INTRODUCTION — Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are severe, mucocutaneous, adverse reactions most commonly induced by medications. Clinical features include fever, prodromal symptoms, mucositis, and extensive epidermal detachment. SJS and TEN are considered a disease continuum and are collectively known as epidermal necrolysis.
The classification of the disease is based on the total body surface area (BSA) involved with skin detachment [1]:
●SJS is defined as skin detachment affecting <10 percent of the BSA (picture 1A-C).
●TEN is defined as skin detachment affecting >30 percent of the BSA (picture 2A-D).
●SJS/TEN overlap is defined as skin detachment involving 10 to 30 percent of the BSA.
We will use the term "Stevens-Johnson syndrome/toxic epidermal necrolysis" to refer collectively to SJS, TEN, and SJS/TEN overlap.
The management, prognosis, and long-term sequelae of SJS/TEN are discussed in this topic. The pathogenesis, clinical manifestations, and diagnosis of SJS/TEN are discussed separately. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis".)
GENERAL PRINCIPLES — The management of SJS/TEN can be broadly classified into two phases [2]:
●Acute phase – During the acute phase, progressive worsening of the cutaneous detachment and mucositis occur for five to seven days, followed by disease arrest and re-epithelization. Due to extensive skin detachment during this phase, the patient is prone to fluid and electrolyte imbalances, sepsis, organ decompensation, and death, with mortality rates of 10 to 50 percent depending on the extent and severity of disease and the patient's age and comorbidities [3,4]. Management during this phase is focused on supportive care and prevention of short- and long-term complications.
●Chronic phase – This phase occurs during the convalescent and recovery stage of SJS/TEN. Physical and psychologic sequelae may occur [5]. Management principles include the screening and treatment of these complications in order to maintain life quality.
INITIAL IN-HOSPITAL EVALUATION — Patients suspected of having SJS/TEN require immediate in-hospital evaluation. As soon as the diagnosis of SJS/TEN has been established, the severity and prognosis of the disease should be rapidly determined to define the appropriate medical setting for management [2,6]. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis", section on 'Diagnosis'.)
Prompt identification and withdrawal of the causative drug — Prompt identification and withdrawal of the offending agent may improve the prognosis. In a 10-year, observational study of 113 patients with TEN or SJS, early withdrawal of the causative drugs with short half-lives reduced the risk of death by 30 percent for each day before the development of blisters and erosions (odds ratio [OR] 0.69, 95% CI 0.53-0.89) [7]. However, in drugs with long half-lives, there was an increased risk of death, independent of early or late withdrawal (OR 4.9, 95% CI 1.3-18.9). This observed difference between short and long half-life medications may be due to the persistence of the causative drug or substrate despite withdrawal. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis", section on 'Drug causality assessment'.)
Causality attribution is generally determined by two principles:
●Temporal sequence – Most causative drugs are started 5 to 28 days (occasionally up to two months) prior to the onset of symptoms [8].
●Drug notoriety – The majority of SJS/TEN cases are caused by a few high-risk medications (table 1) [9-11]. An algorithm for determining drug causality has been formulated (algorithm of drug causality for epidermal necrolysis [ALDEN]) and is useful for the determination of the culprit drug (table 2) [8].
Assessment of the extent of skin detachment — The Lund-Browder chart (figure 1) is commonly used for a rapid estimation of the percentage of the total body surface area (BSA) involved in adults and children. In adult patients, the "rule of nines" can be alternatively used:
●The head represents 9 percent total BSA.
●Each arm represents 9 percent total BSA.
●Each leg represents 18 percent total BSA.
●The anterior and posterior trunk each represent 18 percent total BSA.
Assessment of prognosis — The prognosis of individual patients can be rapidly evaluated on admission by applying prognostic scoring systems. Two scores have been proposed:
●SCORTEN score – The severity of illness score of toxic epidermal necrolysis (SCORTEN) is based upon seven independent clinical and laboratory variables (table 3) [12]. The extent of the cutaneous score has been validated for use on days 1 and 3 of hospitalization [13]. The survival curves based upon the SCORTEN score at admission (figure 2) may be helpful when discussing a patient's prognosis [13]. However, due to progress in the management of patients with SJS/TEN over time, particularly in specialized centers, the SCORTEN score may overestimate the mortality risk in these settings [14,15].
●ABCD-10 – A prognostic scoring system consisting of five variables (age, serum bicarbonate level, cancer, dialysis, involvement of >10 percent BSA [ABCD-10]) has been proposed based on data from a multicenter, United States cohort (n = 370) of adult patients with SJS/TEN admitted between 2000 and 2015 [16]. One point is given each for age 50 years and older, epidermal detachment >10 percent BSA, and serum bicarbonate <20 mmol/L. Two points are given for presence of active cancer, and three points are given for dialysis prior to admission. This translated to a mortality risk of 2.3, 5.4, 12.3, 25.5, 45.7, 67.4, and 83.6 percent for scores of 0, 1, 2, 3, 4, 5, and ≥6, respectively.
Comparisons between the two prognostic scores suggest that the performance of SCORTEN remains superior and should be considered as the default prognostic tool [15,17].
A study evaluating the association between inflammatory markers and in-hospital mortality in 192 patients with SJS/TEN showed that red cell distribution width to hemoglobin ratio (RDW/Hb) had a similar predictive accuracy as SCORTEN [18]. Furthermore, when incorporated into SCORTEN (revision of SCORTEN [Re-SCORTEN]), the prognostic ability was further augmented. In the Re-SCORTEN prognostic score, an additional two points are added to SCORTEN if the RDW/Hb is 1.2 or higher. This translates to a mortality risk of 1.9, 4.3, 8.7, 14, 23, 35.3, 54.5, 73.2, and 86.1 percent for scores of 0, 1, 2, 3, 4, 5, 6, 7, and ≥8, respectively.
Transfer to a referral center — Patients with BSA detachment of ≥10 percent, rapidly progressive disease, or worsening biochemical/organ function should be transferred and managed in specialized centers, such as specialized dermatology units, burn centers, or intensive care units [2,6]. Several studies indicate that prognosis is better for patients promptly transferred to a reference center or burn care unit. It has also been shown that survival is increased for patients treated in reference centers with a high volume of admissions for SJS/TEN.
●In a retrospective, multicenter review of 199 patients with TEN treated at burn care centers in the United States, the overall mortality was 32 percent compared with 51 percent among patients initially cared for in other settings and transferred to a burn center more than one week after disease onset [19].
●In an analysis of the French nationwide hospital discharge database that identified 991 patients with SJS/TEN, the OR of mortality was 0.5 (95% CI 0.3-1.0) for one supplemental admission to centers with experience in treating TEN compared with centers without experience, even after adjusting for individual risk factors [20].
●In a retrospective cohort study, late admission (>5 days after skin detachment) was associated with increased risk of death (OR 2.9, 95% CI 1.0-8.3), bacteremia (OR 3.7, 95% CI 1.2-11.2), and intensive care unit admission (OR 3.3, 95% CI 1.1-9.8) compared with early admission (within four days of skin detachment) [21].
ACUTE MANAGEMENT — The management of SJS/TEN is multidisciplinary and should be driven by a team of clinicians experienced in treating this disease [2,6,22-24].
