Unité d'Exploration Médico-Chirurgicale Oncologique, Institut Paoli-Calmettes, France.
Gastrointest Endosc. 2009 Nov;70(5):907-14. doi: 10.1016/j.gie.2009.05.020. Epub 2009 Jul 29.
Pancreatic endocrine tumors (PETs) differ in clinical behavior and prognosis. Determination of malignant potential through specimens obtained by EUS-FNA can help in the management of these patients.
To determine the value of EUS-FNA for diagnosing PETs and for classifying their underlying malignant potential based on the World Health Organization (WHO) classification.
Single-center, retrospective, cohort study.
Tertiary referral hospital.
This study involved 86 consecutive patients (44 men, mean age 58 +/- 14 years) who had been diagnosed with PETs and submitted to EUS-FNA from January 1999 to August 2008.
EUS-FNA of a pancreatic mass and/or a metastasis site. Immunohistochemistry on microbiopsies or on monolayer cytology was routinely used. The lesions were classified as recommended by the WHO.
EUS-FNA sensitivity and 5-year survival rate.
Overall, in 90% (77 of 86) of patients in this study, PET was diagnosed with EUS-FNA. The sensitivity did not vary with tumor size, type, location, or the presence of hormonal secretion. Of 86 patients, 30 (35%) were submitted to surgical resection. The kappa correlation index between the WHO classification obtained by EUS-FNA and by surgery was 0.38 (P = .003). Major discrepancies were found in the group of patients diagnosed with endocrine tumor of uncertain behavior by EUS-FNA, because 72% turned out to have well-differentiated endocrine carcinoma. Sixteen patients (27%) died during a mean follow-up period of 34 +/- 27 months. The 5-year survival rates were 100% for endocrine tumors, 68% for well-differentiated endocrine carcinomas, and 30% for poorly differentiated endocrine carcinomas (P = .008, log-rank test).
Retrospective design, selection bias, and small sample size.
This largest single-center experience to date demonstrated the accuracy of EUS-FNA in diagnosing and determining the malignant behavior of PETs. EUS-FNA findings predict 5-year survival in patients with PETs.
胰腺内分泌肿瘤（PETs）在临床行为和预后方面存在差异。通过 EUS-FNA 获得的标本确定恶性潜能有助于这些患者的管理。
确定 EUS-FNA 对诊断 PETs 并根据世界卫生组织（WHO）分类对其潜在恶性程度进行分类的价值。
单中心回顾性队列研究。
三级转诊医院。
这项研究涉及 86 例连续患者（44 名男性，平均年龄 58 +/- 14 岁），他们于 1999 年 1 月至 2008 年 8 月期间因胰腺肿块和/或转移部位接受 EUS-FNA 诊断为 PETs。
EUS-FNA 对胰腺肿块和/或转移部位进行。常规使用微生物活检或单层细胞学的免疫组织化学。病变按照 WHO 的建议进行分类。
EUS-FNA 的敏感性和 5 年生存率。
总体而言，在这项研究的 86 例患者中，90%（77 例）通过 EUS-FNA 诊断为 PET。肿瘤大小、类型、位置或激素分泌的存在对敏感性没有影响。86 例患者中有 30 例（35%）接受了手术切除。EUS-FNA 和手术获得的 WHO 分类之间的kappa 相关指数为 0.38（P =.003）。在 EUS-FNA 诊断为不确定行为的内分泌肿瘤患者组中发现了主要差异，因为 72%的患者实际上患有分化良好的内分泌癌。16 例患者（27%）在平均 34 +/- 27 个月的随访期间死亡。5 年生存率为内分泌肿瘤 100%，分化良好的内分泌癌 68%，分化差的内分泌癌 30%（P =.008，对数秩检验）。
回顾性设计、选择偏倚和样本量小。
这是迄今为止最大的单中心经验，证明了 EUS-FNA 在诊断和确定 PETs 恶性行为方面的准确性。EUS-FNA 结果预测了 PETs 患者的 5 年生存率。
