Premium
When to perform lymph node dissection in patients with renal cell carcinoma: a novel approach to the preoperative assessment of risk of lymph node invasion at surgery and of lymph node progression during follow‐up
Author(s) -
Capitanio Umberto,
Abdollah Firas,
Matloob Rayan,
Suardi Nazareno,
Castiglione Fabio,
Di Trapani Ettore,
Capogrosso Paolo,
Gallina Andrea,
Dell'Oglio Paolo,
Briganti Alberto,
Salonia Andrea,
Montorsi Francesco,
Bertini Roberto
Publication year - 2013
Publication title -
bju international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.773
H-Index - 148
eISSN - 1464-410X
pISSN - 1464-4096
DOI - 10.1111/bju.12125
Subject(s) - medicine , lymph node , renal cell carcinoma , nephrectomy , dissection (medical) , nomogram , logistic regression , stage (stratigraphy) , lymph , urology , radiology , odds ratio , surgery , oncology , kidney , pathology , paleontology , biology
Objective To identify preoperatively patients who might benefit from lymph node dissection ( LND ).Patients and Methods We assessed lymph node invasion ( LNI ) at final pathology and lymph node ( LN ) progression during the follow‐up for 1983 patients with RCC , treated with either partial or radical nephrectomy. LN progression was defined as the onset of a new clinically detected lymphadenopathy (>10 mm) in the retroperitoneal lymphatic area. Logistic regression analyses were used to assess the effect of each potential clinical predictor (age, body mass index, tumour side, symptoms, performance status, clinical tumour size, clinical tumour‐node‐metastasis stage, and albumin, calcium, creatinine, haemoglobin and platelet levels) on the outcome of interest. The most parsimonious multivariable predictive model was developed, and discrimination, calibration and net benefit were calculated.Results The prevalence of LNI was 6.1% (120/1983 patients) and during the follow‐up period, 82 patients (4.1%) experienced LN progression. On multivariable analyses, the most informative independent predictors were tumour stage ( cT3–4 vs cT1–2 , odds ratio [ OR ] 1.52, P = 0.05), clinical nodal status [ cN1 vs cN0 , OR 7.09, P < 0.001], metastases at diagnosis ( OR 3.04, P < 0.001) and clinical tumour size ( OR 1.14, P < 0.001). The accuracy of the multivariable model was found to be 86.9%, with excellent calibration and net benefit at decision‐curve analyses.Conclusions By relying on a unique approach, combining the risk of harbouring LNI and/or LN progression during the follow‐up period, we have provided the first clinical presurgery model predicting the need for LND .