Burn Center Care of Patients with Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis




Stevens-Johnson syndrome and toxic epidermal necrolysis are rare, life-threatening, cutaneous drug reactions. Medications are the most common cause, although an infection may be responsible. A link between genetics and certain medications has been established. Clinical diagnosis should be confirmed with biopsy. When the area of epidermal detachment approaches 30%, burn center care is advisable. An ophthalmologist should be consulted to optimize ocular care. Pharmacologic interruption has been sought but there is little consensus on the most appropriate agent and no high-quality studies have been conducted to demonstrate if any of these agents lead to improved survival.


Key points








  • Although patients with Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are now routinely referred to burn centers for definitive care, not all cases of SJS or even SJS-TEN overlap automatically warrant burn center admission. However, the more extensive the degree of exfoliation (especially in TEN), the greater is the need for and benefit from providing care in a burn center setting.



  • Diagnosis of SJS and TEN relies on careful documentation of systemic, cutaneous, and mucosal features obtained from a detailed history and physical examination, combined with histologic confirmation by biopsy.



  • The treatment bundle in the burn center should include cessation of the causative medication; careful airway assessment and protection, if indicated; directed (rather than routine) fluid replacement for hypovolemia; early enteral nutrition; wound coverage with skin substitutes; urgent ophthalmologic consultation; careful surveillance for infection; avoidance of prophylactic antibiotics; and consideration of a pharmacologic intervention to attempt to halt the disease process.



  • Pharmacologic interventions to halt the disease process have not been examined through large high-quality studies and consensus on use of these agents is lacking.






Introduction


Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are severe, potentially life-threatening, adverse cutaneous reactions that most commonly are precipitated by the use of a medication. Although these conditions are rare, with only 1.5 to 1.9 cases occurring per million of the population per year, the associated mortality ranges between 9% for SJS to 30% to 50% for TEN. An important feature of these conditions is varying degrees of detachment of the epidermis, which results in wounds that are analogous to superficial partial thickness burns. Consequently, patients with SJS and TEN are frequently referred to specialized burn units for their care, primarily to optimize wound and dressing management but also to garner all of the other beneficial aspects of multidisciplinary care that modern burn treatment facilities can offer for patients with large cutaneous wounds. Not surprisingly, a multicenter review of TEN found that delayed referral of patients with TEN to a burn center was associated with significantly lower odds of survival. This article reviews all aspects of SJS and TEN but primarily focuses on the important principles of management of patients with these conditions.




Introduction


Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are severe, potentially life-threatening, adverse cutaneous reactions that most commonly are precipitated by the use of a medication. Although these conditions are rare, with only 1.5 to 1.9 cases occurring per million of the population per year, the associated mortality ranges between 9% for SJS to 30% to 50% for TEN. An important feature of these conditions is varying degrees of detachment of the epidermis, which results in wounds that are analogous to superficial partial thickness burns. Consequently, patients with SJS and TEN are frequently referred to specialized burn units for their care, primarily to optimize wound and dressing management but also to garner all of the other beneficial aspects of multidisciplinary care that modern burn treatment facilities can offer for patients with large cutaneous wounds. Not surprisingly, a multicenter review of TEN found that delayed referral of patients with TEN to a burn center was associated with significantly lower odds of survival. This article reviews all aspects of SJS and TEN but primarily focuses on the important principles of management of patients with these conditions.




Nomenclature and classification


SJS and TEN likely represent different degrees of severity of the same disease process. In both conditions, there must be a skin rash that features epidermal detachment as well as erosion of mucosal membranes (mucositis) at 2 or more locations. In SJS, epidermal detachment is limited to less than 10% of the total body surface area (TBSA) whereas in TEN, epidermal loss occurs over greater than 30% of the TBSA. When the epidermal detachment involves 10% to 30% of the TBSA, the condition is referred to as SJS-TEN overlap. In SJS, SJS-TEN overlap, and TEN, the rash features spots but not typical target lesions, although flat atypical target lesions can be seen. The spectrum of SJS-TEN should be distinguished from another distinct condition, erythema multiforme major, which occurs following infection with herpes simplex virus or Mycoplasma pneumoniae , and which features oral mucositis, a rash consisting of typical target lesions, as well as raised atypical target lesions but minimal to no epidermal detachment. A final entity, TEN without spots, features diffuse erythema but no spots or target lesions and epidermal detachment involving greater than 10% of the TBSA. These points of classification are summarized in Table 1 .



