Drug-induced Steatohepatitis Caused by Long-term Use of Topical Steroids for Atopic Dermatitis

Authors: Yuki Tamura, Atsushi Naganuma, Yuhei Suzuki, Sanae Uehara, Takashi Hoshino, Takeshi Hatanaka, Nobuyuki Shibusawa, Akihito Uehara, Akira Ogawa, Satoru Kakizaki, Toshio Uraoka

Abstract

Atopic dermatitis is common in children and often treated with topical corticosteroids (TCs). A boy in his late teens who had been using TCs for atopic dermatitis was diagnosed with liver damage during a health checkup. A medical examination revealed severe steatotic liver disease and elevated liver enzyme levels despite the absence of typical symptoms such as central obesity. After discontinuation of TCs, an improvement in liver enzyme levels was observed, leading to the diagnosis of drug-induced steatohepatitis. This case underscores the potential liver risks associated with prolonged TC use in children, highlighting the need for parental education.

Introduction

Topical corticosteroids (TCs) are widely used to treat skin diseases and are easily available to the general public. However, the long-term use of strong TCs can cause severe side effects, particularly when applied to children and individuals with thin skin. Occasionally, such side effects are caused by inappropriate use of TCs by patients or their families. However, while there have been reports of cases of hypercortisolism, hypothalamic-pituitary-adrenal axis suppression, and adrenal insufficiency caused by TCs, there have been no reports of steatohepatitis caused by TCs (1,2).

We herein report a case of drug-induced steatohepatitis caused by the long-term systemic application of TCs.

Case Report

A 19-year-old boy was diagnosed with liver damage during a health checkup, despite having no complaints. He underwent abdominal ultrasound (US) at a nearby hospital, which revealed severe steatotic liver disease, and was thus referred to our hospital.

At the initial medical examination, his height, body weight, body mass index (BMI), and blood pressure were 170 cm, 70 kg, 26.3 kg/m^2^, and 116/64 mmHg, respectively. Since five years old, he had used systemic TCs, including potent glucocorticoids (GCs), for atopic dermatitis but had no other medical history. No signs of developmental delay or intellectual disability were observed. A physical examination did not reveal a moon face, central obesity, truncal hirsutism, or skin striae on the thigh.

Blood tests (Table 1) showed elevated hepatic and biliary enzymes (aspartate aminotransferase 236 U/L, alanine aminotransferase 492 U/L, total bilirubin 0.63 mg/dL, alkaline phosphatase 132 U/L, γ-glutamyl transferase 93 U/L), but the liver reserve was not compromised (platelet count 34.9×10^4^/μL, albumin 5.2 g/dL, prothrombin time 98.6%). Serological tests for hepatitis A, B, C, and E viruses and cytomegalovirus were negative. The test for serum immunoglobulin against the Epstein-Barr virus was positive but not indicative of virus reactivation. Tests for anti-nuclear and anti-mitochondrial antibodies were negative. No abnormal glucose tolerance was found, and free thyroxine and thyroid-stimulating hormone levels were within the reference ranges. The cortisol levels in the morning were normal.

US showed increased echogenicity of the liver parenchyma compared to the cortex of the right kidney and a marked increase in fine echoes with poor visualization of the intrahepatic vessel borders and posterior right lobe of the liver (Fig. 1a). Computed tomography (CT) revealed that the hepatic vessels had higher attenuation than the liver parenchyma and indicated a liver-to-spleen ratio of 0.02 (3). However, there were no ascites, hepatomegaly, or increased gallbladder wall thickness indicative of acute liver failure (Fig. 1b, c). The visceral adipose tissue (VAT) measured 83.9 cm^2^, and subcutaneous adipose tissue (SAT) was 213.9 cm^2^, resulting in a VAT/SAT ratio of 0.39 (4).

A liver biopsy revealed severe steatosis exceeding 80% of the liver parenchyma (Fig. 2a), a few instances of hepatocyte ballooning (Fig. 2b, arrow), mild lobular inflammation with ＜2 foci per 20× field, and findings corresponding to an NAFLD activity score (NAS score) of 5 (5). Furthermore, mild pericellular fibrosis was observed in zone 3 (Fig. 2d), corresponding to stage 1 and grade 1 according to the Brunt criteria (6).

