Abstract
CONCLUSIONS
SNB appears to have diagnostic accuracy comparable to ePLND, with high sensitivity, specificity, PPV and NPV, and a low FN rate. With a low FP rate (rate of detecting positive nodes outside the ePLND template), SNB may not have any additional diagnostic value over and above ePLND, although SNB appears to increase nodal yield by increasing the number of affected nodes when combined with ePLND. Thus, in high-risk disease it may be prudent to combine ePLND with SNB.
EVIDENCE ACQUISITION
A systematic literature search of Medline, Embase, and the Cochrane Library (1999-2016) was undertaken using PRISMA guidelines. All studies of SNB in men with PCa using PLND as reference standard were included. The primary outcomes were the nondiagnostic rate (NDR), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and false positive (FP) and false negative (FN) rates. Relevant sensitivity analyses based on SN definitions, ePLND as reference standard, and disease risk were undertaken, including a risk of bias (RoB) assessment.
PATIENT SUMMARY
This literature review showed a high diagnostic accuracy for sentinel node biopsy in detecting positive lymph nodes in prostate cancer, but further studies are needed to explore the effect of sentinel node biopsy on complications and oncologic outcome.
CONTEXT
Extended pelvic lymph node dissection (ePLND) is the gold standard for detecting lymph node (LN) metastases in prostate cancer (PCa). The benefit of sentinel node biopsy (SNB), which is the first draining LN as assessed by imaging of locally injected tracers, remains controversial.
EVIDENCE SYNTHESIS
Of 373 articles identified, 21 studies recruiting a total of 2509 patients were eligible for inclusion. Median cumulative percentage (interquartile range) results were 4.1% (1.5-10.7%) for NDR, 95.2% (81.8-100%) for sensitivity, 100% (95.0-100%) for specificity, 100% (87.0-100%) for PPV, 98.0% (94.3-100%) for NPV, 0% (0-5.0%) for the FP rate, and 4.8% (0-18.2%) for the FN rate. The findings did not change significantly on sensitivity analyses. Most studies (17/22) had low RoB for index test and reference standard domains.
OBJECTIVE
To assess the diagnostic accuracy of SNB in PCa.