Secondary Logo

Journal Logo

Original Article

Ophthalmic Complications and Management of Stevens-Johnson Syndrome at a Tertiary Eye Care Centre in South India

Kompella, Viswanadh B MS; Sangwan, Virender S MS; Bansal, Aashish K MS; Garg, Prashant MS; Aasuri, Murali K MD; Rao, Gullapalli N MD

Author Information
Indian Journal of Ophthalmology: Volume 50 - Issue 4 - p 283-286
  • Open


Ferdinand von Hebra is credited with the discovery of erythema multiforme in 1860. 1 Erythema multiforme is an acute, self-limiting eruption of the skin and mucous membranes, characterised by distinctive target lesions with a diagnostic sequence of pathological changes. A severe variant of von Hebra’s erythema multiforme was described by Stevens and Johnson in 1922 2 as acute febrile illness in two boys. They characterised this as a “new eruptive fever with stomatitis and ophthalmia”. Stevens-Johnson syndrome (SJS) involves several mucosal surfaces and internal organs, accompanied by severe constitutional symptoms. It is also called erythema multiforme major.

SJS predominantly involves the oral mucosa and conjunctiva. Various studies have shown conjunctival involvement varying from 49% to 81%. 23 A wide range of factors have been suggested to precipitate SJS but only in a few instances could it be convincingly established. A variety of drugs are thought to cause up to 60% of the cases of SJS, but there is no conclusive evidence except with long-acting sulfonamides. 45 Management of patients with SJS is a challenge and is often frustrating. Acute medical intervention involves both local and systemic measures for care of lids, conjunctiva and cornea. In the later part of the disease process, medical therapy aims at reducing the sequelae of cicatrisation. Surgical therapy aims to correct structural abnormalities.

This retrospective review of patients presenting with SJS at our institute aims to evaluate the presentation characteristics, possible aetiological factors, ocular complications and their management.

Materials and Methods

Medical records of patients with SJS who presented at LV Prasad Eye Institute, Hyderabad, India, between 1987 and 1998 were retrospectively reviewed. The inclusion criteria 1 applied to diagnose SJS were as follows:

  • 1. A serious mucocutaneous illness with characteristic target-like lesions
  • 2. Bullae and extensive areas of necrosis
  • 3. A prominent acute prodromal period
  • 4. Involvement of at least two mucosal sites

In all cases, a detailed clinical history pertaining to the diagnostic criteria was taken or the referring physician’s prescription at the initial episode was sought. The patients who met the above criteria were included in the analysis. The hospital records were reviewed to obtain detailed demographic data. Details of laterality, symmetry, duration of initial insult to time of presentation to the institute and possible aetiological factors were noted.

Use of any oral or systemic drug was considered a possible aetiological factor, if the drug had been taken within two weeks of onset of prodromal symptoms. If the patient showed signs of regression in the clinical condition despite continuing to take this drug, then it was not considered as the possible aetiological agent. All patients were asked to narrate important events that preceded the onset of SJS.

Detailed slitlamp examination included the ocular findings of lids, conjunctiva and cornea (Table 1). Lid abnormalities included lid thickening, discharge, meibomitis, blepharitis, trichaisis, dystrichaisis, entropion and ectropion. Conjunctival pathologies included congestion, xerosis, symblepharon and ankyloblepharon. Corneal abnormalities included superficial punctate keratitis, thinning, scarring, vascularisation, keratinisation and infectious keratitis.

Table 1:
Ocular complications in Stevens-Johnson Syndrome (n=95)

The best corrected visual acuity in the better eye was noted at presentation. All patients received medical treatment with topical tear substitutes topical corticosteroids or topical broad spectrum antibiotics either alone or in combination. Patients who had some structural abnormalities were treated surgically. All patients were informed about the chronicity of ocular disease.


A total of 95 patients were diagnosed with SJS during the 11-year period. The total number of patients seen in the outpatient department during this period was 226, 760. Thus, the prevalence of SJS in our centre was 0.0004% which in our institute translates to approximately one patient seen every 6 weeks. Eight patients were less than 1 year of age, 7 were between 1 and 10 years, 19 patients between 11 and 20 years, 53 patients between 21 and 40 years and 8 patients more than 40 years. There were 40 males (42.10%) and 55 females (57.89%). All patients had bilateral involvement and 93 (97.89%) patients had symmetrical involvement in both eyes.

Since the hospital is a tertiary eye-care centre in South India, patients come from all regions of India. As expected, most of the patients were from South India (n=58; 61.05%) followed by the West (n=15; 15.78%), East (n=13; 13.68) and North (n=9; 9.47%).

The duration from initial onset of symptoms to the time of presentation to the institute varied from 6 days to 18 years. Eight (8.42%) patients presented within 30 days, 28 (29.47%) patients between 2 months to 6 months, 20 (21.05%) patients presented between 6 months and a year and a majority of patients (n=39; 41.05%) presented after one year (Figure 1). Eighty seven (91.51%) patients presented later than 4 weeks.

