- Email: [email protected]
Variability in invasive mediastinal staging for lung cancer: A multicenter regional study
Accepted Manuscript Variability in Invasive Mediastinal Staging for Lung Cancer: A Multi-Center Regional Study Lucas W. Thornblade, MD MPH, Douglas E. Wood, MD, Michael S. Mulligan, MD, Alexander S. Farivar, MD, Michal Hubka, MD, Kimberly E. Costas, MD, Bahirathan Krishnadasan, MD, Farhood Farjah, MD, MPH PII:
S0022-5223(18)30311-8
DOI:
10.1016/j.jtcvs.2017.12.138
Reference:
YMTC 12557
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 17 May 2017 Revised Date:
28 November 2017
Accepted Date: 14 December 2017
Please cite this article as: Thornblade LW, Wood DE, Mulligan MS, Farivar AS, Hubka M, Costas KE, Krishnadasan B, Farjah F, The SCOAP-CERTAIN Collaborative, Variability in Invasive Mediastinal Staging for Lung Cancer: A Multi-Center Regional Study, The Journal of Thoracic and Cardiovascular Surgery (2018), doi: 10.1016/j.jtcvs.2017.12.138. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
TITLE: Variability in Invasive Mediastinal Staging for Lung Cancer: A Multi-Center Regional
2
Study
3
AUTHORS: Lucas W. Thornblade, MD MPH1, Douglas E. Wood, MD1, Michael S. Mulligan,
4
MD1, Alexander S. Farivar, MD2, Michal Hubka, MD3, Kimberly E. Costas, MD4, Bahirathan
5
Krishnadasan, MD5, Farhood Farjah, MD, MPH1 and The SCOAP-CERTAIN Collaborative
6
AUTHOR AFFILIATION:
SC
RI PT
1
1. Department of Surgery, University of Washington, Seattle, WA
8
2. Division of Thoracic Surgery, Swedish Cancer Institute, Seattle, WA
9
3. Department of Thoracic Surgery, Virginia Mason Medical Center, Seattle, WA
M AN U
7
10
4. Division of Thoracic Surgery, Providence Regional Medical Center, Everett, WA
11
5. General Thoracic Surgery, CHI Franciscan Health System, Tacoma, WA CONFLICTS OF INTEREST:
13
None
14
SORCE OF FUNDING:
15
This study was not funded. The University of Washington’s Surgical Outcomes Research Center
16
and Comparative Effectiveness Research Translational Network has received funding from the
17
Agency for Healthcare Research and Quality, Life Sciences Discovery Fund, and National
18
Institutes of Health.
19
CORRESPONDENCE ADDRESS:
20
Farhood Farjah, MD MPH
21
Surgical Outcomes Research Center
AC C
EP
TE D
12
ACCEPTED MANUSCRIPT 2
1107 NE 45th Street, Suite 502
23
Box 354808
24
Seattle, WA 98105
25
Telephone: 206-598-4477
26
Fax: 206-616-9032
27
Email: [email protected]
28
IRB: Date 6/6/16, IRB #52136
29
ARTICLE WORD COUNT: 3449
30
MEETING AFFILIATION: An abstract based on this manuscript was presented orally at the
31
2017 Western Thoracic Surgical Association Annual Meeting
AC C
EP
TE D
M AN U
SC
RI PT
22
ACCEPTED MANUSCRIPT 3
GLOSSARY OF ABBREVIATIONS:
33
ACCP – American College of Chest Physicians
34
ASA – American Society of Anesthesiologists
35
CERTAIN – Comparative Effectiveness Research Translation Network
36
CI – confidence interval
37
CINCO – Collaborative to Improve Native Cancer Outcomes
38
DLCO – Diffusion capacity of the lungs for carbon monoxide
39
EBUS – Endobronchial ultrasound guided nodal aspirate
40
EUS – Endoscopic ultrasound guided nodal aspirate
41
FEV1 – Forced expiratory volume in one second
42
IMS – Invasive mediastinal staging
43
IQR – Interquartile range
44
NCCN – National Comprehensive Cancer Network
45
NSCLC – Non-small cell lung cancer
46
QI – quality improvement
47
SCOAP – Surgical Care Outcomes Assessment Program
48
SD – Standard deviation
49
STS – Society of Thoracic Surgeons
AC C
EP
TE D
M AN U
SC
RI PT
32
ACCEPTED MANUSCRIPT 4
WA - Washington
51
CENTRAL PICTURE LEGEND: Variation in Hospital-Level Rates of Invasive Mediastinal
52
Staging
53
CENTRAL MESSAGE: Rates of invasive mediastinal staging vary significantly across
54
hospitals providing surgical care to lung cancer patients.
55
PERSPECTIVE STATEMENT: Our study demonstrates what many surgeons suspect—lung
56
cancer staging varies significantly across hospitals. This finding provides the critical first step in
57
a line of investigation that seeks to better understand the reasons underlying this highly variable
58
care. Better understanding variability in lung cancer staging may inform quality improvement
59
initiatives or lead to novel staging algorithms.
M AN U
SC
RI PT
50
AC C
EP
TE D
60
ACCEPTED MANUSCRIPT 5
ABSTRACT:
62
Objective(s): Prior studies have reported underutilization of—but not variability in–invasive
63
mediastinal staging in the pre-treatment evaluation of lung cancer patients. We sought to
64
compare rates of invasive mediastinal staging for lung cancer across hospitals participating in a
65
regional quality improvement and research collaborative.
66
Methods: We conducted a retrospective study (2011-2013) of resected lung cancer patients from
67
the Surgical Clinical Outcomes and Assessment Program in Washington State. Invasive
68
mediastinal staging included mediastinoscopy and/or endobronchial/esophageal ultrasound-
69
guided nodal aspiration. We used a mixed-effects model to mitigate the influence of small
70
sample sizes at any one hospital on rates of invasive staging and to adjust for hospital-level
71
differences in the frequency of clinical stage IA disease.
72
Results: A total of 406 patients (mean age 68 years, 69% clinical stage IA, 67% lobectomy)
73
underwent resection at five hospitals (four community, one academic). Invasive staging occurred
74
in 66% of patients (95% confidence interval [CI] 61-71%). CI inspection revealed that two
75
hospitals performed invasive staging significantly more often than the overall average (94%,
76
[95% CI: 89-96%] and 84% [95% CI: 78-88%]), whereas two hospitals performed invasive
77
staging significantly less often than overall average (31% [95% CI: 21-44%] and 17% [95% CI:
78
7-36%]).
79
Conclusions: Rates of invasive mediastinal staging varied significantly across hospitals
80
providing surgical care for lung cancer patients. Future studies that aim to understand the reasons
81
underlying variability in care may inform quality improvement initiatives or lead to the
82
development of novel staging algorithms.
AC C
EP
TE D
M AN U
SC
RI PT
61
ACCEPTED MANUSCRIPT 6
AC C
EP
TE D
M AN U
SC
RI PT
83
ACCEPTED MANUSCRIPT 7
84
INTRODUCTION: An American College of Surgeons Commission on Cancer study described substantially
86
lower than expected rates of invasive mediastinal staging (IMS) for resected lung cancer (27%)
87
in 2001.1 In addition, fewer than half of all patients who underwent mediastinoscopy had lymph
88
node tissue submitted for pathologic evaluation. The late Dr. Carolyn Reed said in her discussion
89
of this paper at the annual meeting of the Society of Thoracic Surgeons (STS) that “The results
90
are an indictment of the present care of patients with non-small cell lung cancer.”1
SC
Since then numerous studies have reported apparent underutilization of IMS in other lung
M AN U
91
RI PT
85
cancer populations using other data sources2–8; however, there are several challenges in drawing
93
firm conclusions about the quality of care based on these studies. First, contemporary guidelines
94
have shifted away from recommending routine IMS.9–11 Second, there are no known datasets
95
with sufficient granularity to directly assess adherence with guideline recommended IMS.
96
Variation in care delivery is a commonly used surrogate measure for suboptimal care.
97
Nonetheless, there may be other reasons for variability in care, such as concerns over the level of
98
evidence supporting guideline recommendations or providers who perceive IMS to be time-
99
consuming, inconvenient, or even dangerous. These speculative reasons form the basis of a long-
EP
TE D
92
standing suspicion of variability in rates of IMS. Interestingly, variability in hospital-level rates
101
of IMS has never been documented.
102
AC C
100
Leveraging a unique data source of academic and community-based thoracic surgical
103
practices in the Puget Sound region of Washington State (WA), we sought to compare hospital-
104
level rates of IMS among resected non-small lung cancer (NSCLC) patients. We hypothesized
105
that there would be significant variation in rates of IMS between hospitals.
106
MATERIALS AND METHODS:
ACCEPTED MANUSCRIPT 8
107 108
Study Design and Population We performed a retrospective cohort study of adult NSCLC patients who underwent resection between July 2011 and December 2013 (n=514). Patients were excluded if they had a
110
prior history of lung cancer (n=43) or were treated with induction therapy (n=59). Data were
111
obtained from the medical record of hospitals/health systems in the Puget Sound region of
112
Washington (WA) State participating in the Collaborative to Improve Native Cancer Outcomes
113
(CINCO) Study.12 Of the 13 CINCO hospitals (Appendix A), seven contributed lung cancer
114
cases. Two hospitals contributing fewer than 10 cases over the duration of the study were
115
excluded from the analysis resulting in the exclusion of 6 additional patients. This study was not
116
considered human subjects research by the University of Washington (UW) Institutional Review
117
Board because both patient and hospital information were de-identified.
118
Data Source
SC
M AN U
Data were obtained from the Surgical Care and Outcomes Assessment Program
TE D
119
RI PT
109
(SCOAP). SCOAP is a physician-led quality improvement system in Washington State. The
121
SCOAP database is a prospective, multicenter clinical registry that contains data on patients’
122
characteristics, perioperative and surgical details, as well as clinical outcomes. SCOAP is
123
administered by the Foundation for Health Care Quality. Research and development in SCOAP
124
is funded in part by grants to the UW’s Surgical Outcomes Research Center and the Comparative
125
Effectiveness Research Translation Network (CERTAIN).13,14 In 2013, CERTAIN developed a
126
thoracic surgery quality improvement module for use by hospitals that participate in SCOAP.
127
Stakeholders include thoracic surgeons, pulmonologists, advanced practice providers, and
128
hospital administrators.12 This group determined the need to collect information about lung
129
cancer staging and added disease-specific variables (e.g. provider documented clinical stage,
AC C
EP
120
ACCEPTED MANUSCRIPT 9
130
details of IMS listed in the next paragraph, use of non-invasive staging modalities, etc.) to the
131
lung cancer module within SCOAP.
132
Invasive Mediastinal Staging Abstractors recorded information from patients’ medical records on the type (e.g.
RI PT
133
mediastinoscopy/mediastinotomy, endobronchial ultrasound-guided nodal aspirate [EBUS];
135
esophageal ultrasound-guided nodal aspirate [EUS]) and timing (before resection, at the time of
136
resection, or both before and at the time of resection) of IMS if it was performed. In addition,
137
abstractors collected data on the number of unique mediastinal nodal stations sampled during
138
IMS. Abstractors did not collect information on the indications for IMS, and therefore it is not
139
possible to determine guideline adherence at the patient-level.
