Narrow Band Imaging Might be a Promising Diagnostic Intervention in Non-Muscle Invasive Bladder Cancer
Introduction
Although the white light cystoscopy (WLC) is a gold standard for detecting bladder cancer, it is associated with a poor diagnostic accuracy in detectingcarcinoma in situand small / flatter lesions. Hence different imaging techniques with a better tumor detection rate have been developed. However, some studies have shown that the newer techniques might have high false positive rates (FPRs) than WLC. Photodynamic diagnosis (PDD) utilizes blue-violet light with intravesical instillation of 5-aminolevulinic acid (5-ALA) or hexaminolevulinic acid (HAL). Narrow band imaging (NBI) is a new image processing technique which avoids the need for intravesical administration and has a higher diagnostic accuracy than WLC as demonstrated by a few studies.
Aim
This is the first systematic review and meta-analysis which evaluated the diagnostic performance of image technique based transurethral resection as compared to WLC.
Method
Study Design
- Studies were identified using PubMed/MEDLINE, Web of Science, EMBASE, Cochrane Library and Central Register of Controlled Trials databases
- Studies included patients diagnosed with non-muscle invasive bladder cancer (NMIBC) or recurrent tumors
- The studies reporting the diagnostic accuracy of PDD with 5-ALA, PDD with HAL or NBI with WLC as a reference standard at the patient or lesion level were included
- Data were pooled using a random effect diagnostic meta-analysis and subgroup analyses were performed
Endpoints
- Pooled sensitivity (SSY)
- Specificity (SPY)
- Diagnostic OR (DOR)
- Area under the receiver operating characteristic curve (AUROC) values per group for NBI, HAL, 5-ALA at the lesion or patient level
Results
- A total of 26 studies comprising of 3979 patients were included. Overall, all studies reported methodology for the index test and reference standard clearly, without a significant source of potential bias. Among them 69% (18/26) of the studies were presented as low or unclear risk of bias across most domains.
- NBI demonstrated the highest diagnostic accuracy as compared to WLC at the lesion level with SSY 0.94, SPY 0.79, DOR 40.09 and AUROC 0.88
- At the patient level too, NBI had the highest DOR 358.71
- The comparison of diagnostic endpoints is shown in table 1.
|
| NBI | HAL | 5-ALA |
| Pooled DOR values | 40.09 | 78.14 | 18.14 |
| AUROC values | 0.88 | 0.94 | 0.82 |
| Pooled SSY | 0.94 | 0.95 | 0.90 |
| Pooled SPY | 0.79 | 0.81 | 0.69 |
- The findings of the subgroup analyses which was conducted on studies with low to moderate risk of bias (ROB) and atleast 100 patients at lesion level were consistent with the overall analysis
- The comparison-adjusted funnel plots of the diagnostic meta-analysis were not suggestive of any publication bias
Conclusion
- The diagnostic accuracy of image technique based transurethral resection such as narrow band imaging (NBI), 5-aminolevulinic acid (5-ALA) and hexaminolevulinic acid (HAL) was superior to that of WLC
- NBI demonstrated the best diagnostic performance outcomes and might be the most promisingdiagnostic technique for non-muscle invasive bladder cancer
BMJ Open. 2019 Oct 17;9(10):e028173. Doi: 10.1136/bmjopen-2018-028173.






