Perinatal/Neonatal Case Presentation

Toxic epidermal necrolysis in a neonate receiving fluconazole


Toxic epidermal necrolysis (TEN) is a rare, life-threatening mucocutaneous condition, which may occur as an adverse reaction to a number of medications. The anitifungal agent, fluconazole, has been associated with TEN in limited reports, mainly in adults with HIV infection. We describe the case of a neonate with liver disease who developed TEN, presumably induced by fluconazole.


Toxic epidermal necrolysis (TEN) may be seen as a severe adverse reaction to various drugs, and carries substantial rates of morbidity and mortality.1 The pathophysiology of severe adverse cutaneous drug reactions remains unknown, but may be related to immunologic or metabolic factors.1,2 Here we report the case of a newborn with liver disease who developed extensive skin and oropharynx sloughing while being treated with intravenous fluconazole for a Candida alibcans urinary tract infection; post-mortem pathology was consistent with TEN and confirmed the suspected primary diagnosis of neonatal hemochromatosis. Prior published cases of TEN associated with fluconazole, and risks for toxicity of fluconazole use in pediatrics, are reviewed.


The patient was a 1580 g female born at 35 5/7 weeks gestation to a 24-year-old mother. Parents were consanguineous and had lost three prior pregnancies to miscarriage. Mother had known protein S deficiency and was treated with low-molecular-weight heparin injections throughout her pregnancy, but otherwise her prenatal course and infectious disease-related laboratory reports were unremarkable. A fetal ultrasound had revealed echogenic bowel and a relatively large, hyperdynamic heart.

The infant was delivered vaginally and while initially vigorous, she became dusky and hypoxic soon after birth, and required oxygen support and admission to the intensive care nursery. During the first 4 days of life (DOL), she developed progressive abdominal distension, bilious gastrointestinal output, increasing respiratory distress and ultimately hypotension requiring vasopressor support. She was found to have various hematologic and coagulation abnormalities, including marked thrombocytopenia, international normalized ratio of 3.8, and direct and total bilirubin of 12.7 and 16.8 mg per 100 ml. Early in her course, she was diagnosed with a methicillin-sensitive Staphylococcus aureus catheter-associated bloodstream infection and was treated with oxacillin from DOL 7 to 24.

Abdominal magnetic resonance imaging revealed multiple non-enhancing nodules, suggestive of nodular sclerosis and cirrhosis, of the liver. Neonatal hemochromatosis was suspected, but a lip mucosal biopsy lacked iron deposition and was not diagnostic. As presumptive treatment (for hemochromatosis), she received 1 g per kg of intraveneous immunogloubulin on days 10, 12 and 15 of life; she also received regular deferoxamine, vitamin E and N-acetylcysteine. Regardless, patient had persistent direct hyperbilirubinemia and abnormal coagulation parameters, without significant transaminitis, throughout her hospitalization.

In addition to frequent platelet and fresh frozen plasma transfusions, patient required ongoing intensive support for other multi-system issues, including high-frequency oscillatory ventilation, and electrolyte management for hyponatremia and glycemic instability. She had severe anasarca and received intermittent steroids and diuretics, dopamine for renal support, and she had an indwelling urinary catheter for fluid management. On DOL 17, the patient was found to have a Candida albicans/dubliensis urinary tract infection and began treatment with fluconazole, first with a loading dose, and then at 10 g per kg intravenously once daily. Persistent funguria (>100 colonies per ml) was detected on repeat urine culture on DOL 22.

In the context of diffuse body and facial edema, and presence of a preceding occipital subgaleal hematoma, patient developed blisters and bullae on her lower abdomen and perineal area on DOL 22. These lesions spontaneously ruptured and drained clear fluid, and cultures grew mixed bacterial species and yeast, consistent with skin colonization. On DOL 26, patient was less edematous and overall showing improvement: vasopressor support was discontinued, and she was transitioned to a conventional ventilator. The previously involved lower abdominal-genital skin was now dry, and she had new, small bullae on her flank; the general impression was that the blisters/bullae were related to skin tension from edema, and friction from contact and manipulation.