Supportive care — Supportive care is the cornerstone of management and includes wound care, fluid and electrolyte management, nutritional support, temperature management, pain control, infection prevention and management, ocular care, and organ support, if needed [2,6,23-25]. Mucosal surfaces of the eyes, mouth, and urogenital tract are frequently affected, and multidisciplinary care is recommended [2,6].
Wound care — The ideal wound management strategy in SJS/TEN has not been determined. Wound care for patients with SJS/TEN varies among centers and can be broadly divided into surgical and conservative approaches [26-29]:
●Surgical approach – In the surgical approach, the devitalized epidermis is removed via operative debridement, manual scrubbing, or the use of whirlpools [30]. In some centers, biologic membranes (eg, allografts, xenografts, biosynthetic dressings) are applied to the dermis following debridement [29].
●Conservative approach – In the conservative approach or "antishear" approach [31], the detached/detachable epidermis is left in situ as a biologic dressing, and additional nonadhesive dressings may be used to promote healing. Gentle cleansing of the skin with sterile water or diluted chlorhexidine should be performed at each wound change. Large bullae may be ruptured with a sterile needle and drained without deroofing the bullae. Petrolatum-impregnated gauze is traditionally used as the primary dressing in many centers. Modern dressings (eg, nonadherent nanocrystalline gauze materials containing silver and biosynthetic dressings) are increasingly used. They may be left in situ for up to seven days, decreasing the frequency of painful dressing changes. However, the use of modern dressings does not seem to impact the healing time [32-34]. Air-fluidized beds are also useful to reduce pressure on the skin and prevent shearing of the skin [2].
Surgical and conservative approaches have not been evaluated in randomized trials. Data from observational studies suggest that they result in comparable outcomes (eg, survival, time to re-epithelization) [31,35].
Fluid and temperature management — Fluid and electrolyte imbalances occur secondarily to increased water loss from the denuded dermis and decreased oral intake in patients with extensive oral mucosal involvement. Strict intake/output monitoring should be instituted. Fluid replacement is essential to prevent end-organ hypoperfusion and shock. Replacement volumes are approximately one-third lower than those needed for burn victims [2,36]. In the first 24 hours, fluid requirements are estimated to be 2 mL/kg/body surface area (BSA) percent affected. Subsequently, fluid should be titrated according to the patient's response and urine output, aiming for urine volume of 0.5 to 1 mL/kg/hour [37].
Room temperature should be maintained between 82.4 to 89.6°F (28 to 32°C) to prevent excessive caloric expenditures due to epidermal loss and to prevent hypothermia [2,23,38]. Heated-air body warmers may also be used.
Nutrition — Nutritional supplementation should be initiated early to support metabolic disturbances and promote healing. The estimated caloric requirement is 20 to 25 kcal/kg per day during the early, catabolic phase of SJS/TEN and 25 to 30 kcal/kg per day during the anabolic, recovery phase of the disease [6].
Oral feeding may be difficult if the oral mucosal involvement is extensive. If required, nasogastric feeding should be initiated as soon as possible and continued throughout the acute phase of SJS/TEN [2,6,23,24]. The passage of a nasogastric tube should be carefully performed to minimize damage to affected mucous membranes.
Pain control — SJS/TEN is an extremely painful condition, and the pain is exacerbated by mobilization and wound care procedures. Every effort should be taken to optimize pain control. (See "Pain control in the critically ill adult patient" and "Paradigm-based treatment approaches for management of burn pain".)
The intensity of pain can be graded by asking the patient to describe the level of pain on a numerical scale of increasing severity (ranging from 0 to 10 (figure 3)). This evaluation may be repeated every four hours during the acute phase. The choice of analgesia is dependent on the level of pain, and the aim of treatment is to keep the level of pain to ≤2 [39]. Additional doses of analgesia may be required during patient mobilization or wound care procedures.
●Mild pain (intensity <4) – Nonopioid analgesics (eg, acetaminophen [paracetamol], ibuprofen).
●Moderate/severe pain (intensity ≥4) – Opioids are the drug of choice due to their potency, rapid onset of action, and dose-dependent degree of sedation. These can be delivered enterally, subcutaneously, via patient-controlled analgesia, or via intravenous infusions. For severe pain, the optimal route of administration is intravenous, which provides faster pain relief and can be titrated to meet the individual needs of the patient [2,39].
Prevention and treatment of infections — Patients with SJS/TEN are at high risk of infection. Up to 30 to 50 percent of patients with SJS/TEN develop bacteremia, and sepsis is the most common cause of death [40-42]. BSA involvement of ≥10 percent, hemoglobin levels ≤10 g/dL, and existing cardiovascular disease are associated with increased risk of developing bacteremia [41].
Prevention, active clinical surveillance, and prompt treatment of infections are key components of SJS/TEN management. Prophylactic systemic antibiotics are not recommended [2,6,41]. (See "Nosocomial infections in the intensive care unit: Epidemiology and prevention".)
●Prevention – Prevention of infection involves the following:
•Sterile handling – Sterile handling is essential. In specialized centers, patients are nursed in reverse isolation procedure rooms [2,6].
•Skin antisepsis – Antiseptic solutions containing octenidine, polyhexanide, or chlorhexidine or silver nitrate preparations may be used for disinfection. Silver sulfadiazine should be avoided if SJS/TEN is suspected to be caused by sulfonamides, but silver nitrate and silver-imbued nanocrystalline gauze materials may be used safely. (See 'Wound care' above.)
●Repeated bacterial cultures – Repeated cultures of the skin, as well as blood, catheters, gastric, and urinary tubes, should be obtained at regular intervals (eg, every 48 hours) to allow early detection of infections [41,42]. Positive skin cultures may predict blood culture positivity from the same organisms. In a French cohort of 98 patients with SJS/TEN involving ≥10 percent of the BSA, the positive and negative predictive values of skin cultures growing Staphylococcus aureus or Pseudomonas aeruginosa were 58 and 89 percent and 50 and 81 percent, respectively [43].
●Surveillance for clinical signs of sepsis – Signs of bacteremia include hypothermia, confusion, hypotension, and reduced urinary output [41,42]. Hypothermia and elevated procalcitonin ≥1 mcg at the time of blood culture have been shown to be predictive of the blood culture positivity [41]. Traditional markers of sepsis (eg, fever, elevated white cell count, erythrocyte sedimentation rates, and C-reactive proteins) are not useful in SJS/TEN [41,42].
●Treatment – Antibiotic choice should be directed by culture results whenever possible [41]. Commonly isolated pathogens include S. aureus and P. aeruginosa [42]. However, other bacteria species, including Enterobacteriaceae, have also been isolated, suggesting that bacteremia occurs both through the skin as well as gut translocation [42]. In patients treated in burn centers and intensive care units, local nosocomial microbiota may additionally influence culture results [41]. (See "Clinical approach to Staphylococcus aureus bacteremia in adults" and "Principles of antimicrobial therapy of Pseudomonas aeruginosa infections" and "Gram-negative bacillary bacteremia in adults".)