Figure 1:
Distribution in months of the time lag between initial episode of SJS and presentation;

Intake of drugs was the most commonly identified possible aetiology (n=55; 57.89%). Sulfonamides, under various trade names, comprised the most frequently identified agent before the onset of SJS (n=27; 49.09%). In five patients (9.09%) penicillin, either by oral or systemic route, was the possible aetiology. Other drugs included antibiotics, anticonvulsants, antipyretics, barbiturates, dapsone and anticancer drugs. We thus identified 20 different drugs that led to SJS in our study patients. Three (3.15%) patients had a prior history of viral infection with no history of drug ingestion. No definitive cause could be ascertained in 37 (38.94%) patients.

The best corrected visual acuity (BCVA) at presentation (Figure 2) in the better eye was as follows: 32 (33.68%) patients had an acuity of 6/12 or better; 17 (17.89%) patients had an acuity between 6/12-6/60; 24 (25.26%) patients had an acuity of 6/60 to counting fingers close to face; 15 (15.78%) patients either had hand movements close to face or light perception (LP) only. While visual acuity could not be assessed in 7 (73.6%) patients, no patient presented without light perception (NLP). In 39 (41.05%) patients, BCVA in the better eye was 6/60 or less.

Figure 2:
Distribution in Best Corrected Visual Acuity in the better eye at the time of presentation.LP = Perception of light; HM = Hand motions; CF = Counting fingers; NLP = No light perception

On slitlamp examination the lid, conjunctiva and corneal complications were seen in 87 (91.5%) patients, 92 (96.84%), and 93 (97.89%) patients respectively. The lid complications included adhesions, lid margin thickening, meibomitis and blepharitis, thick discharge, trichiasis and dystrichiasis, ectropion and entropion. Conjunctival complications included congestion, xerosis, symblepharon, fornicial shortening and ankyloble-pharon. Corneal complications included superficial punctate epithelial keratitis, corneal scarring, superficial and deep vascularisation, keratinisation, infectious keratitis, and corneal thining. Lid oedema/lid margin thickening (n= 51; 53.68%) and meibomitis/blepharitis (n=56; 58.94%) were the most common lid complications. Conjunctival xerosis (n=75; 78.94%) and severe cicatrizing complications (n=57; 60%) were most common conjunctival complications. Superficial punctate keratitis (n=62; 65.26%) and corneal scarring (n=68; 71.57%) were the most common corneal complications. The frequency of complications of lid, conjunctiva and cornea is shown in (Table 1).

All patients presenting with ophthalmic complications received topical medications. This included antibiotics, tear substitute and topical corticosteroids. All patients were either on one, two or all three topical medications at presentation. Twenty-six (27.36%) patients received surgery at our centre. Among them, eight (30.76%) patients received tissue adhesive and bandage contact lens, 3 (11.53%) patients underwent therapeutic penetrating keratoplasty for non-resolving infectious keratitis, 4 (15.38%) patients received paramedian tarsorrhaphy to prevent exposure, 3 (11.53%) patients received electrolysis for trichiasis and one (3.84%) patient received punctal occlusion to restore tears in the eye. Five (19.23%) patients needed extensive symblepharon release, reconstruction of the fornix and amniotic membrane grafting.


Stevens-Johnson syndrome is an important cause of ocular morbidity. Little has changed over several decades in our understanding of the pathology and management of ocular complications. This analysis aimed to study the presenting features, possible aetiological factors, ocular complications and their management in a tertiary eye-care centre. Most patients who develop SJS consult a physician or a dermatologist in the acute phase. They consult ophthalmologists only after the resolution of skin lesions and are later referred to tertiary eye-care centres. This explains the delay in most patients seen at the institute. The rarity of SJS and its complications is demonstrated by the fact that only 95 patients were seen in a span of 11 years, and out of a total 226,760 patients (prevalence 0.0004%) seen during this period at our centre. This is similar to some of the reported small series. 678 In a relatively larger series Power et al 1 documented 37 patients over a period of 34 years.

Although the acute phase of SJS persists for 2-6 weeks, only about 8.42% presented during this period and the rest presented between 2 months to more than one year after this phase. This clearly indicates that most of these patients are managed in other centres for a long time before they are referred to the tertiary eye care centres.

Various studies have shown that drugs are responsible in up to 60% of cases of SJS. [1, 9, 10] In our study drugs were the cause in 55.89% and sulfonamides was the most common (49.09%) aetiological agent. Despite the advent of newer antibiotics, sulfonamides continue to be used by some physicians. This remains a cause of concern. Definite cause could not be ascertained in 37 (38.94%) patients, similar to the observation in another study. 1

After the acute vesicular phase, within hours to days, a concomitant conjunctivitis typically develops. A more severe conjunctival lesion results in pseudomembrane or membranous conjunctivitis 1112 which may often lead to cicatricial conjunctivitis. 13 Later complications result from scarring of lids leading to ectropion, entropion, trichiasis and lagophthalmos. The conjunctiva heals by scarring and may lead to symblepharon or ankyloblepharon. 13 Tear film deficiency is often troublesome and leads to late phase complications 14 which in turn lead to conjunctival and corneal xerosis with ocular surface problems. Late phase corneal complications develop due to corneal exposure leading to punctate epithelial keratitis, 8 recurrent epithelial defects, ingrowth of abnormal new blood vessels (corneal neovascularisation), opacification in the visual axis and blindness. Finally, the impairment of host defense system due to alteration in lid structure, composition of tear film, loss of integrity of epithelium lead to more serious complications such as corneal and scleral infections. Uncontrolled infection may lead to perforation, endophthalmitis and panophthalmitis 115 that may finally need evisceration or enucleation.