On DOL 28, however, patient abruptly developed more wide-spread skin lesions, mainly bullae on her trunk and scalp. Over the course of the day, patient’s blistering coalesced and became more extensive, and a scalp lesion had malodorous, green discharge; she also began to have copious, mucosy stools and pre-renal azotemia with complicated fluid management. Concern for a severe drug reaction prompted discontinuation of fluconazole and deferoxamine; vancomycin and cefepime were started for possible sepsis. By the next day, her skin was found to be sloughing off in sheets, and when she needed to be re-intubated, sloughing of her oral and laryngeal mucosa was also noted. Patient experienced further progressive multi-organ failure, and ultimately became asystolic after a bradycardia-desaturation episode, and died on DOL 30.

Wound, endotracheal-tube and blood cultures obtained in the last 2 days of life grew Acinetobacter lwoffi. At post-mortem analysis, greater than 90% of both the body surface area’s skin and oral mucosa were sloughed. Dermatopathology demonstrated separation of the epidemis, and fibrin and eosinophil deposition in the skin, all consistent with TEN. In addition to iron deposition in the liver parenchyma, extra-hepatic siderosis was found in the thyroid and pancreas, confirming neonatal hemochromatosis.


Steven–Johnson Syndrome (SJS) and TEN are serious conditions involving mucocutaneous sloughing. Generally felt to be on a continuum of severity, TEN classifies cases in which greater than 30% of the skin has undergone epidermal detachment; erosive mucosal lesions are routine, with oral, ocular and genital involvement all are common.1 Disruption of skin and mucosal integrity leads to fluid loss, disrupted thermal regulation and secondary infectious complications. Mortality rates for TEN have historically been 30–40%, with sepsis being the major cause.1

Most cases (approximately 80–95%) of TEN are strongly associated with exposure to specific medications.1 Although anticonvulsants and sulfa drugs have been most reported, over 100 different agents, including several antimicrobials, have been implicated.1 Patient risk factors for development of TEN are not well-understood, but appear to be related to differences in drug metabolism, detoxification and/or immune reactivity.1, 2, 3, 4, 5 Patients with HIV infection have increased rates of many drug-induced skin rashes, including SJS and TEN, with a combined incidence of 1 per 1000 person-years. At the epidermal level, a cell-mediated inflammatory cytotoxic reaction against keratinocytes, which process and may concentrate drug metabolites, has been postulated to underlie the pathophysiology of TEN.4

At the time of writing, there have been three reported cases of highly probable TEN, and eight of SJS, associated with fluconazole in the medical literature.2,3,5, 6, 7, 8 Two of the TEN cases were of young, HIV-infected men, being treated with fluconazole for oral candidiasis, who developed characteristic mucocutaneous findings within 1 week of starting the drug.3,5 The third report was of a 50-year old woman with multiple medical problems and suspected immunocompromise, receiving fluconazole for esophageal candidiasis.2 In the reported cases of fluconazole-induced SJS, nearly all (seven of eight) patients were HIV negative, and the majority were receiving low-dose, oral fluconazole, for either vaginal candidiasis or antifungal prophylaxis.7, 8, 9

Establishing that an adverse reaction was caused by a specific medication is often challenging; nevertheless, we believe that our patient’s TEN was most likely induced by fluconazole. Other drugs she was receiving at the same time were all started well before fluconazole (Figure 1), and none have been associated with severe cutaneous reactions. For example, although aminopenicillins and other β-lactam antibiotics have been linked to SJS and TEN in adults and children, oxacillin has not.1,9

Figure 1

Medications received before mucocutaneous findings. IVIG, intraveneous immunogloubulin.

The emergence of blisters at about 6 days after drug initiation, with first steady and then rapid mucocutaneous involvement occurring over the following days, is consistent with previously described cases of fluconazole-associated TEN.2,3,5 Also, intraveneous immunogloubulin—which patient received for suspected hemochromatosis through the day before starting fluconazole—may ameliorate TEN,5 and potentially could have modified the course of mucocutaneous findings in this case.