Table 1

Classification of exfoliative skin reactions





















































Erythema Multiforme Major SJS SJS-TEN Overlap TEN TEN Without Spots
Mucositis Oral mucositis ≥2 sites ≥2 sites ≥2 sites ≥2 sites
Spots No Yes Yes Yes No
Atypical target lesions Yes, raised Yes, flat Yes, flat Yes, flat No
Typical target lesions Yes No No No No
Epidermal detachment Negligible <10% TBSA 10%–30% TBSA >30% TBSA >10% TBSA
Mortality None 10% 30%–50%

Adapted from Bastuji-Garin S, Rzany B, Stern RS, et al. Clinical classification of cases of toxic epidermal necrolysis, Stevens-Johnson syndrome, and erythema multiforme. Arch Dermatol 1993;129:92–6.




Causes of Stevens-Johnson syndrome and toxic epidermal necrolysis


Although approximately 75% of cases of SJS-TEN result from the use of a drug, it is recognized that viral illnesses, influenza-like illnesses, and M pneumoniae infections may also be causative. A recent case-control study identified that drugs with the highest risk of inducing SJS-TEN included allopurinol, anti-infective sulfonamides (eg, cotrimoxazole), phenytoin, carbamazepine, phenobarbital, lamotrigine (an antiepileptic drug), nevirapine (an anti–human immunodeficiency virus [HIV] drug), and oxicam-type nonsteroidal anti-inflammatory drugs (NSAIDs) (eg, meloxicam). The latency period between initiation of these high-risk drugs and onset of the cutaneous reaction ranged between 4 and 28 days. Commonly prescribed drugs that carry a moderate risk of inducing SJS-TEN included cephalosporins, macrolide and quinolone antibiotics, tetracycline, and acetic acid–type NSAIDs (eg, diclofenac). Frequently prescribed drugs that had no increased risk of causing SJS-TEN included valproic acid (in contrast to most of the other antiepileptics), angiotensin converting enzyme (ACE)-inhibitors, beta blockers, calcium channel blockers, diuretics and oral hypoglycemic drugs that carry a sulfonamide structure, and propionic type NSAIDs (eg, ibuprofen). Although these drugs do not seem to carry a higher risk of causing SJS-TEN, they have been implicated in individual cases of SJS-TEN and, in general, have a much longer latency period than the high-risk drugs (eg, >30 weeks in the case of valproic acid). Unfortunately, there are no reliable in vivo or in vitro tests to identify drug causality in SJS-TEN, although the lymphocyte transformation test and granulysin expression test have been explored. Two essential principles to remember with respect to medication use and causes of SJS-TEN are




  • All medications should be considered as potential suspects, including intermittently used drugs such as vitamins and analgesics, and homeopathic, traditional, or natural products.



  • Drugs may be started to treat the initial prodromal symptoms of SJS-TEN, and must not be mistaken as the causative agent. A careful history noting the time sequence of medication use and appearance of systemic and cutaneous signs and symptoms is mandatory.





Genetic susceptibility


A genetic predisposition to SJS-TEN has long been suspected and, in the last few decades, various relationships between genetic expression of certain human leukocyte antigen (HLA) alleles, specific medications, and racial background have been identified. One of the earliest discovered associations was a weak link between expression of HLA-B*12 and oxicam-induced TEN in patients of European ancestry. More recently, a very strong association between expression of HLA-B*15:02, carbamazepine-induced SJS-TEN, and Han Chinese, Thai, Indian, and Malaysian origin has been observed; as has a very strong association between HLA-B*58:01 expression and allopurinol-induced SJS-TEN, in Han Chinese, Thai, Japanese, Korean, and European populations. Although many other associations have been discovered, the US Food and Drug Administration currently only recommends HLA-B*15:02 screening in patients of Asian ancestry before starting carbamazepine.