140
Analysis
M AN U
141
SC
134
Patient characteristics were summarized by frequencies (categorical variables) and means with standard deviation (SD) or medians with interquartile ranges (IQR) for normally and non-
143
normally distributed continuous variables, respectively. Univariate differences between hospitals
144
are summarized by Χ2 tests (categorical variable) and Analysis of Variance tests or Kruskal-
145
Wallis tests for normally and non-normally distributed continuous variables, respectively. All
146
analyses were performed using Stata v14.2 (StataCorp LLC, College Station, TX).
EP
Confidence interval (CI) inspection was used to compare hospital-level rates of IMS to
AC C
147
TE D
142
148
each other and the overall average rate for the study population. Because small sample sizes at
149
any one hospital can influence the observed rate of IMS and erroneously inflate the magnitude of
150
variation, we used an empirical Bayes shrinkage estimator.15–17 This approach, using a mixed
151
effects logistic regression model, also allowed for adjustment for clinical stage IA tumors.
152
Guidelines allow for omission of IMS in a subset of these patients,11,18 and the frequency of
ACCEPTED MANUSCRIPT 10
clinical stage IA disease may vary across hospitals; therefore, adjustment for clinical stage IA is
154
imperative to avoid confounding bias in comparisons of hospital-level rates of IMS. Finally, this
155
approach allowed for estimation of the proportion of variability explained by differences in the
156
rates of clinical stage IA disease across hospitals (by calculating the ratio of random effects
157
between models). CIs for point estimates following reliability adjustment were based on
158
Empirical Bayes variance estimates and on approximate normality of the posterior facility-level
159
effects. A post-hoc analysis calculated the median odds ratio (MOR) as another measure of
160
between-hospital variation in IMS.19 Missing data analysis revealed no missing values among
161
variables included in risk and reliability adjustment.
SC
M AN U
162
RI PT
153
A pre-specified sensitivity analysis was conducted in two different subgroups in an attempt to evaluate hospital-level variability in IMS rates among populations more likely to be
164
candidates for IMS. Specifically, one subgroup analysis was restricted to patients who underwent
165
preoperative positron emission tomography (PET) on the presumption that the provider had a
166
higher index of suspicion of NSCLC. Another subgroup analysis restricted to patients who
167
underwent an anatomic resection on the presumption that surgeons preferentially perform an
168
anatomic resection for cases that are certain to be NSCLC.
169
171
RESULTS:
AC C
170
EP
TE D
163
Between 2011 and 2013, 406 patients met criteria for this study from across five hospitals
172
in the Puget Sound region of Washington State. Clinical and treatment variables are summarized
173
in Table 1. Preoperative patient factors that varied across hospitals included insurance case-mix
174
(p<0.001), ASA class (p<0.01), and clinical stage IA disease (p<0.001). Age, sex, race,
175
comorbidity index, and lung function did not vary significantly across hospitals (all p>0.05).
ACCEPTED MANUSCRIPT 11
Operative factors (extent and approach to resection) varied significantly across hospitals
177
(p≤0.001), as did pathologic stage (p=0.008), but histology did not (p=0.95). Nine percent of
178
patients were ultimately diagnosed with pathologic N2 disease.
179
RI PT
176
Overall, 268 patients underwent IMS (66%, 95% CI: 61-71%). Mediastinoscopy
accounted for a majority (85%) of IMS procedures, and a majority (64%) of patients underwent
181
IMS on the day of resection. The median number of mediastinal lymph node stations sampled
182
was 3 (range 0-7). Seven percent of patients (95% CI: 4-10%) underwent IMS but did not have
183
lymph node tissue sampled based on pathologic evaluation. Of the 18 patients who did not have
184
lymph node tissue sampled, 11 underwent mediastinoscopy, six underwent EBUS, and one
185
underwent both mediastinoscopy and EBUS. Table 2 summarizes patterns of IMS across
186
hospitals in this study. The type of IMS modality varied across hospitals (p<0.001), as did the
187
timing of IMS (p<0.001). Unadjusted rates of IMS varied significantly across hospitals
188
(p<0.001). Hospital B performed IMS without any lymph node tissue sampled two times more
189
often (14%) than the overall average.
TE D
M AN U
SC
180
Adjusted rates of IMS varied widely ranging from 17% (95% CI: 7-36%) at Hospital E
191
to 94% (95% CI: 89-96%) at Hospital A (Figure 1). Hospitals A, B, and D had markedly higher
192
rates of IMS compared to Hospitals C and E. Hospital A and B performed IMS significantly
193
more often than the overall average, whereas hospitals C and E performed IMS significantly less
194
often than the overall average. Adjusting for clinical stage IA disease explained only 3% of the
195
variability across hospitals, despite large variation in the frequency of clinical stage IA disease
196
across hospitals. Our pre-specified subgroup analyses revealed the same degree of variability in
197
hospital-level rates of IMS among patients who underwent anatomic resection or had a PET
198
(Figures 2-3).
AC C
EP
190
ACCEPTED MANUSCRIPT 12
199
Several post-hoc analyses were conducted. As an alternative measure of variability in hospital-level rates of IMS, we estimated the MOR across hospitals to be 4.99 (95% CI 1.56-
201
7.24) indicating significant between-hospital variance. This value is interpreted as follows: if a
202
randomly selected patient from one center were to transfer care to a hospital with a higher
203
probability of performing IMS, then his or her risk of undergoing IMS would increase by 499%.
204
Table 3 summarizes a post-hoc analysis exploring potential differences in patient characteristics
205
between those who did and did not undergo IMS. Compared with those who did not receive IMS,
206
patients who underwent IMS had a different payer mix (p<0.001) and were more likely to have a
207
higher ASA class (p=0.002), smoke (p=0.002), have a lower predicted FEV1 (p=0.03), have a
208
preoperative PET scan (p<0.001), and have a higher clinical stage (p=0.001). A “fully adjusted”
209
model did not change the conclusions of our primary analysis, and it revealed that patient-level
210
variables accounted for a total of 5% of the variation in rates of IMS across hospitals. Another
211
patient-level stratified analysis showed that patients with clinical stage IB or higher disease had
212
higher rates of IMS compared to those with clinical stage IA (78% versus 61%, p<0.001).
213
Finally, hospitals with an exclusively thoracic surgical-based practice model (A, B, and D) had
214
higher rates of IMS compared to hospitals with a mixed-specialty practice model (82% versus
215
19%, p<0.001).
216
DISCUSSION:
SC
M AN U
TE D
EP
AC C
217
RI PT
200
We hypothesized that there would be significant variation in hospital-level rates of
218
invasive mediastinal staging (IMS) and aimed to compare hospital-specific rates of IMS to each
219
other and to the overall population mean. The key finding of our study is that there is significant
220
variation in rates of IMS across hospitals providing surgical care to NSCLC patients.
221
Surprisingly, patient-level factors, including clinical stage IA disease, explain very little of the
ACCEPTED MANUSCRIPT 13
222
variability. Post-hoc analyses confirm our key finding. Hospitals with a purely thoracic surgical
223
practice (as opposed to a mixed-specialty practice) were associated with higher rates of IMS.
224
Variability in care is often considered a marker of poor quality care. Surgeons may not adhere to guidelines because they are unaware that they exist. Those with a practice solely
226
dedicated to thoracic surgery may be more likely to know guideline recommended indications
227
for IMS. Supporting this hypothesis is the observation that an exclusively thoracic surgical-based
228
hospital practice model was associated with higher rates of IMS. However, a previous study
229
examining the relationship between surgeon specialty and outcomes reported no differences in
230
the utilization of mediastinoscopy among general, cardiothoracic, and thoracic surgeons.20
231
Survey research may better elucidate the frequency of surgeon awareness of guideline
232
recommendations and factors associated with both awareness and adherence.
SC
M AN U
233
RI PT
225
Another possible explanation for variation in IMS is that surgeons may purposefully choose to not adhere to guidelines because they do not accept the recommendations to be valid.
235
A previous study showed that providers may deviate from guidelines when either the level of
236
supporting evidence for guidelines is low or when they perceive the supporting evidence to not
237
be generalizable to their patient population or practice.21 Indeed the level of evidence for
238
guideline recommended indications for IMS is 2.9–11. There are no randomized controlled trials
239
(RCTs) comparing varying intensities of IMS (i.e. routine versus selective versus highly
240
selective IMS). It was recently shown that guideline recommended IMS has a sensitivity and
241
specificity of 100% and 35%, respectively.22,23 Clinically, these estimates mean that existing
242
guideline recommendations perfectly select all patients with true nodal disease to undergo IMS,
243
but they also select many patients without true nodal disease to undergo IMS. The high rate of
244
negative IMS may provide the rationale for why some surgeons do not accept guideline
AC C
EP
TE D
234
ACCEPTED MANUSCRIPT 14
recommendations. A high “true negative” rate of IMS may be perceived as futile, inefficient, or
246
cost-ineffective24, although recent evidence suggests these perceptions may not be correct.25
247
However, underlying the rationale for guidelines directing a majority of lung cancer patients to
248
IMS is the view that the identification of mediastinal nodal disease (N2 or N3, Stage IIIA or
249
IIIB) is important from a patient and treatment selection perspective. Even if the multi-modality
250
options for stage IIIA are equivalent with regards to long-term survival,26 many patients and
251
providers would likely agree that more accurate staging leads to more informed treatment
252
decisions. Nonetheless, some providers may be skeptical of this view and believe that patients
253
are better served by initial surgical resection.
SC
M AN U
254
RI PT
245
Better understanding the reasons underlying hospital-level variability in IMS is an opportunity to improve care. Qualitative research (focus groups, key informant surgeon
256
interviews, etc.) can help identify the reasons for variability in practice patterns. To the extent
257
that a knowledge gap explains variability in IMS, hospital credentialing and privileging policies
258
could be an avenue for evaluating and increasing the knowledge base of surgeons who do not
259
have a dedicated oncologic practice and/or board certification in thoracic surgery.27 Greater
260
access to continuing medical education—such as the GAIN curriculum sponsored by the
261
American College of Chest Physicians, National Comprehensive Cancer Network, and others—
262
is another opportunity to address knowledge gaps.28 Finally, clinical registries (e.g. STS,
263
National Surgical Quality Improvement Program) can facilitate performance feedback in terms
264
of process compliance (i.e. benchmarked rates of IMS).29–31
EP
AC C
265
TE D
255
On the other hand, if the variability in IMS is due to widespread disagreement about the
266
optimal indications for IMS, then more research will be needed to elucidate the optimal IMS
267
strategy. One example of an alternative staging strategy is selective IMS using prediction
ACCEPTED MANUSCRIPT 15
models. Preliminary research suggests that prediction models can, similar to existing practice
269
guidelines, maintain 100% sensitivity while achieving higher specificity.22,32 Clinically, this
270
means that the use of a prediction model is expected to lead to fewer IMS procedures without
271
compromising the ability to detect true nodal disease prior to first treatment. This hypothesis
272
could also be tested through an RCT. Still other alternative staging strategies may arise from
273
qualitative interviews with thoracic surgeons and pulmonologists.