Management of acute respiratory involvement — Up to 40 percent of patients with SJS/TEN may develop acute respiratory complications [44]. These include specific bronchial erosions as well as nonspecific pulmonary manifestations (eg, pneumonia, pulmonary edema, atelectasis). Twenty-five to 38 percent of patients may develop acute respiratory failure requiring ventilatory support [44-46]. Among intubated patients, the mortality rate increases up to over 50 percent [45]. Risk factors for mechanical ventilation include serum bicarbonate <20 mmol/L, serum urea >10 mmol/L, baseline detached BSA >10 percent, white blood cell count >12,000/mm3, hemoglobin <8 g/dL, bacteremia, and shock/organ failure on admission [45-47].
All patients need to be closely monitored for respiratory involvement during the acute phase and for referral to a specialized intensive care unit. Baseline evaluation should include chest radiographs and arterial blood gases. Features suggestive of pulmonary involvement include cough, dyspnea, hypoxemia, or radiologic abnormalities [44].
Bronchoscopy may be warranted to confirm bronchial involvement, evaluate for the presence of pulmonary infection, and prevent atelectasis and airway obstruction through the mechanical removal of sloughed bronchial epithelium [44].
Noninvasive ventilation is contraindicated because of the skin lesions and the risk of obstruction due to laryngeal involvement. (See "Noninvasive ventilation in adults with acute respiratory failure: Benefits and contraindications".)
Oral intubation may be challenging due to the oral/pharyngeal mucosal involvement. Patients needing mechanical ventilation should be referred to an intensive care unit [2].
Management of urogenital involvement — Urogenital involvement occurs in up to 70 percent of patients with SJS/TEN [48]. It may present acutely as erosions of the scrotum/labia and penis/vulva and may result in dysuria and urinary retention. Early urogenital/gynecologic examination should be part of a baseline assessment [49].
Nonadhesive dressings should be applied to urogenital erosions to reduce pain and prevent adhesions. Moderate-potency topical corticosteroids (group 4 (table 4)) can be applied daily until resolution of the acute phase of illness [49,50]. Estrogen cream and barrier creams (eg, petrolatum, zinc oxide) may also promote re-epithelization [49-51]. The use of viscous lidocaine may also reduce mucosal pain [51].
In female patients, additional local vulvar care includes gentle water washes or sitz baths. If vaginal involvement is suspected, an intravaginal potent corticosteroid ointment should be applied, and silicone vaginal dilators can be used to prevent strictures and adhesions [49]. These should be placed prophylactically during the acute phase of the illness and used regularly until complete healing of the erosions has occurred. In severe vaginal involvement, a vagina estrogen ring may be helpful, as it separates the anterior and posterior vaginal walls while providing low-dose estrogen to aid in epithelial healing. Menstrual suppression during the acute phase of illness in female patients of reproductive age is recommended to reduce the risk of vaginal adenosis and endometriosis [49,50]. To prevent periurethral stricture, patients with vulvovaginal involvement should receive a urinary catheter.
In uncircumcised male patients, daily retraction of the foreskin can prevent preputial adhesions [2].
Management of ocular involvement — Acute ocular involvement affects 70 to 90 percent of patients [52-54]. The severity of the acute ocular involvement does not correlate with the score of toxic epidermal necrolysis (SCORTEN) or BSA involved. The development of chronic ocular sequelae is unpredictable. In a French cohort of 159 patients, the severity of the acute ocular involvement was the strongest predictor for chronic ocular sequelae [52]. Thus, effective management of the acute ocular involvement may mitigate and prevent long-term ocular sequelae.
Evaluation — Repeated ophthalmic evaluation of all patients with suspected SJS/TEN should be performed even if ocular involvement is not visible at the onset of the disease, as it may follow cutaneous involvement. Examination should be performed daily until the ocular condition stabilizes [24].
The entire ocular surface should be carefully examined. Examination should include fluorescein staining to document the presence of membranes and loss of surface epithelium. The severity of ocular lesions can be graded using a simple score [55]:
●0 (none) – No ocular involvement
●1 (mild) – Conjunctival hyperemia
●2 (severe) – Either ocular surface epithelial defect or pseudomembrane formation
●3 (very severe) – Both ocular surface epithelial defect and pseudomembrane formation
Ophthalmic therapy — The principles of ophthalmic care are to reduce destructive inflammation at the ocular surface and lid margin; prevention of conjunctival adhesions; infection prophylaxis; and prompt identification and management of blinding complications of corneal exposure, ulceration, and infection [6].
●Saline rinses – Saline rinses can be used to clean the eyes and eyelids as well as to remove mucous and inflammatory debris. Local ocular therapy should be started based upon the severity of eye involvement [56].
●Lubricants – For patients with no apparent eye involvement (grade 0), multiple daily lubrication with preservative-free eye drops (artificial tears) or ointments is indicated.
●Topical corticosteroids/antibiotics – For patients with conjunctival hyperemia (grade 1), ophthalmic preparations containing topical corticosteroids and broad-spectrum antibiotics (eg, moxifloxacin hydrochloride 0.5%) should be applied four to six times per day along with lubricants.
●Adhesion separation – Conjunctival adhesions should be separated daily. Forceps, squint hooks, or scissors can be used.
●Amniotic membrane transplantation – In addition to topical antibiotics, corticosteroids, and lubricants, amniotic membrane transplantation (AMT) should be considered for patients with extensive sloughing of the bulbar conjunctiva and/or pseudomembrane formation (grades 2 and 3) [57]. Amniotic membrane sheets can be applied to the ocular surface without sutures using a symblepharon ring, a technique that avoids general anesthesia [58]. Cryopreserved amniotic membrane is commercially available in the United States and Canada. In the United Kingdom, it can be obtained from the National Health Service (NHS) Blood and Transplant tissue services.
The use of AMT for the treatment of ocular involvement is supported by several case series and one randomized trial [57,59-61]. In this trial including 25 patients (50 eyes), the eyes were randomized to receive AMT or standard medical therapy [59]. The main endpoints were the maintenance of the best corrected visual acuity (BCVA) and stable ocular surface (eg, absence of corneal haze, limbal stem cell deficiency, symblepharon, ankyloblepharon, or lid-related complications). After six months, the eyes treated with AMT had better BCVA, tear film break-up time, and Schirmer test results than eyes treated with medical therapy. Conjunctival hyperemia persisted in 4 percent of the eyes treated with AMT versus 44 percent of those treated with medical therapy. Corneal haze, corneal vascularization and conjunctivalization, and symblepharon occurred in 44, 24, and 16 percent, respectively, of the eyes treated with medical therapy but in none of the eyes treated with AMT.
Role of systemic therapy — There is scarce and conflicting evidence on the role of systemic adjunctive therapies in halting the ocular damage and improving the visual outcome.
●In a Korean, retrospective study, 43 patients with SJS/TEN were allocated into five treatment groups: systemic corticosteroids, intravenous immune globulin (IVIG), systemic steroids and immunoglobulins, pulse corticosteroids, and supportive care only. There were no benefits of systemic immunomodulatory treatments in ocular outcomes measured, such as BCVA and chronic ocular surface complications [62].
●In a retrospective, historically controlled study, 10 patients treated with IVIG (2 g/kg) were compared with 18 historical controls (5 received prednisolone). Outcomes of ocular complications and visual acuity six weeks post-IVIG were similar in both groups [63].