Pathogenesis and pathologic features


The pathogenesis of SJS-TEN is complex and incompletely understood. Current theories all generally involve activation of keratinocyte apoptosis, which leads to extensive full thickness loss of the epidermal layer, and then separation of the nonviable epidermis at the dermal-epidermal junction. The underlying dermis is viable and usually shows only a mild infiltration of lymphocytes. How a drug elicits this response is unknown but it seems to involve an immune reaction featuring activation of CD 8 + cytotoxic T lymphocytes, natural killer cells, and increased production of the proapoptotic protein, granulysin. Serum levels of granulysin seem to correlate with disease severity, suggesting that granulysin is a critical mediator for keratinocyte death. Activation of T cells seems to involve an interaction between the drug, HLA molecules on antigen-presenting cells, and T cell receptors. In addition, keratinocyte apoptosis also seems to be stimulated by the interaction of the Fas ligand (FasL) binding with the Fas receptor on the keratinocyte cell membrane. Raised serum levels of soluble FasL, arising either from the keratinocytes themselves and/or peripheral blood mononuclear cells, have been identified in patients with SJS-TEN, presumably leading to an interaction with keratinocyte Fas to activate keratinocyte apoptosis. Finally, inflammatory cascades involving reactive oxygen species and tumor necrosis factor (TNF)-α also act to intensify stimulation of apoptotic pathways among keratinocytes.




Clinical picture and diagnosis


SJS and TEN appear most commonly between 4 days and 4 weeks after exposure to the drug. There is usually an initial nonspecific prodromal illness lasting 2 or 3 days that may feature malaise and fatigue, fever, sore throat, rhinitis, and irritation or pruritus of the skin and eyes. Patients often take, or are prescribed a medication for these symptoms, which is then erroneously implicated as the causative drug once the cutaneous manifestations appear and the diagnosis of SJS or TEN becomes apparent.


After this prodromal phase, cutaneous lesions start to appear on the central trunk, face, and proximal limbs, eventually spreading to the distal extremities, including the palms and soles. The initial lesions are dusky erythematous, purpuric, and irregularly shaped macules and spots. Some lesions have the appearance of atypical flat target lesions with 2 concentric rings around a necrotic center. These lesions begin to blister and coalesce to form bullae. The epidermis in these involved areas is unstable and tangential digital pressure on the skin causes the epidermis to shear off and detach (the positive but nonspecific Nikolsky sign). Further confluence of the blistered skin and bullae leads formation of larger flaccid bullae and, ultimately, to spontaneous sheet-like detachment of necrotic epidermis, leaving painful raw areas of glistening bright red dermis exposed ( Figs. 1 and 2 ). The percent of the TBSA in which there is imminent or actual epidermal detachment determines where the patient is placed on the SJS-TEN spectrum: less than 10% TBSA in SJS, 10% to 30% TBSA in SJS-TEN overlap, and greater than 30% TBSA in TEN.




Fig. 1


Initial rash featuring dusky purpuric macules and spots with early vesicle and blister formation ( left ). Initial progression of rash with coalescence of lesions, formation of larger blisters and bullae, and early epidermal detachment ( center ). Final progression to extensive sheet-like epidermal detachment ( right ).



Fig. 2


Demonstration of the positive Nikolsky sign in a patient with TEN.


Simultaneously, erosions and sloughing of the mucous membranes in at least 2 sites have also been progressing. The lips and oropharynx are the most commonly involved, followed by the conjunctiva, genital mucosa, and anorectal mucosa. Mucosal sloughing may also involve the esophagus, the remainder of the gastrointestinal tract, and the tracheobronchial surfaces. Aggressive involvement of the lips and oropharynx is common, with severe pain, erosions and bleeding, dysphagia, odynophagia, and drooling. The lips typically become crusted with dried blood. Patients usually are unable to eat or drink because of the severity of this mucositis ( Fig. 3 ). Similarly, ocular involvement is also usually severe, with photophobia, tearing, conjunctivitis, purulent exudates, crusting, and pseudomembranous conjunctival erosions that result in the formation of synechiae between the lids or between the conjunctiva and the eyelids ( Fig. 4 ). Keratitis and corneal ulcerations can occur and may lead to permanent partial or complete visual loss. Erosions involving the glans penis, vulva, or vagina may cause the patient to report pain or burning during micturition.




Fig. 3


Severe oral mucositis in a patient with TEN. This patient was intubated because of the severity of the upper airway mucositis and exfoliation of more than 70% TBSA.