One unexpected finding was that the rate of IMS in this study (66%) was higher than
SC
274
RI PT
268
previously reported (27%).1 Differences across studies may reflect differences in the quality of
276
data collection or delivery of care over time. Another possibility is the Hawthorne Effect given
277
that all hospitals agreed to participate in a regional quality improvement collaborative. However,
278
this explanation is unlikely because the decision to measure utilization of IMS was made by our
279
regional collaborative in 2013, the study period spans 2011 to 2013, and there were no
280
interventions intended to influence rates of IMS. Furthermore, rates of IMS for the entire study
281
population were lower than expected population rates of IMS derived from prior studies (77-
282
79%).22,23,33 In addition, although rates of failed lymph node sampling (7%) were much lower
283
than that previously reported (~47%),1 opportunities for QI remain given that one hospital failed
284
sample lymph node tissue in up to 14% of patients undergoing IMS—a rate double that of other
285
hospitals. Failing to submit or obtain adequate lymph node tissue is an outcome measure that
286
could be monitored by a clinical registry such as the STS General Thoracic Surgery
287
Database.30,31
TE D
EP
AC C
288
M AN U
275
There are important limitations to this study. First there may be concerns about
289
generalizing regional results to the rest of the nation. By design, however, this multicenter study
290
included mostly community hospitals and only a single academic center in an attempt to
ACCEPTED MANUSCRIPT 16
maximize generalizability. We expect hospital-level variability in IMS to be found across the
292
nation, and several members of our study team are currently testing this hypothesis using the
293
STS General Thoracic Surgery Database (even though this database also has generalizability
294
concerns). Second, because we were unable to directly access medical records to determine the
295
indications for or against IMS, we could not measure adherence with guideline recommendations
296
at the patient level. Third, because we were unable to measure central versus peripheral tumors,
297
some of the variation in IMS may be explained by differences in case-mix (defined by central
298
tumors) across hospitals. Fourth, we did not have information on the provider who performed or
299
decided against performing IMS, and therefore we can better understand the contribution of
300
individual providers or provider-specialty to variation in care. Fifth, we may have
301
underestimated rates of IMS. Data abstractors may not have been able to record IMS occurring at
302
a referring hospital/health system (i.e. not participating in our collaborative). However, given the
303
fact that rates of IMS are higher than previously described in community practice, it is unlikely
304
that we significantly underestimated rates of IMS. Finally, we were unable to elucidate the
305
reasons for variable rates of IMS across hospitals. As discussed earlier we plan to conduct
306
qualitative and mixed-methods research to gain a better understanding of the determinants of
307
variability in IMS.
SC
M AN U
TE D
EP
In conclusion, we describe—for the first time—that there is significant variability in rates
AC C
308
RI PT
291
309
of IMS between hospitals. We speculate that this variability might be due to gaps in the quality
310
of care, disagreement over the best indications for IMS, or both. Planned qualitative and mixed-
311
methods studies will likely reveal the reasons underlying variability leading to either quality
312
improvement and educational interventions and/or trials comparing novel IMS strategies to
313
guideline recommended staging.
ACCEPTED MANUSCRIPT 17
AC C
EP
TE D
M AN U
SC
RI PT
314
ACCEPTED MANUSCRIPT 18
315 316
ACKNOWLEDGEMENTS The authors would like to thank the Surgical Care and Outcomes Assessment Program for providing data for this research. Also, special thanks to Rebecca G. Symons, Megan
318
Zadworny, Hao He, PhD, and Katherine Odem-Davis, PhD at the Surgical Outcomes Research
319
Center for their help with data acquisition and statistical consultation. Research reported in this
320
publication was supported by the National Institute of Diabetes and Digestive and Kidney
321
Diseases of the National Institutes of Health under Award Number T32DK070555. The content
322
is solely the responsibility of the authors and does not necessarily represent the official views of
323
the National Institutes of Health.
AC C
EP
M AN U
TE D
324
SC
RI PT
317
ACCEPTED MANUSCRIPT
REFERENCES:
326 327
1.
Little AG, Rusch VW, Bonner JA, et al. Patterns of Surgical Care of Lung Cancer Patients. Ann Thorac Surg. 2005;80(6):2051-2056. doi:10.1016/j.athoracsur.2005.06.071.
328 329 330
2.
Little AG, Gay EG, Gaspar LE, Stewart AK. National survey of non-small cell lung cancer in the United States: Epidemiology, pathology and patterns of care. Lung Cancer. 2007;57(3):253-260. doi:10.1016/j.lungcan.2007.03.012.
331 332 333
3.
Farjah F, Flum DR, Ramsey SD, Heagerty PJ, Symons RG, Wood DE. Multi-Modality Mediastinal Staging for Lung Cancer Among Medicare Beneficiaries. J Thorac Oncol. 2009;4(3):355-363. doi:10.1097/JTO.0b013e318197f4d9.
334 335 336
4.
Lathan CS, Neville BA, Earle CC. The effect of race on invasive staging and surgery in non-small-cell lung cancer. J Clin Oncol. 2006;24(3):413-418. doi:10.1200/JCO.2005.02.1758.
337 338 339
5.
Vest MT, Tanoue L, Soulos PR, et al. Thoroughness of Mediastinal Staging in Stage IIIA Non-small Cell Lung Cancer. J Thorac Oncol. 2012;7(1):188-195. doi:10.1097/JTO.0b013e318236ecbb.
340 341 342
6.
Ost DE, Niu J, Elting LS, Buchholz TA, Giordano SH. Determinants of practice patterns and quality gaps in lung cancer staging and diagnosis. Chest. 2014;145(5):1097-1113. doi:10.1378/chest.13-1628.
343 344 345
7.
Ost DE, Niu J, Elting LS, Buchholz TA, Giordano SH. Quality gaps and comparative effectiveness in lung cancer staging and diagnosis. Chest. 2014;145(2):331-345. doi:10.1378/chest.13-1599.
346 347 348
8.
Faris N, Yu X, Sareen S, et al. Preoperative Evaluation of Lung Cancer in a Community Health Care Setting. Ann Thorac Surg. 2015;100(2):394-400. doi:10.1016/j.athoracsur.2015.03.008.
349 350
9.
Network NCC. Non-Small Cell Lung Cancer (Version 5.2017). https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. Accessed March 18, 2017.
351 352
10.
Silvestri GA, Gonzalez A V., Jantz MA, et al. Methods for Staging Non-small Cell Lung Cancer. Chest. 2013;143(5):e211S-e250S. doi:10.1378/chest.12-2355.
353 354 355
11.
356 357
12.
358 359 360 361
13.
Devine EB, Capurro D, van Eaton E, et al. Preparing Electronic Clinical Data for Quality Improvement and Comparative Effectiveness Research: The SCOAP CERTAIN Automation and Validation Project. EGEMS (Washington, DC). 2013;1(1):1025. doi:10.13063/2327-9214.1025.
362
14.
Flum DR, Alfonso-Cristancho R, Devine EB, et al. Implementation of a “real-world”
EP
TE D
M AN U
SC
RI PT
325
AC C
19
Ettinger DS, Akerley W, Borghaei H, et al. Non–Small Cell Lung Cancer, Version 2.2013. J Natl Compr Cancer Netw . 2013;11(6):645-653. http://www.jnccn.org/content/11/6/645.abstract. CERTAIN. CINCO. http://www.becertain.org/projects/cancer/cinco. Accessed March 14, 2017.
ACCEPTED MANUSCRIPT 20
learning health care system: Washington state’s Comparative Effectiveness Research Translation Network (CERTAIN). Surgery. 2014;155(5):860-866. doi:10.1016/j.surg.2014.01.004.
363 364 365
15.
Dimick JB, Staiger DO, Birkmeyer JD. Ranking hospitals on surgical mortality: The importance of reliability adjustment. Health Serv Res. 2010;45(6 PART 1):1614-1629. doi:10.1111/j.1475-6773.2010.01158.x.
369 370 371
16.
Dimick JB, Ghaferi A a, Osborne NH, Ko CY, Hall BL. Reliability adjustment for reporting hospital outcomes with surgery. Ann Surg. 2012;255(4):703-707. doi:10.1097/SLA.0b013e31824b46ff.
372 373 374
17.
MacKenzie TA, Grunkemeier GL, Grunwald GK, et al. A primer on using shrinkage to compare in-hospital mortality between centers. Ann Thorac Surg. 2015;99(3):757-761. doi:10.1016/j.athoracsur.2014.11.039.
375 376 377
18.
Detterbeck FC, Jantz MA, Wallace M, Vansteenkiste J, Silvestri GA. Invasive mediastinal staging of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest. 2007;132(3 SUPPL.). doi:10.1378/chest.07-1362.
378 379 380
19.
Larsen K, Merlo J. Appropriate assessment of neighborhood effects on individual health: Integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol. 2005;161(1):81-88. doi:10.1093/aje/kwi017.
381 382 383
20.
Farjah F, Flum DR, Varghese TK, Symons RG, Wood DE. Surgeon Specialty and LongTerm Survival After Pulmonary Resection for Lung Cancer. Ann Thorac Surg. 2009;87(4):995-1006. doi:10.1016/j.athoracsur.2008.12.030.
384 385 386
21.
Brouwers MC, Makarski J, Garcia K, et al. A mixed methods approach to understand variation in lung cancer practice and the role of guidelines. Implement Sci. 2014;9(1):36. doi:10.1186/1748-5908-9-36.
387 388 389
22.
Farjah F, Backhus LM, Varghese TK, et al. External validation of a prediction model for pathologic N2 among patients with a negative mediastinum by positron emission tomography. J Thorac Dis. 2015;7(4):576-584. doi:10.3978/j.issn.2072-1439.2015.02.09.
390 391 392
23.
Farjah F, Madtes DK, Wood DE, et al. Vascular endothelial growth factor C complements the ability of positron emission tomography to predict nodal disease in lung cancer. J Thorac Cardiovasc Surg. 2015;150(4):796-803. doi:10.1016/j.jtcvs.2015.08.001.
393 394 395
24.
396 397 398
25.
399 400 401 402
26.
AC C
EP
TE D
M AN U
SC
RI PT
366 367 368
Meyers BF, Haddad F, Siegel BA, et al. Cost-effectiveness of routine mediastinoscopy in computed tomography- and positron emission tomography-screened patients with stage I lung cancer. J Thorac Cardiovasc Surg. 2006;131(4). doi:10.1016/j.jtcvs.2005.10.045. Czarnecka-Kujawa K, Rochau U, Siebert U, et al. Cost-effectiveness of mediastinal lymph node staging in non–small cell lung cancer. J Thorac Cardiovasc Surg. 2017;153(6):15671578. doi:10.1016/j.jtcvs.2016.12.048. Martins RG, D’Amico TA, Loo Jr. BW, et al. The management of patients with stage IIIA non-small cell lung cancer with N2 mediastinal node involvement. J Natl Compr Canc Netw. 2012;10(5):599-613. http://www.ncbi.nlm.nih.gov/pubmed/22570291\nhttp://www.jnccn.org/content/10/5/599.l
ACCEPTED MANUSCRIPT 21
ong.
