Copyright 2016 American Medical Association. All rights reserved.
Letters
RESEARCH LETTER
Five-Year Survival After Endosonography
vs Mediastinoscopy for Mediastinal Nodal Staging
of Lung Cancer
Lung cancer accounts for the highest cancer-related mortal-
ity rate worldwide.
1
Accurate mediastinal nodal staging is cru-
cial in the management of non–small cell lung cancer (NSCLC)
because it directs therapy and has prognostic value.
2,3
The Assessment of Surgical Staging vs Endosonographic
Ultrasound in Lung Cancer (ASTER) trial compared mediasti-
noscopy (surgical staging) with an endosonographic staging
strategy (which combined the use of endobronchial and trans-
esophageal ultrasound fol-
lowed by mediastinoscopy if
negative).
4
The endosono-
graphic strategy was signifi-
cantly more sensitive for diagnosing mediastinal nodal me-
tastases than surgical staging (94% endosonographic strategy
vs 79% surgical strategy).
If mediastinal staging is improved, more patients should
receive optimal treatment and might survive longer. The cur-
rent post hoc analysis evaluated survival in ASTER.
Methods | At inclusion in ASTER, all participants provided
written informed consent; the current analysis was either
approved or waived by the involved ethical committees. Of
241 patients with potentially resectable NSCLC, 123 were ran-
domized to endosonographic staging and 118 to surgical stag-
ing in 4 tertiary referral centers in Leiden (the Netherlands),
Ghent and Leuven (Belgium), and Cambridge (United
Kingdom) between February 2007 and April 2009.
4
Surgical-
pathological staging was the reference standard for mediasti-
nal nodal assessment.
Between June 30, 2015, and October 15, 2015, survival
data were obtained through patient records, death registers,
or contact with general practitioners (trial protocol in the
Supplement).
The proportion of survivors at 5 years for both staging
strategies and odds ratios (ORs) with 95% CIs were calcu-
lated. Kaplan-Meier analysis was performed and hazard
ratios were calculated to compare survival between the strat-
egies, adjusting for mediastinal nodal metastases in a Cox
proportional hazards model. Survival for patients with no
date of death were censored on the date last known to be
alive. The assumption of proportional hazard was tested and
met. Subgroup analysis was performed for patients with
nodal stages N2/N3 and N0/N1. Data were analyzed using
SPSS Statistics (IBM), version 22.0.
Results | Survival data at 5 years were obtained for 237 of 241 pa-
tients (98%); 2 patients in both groups were lost to follow-up.
There were 182 men (77%) with a mean age at randomization
of 65 years (SD, 9). Detailed patient characteristics were previ-
ously reported.
4
The prevalence of mediastinal nodal metasta-
ses was 54% in the endosonographic strategy group and 44%
in the surgical strategy group.
Survival at 5 years was 35% (42 of 121 patients) for the
endosonographic strategy vs 35% (41 of 116 patients) for the
surgical strategy (OR, 0.97 [95% CI, 0.57-1.66]) (Table). The
estimated median survival was 31 months (95% CI, 21-41)
for the endosonographic strategy vs 33 months (95% CI,
23-43) for the surgical strategy (adjusted hazard ratio, 0.98
[95% CI, 0.73-1.32]) (Figure).
In the subgroup with N2/N3 metastases, survival was
17% (11 of 64 patients) in the endosonographic strategy vs
19% (10 of 52 patients) in the surgical strategy (OR, 0.87
[95% CI, 0.34-2.25]). In the subgroup with N0/N1 metasta-
ses, survival was 54% (31 of 57 patients) for the endosono-
Supplemental content at
jama.com
Table. Survival Among Patients With Lung Cancer in the
Endosonographic vs the Surgical (Mediastinoscopy) Staging Strategies
a
Survival at 5 Years
No./Total No. (%) Odds Ratio (95% CI)
Overall
Endosonographic
staging
42/121 (35)
0.97 (0.57-1.66)
Surgical staging 41/116 (35)
N2/N3
Endosonographic
staging
11/64 (17) 0.87 (0.34-2.25)
Surgical staging 10/52 (19)
N0/N1
Endosonographic
staging
31/57 (54)
1.27 (0.62-2.60)
Surgical staging 31/64 (48)
Estimated Survival
Duration,
Median (95% CI), mo
Unadjusted Mortality,
Hazard Ratio
(95% CI)
b
Overall
Endosonographic
staging
31 (21-41)
1.04 (0.77-1.40)
Surgical staging 33 (23-43)
N2/N3
Endosonographic
staging
21 (15-27)
1.04 (0.70-1.55)
Surgical staging 22 (15-27)
N0/N1
Endosonographic
staging
72 (38-106)
0.91 (0.57-1.44)
Surgical staging 57 (30-84)
a
The endosonographic staging strategy combined the use of endobronchial
and transesophageal ultrasound, followed by mediastinoscopy if negative.
b
Adjusted for mediastinal nodal metastases status (N0/1 vs N2/3), the mortality
hazard ratio was 0.98 (95% CI, 0.73-1.32).