In our series, since most patients presented beyond the critical 6-week period, it was expected that a majority had late-onset complications of lids, conjunctiva and cornea leading to blindness (VA ≤ 6/60) in 41.05%. Impairment of various host-defense mechanisms resulted in ocular infection in the form of infectious keratitis in 12 eyes (12.90%).

In the critical acute phase, local measures play an important role. Measures advocated include tear substitutes, frequent irrigation of both superior and inferior conjunctival fornices with preservative-free solutions, removal of pseudomembranes and lysis of symblepharon using a glass rod or symblepharon ring. Topical antibiotics to prevent secondary infections, topical corticosteroid to prevent scar formation and topical cycloplegics can be used to relieve pain, photophobia and ciliary spasm. The integrity of corneal epithelium could be maintained by bandage contact lens, but this should be used with extreme caution. Since most of these patients do not receive meticulous local measures, they tend to develop longterm complications. Thus crucial time is often lost before the patient is referred to a tertiary eye care centre. Surgical therapy at this stage only aims to correct structural defects of lids, conjunctiva and cornea. Corneal transplantation has poor prognosis with cicatricial ocular disorders. 16 Recently we have reported better results with amniotic membrane transplantation for ocular surface reconstruction in selected cases of Stevens-Johnson syndrome. 9

Despite its retrospective nature, this study is useful to understand some of the presenting features, possible aetiological factors, ocular complications and management in SJS. We did not intend to analyse the outcome as most of the patients came for a single consultation. All the same this report represents the first large series from the Indian subcontinent.


1. Power WJ, Ghoraishi M, Merayo-Lloves J, Neves RA, Foster CS. Analysis of acute ophthalmic manifestations of erythema multiforme/Stevens Johnson syndrome/toxic epidermal necrolysis disease spectrum Ophthalmology. 1995;102:1669–76
2. Arstikatis MJ. Ocular aftermath of Stevens Johnson syndrome:review of 33 cases Ophthalmology. 1973;90:37679
3. Rasmussen JE. Erythema multiforme in children. Response to treatment with systemic corticosteroids Br J Dematol. 1976;95:181–86
4. Biancine JR, Macaraeg PV Jr, Lasagna L, Azarnoff DL, Brunk SF, Hvidberg EF, et al Drugs as etiological factors in Stevens Johnson syndrome Am J Ophthalmol. 1968;44:390–405
5. Bottinger LE, Strandberg I, Westernholm B. Drug-induced febrile mucocutaneous syndrome:With a review of literature Acta Med Scand. 1975;198:229–33
6. Yetiv JZ, Bianchieve JR, Ower JA Jr. Etiologic factors in the Stevens Johnson syndrome South Med J. 1980;73:599–602
7. Howard GM. The Stevens Johnson syndrome. Ocular prognosis and treatment Am J Ophthalmol. 1963;55:893–900
8. Wright P, Collin JR. The ocular complications of erythema multiforme (Stevens Johnson syndrome) and their management Trans Ophthalmol Soc UK. 1983;103:338–41
9. Honavar SG, Bansal AK, Sangwan VS, Rao GN. Amniotic membrane transplantation for ocular surface reconstruction in Stevens-Johnson syndrome Ophthalmology. 2000;107:975–79
10. Schopf E, Stuhmer A, Rzany B, Victor N, Zentgraf R, Kapp JF. Toxic epidermal neurolysis and Stevens Johnson syndrome. An epidemiologic study from West Germany Arch Dermatol. 1991;127:839–42
11. Foster CS, Fong LP, Azar D, Kenyon KR. Episodic conjunctival inflammation after Stevens Johnson syndrome Ophthalmology. 1988;95:453–62
12. Prendiville JS, Herbert AA, Greenwald Mj, Esterly MD. Management of Stevens Johnson syndrome and toxic epidermal necrolysis in children J Pediatr. 1989;115:881–87
13. Mondino B. Cicatricial pemphigoid and erythema multiforme Ophthalmology. 1990;97:939–52
14. Baum J. Clinical manifestations of dry eye states Trans Ophthalmol Soc UK. 1985;104:415–23
15. Tabbara K, Shammas H. Bilateral corneal perforations in Stevens Johnson syndrome Can J Ophthalmol. 1975;10:51417
16. Tutkun IT, Akova YA, Foster CS. Penetrating keratoplasty in cicatrizing conjunctival diseases Ophthalmology. 1995;102:576–85

Stevens-Johnson syndrome; erythema multiforme major; complications; management

© 2002 Indian Journal of Ophthalmology | Published by Wolters Kluwer – Medknow