403
27.
Blackmon SH, Cooke DT, Whyte R, et al. The society of thoracic surgeons expert consensus statement: A tool kit to assist thoracic surgeons seeking privileging to use new technology and perform advanced procedures in general thoracic surgery. Ann Thorac Surg. 2016;101(3):1230-1237. doi:10.1016/j.athoracsur.2016.01.061.
408 409 410
28.
CHEST GAIN Regional non-small cell lung cancer summits. http://www.chestnet.org/Education/Advanced-Clinical-Training/CHEST-GAIN-NSCLC. Accessed November 4, 2017.
411 412
29.
ACS NSQIP Data Collection, Analysis, and Reporting. https://www.facs.org/qualityprograms/acs-nsqip? Accessed September 29, 2016.
413 414
30.
Grover FL, Shahian DM, Clark RE, Edwards FH. The STS National Database. Ann Thorac Surg. 2014;97(1):S48-S54. doi:10.1016/j.athoracsur.2013.10.015.
415 416
31.
Jacobs JP, Shahian DM, Prager RL, et al. Introduction to the STS National Database Series. Ann Thorac Surg. 2015;100(6):1992-2000. doi:10.1016/j.athoracsur.2015.10.060.
417 418 419 420
32.
Farjah F, Lou F, Sima C, Rusch VW, Rizk NP. A prediction model for pathologic N2 disease in lung cancer patients with a negative mediastinum by positron emission tomography. J Thorac Oncol. 2013;8(9):1170-1180. doi:10.1097/JTO.0b013e3182992421.
421 422
33.
Mazzone PJ, Vachani A, Chang A, et al. Quality indicators for the evaluation of patients with lung cancer. Chest. 2014;146(3):659-669. doi:10.1378/chest.13-2900.
SC
M AN U
TE D EP AC C
423
RI PT
404 405 406 407
ACCEPTED MANUSCRIPT 22
FIGURE LEGENDS:
425
Figure 1: Rates of Invasive Mediastinal Staging by Hospital Compared with Population Average
426
Figure 2: Rates of invasive mediastinal staging by hospital (among patients undergoing
427
anatomic resection)
428
Figure 3: Rates of invasive mediastinal staging by hospital (among patients with preoperative
429
PET)
SC
RI PT
424
M AN U
430
VIDEO LEGEND:
432
Dr. Farhood Farjah explains the significance of the research findings herein.
AC C
EP
TE D
431
ACCEPTED MANUSCRIPT
Table 1: Characteristics of patients undergoing resection for lung cancer across hospital Hospital All Hospital A B Hospital C n = 406 n = 130 n = 134 n = 72 Mean age, years [SD] 68.2 (10.0) 66.7 (10.6) 68.8 (9.9) 68.2 (10.0)
Hospital D n = 40 69.7 (8.8)
Hospital E n = 30 71.5 (8.5)
pvalue 0.12 0.05
Women, n (%)
228 (57)
70 (54)
84 (64)
38 (53)
16 (40)
20 (67)
Race, n (%) White Black Asian Native American Pacific Islander Other
354 (88) 7 (2) 26 (6) 4 (1) 3 (1) 8 (2)
109 (85) 4 (3) 9 (7) 3 (2) 1 (1) 2 (2)
116 (87) 2 (2) 10 (8) 0 (0) 2 (2) 3 (2)
64 (89) 1 (1) 5 (7) 0 (0) 0 (0) 2 (3)
37 (95) 0 (0) 0 (0) 1 (3) 0 (0) 1 (3)
28 (93) 0 (0) 2 (7) 0 (0) 0 (0) 0 (0)
Insurance, n (%) Commercial Medicare Medicaid Tricare (military) Other Government Uninsured Government & Commercial Mixed Government
145 (36) 77 (19) 13 (3) 2 (<1) 3 (1) 2 (<1) 126 (31) 38 (9)
39 (30) 9 (7) 5 (4) 1 (1) 2 (2) 1 (1) 63 (48) 10 (8)
49 (37) 16 (12) 5 (4) 0 (0) 0 (0) 0 (0) 44 (33) 20 (15)
44 (61) 27 (38) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0)
9 (23) 15 (38) 2 (5) 1 (3) 0 (0) 0 (0) 7 (18) 6 (15)
4 (13) 10 (33) 1 (3) 0 (0) 1 (3) 0 (0) 12 (40) 2 (7)
Charlson comorbidity index, n (%) 0 1 2 3+
227 (56) 141 (35) 28 (7) 10 (2)
73 (56) 42 (32) 12 (9) 3 (2)
69 (51) 50 (37) 11 (8) 4 (3)
47 (65) 21 (29) 1 (1) 3 (4)
21 (53) 15 (38) 4 (10) 0 (0)
17 (57) 13 (43) 0 (0) 0 (0)
ASA class, n (%) II III IV
71 (17) 312 (77) 23 (6)
12 (9) 107 (82) 11 (8)
25 (19) 99 (74) 10 (7)
20 (28) 51 (71) 1 (1)
10 (25) 29 (73) 1 (3)
4 (13) 26 (87) 0 (0)
M AN U
SC
<0.001
TE D
Smoking status, n (%) Former Current Never
RI PT
0.79
0.34
0.01
0.13
235 (60) 96 (25) 60 (16)
78 (63) 31 (25) 15 (12)
79 (60) 39 (30) 14 (11)
42 (63) 11 (16) 14 (21)
20 (53) 8 (21) 10 (26)
16 (53) 7 (23) 7 (23)
83 (69-96)
81 (66-95)
80 (66-94)
84 (75-100)
87 (73-100)
84 (68-103)
0.16
64 (52-77)
64 (55-75)
63 (49-80)
66 (51-78)
66 (62-83)
60 (50-70)
0.31
281 (69) 54 (13) 33 (8) 20 (5) 15 (4) 1 (<1) 2 (<1)
74 (57) 15 (12) 17 (13) 10 (8) 11 (8) 1 (1) 2 (2)
110 (82) 11 (8) 6 (5) 6 (5) 1 (1) 0 (0) 0 (0)
54 (75) 15 (21) 2 (3) 0 (0) 1 (1) 0 (0) 0 (0)
22 (55) 8 (20) 4 (10) 4 (10) 2 (5) 0 (0) 0 (0)
21 (70) 5 (17) 4 (13) 0 (0) 0 (0) 0 (0) 0 (0)
Preoperative PET scan, n (%)
371 (92)
125 (96)
120 (90)
64 (89)
38 (97)
24 (83)
Extent of Resection, n (%) Wedge Segmentectomy Lobectomy Sleeve lobectomy Lobectomy & wedge Bilobectomy Pneumonectomy
35 (9) 7 (2) 272 (67) 2 (<1) 66 (16) 14 (3) 10 (2)
19 (15) 3 (2) 71 (55) 1 (1) 19 (15) 9 (7) 8 (6)
9 (7) 2 (1) 102 (76) 0 (0) 18 (13) 1 (1) 2 (1)
5 (7) 0 (0) 46 (64) 0 (0) 17 (24) 4 (6) 0 (0)
1 (3) 2 (5) 28 (70) 1 (3) 8 (20) 0 (0) 0 (0)
1 (3) 0 (0) 25 (83) 0 (0) 4 (13) 0 (0) 0 (0)
MIS Approach (%)
292 (72)
84 (65)
129 (96)
41 (57)
33 (83)
5 (17)
<0.001
Right sided resection (%)
247 (61)
81 (62)
81 (60)
39 (54)
28 (70)
18 (60)
0.75
% Predicted FEV1, median (IQR)
AC C
Clinical stage, n (%) IA IB IIA IIB IIIA IIIB IV
EP
% Predicted DLCO, median (IQR)
<0.001
0.049 0.001
ACCEPTED MANUSCRIPT
Pathologic Stage, n (%) 0 IA IB IIA IIB IIIA IV
5 (1) 201 (50) 84 (21) 38 (9) 32 (8) 41 (10) 5 (1)
2 (2) 51 (39) 24 (18) 15 (12) 15 (12) 22 (17) 1 (1)
1 (1) 79 (59) 28 (21) 11 (8) 8 (6) 5 (4) 2 (1)
1 (1) 40 (56) 17 (24) 5 (7) 2 (3) 7 (10) 0 (0)
Pathologic Nodal Stage, n (%) pN0 pN1 pN2
317 (79) 47 (12) 36 (9)
93 (73) 18 (14) 17 (13)
110 (83) 17 (13) 5 (4)
55 (79) 7 (10) 8 (11)
0.008 0 (0) 20 (50) 6 (15) 1 (3) 5 (13) 7 (18) 1 (3)
1 (3) 11 (37) 9 (30) 6 (20) 2 (7) 0 (0) 1 (3)
RI PT
0.053
32 (80) 2 (5) 6 (15)
27 (90) 3 (10) 0 (0)
AC C
EP
TE D
M AN U
SC
Histology (%) 0.95 Adenocarcinoma 270 (67) 82 (63) 87 (65) 49 (68) 30 (75) 22 (73) AIS 10 (2) 5 (4) 3 (2) 1 (1) 0 (0) 1 (3) Squamous cell carcinoma 86 (21) 30 (23) 30 (22) 12 (17) 9 (23) 5 (17) Large cell carcinoma 7 (2) 3 (2) 2 (1) 2 (3) 0 (0) 0 (0) NSCLC NOS 22 (5) 6 (5) 9 (7) 6 (8) 0 (0) 1 (3) Other 11 (3) 4 (3) 3 (2) 2 (3) 1 (3) 1 (3) Adenocarcinoma in situ (AIS); American Society of Anesthesiologists (ASA); Forced expiratory volume in one second (FEV1); Diffusion capacity of carbon monoxide (DLCO); Minimally-invasive surgery (video-assisted thoracoscopic surgery or robotic ); Nonsmall cell lung cancer (NSCLC); Not otherwise specified (NOS); Length-of-stay (LOS); Interquartile range (IQR), Positron Emission Tomography (PET)
ACCEPTED MANUSCRIPT
Any invasive mediastinal staging, n (%)
268 (66)
119 (92)
100 (75)
16 (22)
Timing of Invasive Staging*, n (%) Before Resection At the time of resection Both before and at the time of resection
97 (36) 147 (55) 24 (9)
3 (3) 95 (80) 21 (18)
87 (87) 12 (12) 1 (1)
4 (25) 10 (63) 2 (13)
Hospital E n = 30
p-value
<0.001 27 (90) 3 (10) 0 (0) 0 (0) 0 (0) 0 (0)
RI PT
Table 2: Patterns of invasive cancer staging for patients undergoing resection of lung cancer across hospitals Hospital Hospital Hospital Hospital All A B C D n = 406 n = 128 n = 132 n = 70 n = 40 Invasive mediastinal staging, n (%) None 138 (34) 11 (8) 34 (25) 56 (78) 10 (25) Mediastinoscopy/Mediastinotomy only 228 (56) 96 (74) 87 (65) 13 (18) 29 (73) EBUS only 9 (2) 2 (2) 5 (4) 1 (1) 1 (3) EUS only 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) EBUS & Mediastinoscopy/Mediastinotomy 30 (7) 21 (16) 7 (5) 2 (3) 0 (0) EBUS & EUS & 1 (<1) 0 (0) 1 (1) 0 (0) 0 (0) Mediastinoscopy/Mediastinotomy 30 (75)
3 (10)
1 (3) 29 (97) 0 (0)
2 (67) 1 (33) 0 (0)
<0.001
SC
<0.001
AC C
EP
TE D
M AN U
