Supplementary Material for: Diagnostic Accuracy of Papanicolaou Society of Cytopathology System for Reporting Respiratory Cytology: A Systematic Review and Meta-Analysis

Introduction
This study conducts the first meta-analysis to evaluate the diagnostic accuracy and the aggregated risk of malignancy associated with each category of the Papanicolaou Society of Cytopathology (PSC) system for reporting respiratory cytology.
Methods
A systematic search was conducted in PubMed, Scopus, and Web of Science using the keywords “(Lung, Respiratory specimens) AND (Papanicolaou Society of Cytopathology System).” Articles were assessed for risk of bias using the QUADAS-2 tool. After excluding inadequate samples, sensitivity and specificity for various cut-off points. Summary receiver operating characteristic (sROC) curves and diagnostic odds ratios (DOR) were pooled to assess diagnostic accuracy.
Results
Five studies, totaling 3489 cases, were included. Sensitivity and specificity for the "Atypical and higher risk categories" considered positive were 60% (95% CI, 51%-68%) and 87% (95% CI, 81%-92%), respectively. For the "Suspicious for malignancy and higher risk categories" considered positive, sensitivity and specificity were 49% (95% CI, 40%-58%) and 95% (95% CI, 92%-97%), respectively. Sensitivity and specificity for the "Malignant" category considered positive for malignancy were 42% (95% CI, 33%-52%) and 97% (95% CI, 92%-99%), respectively. The pooled Area Under the Curve (AUC) ranged from 68% to 75% for each cut-off.
Conclusion
This meta-analysis underscores the PSC system's accuracy in reporting respiratory cytology. It highlights the diagnostic importance of the "Suspicious" and "Malignant" categories in identifying malignancy, and the utility of the "Atypical" category for initial screening. These findings support the PSC system's role in enhancing diagnostic accuracy and clinical decision-making in respiratory cytology.

引言
本研究开展了首项元分析，旨在评估用于呼吸道细胞学报告的巴帕尼科拉乌细胞病理学会（Papanicolaou Society of Cytopathology, PSC）系统各分类对应的诊断效能，以及其相关的恶性肿瘤综合风险。

方法
本研究在PubMed、Scopus及Web of Science数据库中开展系统检索，检索式为“(肺, 呼吸道标本) AND (巴帕尼科拉乌细胞病理学会系统)”。采用QUADAS-2工具对纳入文献的偏倚风险进行评估。在排除不合格样本后，计算不同截断值下的灵敏度与特异度。通过合并汇总受试者工作特征（Summary Receiver Operating Characteristic, sROC）曲线及诊断比值比（Diagnostic Odds Ratios, DOR）以评估诊断效能。

结果
最终纳入5项研究，共计3489例病例。将“非典型性及更高风险分类”判定为阳性时，其灵敏度为60%（95%置信区间：51%~68%），特异度为87%（95%置信区间：81%~92%）。将“疑似恶性及更高风险分类”判定为阳性时，灵敏度为49%（95%置信区间：40%~58%），特异度为95%（95%置信区间：92%~97%）。将“恶性肿瘤分类”判定为恶性阳性时，灵敏度为42%（95%置信区间：33%~52%），特异度为97%（95%置信区间：92%~99%）。各截断值对应的合并曲线下面积（Area Under the Curve, AUC）介于68%~75%之间。

结论
本项元分析证实了PSC系统在呼吸道细胞学报告中的诊断效能。本研究明确了“疑似恶性”与“恶性肿瘤”分类在恶性肿瘤鉴别中的重要诊断价值，以及“非典型性”分类在初筛中的应用潜力。上述研究结果支持PSC系统在提升呼吸道细胞学诊断效能及辅助临床决策中的应用价值。