EURURO Vol. 72 No. 3

E59 476

Letter to the Editor
Re: Maria Carmen Mir, Ithaar Derweesh,
Francesco Porpiglia, Homayoun Zargar,
Alexandre Mottrie, Riccardo Autorino. Partial
Nephrectomy Versus Radical Nephrectomy for Clinical
T1b and T2 Renal Tumors: A Systematic Review
and Meta-analysis of Comparative Studies.
Eur Urol. 2017;71:606–17
We commend Mir et al for their systematic review (SR)
that attempted to clarify uncertainties in the optimal
surgical treatment (partial nephrectomy [PN] or radical
nephrectomy [RN]) for patients with large (T1b–T2) renal
masses. However, their decision to combine the results of
individual retrospective studies with a high risk of selection
bias and confounding in a meta-analysis (MA) is methodo-
logically flawed andmay result in erroneous andmisleading
conclusions. There are imbalances between the PN and RN
groups in the studies included with regard to age,
comorbidities, tumor size and stage, and biological behav-
ior. These imbalances may have a greater impact on patient
outcome than the choice of PN or RN.
When used appropriately, MA can provide the best
estimate of a treatment effect size from the pooled
weighted averages of the results from individual trials.
MAs of well-conducted randomized controlled trials (RCTs)
are based on the assumption that each trial provides an
unbiased estimate of the treatment effect, so that the overall
combined effect will provide a more precise, unbiased
estimate of the treatment effect.
By contrast, observational studies produce estimates
that may deviate from the truth owing to the effects of
confounding factors, the influence of bias, or both . MAs
of observational studies will simply compound these errors
and provide pooled estimates that may be seriously
misleading, and even wrong. Unfortunately, the results of
such an MA may be interpreted as having more credibility
than it deserves. Instead, a careful analysis of the possible
sources of methodological, clinical, and statistical hetero-
geneity between studies in the SR with a narrative rather
than a quantitative synthesis of the evidence will provide
more insight than calculation of an overall measure of effect
that is likely to be biased . Our concern is especially
relevant as several examples of important selection biases
in assessing the comparative effectiveness of PN versus RN
have recently been presented and may be responsible
for the conflicting evidence between MAs and the RCT on
this topic
High-quality SRs andMAs are vital for clinicians, patients,
guideline developers, and policy makers in order to guide
clinical practice, patient decisions, and health care policies.
For underpinning of treatment recommendations in the
European Association of Urology (EAU) guidelines, the EAU
Guidelines Methods Committee advises against the use of
MA for synthesis of nonrandomized studies. In this case, an
SR with a ‘‘narrative synthesis’’ should be used to summarize
and explore relationships within and between studies.
As acknowledged by the authors , it is clear that the
existing evidence cannot reliably determine the optimum
surgical management for patients with large (>4 cm) renal
masses. Ideally, a carefully designed RCT should be
conducted with clear definition of the eligibility criteria
in terms not only of tumor size but also patient
characteristics (such as baseline renal function and
comorbidities), tumor characteristics (such as anatomic
complexity), and surgical technique (such as resection
technique and type and duration of ischemia), and
measuring clinically and patient important outcomes using
standardized definitions and measures.
Conflicts of interest: The authors have nothing to disclose.
References
[1]
.
[2]
[3]
[4]
.
[5]
[1_TD$DIFF] .
ava ilable at
journal homepage:
DOI of original article: .
0302-2838/# 2017 Published by Elsevier B.V. on behalf of European Association of Urology.