We suspect that primary liver disease contributed to our patient’s risk for developing a drug-related severe cutaneous reaction. Fluconazole metabolism is primarily hepatic. SJS has been reported in patients with alcoholic and auto-immune hepatitis.8

Tests of liver function, such as synthesis of clotting factors, were persistently abnormal from birth in our patient; some alterations in hepatic drug processing and detoxification would also be expected. Although speculative, our patient may have had increased circulation of toxic drug metabolites, which could accumulate and immunoreact in the epidermis.4

To the best of our knowledge, this is the first reported case of fluconazole-associated TEN in pediatrics. Fluconazole is generally well tolerated in children, with the most common issues being gastrointestinal disturbance and reversible hepatotoxicity. Although high-quality, systematic studies are lacking, a comprehensive literature review and meta-analysis on the safety of fluconazole in pediatrics was published in January of 2013: including data from 4209 patients from 90 articles, hepatotoxicity was noted in 9% of all patients and in 12% of neonates.10 No severe cutaneous reactions were documented, and skin-related adverse events (of any nature) were found in about 1%.

Use of antifungal drugs, for treatment and prophylaxis, in pediatrics is increasing, especially in vulnerable neonates and children with immunocompromising conditions. Although serious adverse events are infrequent, this case highlights that fluconazole can be associated with TEN. In patients with underlying liver disease, azoles should be used with caution. In particular, for patients in whom recognition of prodromal features is challenging, such as critically ill neonates, vigilance for skin lesions characteristic of SJS and TEN should be maintained.


  1. 1

    Roujeau JC, Stern R . Medical progress: severe cutaneous reactions to drugs. N Engl J Med 1994; 331: 1272–1285.

    CAS  Article  Google Scholar 

  2. 2

    Ofoma UR, Chapnick EK . Fluconazole induced toxic-epidermal necrolysis: a case report. Cases J 2009; 2: 9071.

    Article  Google Scholar 

  3. 3

    George J, Sharma A, Dixit R, Chabra N, Sharma S . Toxic epidermal necrolysis caused by fluconazole in a patient with human immunodefiency virus infection. J Pharmacol Pharmacother 2012; 3: 276–278.

    CAS  Article  Google Scholar 

  4. 4

    Paquet O, Pierard GE, Quatresooz P . Novel treatments for drug-induced toxic epidermal necrolysis (Lyell’s Syndrome). Int Arch Allergy Immunol 2005; 136: 205–216.

    CAS  Article  Google Scholar 

  5. 5

    Azon-Masoliver A, Vilaplana J . Fluconazole-induced toxic epidermal necrolysis in a patient with human immunodeficiency syndrome. Dermatology 1993; 187: 268–269.

    CAS  Article  Google Scholar 

  6. 6

    Gussenhoven MJ, Haak A, Peereboom-Wynia JD, Van’t Wout JW . Stevens-Johnson syndrome after fluconazole. Lancet 1991; 338: 120.

    CAS  Article  Google Scholar 

  7. 7

    Craythorne E, Creamer D . Stevens–Johnson syndrome due to prophylactic fluconazole in two patients with liver failure. Clin Exp Dermatol. 2009; 34 (7): e389–e390.

    CAS  Article  Google Scholar 

  8. 8

    Pasmatzi E, Monastirli A, Georgiou S, Sqouros G, Tsambaos D . Short-term and low-dose oral fluconazole treatment can cause Steven-Johnson-Syndrome in HIV-negative patients. J Drug Dermatol 2011; 10: 1360.

    Google Scholar 

  9. 9

    Levi N, Bastuji-Garin S, Mockenaupt M, Roujeau JC, Flahault A, Kelly JP et al. Medications as risk factors of Stevens-Johnson syndrome and toxic epidermal necrolysis in children: a pooled analysis. Pediatrics 2009; 123 (2): e297–e304.

    Article  Google Scholar 

  10. 10

    Egunsola O, Adefurin A, Fakis A, Jacqz-Aigran E, Choonara I, Sammons H et al. Safety of fluconazole in paediatrics: a systematic review. Eur J Clin Pharmacol 2013; 69: 1211–1212.

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to S Islam.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Islam, S., Singer, M. & Kulhanjian, J. Toxic epidermal necrolysis in a neonate receiving fluconazole. J Perinatol 34, 792–794 (2014).

Download citation

Further reading


Quick links