目的 基于肝多房棘球蚴病（HAE）病理特征提出4型超声分类体系，探讨其与病理分型的一致性和临床应用价值。方法 回顾性分析2018年1月—2024年12月于新疆医科大学第一附属医院经手术病理确诊为HAE的患者，收集患者的超声资料、病理资料及治疗方式。对患者进行超声检查，记录病灶的位置、最大直径、病灶内部回声、液化坏死区及钙化范围等情况，参照超声检查结果和病灶术后病理特征，将病灶分为浸润型、钙化型、液化型、复杂型4种类型。对比分析病灶超声分型的观察者间一致性和超声分型与病理结果的一致性，一致性评价采用Kappa值。比较4型病灶的超声相关参数之间的差异性，包括最大直径、实性区域面积占比、钙化区域面积占比、液化区域面积占比及病灶周围低回声晕圈等。基于患者最大病灶的分型确定患者类型，同时收集患者的临床治疗措施，分析患者类型和治疗方式的关联性。采用SPSS 26.0软件进行统计学分析。结果 最终纳入HAE患者344例、HAE病灶469个。超声分型结果为浸润型病灶227个、钙化型病灶73个、液化型病灶106个、复杂型病灶63个；病理分型结果为浸润型病灶217个、钙化型病灶81个、液化型病灶110个、复杂型病灶61个。超声分型的总体符合率为94.5%（443/469），各型病灶符合率分别为：钙化型100%（73/73）、液化型97.2%（103/106）、浸润型92.5%（210/227）、复杂型90.5%（57/63）。观察者间一致性Kappa值为0.936，超声分型与病理结果的一致性Kappa值为0.918。4型病灶在实性、钙化、液化成分占比和最大直径的差异均有统计学意义（H = 381.729、237.657、384.372、101.102，均P < 0.05）。76.7%（264/344）的患者行HAE根治性切除术，21.2%（73/344）行自体肝移植，2.0%（7/344）行异体肝移植。术前12.8%（44/344）的患者行姑息治疗，其中经皮肝穿刺胆道减压术16例，25例液化型及3例复杂型患者行脓腔穿刺置管引流。浸润型患者占44.5%（153/344）、钙化型患者占12.5%（43/344）、液化型患者占27.3%（94/344）、复杂型患者占15.7%（54/344）。关联分析结果显示，不同患者类型中脓腔穿刺置管引流术选择率的差异有统计学意义（χ² = 47.32，P < 0.01）。不同患者类型中根治性病灶切除术选择率的差异虽有统计学意义（χ² = 10.47，P < 0.05），但经Bonferroni校正后（α = 0.008 33），任意两型间选择率的差异均未没有统计学意义。结论 4型超声分类体系中HAE超声分型与病理特征高度一致且具备高度可重复性，各分型具备特征性超声参数，该分型方式通过术前识别病灶的主要成分，为HAE的个体化治疗提供影像学依据。

Objective To propose a 4-type ultrasound classification system based on the pathological features of hepatic alveolar echinococcosis (HAE), and to explore its consistency with pathological classification and clinical application value. Methods A retrospective analysis was performed on patients who were pathologically diagnosed with HAE after surgery at the First Affiliated Hospital of Xinjiang Medical University from January 2018 to December 2024. The ultrasound data, pathological data and treatment modalities of the patients were collected. Ultrasound examinations were conducted to record the location, maximum diameter, internal echo of the lesion, liquefactive necrosis area and calcification range. Referring to the ultrasound results and postoperative pathological features of the lesions, the lesions were divided into 4 types: infiltrative type, calcified type, liquefied type and complex type. The inter-observer consistency of ultrasound classification and the consistency between ultrasound classification and pathological results were comparatively analyzed, and the consistency evaluation was performed using the Kappa value. The differences in ultrasound-related parameters among the 4 types of lesions were compared, including the maximum diameter, the proportion of solid area, the proportion of calcified area, the proportion of liquefied area, and the hypoechoic halo around the lesion. The patient type was determined based on the classification of the largest lesion of the patient, and the clinical treatment measures of the patients were collected simultaneously to analyze the correlation between patient types and treatment modalities. Statistical analysis was carried out using SPSS 26.0 software. Results A total of 344 HAE patients and 469 HAE lesions were finally included. The ultrasound classification results showed 227 infiltrative lesions, 73 calcified lesions, 106 liquefied lesions and 63 complex lesions; the pathological classification results showed 217 infiltrative lesions, 81 calcified lesions, 110 liquefied lesions and 61 complex lesions. The overall coincidence rate of ultrasound classification was 94.5% (443/469), and the coincidence rates of each type of lesion were 100% (73/73) for calcified type, 97.2% (103/106) for liquefied type, 92.5% (210/227) for infiltrative type, and 90.5% (57/63) for complex type, respectively. The inter-observer consistency Kappa value was 0.936, and the Kappa value for the consistency between ultrasound classification and pathological results was 0.918. The differences in the proportions of solid, calcified, liquefied components and the maximum diameter among the 4 types of lesions were all statistically significant (H = 381.729, 237.657, 384.372, 101.102, all P < 0.05). 76.7% (264/344) of the patients underwent radical resection for HAE, 21.2% (73/344) underwent autologous liver transplantation, and 2.0% (7/344) underwent allogeneic liver transplantation. Preoperatively, 12.8% (44/344) of the patients received palliative treatment, including 16 cases of percutaneous transhepatic biliary decompression, and 25 cases of liquefied type and 3 cases of complex type underwent puncture drainage of the abscess cavity. The proportions of patients with different types were 44.5% (153/344) for infiltrative type, 12.5% (43/344) for calcified type, 27.3% (94/344) for liquefied type, and 15.7% (54/344) for complex type. Correlation analysis showed that there was a statistically significant difference in the selection rate of abscess cavity puncture and drainage among different patient types (χ² = 47.32, P < 0.01). Although the difference in the selection rate of radical lesion resection among different patient types was statistically significant (χ² = 10.47, P < 0.05), after Bonferroni correction (α = 0.008 33), there was no statistically significant difference in the selection rate between any two types. Conclusion This HAE ultrasound classification is highly consistent with pathological features and demonstrates high reproducibility, with each subtype possessing characteristic ultrasound parameters. This classification method, by identifying the main components of the lesion preoperatively, provides the imaging basis for the individualized treatment of hepatic alveolar echinococcosis.