Although the exact frequency of use was unknown, the patient had been using systemic TCs of various strengths, including very strong, strong, and medium. His home doctor routinely prescribed these TCs without evaluating his skin condition according to his mother's wish. We suspected that the long-term use of TCs was the cause of steatohepatitis and discontinued their use, switching to antihistamines. The elevated hepatic enzymes did not completely normalize but had clearly improved by day 39 (Fig. 3). There was no significant change in the BMI, and the atopic dermatitis remained in remission during this treatment. Therefore, the patient was diagnosed with drug-induced steatohepatitis.

On day 150, with the patient’s BMI at 25.2 kg/m^2^ and hepatic enzyme levels remaining elevated, we suspected a correlation between metabolic dysfunction and liver dysfunction. We implemented a hypocaloric diet and exercise for weight loss, discontinued TCs, and monitored the progress, hoping that the hepatic enzymes would gradually normalize. By day 270, his hepatic enzymes had almost normalized (Fig. 3), and CT and US showed marked improvements in steatohepatitis (Fig. 1f).

Cortisol levels in the morning gradually decreased and reached a low level on day 88 without any clinical symptoms, leading to suspicion of possible adrenal insufficiency. We carefully followed up the patient, and his serum cortisol levels improved to within the normal range by day 151 and remained stable on day 270 (Fig. 3). The adrenal function was not suppressed after discontinuation of TCs without tapering.

Discussion

The basic therapy for atopic dermatitis includes hydrating and barrier-stabilizing topical treatment, anti-inflammatory topical agents (such as corticosteroids and calcineurin inhibitors; e.g., tacrolimus and pimecrolimus), and avoiding specific and nonspecific provocation factors (7-9). However, TCs are still the first-line anti-inflammatory treatment option for atopic dermatitis. Although the incidence of reported side effects from TC use is low, long-term use of TCs, especially on thin skin areas (face, inside of the thighs, upper arms, and breasts) and high-potency agents, can cause cutaneous side effects (10). TCs, particularly potent and highly potent agents, can be absorbed in sufficient amounts to cause systemic side effects (11). Children are more susceptible than adults to such effects due to their greater body surface-to-weight ratio (8,9).

While systemic side effects of TCs are rare, there are reports of hypothalamic-pituitary-adrenal axis suppression and adrenal insufficiency caused by TCs (1,2,12). A study by Güven reviewed 14 children, ranging from 0.19 to 11.89 years old, who were diagnosed with iatrogenic Cushing’s syndrome due to hypothalamic-pituitary-adrenal axis suppression caused by potent GCs administered orally and topically (1). US was performed in 11 patients, and 6 of whom had hepatic steatosis. Two of these six patients were exposed to high-dose GCs via the transdermal route, and one patient was exposed through both the transdermal and oral routes. The presence of steatohepatitis was unknown because no liver biopsies were performed. However, hepatic steatosis was detected even in young infants exposed to potent GCs (Table 2).

GCs can lead to the development of steatotic liver disease. The pathways of lipid deposition include an increased appetite and high caloric intake, elevated blood glucose levels due to GC-induced stimulation of gluconeogenesis, and stimulation of de novo lipogenesis, which is augmented by high glucose and insulin levels and by GCs themselves. In addition, there is an increased release of free fatty acids from adipose stores and an enhanced uptake of these acids by the liver (13). Even 24-h treatment with oral GCs results in significant alterations in glucose homeostasis (14).

In the present case, metabolic dysfunction-associated steatotic liver disease was suspected because the patient’s BMI was 26.2 kg/m^2^, but after discontinuing TCs, his elevated hepatic enzymes clearly improved without weight loss. Therefore, the patient was diagnosed with drug-induced steatohepatitis. In reported cases of steatotic liver disease, the hypothalamic-pituitary-adrenal axis was suppressed. However, in this case, although adrenal insufficiency was suspected, it was not diagnosed. We believe that long-term application of TCs (14 years) may have contributed to this observation. The patient had been using various TCs all over his body, mainly due to mother’s wishes, despite the remission of his atopic dermatitis. The possibility of steatotic liver disease or steatohepatitis due to the inappropriate use of TCs should be recognized.

Conclusion

Long-term application of potent TCs can cause steatohepatitis without typical clinical symptoms, such as central obesity and moon face, which are commonly associated with side effects of TCs. Our case highlights the need for vigilance in monitoring less obvious but serious complications associated with long-term TC use. Furthermore, clinicians should warn parents, especially mothers, about the risk of inappropriate TC use when prescribing them to children.

The patient gave his written informed consent for publication of the details of his case.

The authors state that they have no Conflict of Interest (COI).