Qiu Ying, Luo Ying, Geng Xinqian, Li Yujian, Feng Yunhua, Yang Ying
Department of Endocrinology, The Affiliated Hospital of Yunnan University, Kunming, China.
School of Medical, Kunming Medical University, Kunming, China.
Front Endocrinol (Lausanne). 2024 Dec 17;15:1510433. doi: 10.3389/fendo.2024.1510433. eCollection 2024.
Glucocorticoid-induced adrenal insufficiency (GIAI) is a hypothalamic-pituitary-adrenal (HPA) axis dysfunction caused by long-term use of exogenous steroids. Adrenal crisis (AC) is an acute complication of GIAI and one of the reasons for the increased risk of death. This study aims to analyze the clinical characteristics of GIAI patients with AC and explore the related risk factors.
Clinical data of adult GIAI patients treated at our hospital between January 1, 2014, and December 31, 2023 were included. The demographic characteristics, clinical characteristics, laboratory tests and comorbidities of the patients were collected. Univariate and multivariate regression analyses were used to explore the variables related to the occurrence of AC, and prediction models were constructed.
51 patients (13.75%) developed AC during hospitalization. Mortality was significantly higher in patients with AC than in those without AC. Multivariate logistic regression analysis showed that infection, psychiatric symptoms, serum sodium, albumin, neutrophil-lymphocyte ratio (NLR) and eosinophil-lymphocyte ratio (ELR) were independent risk factors for AC. Among the prediction models constructed by machine learning algorithms, logistic regression model had the best prediction effect.
This study investigated the clinical characteristics of AC in GIAI patients. NLR and ELR may be effective predictors of AC in GIAI patients, and combined with other clinically significant indicators, an effective prediction model was constructed. Logistic regression model had the best performance in predicting AC in GIAI patients.
糖皮质激素诱导的肾上腺功能不全（GIAI）是长期使用外源性类固醇引起的下丘脑 - 垂体 - 肾上腺（HPA）轴功能障碍。肾上腺危象（AC）是GIAI的一种急性并发症，也是死亡风险增加的原因之一。本研究旨在分析发生AC的GIAI患者的临床特征，并探索相关危险因素。
纳入2014年1月1日至2023年12月31日在我院接受治疗的成年GIAI患者的临床资料。收集患者的人口统计学特征、临床特征、实验室检查和合并症。采用单因素和多因素回归分析探索与AC发生相关的变量，并构建预测模型。
51例患者（13.75%）在住院期间发生AC。AC患者的死亡率显著高于未发生AC的患者。多因素逻辑回归分析显示，感染、精神症状、血清钠、白蛋白、中性粒细胞与淋巴细胞比值（NLR）和嗜酸性粒细胞与淋巴细胞比值（ELR）是AC的独立危险因素。在通过机器学习算法构建的预测模型中，逻辑回归模型具有最佳预测效果。
本研究调查了GIAI患者AC的临床特征。NLR和ELR可能是GIAI患者AC的有效预测指标，并结合其他具有临床意义的指标构建了有效的预测模型。逻辑回归模型在预测GIAI患者AC方面表现最佳。