Fig. 4


Florid conjunctivitis with exudate and strands between upper and lower eyelids. A glass rod is inserted into the medial lower conjunctival fornix to be swept across the fornix to break up these bands.


By the time a patient presents to the burn unit, systemic manifestations may be well advanced. Often the patient is dehydrated and malnourished. Significant upper airway mucositis may compromise the airway and around one-fourth of TEN cases may have early respiratory dysfunction featuring cough, trachea-bronchial mucosal involvement, and dyspnea. Gastrointestinal mucosal involvement is much less common but when it occurs may present with diarrhea, or bleeding or excretion of sloughing bowel mucosal epithelium. Early acute kidney injury may be manifested by elevation in the serum creatinine and oliguria. Hematuria and microalbuminuria may also be present. On presentation, laboratory abnormalities may include anemia, neutropenia, thrombocytopenia, and elevated hepatic enzymes. In cases in which there have been extensive areas of epidermal loss, the patient may develop problems related to fluid loss, hypothermia, and invasive infection. Not uncommonly, hospital-acquired infections related to invasive care with urinary catheters, central lines, and endotracheal intubation eventually appear, leading to sepsis and multiorgan failure, which is the main cause of death in patients with TEN.


The diagnosis of SJS-TEN rests on 2 factors: (1) presence of systemic, cutaneous, and mucosal features consistent with SJS-TEN based on a detailed history and physical examination and (2) histologic confirmation. A clear identification of the prodromal illness followed by onset of skin and mucosal changes, a positive Nikolsky sign, and a clear documentation of the percent body surface area where there is actual or imminent epidermal slough only, using the rule of nines and/or a Lund and Browder burn diagram are considered the minimum elements needed to make the clinical diagnosis. Several punch biopsies should be obtained from representative areas where the epidermis is still intact for both immediate frozen section and fixed section analysis. Histologic examination will show complete necrosis of all epidermal layers with a cleavage plane separating the epidermis from the dermis. Direct immunofluorescence should be performed to prove that there is no immunoglobulin or complement deposition in the epidermis, which are features more in keeping with other autoimmune blistering disorders, such as paraneoplastic pemphigus and bullous pemphigoid.


The differential diagnosis includes staphylococcal scalded skin syndrome (SSSS), autoimmune blistering diseases (paraneoplastic pemphigus, pemphigus vulgaris, bullous pemphigoid), Acute generalized exanthematous pustulosis (AGEP), and generalized bullous fixed drug eruption (GBFDE). Mucositis is absent in SSSS and frozen section biopsy will show that the epidermal separation arises from an intraepidermal cleavage plane rather than a dermoepidermal split as in SJS-TEN. Intraepidermal cleavage is also featured in AGEP. In GBFDE, blistering is limited and mucositis is minimal and there are usually no systemic symptoms. Autoimmune blistering disorders may need to be diagnosed by immunofluorescence testing as previously described. Finally, an acute graft versus host reaction may closely mimic SJS-TEN but would typically be seen within 2 weeks of an allogeneic hematopoietic stem cell transplant.




When to refer to a burn center


Early referral of patients with TEN to a burn center seems to be beneficial in many respects. The obvious reason to consider burn center transfer is to optimize wound care and dressing application to extensive and complex anatomic areas. Burn centers also may provide availability of skin substitutes for temporary coverage of large wounds. Strict adherence to isolation and barrier precautions, multidisciplinary care involving burn-critical care nurses, dietitians, and rehabilitation therapists, and the capability of environmental temperature regulation and specialty pressure reduction beds make burn units the ideal place to care for patients with SJS-TEN.


However, there has been a recent increasing trend to inappropriately refer patients with rashes, nonexfoliative skin conditions, and patients with mild SJS to burn centers. Clearly, the decision to involve a burn center must be based on an accurate estimation of the extent of epidermal detachment. Detachment of 10% or less of the TBSA does not automatically warrant transfer, whereas cases of TEN with detachment greater than 30% of the TBSA should be transferred as soon as possible. In between these 2 extremes is a gray zone in which referral should be considered on an individual basis.


Nov 17, 2017 | Posted by in General Surgery | Comments Off on Burn Center Care of Patients with Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access