Mariappan Paramananthan, Lavin Victoria, Phua Chu Qin, Khan Shahid Aziz Anwar, Donat Roland, Smith Gordon
Edinburgh Urological Cancer Group, Department of Urology, Western General Hospital, Edinburgh, United Kingdom.
Edinburgh Urological Cancer Group, Department of Urology, Western General Hospital, Edinburgh, United Kingdom.
Urology. 2017 Nov;109:134-139. doi: 10.1016/j.urology.2017.08.007. Epub 2017 Aug 14.
To assess urologists' ability to predict the grade and stage of new bladder cancers from the cystoscopic features alone.
We conducted a prospective clinical study on consecutive patients who underwent transurethral resection of bladder tumor (TURBT) for new bladder cancers. Using only cystoscopic tumor morphology at the time of initial TURBT, 3 urology consultants predicted the grade and stage, recording these on a proforma along with tumor features. Predictions were compared with assessments by uropathologists, blinded to the clinical prediction. We analyzed the accuracy in (1) predicting the exact grade and stage of the cancer; (2) differentiating between low- and high-grade cancers; and (3) discerning between Ta, T1, and T2 cancers.
Of 248 patients, 224 were suitable for analysis. The positive predictive values for low- and high-grade cancers were 85.8% and 71.3%, respectively. The overall likelihood of a consultant predicting high-grade cancers as being low grade was 16/83 (19.3%). When tumors were large (>30 mm), this likelihood dropped significantly to 7.3% (4/55) (odds ratio = 3.1, 95% confidence interval = 1.0-9.7, P = .04). Non-muscle-invasive and muscle-invasive cancers were predicted accurately in 93.4% and 85.2% patients, respectively. Six of 161 (3.7%) tumors predicted to be non-muscle-invasive bladder cancer were actually muscle invasive on histology.
For clinical purposes, in newly presenting patients with bladder cancer, urologists appear to reliably predict lower grade and muscle-invasive disease, confirming widely held belief. This allows for appropriate and efficient use of surgical expertise, available technology, and selection of participants for clinical trials on the basis of prehistology risk categories.
评估泌尿外科医生仅根据膀胱镜检查特征预测新发膀胱癌分级和分期的能力。
我们对因新发膀胱癌接受经尿道膀胱肿瘤切除术（TURBT）的连续患者进行了一项前瞻性临床研究。3位泌尿外科会诊医生仅根据初次TURBT时的膀胱镜肿瘤形态预测分级和分期，并将这些信息连同肿瘤特征记录在一份表格上。将预测结果与泌尿病理学家的评估结果进行比较，泌尿病理学家对临床预测情况不知情。我们分析了以下方面的准确性：（1）预测癌症的准确分级和分期；（2）区分低级别和高级别癌症；（3）辨别Ta、T1和T2期癌症。
248例患者中，224例适合分析。低级别和高级别癌症的阳性预测值分别为85.8%和71.3%。会诊医生将高级别癌症预测为低级别癌症的总体可能性为16/83（19.3%）。当肿瘤较大（>30毫米）时，这种可能性显著降至7.3%（4/55）（优势比=3.1，95%置信区间=1.0-9.7，P=0.04）。非肌层浸润性和肌层浸润性癌症分别在93.4%和85.2%的患者中被准确预测。161例（3.7%）被预测为非肌层浸润性膀胱癌的肿瘤，组织学检查实际为肌层浸润性。
出于临床目的，对于新诊断的膀胱癌患者，泌尿外科医生似乎能够可靠地预测低级别和肌层浸润性疾病，这证实了普遍的观点。这有助于合理有效地利用手术专业知识、现有技术，并根据组织学检查前的风险类别选择临床试验参与者。