●In another study, five patients with SJS or TEN with ocular complications during the acute stage were treated within the first four days from disease onset with intravenous pulses of methylprednisolone 500 or 1000 mg per day for three to four days along with 0.1% betamethasone solution, 0.1% betamethasone eye ointment, or both five to eight times daily [64]. At one year, the BCVA was 20/20 or better in all eyes. Epithelial defect, loss of limbal stem cells, conjunctivalization, neovascularization, opacification, or keratinization were absent in all eyes. Only mild, superficial, punctate keratopathy was noted in five eyes.
●A retrospective study of 85 patients with SJS/TEN and severe ocular involvement found that corticosteroid pulse therapy given within four days of disease onset (n = 36) was more effective in preventing long-term ocular sequelae than corticosteroid pulse therapy given after five days of disease onset, corticosteroids given as nonpulse therapy, or no corticosteroids (n = 49) [65]. A BCVA in the worse eye of ≥1 was found in 52.8 percent of the eyes in the corticosteroid pulse therapy group compared with 14.3 percent in the other group. In addition, there was a reduction in chronic sequelae affecting the cornea, conjunctiva, and eyelid.
●In a small study, seven patients who received systemic cyclosporine at a median dose of 2.5 mg/kg/day over 12±8 days were compared with a control group of seven patients who did not receive cyclosporine. Visual outcomes of BCVA and a chronic ocular surface complication score were not significantly different between the eyes of patients who received cyclosporine versus those who did not receive cyclosporine [66].
Additional supportive measures — Similar to other patients in intensive care units, patients with SJS/TEN are at risk of additional complications, including stress-related peptic ulcer disease; disseminated intravascular coagulation, particularly in those with sepsis; venous thromboembolism in those who are immobile; hyper- or hypoglycemia; and neutropenia.
●Stress ulcer prophylaxis – Stress ulcer prophylaxis with proton pump inhibitor may be necessary in those who are unable to feed enterally [6]. (See "Stress ulcers in the intensive care unit: Diagnosis, management, and prevention".)
●Disseminated intravascular coagulation – In patients with disseminated intravascular coagulation and active bleeding, blood component replacement (eg, cryoprecipitate, fresh frozen plasma, platelets, red blood cells) may be necessary [67,68]. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults".)
●Prevention of deep vein thrombosis – Deep vein thrombosis prevention is encouraged with compression stockings, calf pumps, and anticoagulation, if there are no contraindications [6]. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)".)
●Hyperglycemia – Glycemic levels should be monitored frequently and managed according to intensive care unit principles. (See "Glycemic control in critically ill adult and pediatric patients".)
●Neutropenia – Neutropenia is known to occur during the acute phase of the disease. The use of granulocyte colony-stimulating factor (G-CSF) has been advocated by some, as it may mitigate the risk of infection as well as aid in wound re-epithelization [69]. (See "Management of the adult with non-chemotherapy-induced neutropenia".)
ADJUNCTIVE PHARMACOLOGIC THERAPIES
General considerations — There is no established pharmacologic treatment for SJS/TEN. A variety of systemic immunosuppressive or immunomodulating agents, such as systemic corticosteroids, intravenous immune globulin (IVIG) [70], cyclosporine [71], plasmapheresis [72], and anti-tumor necrosis factor (TNF) agents [73], have been utilized for the treatment of SJS/TEN.
The evidence of benefit for such interventions is limited, and none of them can be conclusively recommended. However, the results of several meta-analyses suggest a potential benefit for (see 'Evidence overview' below):
●Systemic corticosteroids
●Combination of IVIG and systemic corticosteroids
When considering the use of any of these agents, it is important to remember that the progression of SJS/TEN in the acute phase is self-limiting, lasting approximately eight days from the onset of symptoms to maximal skin detachment [3]. Thus, the initiation of any immunomodulatory agent outside this window is unlikely to modify disease progression.
Evidence overview — Only two randomized trials have evaluated treatments for SJS/TEN [73,74].
●A 1998 trial comparing thalidomide with supportive care was prematurely terminated due to excess death in the thalidomide group [74]. Ten of 12 patients who received thalidomide 400 mg daily for five days died compared with 3 of 10 patients in the placebo group. (See 'Harmful agents' below.)
●A 2018 randomized trial that included 93 patients compared etanercept 25 or 50 mg twice a week with prednisolone 1 to 1.5 mg/kg per day [73]. There was no statistically significant difference in mortality between the corticosteroid group and the etanercept group (16.3 versus 8.3 percent), although mortality in both groups was lower than that predicted by the score of toxic epidermal necrolysis (SCORTEN; 20.3 and 17.7 percent for the corticosteroid and etanercept groups, respectively). (See 'Tumor necrosis factor inhibitors' below.)
Evidence on the efficacy of immunomodulatory treatments is largely derived from observational studies and has been summarized in several meta-analyses [75-79].
●A 2017 meta-analysis of 27 studies (1209 patients) found that systemic corticosteroids and cyclosporine, in addition to supportive treatment, were associated with a decreased risk of death compared with supportive treatment alone (odds ratio [OR] 0.67, 95% CI 0.46-0.97 and OR 0.1, 95% CI 0.0-0.4, respectively). Treatment with IVIG did not improve survival [78].
●A 2021 meta-analysis of data from 38 studies published between 2000 and 2019 evaluating the efficacy of systemic corticosteroids, supportive care, IVIG plus systemic corticosteroids, cyclosporine, and etanercept in reducing mortality as predicted by SCORTEN found that only cyclosporine (standardized mortality rate [SMR] -0.88, 95% CI -1.47 to 0.29), IVIG plus corticosteroids (SMR -0.56, 95% CI -0.94 to -0.19), and etanercept (SMR -0.95, 95% CI -1.82 to -0.07) were found to have survival benefits [75].
●In a 2021 network meta-analysis of 67 studies (2079 patients), no treatment was superior to supportive care in reducing the mortality rate. However, when considering the mortality rates predicted by SCORTEN, combination therapy with IVIG and corticosteroids was the only treatment associated with survival benefits (SCORTEN-based SMR 0.53, 95% CI 0.31-0.93) [76].
●In another 2021 meta-analysis (24 studies, 979 patients), direct pairwise comparison showed that cyclosporine was associated with a lower mortality risk compared with supportive care (OR 0.32, 95% CI 0.13-0.82) and IVIG (OR 0.08, 95% CI 0.01-0.54). Systemic corticosteroids plus IVIG was also associated with a reduction in mortality compared with IVIG alone (OR 0.16, 95% CI 0.04-0.61). On network analysis, the risk of death was reduced for cyclosporine compared with supportive care (OR 0.19, 95% CI 0.05-0.59). Hierarchy ranking of treatments showed that cyclosporine was the most effective treatment, followed by corticosteroids plus IVIG and etanercept [77].
●A 2022 Cochrane review that included nine studies (308 patients) concluded that treatment with etanercept may result in a reduction in mortality compared with systemic corticosteroids, although evidence was of low certainty. Evidence of benefit was very uncertain for corticosteroids versus no corticosteroids, IVIG versus no IVIG, and cyclosporine versus IVIG [79].
It should be noted that the quality of the studies included in these meta-analyses was generally poor and that nearly all were retrospective cohort studies. The studies were also heterogeneous, with high variability in population, type of treatment and dosing, timing and duration of treatment, and site of care.