1110 JAMA September 13, 2016 Volume 316, Number 10 (Reprinted) jama.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/09/2022
Copyright 2016 American Medical Association. All rights reserved.
graphic strategy vs 48% (31 of 64 patients) for the surgical
strategy (OR, 1.27 [95% CI, 0.62-2.60]).
Discussion | No survival difference was found 5 years follow-
ing randomization to an endosonographic or surgical stag-
ing strategy for patients with NSCLC. Since the original
results of ASTER were published, clinical guidelines on lung
cancer management underwent major revisions and now
advocate endosonography instead of mediastinoscopy as
the initial step for mediastinal nodal staging.
2,3
The endo-
sonographic strategy is more accurate, less invasive, and
reduces unnecessary thoracotomies.
4
Data from a recent randomized trial show prolonged
survival in patients who underwent endosonography com-
pared with conventional staging.
5
However, most patients
in the latter group underwent bronchoscopy instead of
mediastinoscopy.
Why did improved mediastinal staging not lead to
improved survival? Missing data occurred in less than 2%
and therefore are an unlikely source of bias. However,
ASTER was powered to detect a difference in diagnostic sen-
sitivity, not survival, as reflected by the wide confidence
intervals. If a survival difference between the strategies
exists, it is likely to be small and a larger sample size may be
needed to detect it. However, randomized trials to detect a
survival difference based on staging strategy are not likely
to be conducted as the endosonographic strategy is now
advised in clinical guidelines.
2,3
Jolanda C. Kuijvenhoven, MD
Daniël A. Korevaar, MD
Kurt G. Tournoy, MD, PhD
Thomas L. A. Malfait, MD
Christophe Dooms, MD, PhD
Robert C. Rintoul, FRCP, PhD
Jouke T. Annema, MD, PhD
Author Affiliations: Depar tment of Respiratory Medicine, Academic Medical
Center, Amsterdam, the Netherlands (Kuijvenhoven, Annema); Department of
Clinical Epidemiology, Biostatistics, and Bioinformatics, Academic Medical
Center, Amsterdam, the Netherlands (Korevaar); Department of Respiratory
Medicine, Onze-Lieve-Vrouw Hospital, Aalst, Belgium (Tournoy); Department of
Respiratory Medicine, University Hospital Ghent, Ghent, Belgium (Malfait);
Department of Respiratory Medicine, Leuven University Hospitals, Leuven,
Belgium (Dooms); Papworth Clinical Trials Unit Collaboration, Papworth
Hospital, Cambridge, United Kingdom (Rintoul).
Corresponding Author: Jouke T. Annema, MD, PhD, Department of Respiratory
Medicine, Academic Medical Center, Meibergdreef 9, 1105 AZ, Amsterdam,
the Netherlands (j.t.annema@amc.uva.nl).
Author Contributions: Dr Kuijvenhoven had full access to all of the data in the
study and takes responsibility for the integrity of the data and the accuracy of
the data analysis.
Concept and design: Kuijvenhoven, Tournoy, Annema.
Acquisition, analysis, or interpretation of data: All Authors.
Drafting of the manuscript: Kuijvenhoven, Korevaar, Tournoy, Annema.
Critical revision of the manuscript for important intellectual content: Tournoy,
Malfait, Dooms, Rintoul, Annema.
Statistical analysis: Kuijvenhoven, Korevaar, Tournoy.
Administrative, technical, or material support: Malfait.
Study supervision: Tournoy, Annema.
No additional contributions: Dooms, Rintoul.
Conflict of Interest Disclosures: All authors have completed and submitted the
ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Annema reports
receiving material and financial support for educational endobronchial and
esophageal ultrasound courses from Hitachi, Pentax, COOK and Symbionix to the
department of respiratory medicine at his institution. Dr Rintoul reports receiving
financial support for educational endobronchial and esophageal ultrasound
courses from Olympus to his institution. No other disclosures were reported.
