背景及目的：原发性醛固酮增多症（primary aldosteronism，PA）的分型诊断是决定治疗方式的关键。肾上腺静脉采血（adrenal vein sampling，AVS）是现有指南推荐的分型诊断的金标准，但因其存在一定的技术难度及要求，短时间内无法广泛开展。本研究旨在建立诊断单侧性原发性醛固酮增多症（unilateral primary aldosteronism，UPA）的预测模型，以减少临床上不必要的AVS。

研究方法：本研究采用回顾性分析的方法，收集了南京鼓楼医院内分泌科2018年7月至2021年8月的确诊为PA并行AVS或单侧肾上腺切除手术的患者数据，经过拟定的纳排标准筛选，最终共纳入183例PA患者。其中，根据手术治疗后的预后结局或AVS结果诊断UPA患者127例，双侧性原发性醛固酮增多症（bilateral primary aldosteronism，BPA）患者56例。并通过单因素分析及多元logistic回归分析确定预测单侧性醛固酮优势分泌的相关因素。最终选择年龄、肾上腺CT影像学表现、血浆醛固酮浓度（plasma aldosterone concentration，PAC）及自发性低钾血症作为主要参数建立诊断UPA的预测模型，确定模型的诊断效率，并与既往研究所得的预测模型进行比较。

研究结果：与BPA组比较，UPA组的患者年龄更小（48±12 vs. 53±10，P=0.008）、高血压病程更短[7（2，12.75）vs. 10（5，13），P=0.034]、肾上腺CT示单侧典型腺瘤的占比更高（46.5% vs. 23.2%，P=0.003）。此外，肾上腺CT检查结果示单侧典型腺瘤的PA患者，仅根据CT结果诊断UPA的准确率约为82%。行手术治疗的PA患者术后取得临床完全缓解结局的占比约为51.2%，取得临床部分缓解结局的占比约为43.3%，取得生化完全缓解结局的占比约为92.9%。进一步行多元logistic回归分析发现肾上腺CT示单侧典型腺瘤（OR=2.666，P=0.010）、年龄（OR=0.967，P=0.030）为存在单侧醛固酮优势分泌的相关因素，再结合PAC及是否合并自发性低钾血症进行排列组合，最终建立诊断UPA的预测模型：年龄＜40岁，PAC＞15 ng/dl，肾上腺CT检查结果为单侧典型腺瘤以及合并自发性低钾血症。该模型在本研究纳入人群中的特异性为1，敏感性为0.14。

研究结论：根据相关临床特征可以建立单侧性原醛症的预测模型。本研究显示年龄＜40岁，PAC＞15 ng/dl，肾上腺CT表现为单侧典型腺瘤以及合并自发性低钾血症的PA患者可诊断UPA，建议直接选择手术治疗。该预测模型可使14%的PA患者避免非必要的AVS，节省医疗资源，具有一定临床应用的价值。

Objective: The subtype classification of primary aldosteronism (PA) is crucial in determining the treatment approach. Adrenal vein sampling (AVS) is the gold standard test for subtype classification, which has limited applicability due to its technical difficulty and requirements. The objective of this study is to establish a predictive model for diagnosing unilateral primary aldosteronism (UPA) to help avoid unnecessary AVS.

Methods: Data obtained from patients diagnosed with PA at the Endocrinology Department of Nanjing Drum Tower Hospital from July 2018 to August 2021 were retrospectively analyzed. A total of 183 PA patients who underwent AVS or unilateral adrenalectomy were included. Based on postoperative outcomes and AVS results, 127 were diagnosed with UPA, and 56 with bilateral primary aldosteronism (BPA). Single-factor analysis and multiple logistic regression analysis were utilized to identify factors related to unilateral excessive aldosterone secretion. A predictive model for diagnosing UPA was established based on these factors: age, adrenal CT findings, plasma aldosterone concentration (PAC), and spontaneous hypokalemia, with a comparison against previous models.

Results: In comparison to the BPA group, the UPA group exhibited a younger age (48±12 vs. 53±10；P=0.008), shorter hypertension duration [7 (2，12.75) vs. 10 (5，13) ；P=0.034], and a higher proportion of unilateral adenoma on adrenal CT (46.5% vs. 23.2%；P=0.003). The accuracy rate of diagnosing UPA based on CT findings which revealed unilateral adenoma was approximately 82%. Among patients who underwent unilateral adrenalectomy, complete clinical success was achieved in approximately 51.2% of cases, while partial clinical success was attained in 43.3%. Complete biochemical success was observed in 92.9% of cases. Subsequent multiple logistic regression analysis revealed two significant factors associated with unilateral excessive aldosterone secretion: unilateral adenoma on adrenal CT (OR = 2.666, p = 0.010) and age (OR = 0.967, p = 0.030). A diagnostic model aimed at predicting UPA was formulated by integrating PAC and spontaneous hypokalemia, as the following criteria: age younger than 40 years, PAC > 15 ng/dl, presence of unilateral adenoma on adrenal CT, and occurrence of spontaneous hypokalemia. In this study cohort, the model demonstrated a specificity of 1 and a sensitivity of 0.14.