Number of nodal stations sampled* Mean (SD) 3.0 (1.2) 3.2 (1.0) 2.7 (1.4) 2.9 (1.3) 3.2 (1.4) 2.7 (0.6) 0.05 Median (range) 3 (0-7) 3 (0-7) 3 (0-5) 3 (0-5) 3 (0-5) 3 (2-3) 0.34 No nodes sampled, n (%) 18 (7) 1 (1) 14 (14) 1 (6) 2 (6) 0 (0) 0.004 Endobronchial ultrasound guided nodal aspirate (EBUS); Esophageal ultrasound guided nodal aspirate (EUS); Standard deviation (SD); Interquartile range (IQR); Collaborative to Improve Native Cancer Outcomes (CINCO) *Calculated among patients who underwent invasive mediastinal staging
ACCEPTED MANUSCRIPT
Table 3: Characteristics of patients receiving and not receiving invasive staging. All Invasive staging n = 406 n=265 Mean age, years [SD] 68.2 (10.0) 68.0 (10.4)
No invasive staging n=137 68.8 (9.3)
p-value
0.95
Women, n (%)
228 (57)
150 (56)
78 (57)
Race, n (%) White Black Asian Native American Pacific Islander Other
354 (88) 7 (2) 26 (6) 4 (1) 3 (1) 8 (2)
231 (87) 6 (2) 17 (6) 4 (2) 3 (1) 4 (2)
123 (90) 1 (1) 9 (7) 0 (0) 0 (0) 4 (3)
Insurance, n (%) Commercial Medicare Medicaid Tricare (military) Other Government Uninsured Government & Commercial Mixed Government
145 (36) 77 (19) 13 (3) 2 (<1) 3 (1) 2 (<1) 126 (31) 38 (9)
89 (33) 35 (13) 11 (4) 2 (1) 3 (1) 1 (<1) 97 (36) 30 (11)
46 (41) 42 (30) 2 (1) 0 (0) 0 (0) 1 (1) 29 (21) 8 (6)
0.48
RI PT
0.33
ASA class, n (%) II III IV
71 (17) 312 (77) 23 (6)
Smoking status, n (%) Former Current Never
235 (60) 96 (25) 60 (16)
M AN U
227 (56) 141 (35) 28 (7) 10 (2)
148 (55) 90 (34) 22 (8) 8 (3)
79 (57) 51 (37) 6 (4) 2 (1)
34 (13) 217 (81) 17 (6)
37 (27) 95 (69) 6 (4)
TE D
Charlson comorbidity index, n (%) 0 1 2 3+
SC
<0.001
0.36
0.002
0.002
159 (62) 70 (27) 28 (11)
76 (57) 26 (19) 32 (24)
80 (69-94)
85 (72-100)
0.03
63 (51-76)
66 (56-80)
0.15
281 (69) 54 (13) 33 (8) 20 (5) 15 (4) 1 (<1) 2 (<1)
170 (63) 36 (13) 25 (9) 20 (7) 14 (5) 1 (<1) 2 (1)
111 (80) 18 (13) 8 (6) 0 (0) 1 (1) 0 (0) 0 (0)
Preoperative PET scan, n (%)
371 (92)
255 (96)
116 (85)
Extent of Resection, n (%) Wedge Segmentectomy Lobectomy Sleeve lobectomy Lobectomy & wedge Bilobectomy Pneumonectomy
35 (9) 7 (2) 272 (67) 2 (<1) 66 (16) 14 (3) 10 (2)
20 (7) 6 (2) 180 (67) 0 (0) 40 (15) 12 (4) 10 (4)
15 (11) 1 (1) 92 (67) 2 (1) 26 (19) 2 (1) 0 (0)
MIS Approach (%)
292 (72)
201 (75)
91 (66)
83 (69-96)
% Predicted DLCO, median (IQR)
64 (52-77)
AC C
Clinical stage, n (%) IA IB IIA IIB IIIA IIIB IV
EP
% Predicted FEV1, median (IQR)
0.001
<0.001 0.02
0.05
ACCEPTED MANUSCRIPT
Right sided resection (%)
247 (61)
163 (61)
84 (61)
Pathologic Stage, n (%) 0 IA IB IIA IIB IIIA IV
5 (1) 201 (50) 84 (21) 38 (9) 32 (8) 41 (10) 5 (1)
1 (<1) 121 (45) 56 (21) 26 (10) 26 (10) 34 (13) 4 (1)
4 (3) 80 (58) 28 (20) 12 (9) 6 (4) 7 (5) 1 (1)
0.01
RI PT
Pathologic Nodal Stage, n (%) pN0 pN1 pN2
0.77
0.003
317 (79) 47 (12) 36 (9)
198 (74) 37 (14) 31 (12)
119 (89) 10 (7) 5 (4)
AC C
EP
TE D
M AN U
SC
Histology (%) 0.049 Adenocarcinoma 270 (67) 174 (65) 96 (70) AIS 10 (2) 5 (2) 5 (4) Squamous cell carcinoma 86 (21) 67 (25) 19 (14) Large cell carcinoma 7 (2) 5 (2) 2 (1) NSCLC NOS 22 (5) 10 (4) 12 (9) Other 11 (3) 7 (3) 4 (3) Adenocarcinoma in situ (AIS); American Society of Anesthesiologists (ASA); Forced expiratory volume in one second (FEV1); Diffusion capacity of carbon monoxide (DLCO); Minimally-invasive surgery (video-assisted thoracoscopic surgery or robotic ); Non-small cell lung cancer (NSCLC); Not otherwise specified (NOS); Length-of-stay (LOS); Interquartile range (IQR), Positron Emission Tomography (PET)
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1
SUPPLEMENTAL MATERIAL:
2
Appendix A—Participating Hospitals
3
MultiCare Allenmore Hospital
4
MultiCare Good Samaritan Hospital
5
Harborview Medical Center
6
Peace Health St. Joseph Medical Center
7
Providence Regional Medical Center Everett
8
St. Joseph Medical Center
9
St. Francis Hospital
M AN U
St. Anthony Hospital
11
St. Clare Hospital
12
St. Elizabeth Hospital
13
MultiCare Tacoma General Hospital
14
University of Washington Medical Center
15
Virginia Mason Medical Center
AC C
EP
TE D
10
16
SC
RI PT
1
S0022-5223(18)30311-8
DOI:
10.1016/j.jtcvs.2017.12.138
Reference:
YMTC 12557
To appear in:
The Journal of Thoracic and Cardiovascular Surgery
Received Date: 17 May 2017 Revised Date:
28 November 2017
Accepted Date: 14 December 2017
Please cite this article as: Thornblade LW, Wood DE, Mulligan MS, Farivar AS, Hubka M, Costas KE, Krishnadasan B, Farjah F, The SCOAP-CERTAIN Collaborative, Variability in Invasive Mediastinal Staging for Lung Cancer: A Multi-Center Regional Study, The Journal of Thoracic and Cardiovascular Surgery (2018), doi: 10.1016/j.jtcvs.2017.12.138. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
TITLE: Variability in Invasive Mediastinal Staging for Lung Cancer: A Multi-Center Regional
2
Study
3
AUTHORS: Lucas W. Thornblade, MD MPH1, Douglas E. Wood, MD1, Michael S. Mulligan,
4
MD1, Alexander S. Farivar, MD2, Michal Hubka, MD3, Kimberly E. Costas, MD4, Bahirathan
5
Krishnadasan, MD5, Farhood Farjah, MD, MPH1 and The SCOAP-CERTAIN Collaborative
6
AUTHOR AFFILIATION:
SC
RI PT
1
1. Department of Surgery, University of Washington, Seattle, WA
8
2. Division of Thoracic Surgery, Swedish Cancer Institute, Seattle, WA
9
3. Department of Thoracic Surgery, Virginia Mason Medical Center, Seattle, WA
M AN U
7
10
4. Division of Thoracic Surgery, Providence Regional Medical Center, Everett, WA
11
5. General Thoracic Surgery, CHI Franciscan Health System, Tacoma, WA CONFLICTS OF INTEREST:
13
None
14
SORCE OF FUNDING:
15
This study was not funded. The University of Washington’s Surgical Outcomes Research Center
16
and Comparative Effectiveness Research Translational Network has received funding from the
17
Agency for Healthcare Research and Quality, Life Sciences Discovery Fund, and National
18
Institutes of Health.
19
CORRESPONDENCE ADDRESS:
20
Farhood Farjah, MD MPH
21
Surgical Outcomes Research Center
AC C
EP
TE D
12
ACCEPTED MANUSCRIPT 2
1107 NE 45th Street, Suite 502
23
Box 354808
24
Seattle, WA 98105
25
Telephone: 206-598-4477
26
Fax: 206-616-9032
27
Email: [email protected]
28
IRB: Date 6/6/16, IRB #52136
29
ARTICLE WORD COUNT: 3449
30
MEETING AFFILIATION: An abstract based on this manuscript was presented orally at the
31
2017 Western Thoracic Surgical Association Annual Meeting
AC C
EP
TE D
M AN U
SC
RI PT
22
ACCEPTED MANUSCRIPT 3
GLOSSARY OF ABBREVIATIONS:
33
ACCP – American College of Chest Physicians
34
ASA – American Society of Anesthesiologists
35
CERTAIN – Comparative Effectiveness Research Translation Network
36
CI – confidence interval
37
CINCO – Collaborative to Improve Native Cancer Outcomes
38
DLCO – Diffusion capacity of the lungs for carbon monoxide
39
EBUS – Endobronchial ultrasound guided nodal aspirate
40
EUS – Endoscopic ultrasound guided nodal aspirate
41
FEV1 – Forced expiratory volume in one second
42
IMS – Invasive mediastinal staging
43
IQR – Interquartile range
44
NCCN – National Comprehensive Cancer Network
45
NSCLC – Non-small cell lung cancer
46
QI – quality improvement
47
SCOAP – Surgical Care Outcomes Assessment Program
48
SD – Standard deviation
49
STS – Society of Thoracic Surgeons
AC C
EP
TE D
M AN U
SC
RI PT
32
ACCEPTED MANUSCRIPT 4
WA - Washington
51
CENTRAL PICTURE LEGEND: Variation in Hospital-Level Rates of Invasive Mediastinal
52
Staging
53
CENTRAL MESSAGE: Rates of invasive mediastinal staging vary significantly across
54
hospitals providing surgical care to lung cancer patients.
