The case report illustrates a TEN patient with good visual and structural outcome after early bilateral amniotic membrane transplantation for severe ocular surface inflammation. To our knowledge, this is the first report demonstrating the effectiveness of early AMT in a Chinese pediatric patient with TEN and adds to the growing support for early ocular intervention in patients with ocular involvement of SJS/TEN spectrum disease.
Ocular surface disease is a common manifestation of toxic epidermal necrolysis (TEN), with 50-67% of patients developing ocular disease, usually presenting as acute membranous or pseudomembranous conjunctivitis with sloughing of the epithelia [14
]. On a cellular level, this is characterized by dermal-epidermal separation as well as epidermal apoptosis and necrosis. The underlying pathogenesis is suspected to be due to disruption of the immune system. 4-20% of patients with TEN suffer from severe ocular surface inflammation, although the severity of eye involvement does not always correspond to the severity of skin involvement or systemic illness [3
During the chronic phase of SJS/TEN, ocular surface disease is severely debilitating and is a product of conjunctival scarring, symblepharon, dry eye, conjunctivalization of cornea, corneal scar with vascularization, and eyelid malformations [3
]. While ocular surface reconstruction in the chronic phase has been reported, it is a major surgical challenge with limited success. More promising is a paradigm shift towards disease modulation during the acute inflammatory stage. Recent literature reports have shown that early intervention with AMT in the acute phase of SJS/TEN can minimize ocular surface damage [5
The amniotic membrane consists of the inner layer of the placenta. It consists of avascular cuboidal/columnar epithelia attached to a basement membrane and secretes certain growth factors, including epidermal growth factor, keratinocyte growth factor and hepatocyte growth factor, which may be useful in healing of the ocular surface [19
]. When transplanted, the membrane acts as a biological bandage contact lens, providing a barrier between the denuded mucosal surfaces, thus preventing adhesion and symblephara formation. Furthermore, the amniotic membrane itself has been shown to have anti-inflammatory and anti-fibrotic effects and provides a physical scaffold for epithelial cells to migrate [20
]. In the treatment of persistent cornea epithelial defects, there is no significant difference in re-epithelialization rates between eyes with membranes transplanted epithelial side up (overlay) or those with membranes transplanted epithelial side down (inlay) [22
]. Therefore AMT can theoretically facilitate epithelialization while reducing inflammation and scarring during the acute inflammatory phase of TEN.
In the landmark report by John et al. in 2002, the authors first described the use of AMT epithelial side up (overlay) in the acute phase of TEN in four eyes of two patients suffering from severe ocular involvement: a 6-year old boy and an 8-year old girl. Furthermore the authors were the first to describe AMT over the denuded external eyelid surface and lid margin [5
]. The aim of treatment was to perform AMT within 2 weeks of disease onset. To avoid amblyopia, the authors excised the membrane covering the cornea at the end of the surgery. To protect the cornea, they placed a bandage soft contact lens (BSCL) over it. Further case reports by Kobayashi et al., Atzori et al. and Tandon et al. confirmed beneficial effects of early AMT on visual and ocular outcomes [6
]. Kobayashi et al. did not excise the membrane over the cornea, despite their patient being 6-years old. They found that the presence of the membrane did not prevent the patient from seeing and also did not prevent the medical team from performing fluorescein staining. The group also performed AMT much earlier, on day 5 of illness. Atzori et al. were the first to show that early AMT benefited elderly patients as well, with good long-term visual outcome in a 68-year old woman.
One major limitation of AMT is that it involves a challenging surgical procedure to be performed on a gravely ill patient under general anesthesia. Alternatives include placing a Prokera device; a symblepharon ring with a sheet of amniotic membrane clipped to it, or placing the amniotic membrane unsutured over the cornea and bulbar conjunctiva. A problem of both of these alternatives is that they do not fully cover all the denuded ocular surfaces. Shammas et al. showed that in eyes which underwent Prokera or unsutured AMT alone, ocular scarring was much more severe than those that underwent sutured AMT [9
]. While Prokera devices may promote epithelialization and can be done in a clinic setting under topical anesthesia, they are less effective in preventing symblephara formation and eyelid malformations. The limitations of the Prokera device in this setting was further confirmed by several case reports [23
The first case series on early AMT use in acute TEN was reported by Gregory in 2011. 10 consecutive patients with severe ocular involvement of SJS or TEN underwent AMT within the first 10 days of illness. All patients achieved a long-term visual acuity of 20/30 or more, with 90% achieving 20/20 vision [10
]. Repeat AMT every 10-14 days was helpful in those with persistent epithelial defects after initial transplantation. 2 out of 3 patients than had inferior visual outcomes had AMT done on day 10, thus the study recommended earlier treatment within the first week of illness. This was supported by a retrospective case-control series by Hsu et al in 2012, which also showed that the best visual outcomes were in patients who underwent AMT within 5 days of disease onset and that Prokera alone was inadequate to minimize long-term ocular cicatricial complications [11
]. A recent study by Kim et al looked into 51 consecutive patients with SJS or TEN. Amongst them, patients younger than 18 years of age had significantly more severe ocular involvement. Although AMT was shown to have beneficial effects on the ocular surface, this study crucially found that early intervention with intravenous immunoglobulin (IVIG) and corticosteroids led to significantly improved long-term visual outcomes [12
]. Hence good systemic treatment by a physician is just as important in the management of acute ocular inflammation.
Most recently, a prospective case series by Lopez-Garcia et al. on AMT for moderate and severe ocular involvement of TEN showed that the observed clinical improvement in ocular surface condition and visual acuity corresponded with a reduction in squamous metaplasia in corneal and conjunctival cells and increase in goblet cell density over 1 year [13
]. This is the first study to demonstrate the beneficial effects of AMT were linked to histological improvements of the ocular surface.
When applying this evidence into practice, it is important to also note that there is a lack of standardization of amniotic membrane quality. There is considerable inter- and intra-donor variation in the quality of amniotic membrane. Furthermore the time from harvesting to use is different in each center, depending on availability. At our center, amniotic membrane is harvested annually and then cryo-preserved at -80°C for up to one year. It has been shown that the biologically active growth factors are depleted over time. Hence clinical efficacy from the mentioned studies may not be universal.
In conclusion, results from this case report add to the growing evidence supporting early AMT in minimizing ocular surface damage and preventing severe vision loss in patients acute TEN. To our knowledge this is the first case written on a Chinese pediatric patient with severe ocular involvement of acute TEN. Effective management is dependent on early and active communication between the patient’s family, pediatricians, intensive care physicians and ophthalmologists.