EURURO Vol. 71 No. 4

603 692

ePLND, with high sensitivity, specificity, PPV, and NPV, and
a low FN rate. This observation, coupled with the relatively
high risk of complications associated with ePLND does
encourage a shift towards adopting SNB for certain groups
of patients, in particular in low- and intermediate-risk
disease. In these patient groups, the SN is often the only
tumor-bearing node, and some have therefore suggested
removal of only the SN in these cases . Kluth et al
stated that for patients in low- and intermediate-risk
groups, any LNs examined resulted in a clinical nodal
staging score of >90%, whereas patients in the high-risk
group needed more than ten LNs examined to achieve a
probability of 80% of being truly LN-negative. With the need
to examine fewer LNs at pathology, and by picking LNs with
an increased chance of containing metastases, SNB alone
may be sufficient to predict ‘‘true’’ nodal status. With SNB
alone, Winter et al detected more LN-positive patients
than expected according to the Briganti nomogram
. This resulted in development of the first nomogram
for radioisotope-guided SN dissection to predict the
probability of LN involvement in prostate cancer . Nev-
ertheless, uncertainty remains regarding whether the same
oncologic outcomes can be achieved with SNB alone, but
with fewer complications, compared to ePLND.
3.7.2.2. Therapeutic potential. Apart from diagnostic benefits,
SNB may potentially have a therapeutic effect via removal
of more metastatic LNs when compared with ePLND-only
dissection. To explore this theory, we performed a
sensitivity analysis for patients who only had metastases
in SNs in all the studies included. Overall, the SN(s) were
the only metastatic site(s) in 73% of LN-positive patients.
Weckermann et al found that when SNs were positive,
the incidence of positive non-SNs was high in patients with
high-risk PCa. They recommended performing SNB togeth-
er with an ePLND to remove all pelvic LN metastases in
high-risk patients. We also assessed patients with LN
metastases found outside the ePLND template regardless of
whether the ePLND was positive or negative. For this
scenario the FP rate increased to 4.9% (IQR 0–10.2%) with
sensitivity of 95.7% (IQR 86.1–100%). This indicates that for
one in 20 patients who undergo ePLND, metastatic LNs
would have been left behind without SNB. How this affects
oncologic outcomes is unknown.
Even a well-performed, meticulous ePLND does not
ensure complete accuracy with regard to nodal status. Kluth
et al . found that the probability of missing LN
metastases decreased as the number of LNs examined
increased. When at least 15 LNs were examined, the
sensitivity of ePLND exceeded 80%. Depending on preoper-
ative characteristics such as higher grade, stage, and
prostate-specific antigen, more LNs needed to be removed
to predict true nodal status. Even with an ePLND in these
cases, a 90% probability of being free from nodal metastases
could not be achieved . In this context, our calculated
sensitivity of 95.2% for SN detection is promising.
3.7.2.3. Implications for future research. In general, the SN is
defined as the initial LN(s) that directly drain(s) lymph from
the site of the primary lesion . A good understanding of
the theoretical definition of SN does not mean there is
consensus regarding the clinical definition of SN. The
various definitions in the studies included affected DTA
outcomes. Currently there is no consensus on the most
appropriate definition or detection technique for SN.
Therefore, there is an urgent need to develop consensus
regarding SN definitions, thresholds, and detecting and
ways of performing SNB.
The extent of PLND still remains a matter of debate.
Different descriptions, such as limited, standard, extended,
and superextended PLND are used, but even then there is
heterogeneity regarding the templates that these refer to. In
addition, the interpretation of preoperative and intraoper-
ative locations of (sentinel) LNs is subjective, allowing
surgeons to assign the same node to different locations
. This means that not only the SNBs were variable but
the (e)PLND templates also varied in the studies included.
This issue affects the calculated primary outcomes for DTA.
It is also likely to influence the number of tumor-positive
SNs identified outside the PLND templates. Standardization
of the SNB approach and a universally accepted classifica-
tion of regions of ePLND template will aid future studies
focused on evaluation of these techniques.
3.7.3. Strengths and limitations
The main strength of this review is the robust, systematic,
and transparent approach taken to assess and appraise the
evidence base, including use of the Cochrane review
methodology, assessment of RoB using QUADAS-2, and
adherence to PRISMA and STARD guidelines. A limiting
factor is the significant inconsistency and heterogeneity in
definitions and DTA outcome measurements, and the
variable quality of the studies included. Extraction of the
required data was hampered by variable ways of reporting
outcomes. To assess the DTA of SNB, studies that only
reported findings on node level but not patient level had to
be excluded. For future trials, reporting results at patient
level is important for assessment of DTA.
Complications of ePLND are an important reason for SNB.
In the current analysis we focused on diagnostic accuracy.
None of the studies included reported on complications
associated exclusively with SNB. Although no data regarding
safety and complications relating solely to SNB can be
derived from the review, there is evidence from the
literature that the procedure is not associated with any
systemic toxicity, on the basis of wide usage of SNB in other
conditions such as penile cancer and breast cancer . In
an earlier published review, we addressed complications
. In most of the studies, the authors refrained from
reporting complications related to SNB. Because we included
only studies with a reference standard (ie, [e]PLND),
extraction of complications related to only SNB was
impossible. Further research on this topic is needed.
3.7.4. Comparison to other reviews and meta-analyses
Our systematic search revealed only one meta-analysis of
SN mapping in PCa The authors concluded that SNB
could prevent unnecessary PLND in PCa, with a pooled
E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 5 9 6 – 6 0 5 603