55
PERSPECTIVE STATEMENT: Our study demonstrates what many surgeons suspect—lung
56
cancer staging varies significantly across hospitals. This finding provides the critical first step in
57
a line of investigation that seeks to better understand the reasons underlying this highly variable
58
care. Better understanding variability in lung cancer staging may inform quality improvement
59
initiatives or lead to novel staging algorithms.
M AN U
SC
RI PT
50
AC C
EP
TE D
60
ACCEPTED MANUSCRIPT 5
ABSTRACT:
62
Objective(s): Prior studies have reported underutilization of—but not variability in–invasive
63
mediastinal staging in the pre-treatment evaluation of lung cancer patients. We sought to
64
compare rates of invasive mediastinal staging for lung cancer across hospitals participating in a
65
regional quality improvement and research collaborative.
66
Methods: We conducted a retrospective study (2011-2013) of resected lung cancer patients from
67
the Surgical Clinical Outcomes and Assessment Program in Washington State. Invasive
68
mediastinal staging included mediastinoscopy and/or endobronchial/esophageal ultrasound-
69
guided nodal aspiration. We used a mixed-effects model to mitigate the influence of small
70
sample sizes at any one hospital on rates of invasive staging and to adjust for hospital-level
71
differences in the frequency of clinical stage IA disease.
72
Results: A total of 406 patients (mean age 68 years, 69% clinical stage IA, 67% lobectomy)
73
underwent resection at five hospitals (four community, one academic). Invasive staging occurred
74
in 66% of patients (95% confidence interval [CI] 61-71%). CI inspection revealed that two
75
hospitals performed invasive staging significantly more often than the overall average (94%,
76
[95% CI: 89-96%] and 84% [95% CI: 78-88%]), whereas two hospitals performed invasive
77
staging significantly less often than overall average (31% [95% CI: 21-44%] and 17% [95% CI:
78
7-36%]).
79
Conclusions: Rates of invasive mediastinal staging varied significantly across hospitals
80
providing surgical care for lung cancer patients. Future studies that aim to understand the reasons
81
underlying variability in care may inform quality improvement initiatives or lead to the
82
development of novel staging algorithms.
AC C
EP
TE D
M AN U
SC
RI PT
61
ACCEPTED MANUSCRIPT 6
AC C
EP
TE D
M AN U
SC
RI PT
83
ACCEPTED MANUSCRIPT 7
84
INTRODUCTION: An American College of Surgeons Commission on Cancer study described substantially
86
lower than expected rates of invasive mediastinal staging (IMS) for resected lung cancer (27%)
87
in 2001.1 In addition, fewer than half of all patients who underwent mediastinoscopy had lymph
88
node tissue submitted for pathologic evaluation. The late Dr. Carolyn Reed said in her discussion
89
of this paper at the annual meeting of the Society of Thoracic Surgeons (STS) that “The results
90
are an indictment of the present care of patients with non-small cell lung cancer.”1
SC
Since then numerous studies have reported apparent underutilization of IMS in other lung
M AN U
91
RI PT
85
cancer populations using other data sources2–8; however, there are several challenges in drawing
93
firm conclusions about the quality of care based on these studies. First, contemporary guidelines
94
have shifted away from recommending routine IMS.9–11 Second, there are no known datasets
95
with sufficient granularity to directly assess adherence with guideline recommended IMS.
96
Variation in care delivery is a commonly used surrogate measure for suboptimal care.
97
Nonetheless, there may be other reasons for variability in care, such as concerns over the level of
98
evidence supporting guideline recommendations or providers who perceive IMS to be time-
99
consuming, inconvenient, or even dangerous. These speculative reasons form the basis of a long-
EP
TE D
92
standing suspicion of variability in rates of IMS. Interestingly, variability in hospital-level rates
101
of IMS has never been documented.
102
AC C
100
Leveraging a unique data source of academic and community-based thoracic surgical
103
practices in the Puget Sound region of Washington State (WA), we sought to compare hospital-
104
level rates of IMS among resected non-small lung cancer (NSCLC) patients. We hypothesized
105
that there would be significant variation in rates of IMS between hospitals.
106
MATERIALS AND METHODS:
ACCEPTED MANUSCRIPT 8
107 108
Study Design and Population We performed a retrospective cohort study of adult NSCLC patients who underwent resection between July 2011 and December 2013 (n=514). Patients were excluded if they had a
110
prior history of lung cancer (n=43) or were treated with induction therapy (n=59). Data were
111
obtained from the medical record of hospitals/health systems in the Puget Sound region of
112
Washington (WA) State participating in the Collaborative to Improve Native Cancer Outcomes
113
(CINCO) Study.12 Of the 13 CINCO hospitals (Appendix A), seven contributed lung cancer
114
cases. Two hospitals contributing fewer than 10 cases over the duration of the study were
115
excluded from the analysis resulting in the exclusion of 6 additional patients. This study was not
116
considered human subjects research by the University of Washington (UW) Institutional Review
117
Board because both patient and hospital information were de-identified.
118
Data Source
SC
M AN U
Data were obtained from the Surgical Care and Outcomes Assessment Program
TE D
119
RI PT
109
(SCOAP). SCOAP is a physician-led quality improvement system in Washington State. The
121
SCOAP database is a prospective, multicenter clinical registry that contains data on patients’
122
characteristics, perioperative and surgical details, as well as clinical outcomes. SCOAP is
123
administered by the Foundation for Health Care Quality. Research and development in SCOAP
124
is funded in part by grants to the UW’s Surgical Outcomes Research Center and the Comparative
125
Effectiveness Research Translation Network (CERTAIN).13,14 In 2013, CERTAIN developed a
126
thoracic surgery quality improvement module for use by hospitals that participate in SCOAP.
127
Stakeholders include thoracic surgeons, pulmonologists, advanced practice providers, and
128
hospital administrators.12 This group determined the need to collect information about lung
129
cancer staging and added disease-specific variables (e.g. provider documented clinical stage,
AC C
EP
120
ACCEPTED MANUSCRIPT 9
130
details of IMS listed in the next paragraph, use of non-invasive staging modalities, etc.) to the
131
lung cancer module within SCOAP.
132
Invasive Mediastinal Staging Abstractors recorded information from patients’ medical records on the type (e.g.
RI PT
133
mediastinoscopy/mediastinotomy, endobronchial ultrasound-guided nodal aspirate [EBUS];
135
esophageal ultrasound-guided nodal aspirate [EUS]) and timing (before resection, at the time of
136
resection, or both before and at the time of resection) of IMS if it was performed. In addition,
137
abstractors collected data on the number of unique mediastinal nodal stations sampled during
138
IMS. Abstractors did not collect information on the indications for IMS, and therefore it is not
139
possible to determine guideline adherence at the patient-level.
140
Analysis
M AN U
141
SC
134
Patient characteristics were summarized by frequencies (categorical variables) and means with standard deviation (SD) or medians with interquartile ranges (IQR) for normally and non-
143
normally distributed continuous variables, respectively. Univariate differences between hospitals
144
are summarized by Χ2 tests (categorical variable) and Analysis of Variance tests or Kruskal-
145
Wallis tests for normally and non-normally distributed continuous variables, respectively. All
146
analyses were performed using Stata v14.2 (StataCorp LLC, College Station, TX).
EP
Confidence interval (CI) inspection was used to compare hospital-level rates of IMS to
AC C
147
TE D
142
148
each other and the overall average rate for the study population. Because small sample sizes at
149
any one hospital can influence the observed rate of IMS and erroneously inflate the magnitude of
150
variation, we used an empirical Bayes shrinkage estimator.15–17 This approach, using a mixed
151
effects logistic regression model, also allowed for adjustment for clinical stage IA tumors.
152
Guidelines allow for omission of IMS in a subset of these patients,11,18 and the frequency of
ACCEPTED MANUSCRIPT 10
clinical stage IA disease may vary across hospitals; therefore, adjustment for clinical stage IA is
154
imperative to avoid confounding bias in comparisons of hospital-level rates of IMS. Finally, this
155
approach allowed for estimation of the proportion of variability explained by differences in the
156
rates of clinical stage IA disease across hospitals (by calculating the ratio of random effects
157
between models). CIs for point estimates following reliability adjustment were based on
158
Empirical Bayes variance estimates and on approximate normality of the posterior facility-level
159
effects. A post-hoc analysis calculated the median odds ratio (MOR) as another measure of
160
between-hospital variation in IMS.19 Missing data analysis revealed no missing values among
161
variables included in risk and reliability adjustment.
SC
M AN U
162
RI PT
153
A pre-specified sensitivity analysis was conducted in two different subgroups in an attempt to evaluate hospital-level variability in IMS rates among populations more likely to be
164
candidates for IMS. Specifically, one subgroup analysis was restricted to patients who underwent
165
preoperative positron emission tomography (PET) on the presumption that the provider had a
166
higher index of suspicion of NSCLC. Another subgroup analysis restricted to patients who
167
underwent an anatomic resection on the presumption that surgeons preferentially perform an
168
anatomic resection for cases that are certain to be NSCLC.
169
171
RESULTS:
AC C
170
EP
TE D
163
Between 2011 and 2013, 406 patients met criteria for this study from across five hospitals
172
in the Puget Sound region of Washington State. Clinical and treatment variables are summarized
173
in Table 1. Preoperative patient factors that varied across hospitals included insurance case-mix
174
(p<0.001), ASA class (p<0.01), and clinical stage IA disease (p<0.001). Age, sex, race,
175
comorbidity index, and lung function did not vary significantly across hospitals (all p>0.05).
ACCEPTED MANUSCRIPT 11
Operative factors (extent and approach to resection) varied significantly across hospitals
177
(p≤0.001), as did pathologic stage (p=0.008), but histology did not (p=0.95). Nine percent of
178
patients were ultimately diagnosed with pathologic N2 disease.
179
RI PT
176
Overall, 268 patients underwent IMS (66%, 95% CI: 61-71%). Mediastinoscopy
accounted for a majority (85%) of IMS procedures, and a majority (64%) of patients underwent
181
IMS on the day of resection. The median number of mediastinal lymph node stations sampled
182
was 3 (range 0-7). Seven percent of patients (95% CI: 4-10%) underwent IMS but did not have
183
lymph node tissue sampled based on pathologic evaluation. Of the 18 patients who did not have
184
lymph node tissue sampled, 11 underwent mediastinoscopy, six underwent EBUS, and one
185
underwent both mediastinoscopy and EBUS. Table 2 summarizes patterns of IMS across
186
hospitals in this study. The type of IMS modality varied across hospitals (p<0.001), as did the
187
timing of IMS (p<0.001). Unadjusted rates of IMS varied significantly across hospitals
188
(p<0.001). Hospital B performed IMS without any lymph node tissue sampled two times more
189
often (14%) than the overall average.