Most studies utilized SCORTEN as a comparator or internal control to determine the observed/expected death ratio in the patient cohort. However, it has been shown that SCORTEN may overestimate the expected mortality, resulting in a positively biased estimate of the efficacy of any intervention [15]. In addition, the "center effect" (ie, better outcomes seen in reference centers/specialized centers due to improvements in supportive care and treatment protocols) cannot be excluded in many studies [20].
Systemic corticosteroids — Varying regimens of systemic corticosteroids have been used for the treatment of SJS/TEN. These include prednisolone 60 to 250 mg/day for 2 to 12 days, intravenous dexamethasone 1.5 mg/kg/day for three days, and intravenous methylprednisolone 250 to 1000 mg/day for three days [80].
However, the efficacy of systemic corticosteroids has not been proven. Thus, routine use of systemic corticosteroids cannot be recommended.
●Early, observational studies indicated a higher frequency of complications and increased mortality for patients with TEN treated with corticosteroids in burn units [81,82].
●An initial analysis of French and German patients recruited in the EuroSCAR study (n = 281) comparing 159 patients who received variable doses of corticosteroids (ranging from 60 to 250 mg per day of prednisone equivalents), 75 patients who received IVIG (40 in association with corticosteroids), and 87 patients who received supportive care alone showed that the mortality rate was 18 percent in the corticosteroid group, 25 percent in the IVIG group, and 25 percent in the supportive care group [83]. The OR of death for patients treated with corticosteroids compared with patients treated with supportive care alone was 0.6 (95% CI 0.3-1.0), suggesting a potential benefit.
●However, in a subsequent, updated review of patients recruited in the RegiSCAR study for whom treatment details were available (n = 442), there was no observed benefit or harm with corticosteroids (hazard ratio [HR] 1.3, 95% CI 0.8-1.9) [4].
●A 2017 meta-analysis using individual data from 1209 patients, of whom 367 received systemic corticosteroids in addition to supportive treatment, found that corticosteroid treatment was associated with a modest decreased risk of death compared with supportive treatment alone (OR 0.67, 95% CI 0.46-0.97) [78].
●A 2018 randomized trial compared the use of etanercept 25 or 50 mg two times per week (48 patients) versus intravenous prednisolone 1 to 1.5 mg/kg/day (43 patients). The mortality rate in the etanercept group was lower than that in the corticosteroid group (8.3 versus 16.3 percent, respectively), although the difference was not statistically significant [73]. The number of observed deaths in both the corticosteroid and etanercept arms was lower than that predicted by SCORTEN.
●In a systematic review of 31 pediatric case series with 128 patients, no deaths were reported among the 20 patients who received either prednisolone/prednisone (1 mg/kg/day) or methylprednisolone (4 mg/kg/day) for five to seven days [84]. Complications occurred in five patients (mild skin infections in three children and bronchiolitis in two children).
●In a retrospective study including 898 pediatric patients hospitalized with a primary diagnosis of SJS or TEN and treated with glucocorticoids only (18.6 percent), IVIG only (25.5 percent), IVIG plus glucocorticoids (17 percent), or supportive care only, there was no association between the use of systemic steroids, IVIG, or IVIG and steroids during the first two days of hospitalization and decreased length of stay or need for mechanical ventilation [85]. The overall hospital mortality was 0.56 percent (0.13 percent for SJS and 3.2 percent for TEN).
Cyclosporine — There is increasing evidence from multiple case series and meta-analyses suggesting that the cyclosporine administered during the acute phase may be associated with reduced mortality [71,76,78,86-90]. The use of cyclosporine early in the course of the disease has consequently been advocated by some experts [91]. We suggest using cyclosporine 3 to 5 mg/kg/day as adjunctive therapy within 24 to 48 hours of symptom onset.
The mechanism of action of cyclosporine in SJS/TEN involves the inhibition of T cell activation, therefore preventing the production and release of cytokines, which play a critical role in the pathogenesis and propagation of SJS/TEN by cytotoxic T cells and natural killer cells. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis", section on 'Pathogenesis'.)
●An open-label trial conducted at a French center for SJS/TEN showed that the use of cyclosporine at 3 mg/kg/day tapered over a month may be associated with survival benefits. Utilizing SCORTEN, four deaths were predicted, although none were observed [71].
●In a study of 71 patients in Spain with SJS/TEN, of whom 49 were treated with cyclosporine and 22 were treated with other therapies, the observed mortality rates were 10 and 32 percent, respectively [89]. The expected mortality rates according to the SCORTEN at admission were 24 percent in the cyclosporine group (observed-to-expected mortality rate ratio [MRR] 0.42, 95% CI 0.14-0.99) and 29 percent in the other therapies group (observed-to-expected MRR 1.09, 95% CI 0.44-2.25).
●Similar findings have been demonstrated in Canada, Singapore, and India [88,90,92].
●In a 2017 meta-analysis including five observational studies with a total of 134 patients with SJS/TEN who were treated with cyclosporine, the observed-to-expected MRR was 0.41 (95% CI 0.21-0.80) [89].
●A 2018 meta-analysis of nine observational studies including 255 patients with TEN found that treatment with cyclosporine was associated with an approximately 70 percent reduction in mortality risk (pooled SMR 0.32, 95% CI 0.12-0.52) compared with supportive care and other treatments, including cyclophosphamide, corticosteroids, IVIG, and plasmapheresis [93].
●However, in a French retrospective cohort of 174 patients with SJS/TEN hospitalized between 2005 and 2016, whereby 95 patients received cyclosporine and 79 patients received supportive care, there was no difference in primary outcome (progression of skin detachment on day 5) and secondary outcomes (proportion of patients with cutaneous re-epithelialization starting on day 5, overall in-hospital mortality) when analyzed via traditional exposed/unexposed or propensity-matched methodology [94].
Intravenous immune globulin — Intravenous immune globulin (IVIG; given at a dose of 2 to 4 g/kg over two to five days) was initially proposed as a treatment for SJS/TEN based on the hypothesis that Fas ligand (FasL) was the main mediator of widespread keratinocyte apoptosis in TEN and on the finding that high-dose IVIG was able to antagonize FasL effects [95]. However, it is now widely accepted that granulysin, a cytotoxic protein found in cytotoxic T cells, is the most important mediator [96]. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis".)
There is little evidence:
●A 2012 systematic review and meta-analysis of 17 observational studies including 113 patients treated with IVIG and 130 patients treated with supportive care only found no difference in the risk of death between the two groups (OR 1.00, 95% CI 0.58-1.75) [97]. However, a subgroup analysis showed a statistically nonsignificant reduction in the mortality risk for patients treated with high-dose IVIG (total dose ≥2 g/kg) compared with those treated with <2 g/kg (OR 0.49, 95% CI 0.11-2.30).
●In a subsequent, single-institution study of 64 cases of TEN and SJS/TEN overlap (mean SCORTEN 2.6) treated with IVIG, the overall mortality rate was 31 percent (SCORTEN-predicted mortality 28 percent, SMR 1.1, 95% CI 0.6-1.6) [98]. Mortality was similar among patients treated with <3 g/kg IVIG and those treated with ≥3 g/kg IVIG (31 and 26 percent, respectively).