Funding/Support: The ASTER trial was supported by local support for data
collection at Ghent University Hospital, the Zorgprogramma Oncologie Gent,
and by the National Institute for Health Research Cambridge Biomedical
Research Centre (Dr Rintoul). Data collection in Papworth Hospital was
supported by the UK National Health Service R&D Health Technology
Assessment Program (project No. 06/302/216). No specific funding
was sought for this post hoc analysis.
Role of the Funder/Sponsor: The funders of the original ASTER trial had no role
in the design and conduct of this post hoc analysis; collection, management,
analysis, and interpretation of the data; preparation, review, or approval of the
manuscript; and decision to submit the manuscript for publication.
Trial Registration: clinicaltrials.gov Identifier: NCT00432640
1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer
statistics, 2012. CA Cancer J Clin. 2015;65(2):87-108.
2. Vilmann P, Clementsen PF, Colella S, et al. Combined endobronchial and
oesophageal endosonography for the diagnosis and staging of lung cancer. Eur
Respir J. 2015;46(1):40-60.
Figure. Survival Among Patients With Lung Cancer in the Endosonographic vs Surgical Staging Strategies
1.0
0.8
0.6
0.4
0.2
0
0
118
121
100
1
0
Cumulative Survival
Months After Randomization
No. at risk
Surgical staging
Endosonographic staging
60
40
41
80
23
22
40
52
52
20
77
76
Surgical staging
Endosonographic staging
Adjusted for mediastinal nodal
metastases status (N0/N1
vs N2/N3) (adjusted hazard ratio,
0.98 [95% CI, 0.73-1.32]).
The median duration of follow-up
was 33 months (interquartile range
[IQR], 13-76) for surgical staging
and 31 months (IQR, 13-75) for
endosonographic staging.
Letters
jama.com (Reprinted) JAMA September 13, 2016 Volume 316, Number 10 1111
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/09/2022
Copyright 2016 American Medical Association. All rights reserved.
3. Silvestri GA, Gonzalez AV, Jantz MA, et al. Methods for staging non-small cell
lung cancer. Chest. 2013;143(5 suppl):e211S-250S.
4. Annema JT, van Meerbeeck JP, Rintoul RC, et al. Mediastinoscopy vs
endosonography for mediastinal nodal staging of lung cancer. JAMA. 2010;304
(20):2245-2252.
5. Navani N, Nankivell M, Lawrence DR, et al. Lung cancer diagnosis
and staging with endobronchial ultrasound-guided transbronchial needle
aspiration compared with conventional approaches. Lancet Respir Med. 2015;3
(4):282-289.
COMMENT & RESPONSE
Sodium Excretion, Cardiovascular Disease,
and Chronic Kidney Disease
To the Editor In the study by Mills and colleagues,
1
high uri-
nary sodium excretion was associated with increased car-
diovascular disease (CVD) risk in patients with chronic kid-
ney disease (CKD). Patients were divided into 4 groups
based on quartiles of calibrated urinary sodium excretion
(<2894 mg/24 hours; 2894-3649 mg/24 hours; 3650-4547
mg/24 hours; and ≥4548 mg/24 hours) and were followed
up for a median of 6.8 years. The cumulative incidence of
CVD for each group from lowest to highest urinary sodium
excretion was 18.4%, 16.5%, 20.6%, and 29.8%, respec-
tively. After multivariable adjustment, no significant asso-
ciation was found between urinary potassium excretion and
CVD events.
The authors did not mention whether there was an
interaction between sodium and potassium excretion for
the composite outcome measure.
2
A urinary sodium to
potassium excretion ratio might yield a different association
with CVD risk.
3
Also, they did not evaluate CVD mortality risk in their
study. In a study of patients with established CVD or diabe-
tes mellitus, O’Donnell and colleagues
2
found an increased
risk of CVD with urinary sodium excretion of more than
7000 mg/24 hours and, surprisingly, an increased risk of
cardiovascular mortality at urinary sodium excretion of less
than 3000 mg/24 hours. Additionally, higher urinary potas-
sium excretion rates were associated with a decreased risk
of stroke. Although Mills and colleagues did not evaluate
cardiovascular mortality and their study population was dif-
ferent from the patients in the study by O’Donnell and col-
leagues, the results of increasing CVD risk with higher
sodium excretion are similar. Further studies are needed
before using these findings in the management of such
patients.