TE D
M AN U
SC
180
Adjusted rates of IMS varied widely ranging from 17% (95% CI: 7-36%) at Hospital E
191
to 94% (95% CI: 89-96%) at Hospital A (Figure 1). Hospitals A, B, and D had markedly higher
192
rates of IMS compared to Hospitals C and E. Hospital A and B performed IMS significantly
193
more often than the overall average, whereas hospitals C and E performed IMS significantly less
194
often than the overall average. Adjusting for clinical stage IA disease explained only 3% of the
195
variability across hospitals, despite large variation in the frequency of clinical stage IA disease
196
across hospitals. Our pre-specified subgroup analyses revealed the same degree of variability in
197
hospital-level rates of IMS among patients who underwent anatomic resection or had a PET
198
(Figures 2-3).
AC C
EP
190
ACCEPTED MANUSCRIPT 12
199
Several post-hoc analyses were conducted. As an alternative measure of variability in hospital-level rates of IMS, we estimated the MOR across hospitals to be 4.99 (95% CI 1.56-
201
7.24) indicating significant between-hospital variance. This value is interpreted as follows: if a
202
randomly selected patient from one center were to transfer care to a hospital with a higher
203
probability of performing IMS, then his or her risk of undergoing IMS would increase by 499%.
204
Table 3 summarizes a post-hoc analysis exploring potential differences in patient characteristics
205
between those who did and did not undergo IMS. Compared with those who did not receive IMS,
206
patients who underwent IMS had a different payer mix (p<0.001) and were more likely to have a
207
higher ASA class (p=0.002), smoke (p=0.002), have a lower predicted FEV1 (p=0.03), have a
208
preoperative PET scan (p<0.001), and have a higher clinical stage (p=0.001). A “fully adjusted”
209
model did not change the conclusions of our primary analysis, and it revealed that patient-level
210
variables accounted for a total of 5% of the variation in rates of IMS across hospitals. Another
211
patient-level stratified analysis showed that patients with clinical stage IB or higher disease had
212
higher rates of IMS compared to those with clinical stage IA (78% versus 61%, p<0.001).
213
Finally, hospitals with an exclusively thoracic surgical-based practice model (A, B, and D) had
214
higher rates of IMS compared to hospitals with a mixed-specialty practice model (82% versus
215
19%, p<0.001).
216
DISCUSSION:
SC
M AN U
TE D
EP
AC C
217
RI PT
200
We hypothesized that there would be significant variation in hospital-level rates of
218
invasive mediastinal staging (IMS) and aimed to compare hospital-specific rates of IMS to each
219
other and to the overall population mean. The key finding of our study is that there is significant
220
variation in rates of IMS across hospitals providing surgical care to NSCLC patients.
221
Surprisingly, patient-level factors, including clinical stage IA disease, explain very little of the
ACCEPTED MANUSCRIPT 13
222
variability. Post-hoc analyses confirm our key finding. Hospitals with a purely thoracic surgical
223
practice (as opposed to a mixed-specialty practice) were associated with higher rates of IMS.
224
Variability in care is often considered a marker of poor quality care. Surgeons may not adhere to guidelines because they are unaware that they exist. Those with a practice solely
226
dedicated to thoracic surgery may be more likely to know guideline recommended indications
227
for IMS. Supporting this hypothesis is the observation that an exclusively thoracic surgical-based
228
hospital practice model was associated with higher rates of IMS. However, a previous study
229
examining the relationship between surgeon specialty and outcomes reported no differences in
230
the utilization of mediastinoscopy among general, cardiothoracic, and thoracic surgeons.20
231
Survey research may better elucidate the frequency of surgeon awareness of guideline
232
recommendations and factors associated with both awareness and adherence.
SC
M AN U
233
RI PT
225
Another possible explanation for variation in IMS is that surgeons may purposefully choose to not adhere to guidelines because they do not accept the recommendations to be valid.
235
A previous study showed that providers may deviate from guidelines when either the level of
236
supporting evidence for guidelines is low or when they perceive the supporting evidence to not
237
be generalizable to their patient population or practice.21 Indeed the level of evidence for
238
guideline recommended indications for IMS is 2.9–11. There are no randomized controlled trials
239
(RCTs) comparing varying intensities of IMS (i.e. routine versus selective versus highly
240
selective IMS). It was recently shown that guideline recommended IMS has a sensitivity and
241
specificity of 100% and 35%, respectively.22,23 Clinically, these estimates mean that existing
242
guideline recommendations perfectly select all patients with true nodal disease to undergo IMS,
243
but they also select many patients without true nodal disease to undergo IMS. The high rate of
244
negative IMS may provide the rationale for why some surgeons do not accept guideline
AC C
EP
TE D
234
ACCEPTED MANUSCRIPT 14
recommendations. A high “true negative” rate of IMS may be perceived as futile, inefficient, or
246
cost-ineffective24, although recent evidence suggests these perceptions may not be correct.25
247
However, underlying the rationale for guidelines directing a majority of lung cancer patients to
248
IMS is the view that the identification of mediastinal nodal disease (N2 or N3, Stage IIIA or
249
IIIB) is important from a patient and treatment selection perspective. Even if the multi-modality
250
options for stage IIIA are equivalent with regards to long-term survival,26 many patients and
251
providers would likely agree that more accurate staging leads to more informed treatment
252
decisions. Nonetheless, some providers may be skeptical of this view and believe that patients
253
are better served by initial surgical resection.
SC
M AN U
254
RI PT
245
Better understanding the reasons underlying hospital-level variability in IMS is an opportunity to improve care. Qualitative research (focus groups, key informant surgeon
256
interviews, etc.) can help identify the reasons for variability in practice patterns. To the extent
257
that a knowledge gap explains variability in IMS, hospital credentialing and privileging policies
258
could be an avenue for evaluating and increasing the knowledge base of surgeons who do not
259
have a dedicated oncologic practice and/or board certification in thoracic surgery.27 Greater
260
access to continuing medical education—such as the GAIN curriculum sponsored by the
261
American College of Chest Physicians, National Comprehensive Cancer Network, and others—
262
is another opportunity to address knowledge gaps.28 Finally, clinical registries (e.g. STS,
263
National Surgical Quality Improvement Program) can facilitate performance feedback in terms
264
of process compliance (i.e. benchmarked rates of IMS).29–31
EP
AC C
265
TE D
255
On the other hand, if the variability in IMS is due to widespread disagreement about the
266
optimal indications for IMS, then more research will be needed to elucidate the optimal IMS
267
strategy. One example of an alternative staging strategy is selective IMS using prediction
ACCEPTED MANUSCRIPT 15
models. Preliminary research suggests that prediction models can, similar to existing practice
269
guidelines, maintain 100% sensitivity while achieving higher specificity.22,32 Clinically, this
270
means that the use of a prediction model is expected to lead to fewer IMS procedures without
271
compromising the ability to detect true nodal disease prior to first treatment. This hypothesis
272
could also be tested through an RCT. Still other alternative staging strategies may arise from
273
qualitative interviews with thoracic surgeons and pulmonologists.
One unexpected finding was that the rate of IMS in this study (66%) was higher than
SC
274
RI PT
268
previously reported (27%).1 Differences across studies may reflect differences in the quality of
276
data collection or delivery of care over time. Another possibility is the Hawthorne Effect given
277
that all hospitals agreed to participate in a regional quality improvement collaborative. However,
278
this explanation is unlikely because the decision to measure utilization of IMS was made by our
279
regional collaborative in 2013, the study period spans 2011 to 2013, and there were no
280
interventions intended to influence rates of IMS. Furthermore, rates of IMS for the entire study
281
population were lower than expected population rates of IMS derived from prior studies (77-
282
79%).22,23,33 In addition, although rates of failed lymph node sampling (7%) were much lower
283
than that previously reported (~47%),1 opportunities for QI remain given that one hospital failed
284
sample lymph node tissue in up to 14% of patients undergoing IMS—a rate double that of other
285
hospitals. Failing to submit or obtain adequate lymph node tissue is an outcome measure that
286
could be monitored by a clinical registry such as the STS General Thoracic Surgery
287
Database.30,31
TE D
EP
AC C
288
M AN U
275
There are important limitations to this study. First there may be concerns about
289
generalizing regional results to the rest of the nation. By design, however, this multicenter study
290
included mostly community hospitals and only a single academic center in an attempt to
ACCEPTED MANUSCRIPT 16
maximize generalizability. We expect hospital-level variability in IMS to be found across the
292
nation, and several members of our study team are currently testing this hypothesis using the
293
STS General Thoracic Surgery Database (even though this database also has generalizability
294
concerns). Second, because we were unable to directly access medical records to determine the
295
indications for or against IMS, we could not measure adherence with guideline recommendations
296
at the patient level. Third, because we were unable to measure central versus peripheral tumors,
297
some of the variation in IMS may be explained by differences in case-mix (defined by central
298
tumors) across hospitals. Fourth, we did not have information on the provider who performed or
299
decided against performing IMS, and therefore we can better understand the contribution of
300
individual providers or provider-specialty to variation in care. Fifth, we may have
301
underestimated rates of IMS. Data abstractors may not have been able to record IMS occurring at
302
a referring hospital/health system (i.e. not participating in our collaborative). However, given the
303
fact that rates of IMS are higher than previously described in community practice, it is unlikely
304
that we significantly underestimated rates of IMS. Finally, we were unable to elucidate the
305
reasons for variable rates of IMS across hospitals. As discussed earlier we plan to conduct
306
qualitative and mixed-methods research to gain a better understanding of the determinants of
307
variability in IMS.
SC
M AN U
TE D
EP
In conclusion, we describe—for the first time—that there is significant variability in rates
AC C
308
RI PT
291
309
of IMS between hospitals. We speculate that this variability might be due to gaps in the quality
310
of care, disagreement over the best indications for IMS, or both. Planned qualitative and mixed-
311
methods studies will likely reveal the reasons underlying variability leading to either quality
312
improvement and educational interventions and/or trials comparing novel IMS strategies to
313
guideline recommended staging.
ACCEPTED MANUSCRIPT 17
AC C
EP
TE D
M AN U
SC
RI PT
314
ACCEPTED MANUSCRIPT 18
315 316
ACKNOWLEDGEMENTS The authors would like to thank the Surgical Care and Outcomes Assessment Program for providing data for this research. Also, special thanks to Rebecca G. Symons, Megan
318
Zadworny, Hao He, PhD, and Katherine Odem-Davis, PhD at the Surgical Outcomes Research
319
Center for their help with data acquisition and statistical consultation. Research reported in this
320
publication was supported by the National Institute of Diabetes and Digestive and Kidney
321
Diseases of the National Institutes of Health under Award Number T32DK070555. The content
322
is solely the responsibility of the authors and does not necessarily represent the official views of
323
the National Institutes of Health.
AC C
EP
M AN U
TE D
324
SC
RI PT
317
ACCEPTED MANUSCRIPT
REFERENCES:
326 327
1.
