Department of Urology, Fudan University Shanghai Cancer Center, 270 Dong An Road, Shanghai 200032, China.
Int Urol Nephrol. 2012 Apr;44(2):451-7. doi: 10.1007/s11255-011-0036-5. Epub 2011 Jul 27.
We investigated whether narrow-band imaging (NBI) was superior to white light imaging (WLI) for detecting primary non-muscle invasive bladder cancer (NMIBC) in a randomized imaging sequence modality, as the increased detection rate by NBI maybe result from the "second look" inspection of the bladder.
Between February 2009 and May 2010, NBI and WLI flexible cystoscopy were prospectively performed on 78 patients with primary suspected NMIBC. According to randomization protocol, the bladder was mapped using WLI then NBI or vice versa within the same observation time. Suspicious lesions together with controlled normal-appearing mucosa were biopsied and examined by a pathologist blinded to the imaging sequence. The tumors detected on patient and tumor level, sensitivity, specificity, and accuracy were compared in both imaging.
Sixty-nine of 78 (88.5%) patients were diagnosed with bladder cancer including 211 tumors totally. Of these, 36 (17.1%) tumors in 13 patients were detected by NBI only, while 4 (1.9%) tumors in 3 patients found by WLI only. NBI identified significantly more additional tumors than WLI (P < 0.01). The sensitivity of WLI versus NBI for detecting primary bladder tumors was 77.7 versus 92.9% (P < 0.0001), the specificity 82.7 versus 73.5% (P > 0.05), and the accuracy 79.3 versus 86.7% (P < 0.05). The sensitivity of WLI versus NBI for detecting carcinoma in situ (CIS) was 68.3 versus 87.8% (P < 0.05), the specificity 82.9 versus 77.1% (P > 0.05), and the accuracy 75 versus 82.9% (P > 0.05).
The "second look" did not compromise the superiority of NBI over standard WLI flexible cystoscopy for detecting primary NMIBC including CIS lesions.
我们研究了在随机成像序列模式下，窄带成像（NBI）是否优于白光成像（WLI）用于检测原发性非肌肉浸润性膀胱癌（NMIBC），因为 NBI 增加的检测率可能是由于膀胱的“二次观察”检查。
2009 年 2 月至 2010 年 5 月，对 78 例原发性疑似 NMIBC 患者前瞻性进行 NBI 和 WLI 软式膀胱镜检查。根据随机分组方案，在相同的观察时间内，使用 WLI 对膀胱进行映射，然后使用 NBI 或反之亦然。可疑病变和受控制的正常外观黏膜均由病理学家进行活检和检查，病理学家对成像序列不知情。在患者和肿瘤水平、敏感性、特异性和准确性方面比较了两种成像方法的肿瘤检测结果。
78 例患者中有 69 例（88.5%）诊断为膀胱癌，共 211 个肿瘤。其中，13 例患者中有 36 个（17.1%）肿瘤仅通过 NBI 检测到，而 3 例患者中有 4 个（1.9%）肿瘤仅通过 WLI 检测到。NBI 比 WLI 显著更多地识别出额外的肿瘤（P < 0.01）。WLI 与 NBI 检测原发性膀胱肿瘤的敏感性分别为 77.7%和 92.9%（P < 0.0001），特异性分别为 82.7%和 73.5%（P > 0.05），准确性分别为 79.3%和 86.7%（P < 0.05）。WLI 与 NBI 检测原位癌（CIS）的敏感性分别为 68.3%和 87.8%（P < 0.05），特异性分别为 82.9%和 77.1%（P > 0.05），准确性分别为 75%和 82.9%（P > 0.05）。
“二次观察”并没有影响 NBI 相对于标准 WLI 软式膀胱镜检查在检测原发性 NMIBC （包括 CIS 病变）方面的优势。
