Kumari Neha, Kalonia Tushar, Malik Akanksha, Kumar Arvind, Rao Shalinee
Department of Pathology, All India Institute of Medical Sciences, Rishikesh, Uttarakhand, India.
Indian J Endocrinol Metab. 2021 Sep-Oct;25(5):402-409. doi: 10.4103/ijem.ijem_389_21. Epub 2022 Jan 12.
Fine-needle aspiration cytology remains the preliminary test for diagnosing papillary thyroid carcinoma (PTC). Numerous features are established to arrive at the diagnosis. However, few cases pose a challenge to correctly diagnose PTC. Our study aims to elicit the combination of features to aid in the diagnosis of such cases.
Cytology smears of histologically proven cases of PTC and benign diagnoses were included as case (n = 36) and control group (n = 38), respectively. Features including papillae with cores, 3-D caps, nuclear grooves (NG), intranuclear cytoplasmic inclusions (INCI), giant cells, macrophages, cellular swirls, psammoma bodies, pale chromatin, nuclear overlapping, nuclear enlargement, and metaplastic cells were assessed. Statistic tests including Independent t test/Mann-Whitney Test and Chi-Square test/Fisher's Exact test were used. Receiver operating characteristic curve was used to assess the cut-off point of many cytological features to predict PTC. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of cytological features was calculated to predict PTC.
Presence of five or more cytological features (papillae with cores, cellular swirls, NG, INCI, and psammoma bodies) together could diagnose PTC (PPV) in 78.95% of the cases, with a NPV of 83.33%. Diagnostic accuracy of these five features combined was 81.08%. Papillae with cores and nuclear grooving were the most sensitive cytological features, whereas INCI followed by cellular swirls and NG were the most specific features.
Relying on a combination of the most sensitive and specific features rather than any one cytological feature can help reduce the misdiagnoses in PTC.
细针穿刺细胞学检查仍是诊断甲状腺乳头状癌（PTC）的初步检查方法。已有许多特征用于确诊。然而，少数病例在正确诊断PTC方面存在挑战。我们的研究旨在找出有助于诊断此类病例的特征组合。
组织学确诊的PTC病例和良性诊断病例的细胞学涂片分别作为病例组（n = 36）和对照组（n = 38）。评估的特征包括有轴心的乳头、三维帽、核沟（NG）、核内胞质包涵体（INCI）、巨细胞、巨噬细胞、细胞漩涡、砂粒体、淡染色质、核重叠、核增大和化生细胞。使用了包括独立t检验/曼-惠特尼检验以及卡方检验/费舍尔精确检验在内的统计检验。采用受试者工作特征曲线评估多种细胞学特征预测PTC的临界点。计算了细胞学特征预测PTC的敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）。
同时存在五种或更多细胞学特征（有轴心的乳头、细胞漩涡、NG、INCI和砂粒体）可在78.95%的病例中诊断PTC（PPV），NPV为83.33%。这五种特征联合的诊断准确率为81.08%。有轴心的乳头和核沟是最敏感的细胞学特征，而INCI其次是细胞漩涡和NG是最具特异性的特征。
依靠最敏感和最具特异性特征的组合而非任何单一细胞学特征有助于减少PTC的误诊。