●A 2015 meta-analysis of 13 studies of patients who received IVIG for SJS or TEN, the severity of which was determined using SCORTEN, found an overall SMR of 0.81 (95% CI 0.62-1.08), with a nonsignificant difference in SMR between patients treated with IVIG and those who did not receive IVIG (SMR difference 0.32, 95% CI -0.77 to 0.12) [99]. However, a meta-regression performed to determine the effect of IVIG doses on SMR found significant reductions in SMR with increasing IVIG doses (≥2 g/kg).
●A 2017 meta-analysis using individual data from 1209 patients, of whom 215 received IVIG in addition to supportive treatment, found that IVIG treatment was not associated with a decreased risk of death compared with supportive treatment alone (OR 0.99, 95% CI 0.64-1.54) [78].
Adverse effects of IVIG include renal, hematologic, and thrombotic complications. The risk for serious complications is increased in patients receiving high-dose IVIG, in older patients, and in those with pre-existing renal or cardiovascular disorders. (See "Intravenous immune globulin: Adverse effects".)
Intravenous immune globulin plus systemic corticosteroids — Systemic corticosteroids and intravenous immune globulin (IVIG) have been administered in combination to patients with SJS/TEN, but the data remain limited.
●In a United States, retrospective, multicenter study including 377 hospitalized patients with SJS/TEN, the SMR among patients receiving both systemic corticosteroids (mean daily dose 148 mg prednisone equivalent) and IVIG (mean dose 1 g/kg per day for three days) was lower (SMR 0.52, 95% CI 0.21-0.79) than among patients receiving corticosteroids only (SMR 0.72, 95% CI 0.48-0.89), IVIG only (SMR 0.79, 95% CI 0.55-0.92), or supportive care alone (SMR 0.70, 95% CI 0.47-0.87) [100].
●In an observational, propensity-matched, retrospective study from China, 93 patients who received combination therapy were compared with 52 patients on corticosteroid monotherapy [101]. There was no difference in the mortality rate between the two groups. However, compared with corticosteroid monotherapy, combination therapy was associated with a shorter mean hospitalization time (-3.37 days) and a lower rate of skin infections.
●In a prospective, nonrandomized study, 36 patients with TEN received low-dose IVIG (0.2 to 0.5 g/kg) plus intravenous dexamethasone (0.1 to 0.3 mg/kg per day tapered in one to two weeks) or dexamethasone alone [102]. One death occurred in the combined therapy group, and three deaths occurred in the corticosteroid-only group. The SCORTEN-predicted deaths were five in both groups.
●In the EuroSCAR cohort study, 35 patients were treated with IVIG alone, and 40 patients were treated with IVIG plus systemic corticosteroids [83]. The dose of IVIG ranged from 0.7 to 2.3 g/kg and was given in one to seven days. The mortality rate was 34 percent in the group treated with IVIG alone and 18 percent in the group treated with IVIG and corticosteroids.
●In a retrospective study of 55 patients with TEN, 39 patients were treated with IVIG 0.4 g/kg per day for five days plus methylprednisolone 1.5 mg/kg per day for three to five days, and 22 patients were treated with methylprednisolone alone [103]. The mortality rate was 13 percent among patients treated with combination therapy (5 of 39) and 23 percent among those treated with corticosteroids alone (5 of 22).
●A retrospective study compared 65 patients over a 14-year period. Forty-five patients were treated with corticosteroids (methylprednisolone 1 to 1.5 mg/kg/day), and 20 patients were treated with IVIG (2 g/kg over five days) and corticosteroids (methylprednisolone 1 to 1.5 mg/kg/day). There was no statistical difference in mortality between the two groups. The SMR was 1.16 (95% CI 0.56-2.13) in the corticosteroid group and 0.85 (95% CI 0.18-2.5) in the combination group [104].
●Another similar, retrospective, Chinese study compared 82 patients who were treated with corticosteroids (n = 58) or IVIG and corticosteroids (n = 24). The SMR was 0.48 (95% CI 0.08-1.92) in the corticosteroids group and 0.57 (95% CI 0.32-1.91) in the combination group. There was no significant difference in SMR between the two groups [105].
Tumor necrosis factor inhibitors — In a few case reports, a single infusion of 5 mg/kg of the tumor necrosis factor (TNF)-alpha inhibitor infliximab halted the progression of skin detachment and induced a rapid re-epithelization of the denuded skin [106-109]. Etanercept, given in a single 50 mg subcutaneous injection, was used successfully in a small number of patients [110-112].
The efficacy of etanercept for the treatment of SJS/TEN was subsequently evaluated in a randomized, unblinded trial including 91 patients with a diagnosis of SJS/TEN [73]. Patients received etanercept 25 or 50 mg subcutaneously twice weekly or intravenous prednisolone 1 to 1.5 mg/kg per day. Seventy-one patients completed the study and were included in the analysis. The mortality rates were 8.3 and 16.3 percent in the etanercept and glucocorticoid groups, respectively. Although the difference was not statistically significant, the observed mortality rates in both groups were lower than those predicted by SCORTEN at admission (17.7 and 20.3 percent for the etanercept and steroid groups, respectively). The median time to complete re-epithelization in patients with ≥10 percent body surface area (BSA) detachment was shorter in the etanercept group than in the corticosteroid group (14 versus 19 days, respectively). Serious adverse events (including sepsis, respiratory failure, and upper gastrointestinal bleeding) occurred in 13 percent of patients in the etanercept group and in 27 percent of patients in the corticosteroid group but, except for gastrointestinal bleeding, were deemed to be unrelated to treatments.
Although this study adds further evidence to support the use of etanercept for SJS/TEN, additional studies are needed to confirm these findings and determine the optimal dose and duration of treatment.
Tumor necrosis factor inhibitors plus corticosteroids — Following initial experience with anti-tumor necrosis factor (TNF) inhibitors, there have been two further studies evaluating combinations of etanercept and systemic corticosteroids in SJS/TEN.
●In a retrospective study of 242 patients with SJS/TEN from Taiwan and China, 196 patients (81 percent) received methylprednisolone monotherapy (average dose of 2 mg/kg/day), 25 patients received etanercept at a dose of 25 or 50 mg twice a week combined with systemic corticosteroids (1.2 mg/kg/day), and 21 patients were treated with IVIG and systemic corticosteroids (1.5 mg/kg/day) [113]. At baseline, those treated with combination therapy had higher mean SCORTEN values, with predicted mortality rates based on SCORTEN of 9.3, 16, and 14.4 percent in the methylprednisolone, methylprednisolone plus IVIG, and methylprednisolone plus etanercept groups, respectively. The observed mortality rate was 6.6 percent (SMR 0.7, 95% CI -0.8-2.6) in the methylprednisolone group, 4.8 percent (SMR 0.3, 95% CI -0.7 to 6.2) in the methylprednisolone plus IVIG group, and 0 percent (SMR 0, 95% CI -1.8 to 3.59) in the methylprednisolone plus etanercept group. Combination of methylprednisolone and etanercept also showed a reduced mean skin healing time of 12 days versus 13 days (methylprednisolone) and 13.5 days (methylprednisolone plus IVIG).
