who ever had a dirty face (OR 0.58, 0.47 to 0.72). Fly control using insecticide: We found no systematic review and no completed RCTs. We found one pilot study for an RCT comparing spraying of deltamethrin for three months compared with no intervention in two pairs of villages.6 One pair received the intervention or none in the wet season, one pair received the intervention or none in the dry season. There was a total of 191 children under 10 years in the control villages and 223 children in the intervention villages. The number of new cases of trachoma (WHO classification) at three months was 75% lower in the children living in the intervention villages than in the comparison group (relative risk (RR) 0.25, 0.09 to 0.64).
Interventions to prevent scarring trachoma by reducing active trachoma: results of randomised controlled trials of antibiotics compared with no treatment placebo, or a monthly vitamin tablet in people with active trachoma
Harms
No harms were identified.
Comment
The cluster randomisation used in both RCTs limits the power to detect differences between groups and makes it difficult to interpret the results for individual children.8
Option Antibiotics
Summary One unpublished systematic review found limited evidence from low powered RCTs that antibiotics reduced active trachoma at three and 12 months compared with no treatment. The same review found limited evidence from three RCTs that oral azithromycin reduced active trachoma compared with topical tetracycline after three months. These RCTs were low powered or of unusual design and at 12 months the difference between treatments had disappeared.
Benefits
Versus placebo or no treatment: We found one unpublished systematic review (search date 1999).9 It identified eight RCTs that compared topical and oral antibiotics with no treatment, placebo, or a monthly vitamin tablet in people with active trachoma. All reported outcomes at three months. Five trials reported outcomes at 12 months (999 people in the treatment group, 76% completed the study; 832 people in the control group, 71% completed the study; table 1). The review found that antibiotics reduced active trachoma at three months (OR for active trachoma based on clinical assessment, antibiotics v control 0.43, 0.35 to 0.52) and at 12 months (OR 0.47, 0.38 to 0.60). It also found that antibiotics reduced Chlamydia trachomatis infection rates based on bacteriological testing compared with control at 12 months (1 RCT; OR 0.27, 0.11 to 0.67). At three months the infection rate was also reduced but not significantly (3 RCTs; OR for infection, antibiotic v control 0.63, 0.38 to 1.05). Oral versus topical antibiotics: We found one unpublished systematic review of three RCTs comparing rates of active trachoma and bacteriological infection after oral azithromycin or topical tetracycline (table 2).9 All trials had problems with methods that limit the results (see comment below). At three months, oral azithromycin compared with topical tetracycline reduced active trachoma (3 RCTs, 6226 people; OR 0.82, 0.71 to 0.94) and infection rates (OR 0.49, 0.39 to 0.62). At 12 months, the review found no significant difference between treatments for active trachoma (1 RCT, 5573 people; OR 0.88, 0.77 to 1.01), but found that oral azithromycin reduced bacteriologically defined infection compared with topical tetracycline (OR 0.70, 0.57 to 0.87).
Interventions to prevent scarring trachoma by reducing active trachoma: results of randomised controlled trials comparing rates of active trachoma and bacteriological infection after oral azithromycin or topic al tetracycline
Harms
No harms were reported.
Comment
Versus placebo or no treatment: The trials were undertaken in various settings; most were in children attending boarding schools. Several unusual study designs were used—for example, family based treatment. The trials were all of moderate or poor quality and many had no intention to treat analysis. Antibiotic treatments included topical and oral doses. Oral versus topical treatment: Two of the RCTs were small (total 224 participants) and low powered. The third trial compared mass treatment, in which people were treated irrespective of disease status and were randomly allocated by village (cluster randomisation). Correlation analysis found some similarity between individuals within a cluster, thus limiting the validity of results. We found no evidence about the development of bacterial resistance.
Question Effects of surgical treatments for scarring trachoma (entropion and trichiasis)
Summary We found no good evidence on the effects of surgery to improve visual acuity in people with scarring trachoma. In people with trichiasis who have six or more lashes in contact with the globe, we found limited evidence from two RCTs that bilamellar tarsal rotation relieved entropion and trichiasis more often than tarsal advance and rotation. In people with trichiasis who had five lashes or fewer in contact with the globe, one RCT found that tarsal rotation relieved entropion and trichiasis more often than electrolysis or cryoablation. In these trials, one experienced surgeon performed most of the operations.