Mehmet Hursitoglu, MD
Author Affiliation: Department of Internal Medicine, Bakirkoy Dr Sadi Konuk
Training and Research Hospital, Istanbul, Turkey.
Corresponding Author: Mehmet Hursitoglu, MD, Department of Internal
Medicine, Bakirkoy Dr Sadi Konuk Training and Research Hospital, Bakirkoy,
Istanbul 34100, Turkey (hursitoglum@yahoo.com).
Conflict of Interest Disclosures: The author has completed and submitted the
ICMJE Form for Disclosure of Potential Conflicts of Interest and none were
reported.
1. Mills KT, Chen J, Yang W, et al; Chronic Renal Insufficiency Cohort (CRIC)
Study Investigators. Sodium excretion and the risk of cardiovascular disease in
patients with chronic kidney disease. JAMA. 2016;315(20):2200-2210.
2. O’Donnell MJ, Yusuf S, Mente A, et al. Urinary sodium and potassium
excretion and risk of cardiovascular events. JAMA. 2011;306(20):2229-2238.
3. Hedayati SS, Minhajuddin AT, Ijaz A, et al. Association of urinary
sodium/potassium ratio with blood pressure: sex and racial differences. Clin J
Am Soc Nephrol. 2012;7(2):315-322.
To the Editor Mills and colleagues
1
found that among patients
with CKD, higher urinary sodium excretion was associated with
increased risk of CVD. Analyses were adjusted for important
covariates for CVD. In all the models, a significantly in-
creased risk of CVD was documented in patients with the high-
est quartile of sodium excretion.
One of the variables that was not included in the statisti-
cal analysis was urinary protein excretion. In Table 1 in the
article, where the characteristics of patients were described,
urinary protein excretion was correlated with urinary sodium
excretion. This correlation has been described previously by
the same group in the same cohort of patients.
2
It is widely accepted that urinary albumin excretion is an
independent predictor of cardiovascular morbidity and mor-
tality in patients with CKD and in the general population.
3,4
Therefore, considering the interaction between urinary
sodium excretion and proteinuria, urinary protein excretion
should be included in the statistical analysis.
Jaume Almirall, MD
Author Affiliation: Department of Nephrology, Parc Tauli Sabadell, Hospital
Universitari, Barcelona, Spain.
Corresponding Author: Jaume Almirall, MD, Nephrology Department ,
Parc Tauli Sabadell, Hospital Universitari, Sabadell, Barcelona 08208, Spain
(jalmirall@tauli.cat).
Conflict of Interest Disclosures: The author has completed and submitted the
ICMJE Form for Disclosure of Potential Conflicts of Interest and none were
reported.
1. Mills KT, Chen J, Yang W, et al; Chronic Renal Insufficiency Cohort (CRIC)
Study Investigators. Sodium excretion and the risk of cardiovascular disease in
patients with chronic kidney disease. JAMA. 2016;315(20):2200-2210.
2. Weir MR, Townsend RR, Fink JC, et al. Urinary sodium is a potent correlate of
proteinuria: lessons from the chronic renal insufficiency cohort study. Am J
Nephrol. 2012;36(5):397-404.
3. Hemmelgarn BR, Manns BJ, Lloyd A, et al; Alberta Kidney Disease Network.
Relation between kidney function, proteinuria, and adverse outcomes. JAMA.
2010;303(5):423-429.
4. Hillege HL, Fidler V, Diercks GF, et al; Prevention of Renal and Vascular End
Stage Disease (PREVEND) Study Group. Urinary albumin excretion predicts
cardiovascular and noncardiovascular mortality in general population. Circulation.
2002;106(14):1777-1782.
In Reply In response to Dr Hursitoglu, we found that the uri-
nary sodium to potassium excretion ratio was not signifi-
cantly associated with CVD in our study (P for trend = .11).
This is likely due to the lack of an inverse association
between urinary potassium and CVD among patients with
CKD. We additionally adjusted for urinary potassium excre-
tion in a multivariable model and the results were not sig-
nificantly changed (Table).
We have previously reported that urinary sodium excre-
tion was positively and significantly associated with all-
cause mortality in patients with CKD.
1
However, cause-
specific mortality data are not yet available in our study.
Letters
1112 JAMA September 13, 2016 Volume 316, Number 10 (Reprinted) jama.com
Copyright 2016 American Medical Association. All rights reserved.
Downloaded From: https://jamanetwork.com/ on 08/09/2022