●In a cohort of 25 patients with SJS/TEN (17 with SJS, 1 with SJS/TEN, and 7 with TEN), 10 patients received methylprednisolone, and 15 patients received etanercept plus methylprednisolone [114]. At baseline, there were no significant differences in terms of age, SCORTEN, and disease classification. There were no deaths in both groups. The combination therapy shortened skin healing time (median 12 days) compared with the corticosteroid monotherapy group (median 16 days).
Plasmapheresis — The use of plasmapheresis alone or in combination with other treatments has been reported in several small case series of TEN with variable results [115-118]. Removal of a toxin, such as a drug, drug metabolite, or other cytotoxic mediator, has been proposed as a mechanism of action. (See "Therapeutic apheresis (plasma exchange or cytapheresis): Indications and technology".)
In a Japanese retrospective study comparing the outcomes of patients with SJS/TEN who did not respond to pulse corticosteroid therapy and were subsequently treated with plasmapheresis (n = 53) or IVIG (n = 213), the mortality rates were similar in the two groups (18.3 and 19.5 percent, respectively) [119]. However, the length of hospital stays and costs were greater for patients undergoing plasmapheresis.
Harmful agents — Thalidomide, a potent inhibitor of TNF-alpha, should not be used for the treatment of SJS/TEN. A randomized, placebo-controlled trial of thalidomide in patients with TEN was prematurely stopped because of excess mortality among patients given thalidomide [74].
PROGNOSIS
Mortality — The overall mortality rate among patients with SJS and TEN is approximately 25 percent, ranging from 10 percent for SJS to >30 percent for TEN [4]. In the United States, the mean adjusted mortality rates between 2009 and 2012 were 4.8 percent for SJS, 19.4 percent for SJS/TEN, and 14.8 percent for TEN [120]. Mortality is considerably lower in children, ranging from 0 to 9.5 percent among centers [121-123], although a very high mortality rate (39 percent) has been reported in infants with TEN [124].
Sepsis, acute respiratory distress syndrome, disseminated intravascular coagulation, and multiple organ failure are the most common causes of in-hospital death [40,41,45,67,125]. Patients who develop acute renal failure requiring renal replacement therapy also have an increased mortality risk [126]. Older age (>70 years) and presence of comorbidities (eg, liver cirrhosis, metastatic cancer) are also associated with an increased risk of death [127].
Disease severity (ie, the extent of cutaneous involvement) is the main risk factor for mortality within 90 days of disease onset [4]. However, the risk of death remains high beyond 90 days and up to one year after the onset of the reaction and is mainly influenced by the patient's age and presence of comorbidities. Among 460 patients with SJS/TEN included in the European RegiSCAR study between 2003 and 2007, death rates of 23 and 34 percent were reported at six weeks and one year after discharge, respectively [4].
Recurrence — The likelihood of recurrence of SJS/TEN is unknown. Recurrence is generally rare in adults unless there is re-exposure to the same drug/structurally similar drug [128,129]. In two studies based on the International Classification of Diseases (ICD) coding and hospital records, recurrence of up to 18 percent occurred in children [130], and recurrence of up to 7 percent has been reported in adults (16 recurrent episodes per 1000 person-years) [131]. However, in a cohort of 106 validated cases of SJS/TEN followed up in a reference center, there were two recurrences (one from Mycoplasma infection and another due to re-exposure to the causative drug; 4.2 recurrent episodes per 1000 person-years) [128]. These differences in recurrence rates may be due to the methodology of utilizing electronic disease coding (eg, ICD coding), which may allow for the inclusion of cases of erythema multiforme and bullous fixed drug eruptions, which recur more frequently.
In children, up to 50 percent of SJS/TEN cases do not have a drug trigger and are believed to be due to infections (eg, Mycoplasma) or idiopathic [130]. The high recurrence in children may be due to the higher incidence of infections as a trigger.
Long-term sequelae — Long-term sequelae include cutaneous, mucosal, ocular, visceral, and psychologic complications, which are increasingly reported as survival improves [5,132,133].
Skin, hair, and nails — Cutaneous sequelae are common, affecting up to 80 percent of survivors, and include postinflammatory hypo- or hyperpigmentation, scarring, eruptive nevi, telogen effluvium, and chronic pruritus [5,133]. Scarring may occur when superinfection complicates re-epithelialization or as a consequence of surgical debridement of necrotic skin or skin grafting. Nail changes occur in up to 70 percent of survivors and include Beau lines and onychomadesis [134]. (See "Overview of nail disorders", section on 'Transverse grooves (Beau lines)' and "Overview of nail disorders", section on 'Onychomadesis'.)
Eyes — Ocular sequelae develop in approximately 50 to 90 percent of patients with acute ocular involvement [130]. These sequelae include dry eye, photophobia, ingrown eyelashes (trichiasis), neovascularization of the cornea, lid margin changes (eg, posterior migration of the mucocutaneous junction and keratinization), keratitis, symblepharon, and corneal scarring leading to visual impairment and, rarely, blindness [52-56,135].
Ocular sequelae are mainly due to functional alteration of the conjunctival epithelium, decreased lacrimal film, ocular surface failure due to loss of corneal stem cells and recurrent episodic inflammation, scleritis, and progressive conjunctival cicatrization [6].
The onset of ocular involvement may be delayed. In a study of 54 patients with SJS or TEN, only 11 percent of patients had severe ocular involvement detected during the acute phase, but nearly 90 percent developed chronic ocular sequelae six months or later after the initial presentation [136]. Late complications were severe in 17 percent of the patients, including trichiasis, mucocutaneous junction abnormalities, and extensive corneal opacities.
Management of long-term ocular sequelae involves the use of artificial tears and scleral lenses [137]. Systemic immunosuppression has been used to suppress conjunctival inflammation with variable results. Surgical interventions to restore ocular surface include limbal stem cell transplantation, amniotic membrane transplantation (AMT), cultivated oral mucosal epithelial transplantation, and corneal transplant.
Oral and dental — Oral and dental sequelae have been observed in survivors of SJS/TEN. These include mouth discomfort, xerostomia, gingival inflammation and synechiae, caries, and periodontal disease [138]. Severe dental growth abnormalities (eg, dental agenesia, root dysmorphia, root-building abortion, microdontia) may occur in survivors who suffered from SJS/TEN in childhood [138].
Urogenital — Long-term consequences include adhesion formation, scarring, urethral stenosis, and strictures. In females, specific urogenital sequelae affect up to 28 percent of survivors. These include introital stenosis, vaginal dryness, dyspareunia, urinary retention, hematocolpos, and vaginal adenosis [48-51,139-141]. In males, strictures and adhesions can lead to phimosis.
Pulmonary — Long-term pulmonary complications include chronic bronchitis/bronchiolitis with obstructive changes (including bronchiolitis obliterans and organizing pneumonia [formerly called bronchiolitis obliterans organizing pneumonia]), bronchiectasis, and obstructive disorders [5,142,143]. Pulmonary sequelae may also occur in asymptomatic patients. A series of 32 patients with SJS/TEN underwent pulmonary function tests (PFTs) at a median time of three months after the acute episode. Although only two patients had dyspnea, abnormal PFTs were demonstrated in more than one-half of the cases [144]. The most frequent PFT abnormality was a reduced carbon monoxide (CO) diffusion, which persisted at 12 months in 8 of 10 patients. (See "Overview of bronchiolar disorders in adults" and "Clinical manifestations and diagnosis of bronchiectasis in adults".)