Benefits
We found no systematic review but found two RCTs comparing surgical interventions for entropion and trichiasis. 10 11 In both trials, one experienced surgeon performed most of the operations. Both trials defined operative success as no lashes in contact with the globe in primary position of gaze, and complete lid closure with gentle voluntary effort. Major trichiasis: See glossary. The first trial compared five surgical techniques (bilamellar tarsal rotation, eversion splinting, tarsal advance, tarsal grooving, and tarsal advance and rotation) in 165 Omani villagers with major trichiasis (as defined for these trials). There were a total of 165 eyelids.10 The trial found very limited evidence that bilamellar tarsal rotation was more successful than other surgical treatments. However, analysis was not by intention to treat, and the power of the trial was very low. The second RCT compared tarsal rotation with tarsal advance and rotation in Omani villagers with major trichiasis (total of 200 eyelids).11 It found that bilamellar tarsal rotation was more successful than tarsal advance and rotation, after 25 months' follow up (hazard ratio (HR) for failure following tarsal advance and rotation v bilamellar tarsal rotation 3.1, 1.9 to 5.2). Minor trichiasis: See glossary. The second RCT compared tarsal rotation, cryoablation, and electrolysis in people with minor trichiasis (as defined for this trial; total of 172 eyelids).11 It found that tarsal rotation was more successful than the other treatments after 25 months' follow up (HR of failure, electrolysis v tarsal rotation 6.1, 2.9 to 12.8; HR of failure, cryoablation v tarsal rotation 7.5, 3.6 to 15.4).
Summary of harms following surgery for scarring trachoma. Values are numbers of patients in trial
Harms
Adverse outcomes of interventions were corneal exposure, ulceration, phthisis bulbi, and severe recurrent trichiasis. 10 12 In these two trials, major trichiasis and defective closure after surgical procedures for scarring trachoma were more common after eversion splinting, tarsal advance, and tarsal grooving than after bilamellar tarsal rotation and tarsal advance and rotation. Cryoablation of the eyelashes can cause necrosis of the lid margin, corneal ulcers, and in the RCT in which cryoablation was used11 it was the only procedure associated with onset of phthisis (two cases out of 57). Further details of harms are summarised in table 3.
Comment
The pragmatic definitions of major and minor trichiasis are limited to use in these trials. The first RCT was very low powered, and was not analysed by intention to treat.10 In both RCTs, one experienced operator performed most of the surgery. The evidence of both benefits and harms may not be applicable to different operators, or where the quality of surgical equipment does not match that in the trials.
Glossary
Bilamellar tarsal rotation—The upper lid is cut full thickness horizontally in a line parallel and 3 mm from the eyelid margin and running from just lateral to the lacrimal punctum to the lateral canthus. Everting sutures are then placed through all layers of the lid to prevent the margin from turning inwards.
Eversion splinting—The lid margin is split posterior to the lashes; the eversion of the anterior section is maintained by sutures tied over a roll of paraffin gauze.
Major trichiasis—Lid closure complete; six or more lashes in contact with eyeball.
Minor trichiasis—Lid closure complete; one to five lashes in contact with eyeball.
Phthisis bulbi—A disorganised, shrunken eye which has no perception of light.
Tarsal advance—The lid margin is split posterior to the lashes. The skin, lashes, and orbicularis are freed from the tarsal plate and retracted away from the cornea and are sutured back on to the tarsal plate, leaving a bare area of tarsus to act as the lid margin.
Tarsal advance and rotation—The upper lid is everted over a speculum. The tarsal plate is fractured parallel to and 3 mm from the lid margin. In this operation the skin and orbicularis are not cut. The short portion of tarsal plate attached to the lid margin is then rotated through 180 degrees and sutured into place to form the new lid margin.
Tarsal grooving—A wedge of skin, orbicularis, and tarsus is removed parallel to the lid margin. Sutures through all layers act to evert the lid margin.
Footnotes
-
Competing interests None declared.
If you are interested in being a contributor or peer reviewer for Clinical Evidence, please visit the Clinical Evidence website (http://www.clinicalevidence.com/)Clinical Evidence is published by BMJ Publishing Group. The fifth issue is available now, and Clinical Evidence will be updated and expanded each month on the website. Individual subscription rate, issues 5 and 6 £75/$110; institutional rate £160/$240; student rate £55/$80. For more information including how to subscribe, please visit the Clinical Evidence website at http://www.clinicalevidence.com/