Chronic pain — Chronic pain at mucocutaneous sites of involvement has been reported in approximately one-third of patients [145].
Psychiatric — Long-term psychosocial complications and psychiatric disorders (eg, anxiety, depression) and decreased health-related quality of life have been reported in survivors of SJS/TEN [133,146,147]. Post-traumatic stress disorder affects 20 percent of survivors six months after the initial episode. Depression and anxiety may affect up to 30 to 50 percent and 40 percent of survivors, respectively. More than 60 percent of respondents were fearful of taking newly prescribed medications, and 30 percent of patients have avoided taking prescribed medications for their diagnosed medical conditions. Close to 30 percent of survivors were unable to work following their acute episode [133].
FUTURE DRUG AVOIDANCE
Patient education — Patients who survive SJS/TEN attributed to a medication must be educated about future avoidance and understand that re-exposure to the culprit drug may be fatal. Relevant information should be inscribed on a medical information bracelet or necklace and worn at all times. Electronic medical records, "allergy cards," and alerts should also be updated accordingly. Patients should learn all the various names for the causative medication, be able to report that they have a history of SJS or TEN, and be able to recognize closely related medications.
Future use of related medications — Recurrent SJS/TEN with readministration of the causative drug or closely, chemically related agents has been widely documented in multiple case reports [128,148,149]. The risk of recurrent SJS/TEN with structurally distinct agents (within the same therapeutic class of drugs) is unknown but probably very low. As an example, patients with past SJS/TEN due to sulfamethoxazole should avoid other anti-infectious sulfonamides (eg, sulfadiazine, sulfapyridine) but can use thiazide diuretics or sulfonylurea-derived oral antidiabetics [150].
Patients with SJS/TEN caused by "aromatic" anticonvulsants (eg, phenytoin, carbamazepine, phenobarbital) should be informed that other agents in the same class may cause similar reactions in both the patient and their family members (table 1) [151,152]. Although there are no data about the safety of nonaromatic anticonvulsants (eg, valproate, succinimides, benzodiazepines, gabapentin) for patients with a history of SJS/TEN induced by aromatic anticonvulsants, the risk of SJS/TEN recurrence after exposure to a different class of anticonvulsants is probably extremely low and certainly lower than the risk related to the absence of antiepileptic treatment [10]. The pharmacology of antiepileptic agents is reviewed in detail elsewhere. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)
Because human leukocyte antigen groups are associated with carbamazepine-, phenytoin-, and allopurinol-induced SJS/TEN, family members of patients should be alerted against the use of the same medication. (See "Stevens-Johnson syndrome and toxic epidermal necrolysis: Pathogenesis, clinical manifestations, and diagnosis".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Drug allergy and hypersensitivity".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Stevens-Johnson syndrome and toxic epidermal necrolysis (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Initial in-hospital evaluation – Patients suspected to have Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) require immediate in-hospital evaluation.
•Identification and withdrawal of the offending drug – Prompt withdrawal of the offending drug (table 1) may improve the prognosis. The algorithm of drug causality for epidermal necrolysis (ALDEN (table 2)) can help identify the offending agent. (See 'Prompt identification and withdrawal of the causative drug' above.)
•Assessment of severity and prognosis – The severity and prognosis of the disease should be rapidly determined by using the score of toxic epidermal necrolysis (SCORTEN) to define the appropriate medical setting for management. (See 'Assessment of the extent of skin detachment' above and 'Assessment of prognosis' above.)
•Transfer to a referral center – Patients with body surface area (BSA) detachment of ≥10 percent, rapidly progressive disease, or worsening biochemical/organ function should be transferred and managed in specialized centers, such as specialized dermatology units, burn centers, or intensive care units. (See 'Transfer to a referral center' above.)
●Supportive care – Supportive care is the mainstay of treatment and includes wound care, fluid and electrolyte management, nutritional support, temperature management, pain control, prevention and treatment of infection, ocular care, and organ support if needed. (See 'Supportive care' above.)
•Wound care – The optimal approach to wound care has not been determined. Success has been reported with both repeated debridement of exfoliating skin and "antishear" wound care, in which the necrotic skin is left in place to act as a biologic dressing. (See 'Wound care' above.)
•Ocular care – Ocular lesions require immediate attention and care. Specific interventions aimed at preventing long-term ocular sequelae in severe cases with extensive sloughing of the bulbar conjunctiva include the use of preservative-free corticosteroid eye drops, coverage of the eye surface with cryopreserved amniotic membrane, and use of scleral spacers. (See 'Management of ocular involvement' above.)
•Infection prevention – Because sepsis is the major cause of death, key infection control measures include sterile handling, topical antiseptic agents, and surveillance cultures of possible sites of superinfection. Prophylactic systemic antibiotics are not utilized by most burn centers, although antimicrobials should be administered at the first sign of infection, and choice of agent should be guided by specific culture data. (See 'Prevention and treatment of infections' above.)
•Monitoring of respiratory involvement – All patients need close monitoring for respiratory involvement and referral to an intensive care unit if they develop features suggestive of pulmonary involvement (eg, cough, dyspnea, hypoxemia, radiologic abnormalities). (See 'Management of acute respiratory involvement' above.)
●Adjunctive pharmacologic therapies
•Beyond supportive care, we suggest cyclosporine rather than other systemic adjunctive therapies for patients with SJS/TEN. Oral cyclosporine at the dose of 3 to 5 mg/kg/day should be given early in the course of the disease (ie, within 24 to 48 hours of symptom onset) (Grade 2C). For patients in whom the use of cyclosporine is contraindicated, a single dose of the tumor necrosis factor (TNF) inhibitor etanercept is an alternative option. (See 'Cyclosporine' above and 'Tumor necrosis factor inhibitors' above.)
•Although frequently used, the role of systemic corticosteroids in the management of SJS/TEN remains uncertain, and treatment modality (eg, oral versus intravenous "pulse" administration), dose, timing, and duration have not been determined. Thus, routine use of systemic corticosteroids cannot be recommended. (See 'Systemic corticosteroids' above.)
•We suggest not using intravenous immune globulin (IVIG) for SJS/TEN as monotherapy (Grade 2C). In several meta-analyses, no clear survival advantage has been found for patients with SJS/TEN treated with IVIG. The role of IVIG in combination with systemic steroids needs to be further investigated. (See 'Intravenous immune globulin' above.)
●Prognosis – The overall mortality rate among patients with SJS/TEN is approximately 25 percent, ranging from approximately 10 percent for SJS to >30 percent for TEN. Sepsis, acute respiratory distress syndrome, and multiple organ failure are the most common causes of in-hospital death. Long-term sequelae involving the skin and eyes are common among survivors. (See 'Mortality' above and 'Long-term sequelae' above.)
●Future drug avoidance – Patients who survive SJS/TEN attributed to a medication must be educated about future avoidance of the causative drug and other chemically related agents and understand that re-exposure may be fatal. Relevant information should be inscribed on a medical information bracelet or necklace and worn at all times. Electronic medical records, "allergy cards," and alerts should also be updated accordingly.
Do you want to add Medilib to your home screen?