Little AG, Rusch VW, Bonner JA, et al. Patterns of Surgical Care of Lung Cancer Patients. Ann Thorac Surg. 2005;80(6):2051-2056. doi:10.1016/j.athoracsur.2005.06.071.
328 329 330
2.
Little AG, Gay EG, Gaspar LE, Stewart AK. National survey of non-small cell lung cancer in the United States: Epidemiology, pathology and patterns of care. Lung Cancer. 2007;57(3):253-260. doi:10.1016/j.lungcan.2007.03.012.
331 332 333
3.
Farjah F, Flum DR, Ramsey SD, Heagerty PJ, Symons RG, Wood DE. Multi-Modality Mediastinal Staging for Lung Cancer Among Medicare Beneficiaries. J Thorac Oncol. 2009;4(3):355-363. doi:10.1097/JTO.0b013e318197f4d9.
334 335 336
4.
Lathan CS, Neville BA, Earle CC. The effect of race on invasive staging and surgery in non-small-cell lung cancer. J Clin Oncol. 2006;24(3):413-418. doi:10.1200/JCO.2005.02.1758.
337 338 339
5.
Vest MT, Tanoue L, Soulos PR, et al. Thoroughness of Mediastinal Staging in Stage IIIA Non-small Cell Lung Cancer. J Thorac Oncol. 2012;7(1):188-195. doi:10.1097/JTO.0b013e318236ecbb.
340 341 342
6.
Ost DE, Niu J, Elting LS, Buchholz TA, Giordano SH. Determinants of practice patterns and quality gaps in lung cancer staging and diagnosis. Chest. 2014;145(5):1097-1113. doi:10.1378/chest.13-1628.
343 344 345
7.
Ost DE, Niu J, Elting LS, Buchholz TA, Giordano SH. Quality gaps and comparative effectiveness in lung cancer staging and diagnosis. Chest. 2014;145(2):331-345. doi:10.1378/chest.13-1599.
346 347 348
8.
Faris N, Yu X, Sareen S, et al. Preoperative Evaluation of Lung Cancer in a Community Health Care Setting. Ann Thorac Surg. 2015;100(2):394-400. doi:10.1016/j.athoracsur.2015.03.008.
349 350
9.
Network NCC. Non-Small Cell Lung Cancer (Version 5.2017). https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. Accessed March 18, 2017.
351 352
10.
Silvestri GA, Gonzalez A V., Jantz MA, et al. Methods for Staging Non-small Cell Lung Cancer. Chest. 2013;143(5):e211S-e250S. doi:10.1378/chest.12-2355.
353 354 355
11.
356 357
12.
358 359 360 361
13.
Devine EB, Capurro D, van Eaton E, et al. Preparing Electronic Clinical Data for Quality Improvement and Comparative Effectiveness Research: The SCOAP CERTAIN Automation and Validation Project. EGEMS (Washington, DC). 2013;1(1):1025. doi:10.13063/2327-9214.1025.
362
14.
Flum DR, Alfonso-Cristancho R, Devine EB, et al. Implementation of a “real-world”
EP
TE D
M AN U
SC
RI PT
325
AC C
19
Ettinger DS, Akerley W, Borghaei H, et al. Non–Small Cell Lung Cancer, Version 2.2013. J Natl Compr Cancer Netw . 2013;11(6):645-653. http://www.jnccn.org/content/11/6/645.abstract. CERTAIN. CINCO. http://www.becertain.org/projects/cancer/cinco. Accessed March 14, 2017.
ACCEPTED MANUSCRIPT 20
learning health care system: Washington state’s Comparative Effectiveness Research Translation Network (CERTAIN). Surgery. 2014;155(5):860-866. doi:10.1016/j.surg.2014.01.004.
363 364 365
15.
Dimick JB, Staiger DO, Birkmeyer JD. Ranking hospitals on surgical mortality: The importance of reliability adjustment. Health Serv Res. 2010;45(6 PART 1):1614-1629. doi:10.1111/j.1475-6773.2010.01158.x.
369 370 371
16.
Dimick JB, Ghaferi A a, Osborne NH, Ko CY, Hall BL. Reliability adjustment for reporting hospital outcomes with surgery. Ann Surg. 2012;255(4):703-707. doi:10.1097/SLA.0b013e31824b46ff.
372 373 374
17.
MacKenzie TA, Grunkemeier GL, Grunwald GK, et al. A primer on using shrinkage to compare in-hospital mortality between centers. Ann Thorac Surg. 2015;99(3):757-761. doi:10.1016/j.athoracsur.2014.11.039.
375 376 377
18.
Detterbeck FC, Jantz MA, Wallace M, Vansteenkiste J, Silvestri GA. Invasive mediastinal staging of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest. 2007;132(3 SUPPL.). doi:10.1378/chest.07-1362.
378 379 380
19.
Larsen K, Merlo J. Appropriate assessment of neighborhood effects on individual health: Integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol. 2005;161(1):81-88. doi:10.1093/aje/kwi017.
381 382 383
20.
Farjah F, Flum DR, Varghese TK, Symons RG, Wood DE. Surgeon Specialty and LongTerm Survival After Pulmonary Resection for Lung Cancer. Ann Thorac Surg. 2009;87(4):995-1006. doi:10.1016/j.athoracsur.2008.12.030.
384 385 386
21.
Brouwers MC, Makarski J, Garcia K, et al. A mixed methods approach to understand variation in lung cancer practice and the role of guidelines. Implement Sci. 2014;9(1):36. doi:10.1186/1748-5908-9-36.
387 388 389
22.
Farjah F, Backhus LM, Varghese TK, et al. External validation of a prediction model for pathologic N2 among patients with a negative mediastinum by positron emission tomography. J Thorac Dis. 2015;7(4):576-584. doi:10.3978/j.issn.2072-1439.2015.02.09.
390 391 392
23.
Farjah F, Madtes DK, Wood DE, et al. Vascular endothelial growth factor C complements the ability of positron emission tomography to predict nodal disease in lung cancer. J Thorac Cardiovasc Surg. 2015;150(4):796-803. doi:10.1016/j.jtcvs.2015.08.001.
393 394 395
24.
396 397 398
25.
399 400 401 402
26.
AC C
EP
TE D
M AN U
SC
RI PT
366 367 368
Meyers BF, Haddad F, Siegel BA, et al. Cost-effectiveness of routine mediastinoscopy in computed tomography- and positron emission tomography-screened patients with stage I lung cancer. J Thorac Cardiovasc Surg. 2006;131(4). doi:10.1016/j.jtcvs.2005.10.045. Czarnecka-Kujawa K, Rochau U, Siebert U, et al. Cost-effectiveness of mediastinal lymph node staging in non–small cell lung cancer. J Thorac Cardiovasc Surg. 2017;153(6):15671578. doi:10.1016/j.jtcvs.2016.12.048. Martins RG, D’Amico TA, Loo Jr. BW, et al. The management of patients with stage IIIA non-small cell lung cancer with N2 mediastinal node involvement. J Natl Compr Canc Netw. 2012;10(5):599-613. http://www.ncbi.nlm.nih.gov/pubmed/22570291\nhttp://www.jnccn.org/content/10/5/599.l
ACCEPTED MANUSCRIPT 21
ong.
403
27.
Blackmon SH, Cooke DT, Whyte R, et al. The society of thoracic surgeons expert consensus statement: A tool kit to assist thoracic surgeons seeking privileging to use new technology and perform advanced procedures in general thoracic surgery. Ann Thorac Surg. 2016;101(3):1230-1237. doi:10.1016/j.athoracsur.2016.01.061.
408 409 410
28.
CHEST GAIN Regional non-small cell lung cancer summits. http://www.chestnet.org/Education/Advanced-Clinical-Training/CHEST-GAIN-NSCLC. Accessed November 4, 2017.
411 412
29.
ACS NSQIP Data Collection, Analysis, and Reporting. https://www.facs.org/qualityprograms/acs-nsqip? Accessed September 29, 2016.
413 414
30.
Grover FL, Shahian DM, Clark RE, Edwards FH. The STS National Database. Ann Thorac Surg. 2014;97(1):S48-S54. doi:10.1016/j.athoracsur.2013.10.015.
415 416
31.
Jacobs JP, Shahian DM, Prager RL, et al. Introduction to the STS National Database Series. Ann Thorac Surg. 2015;100(6):1992-2000. doi:10.1016/j.athoracsur.2015.10.060.
417 418 419 420
32.
Farjah F, Lou F, Sima C, Rusch VW, Rizk NP. A prediction model for pathologic N2 disease in lung cancer patients with a negative mediastinum by positron emission tomography. J Thorac Oncol. 2013;8(9):1170-1180. doi:10.1097/JTO.0b013e3182992421.
421 422
33.
Mazzone PJ, Vachani A, Chang A, et al. Quality indicators for the evaluation of patients with lung cancer. Chest. 2014;146(3):659-669. doi:10.1378/chest.13-2900.
SC
M AN U
TE D EP AC C
423
RI PT
404 405 406 407
ACCEPTED MANUSCRIPT 22
FIGURE LEGENDS:
425
Figure 1: Rates of Invasive Mediastinal Staging by Hospital Compared with Population Average
426
Figure 2: Rates of invasive mediastinal staging by hospital (among patients undergoing
427
anatomic resection)
428
Figure 3: Rates of invasive mediastinal staging by hospital (among patients with preoperative
429
PET)
SC
RI PT
424
M AN U
430
VIDEO LEGEND:
432
Dr. Farhood Farjah explains the significance of the research findings herein.
AC C
EP
TE D
431
ACCEPTED MANUSCRIPT
Table 1: Characteristics of patients undergoing resection for lung cancer across hospital Hospital All Hospital A B Hospital C n = 406 n = 130 n = 134 n = 72 Mean age, years [SD] 68.2 (10.0) 66.7 (10.6) 68.8 (9.9) 68.2 (10.0)
Hospital D n = 40 69.7 (8.8)
Hospital E n = 30 71.5 (8.5)
pvalue 0.12 0.05
Women, n (%)
228 (57)
70 (54)
84 (64)
38 (53)
16 (40)
20 (67)
Race, n (%) White Black Asian Native American Pacific Islander Other
354 (88) 7 (2) 26 (6) 4 (1) 3 (1) 8 (2)
109 (85) 4 (3) 9 (7) 3 (2) 1 (1) 2 (2)
116 (87) 2 (2) 10 (8) 0 (0) 2 (2) 3 (2)
64 (89) 1 (1) 5 (7) 0 (0) 0 (0) 2 (3)
37 (95) 0 (0) 0 (0) 1 (3) 0 (0) 1 (3)
28 (93) 0 (0) 2 (7) 0 (0) 0 (0) 0 (0)
Insurance, n (%) Commercial Medicare Medicaid Tricare (military) Other Government Uninsured Government & Commercial Mixed Government
145 (36) 77 (19) 13 (3) 2 (<1) 3 (1) 2 (<1) 126 (31) 38 (9)
39 (30) 9 (7) 5 (4) 1 (1) 2 (2) 1 (1) 63 (48) 10 (8)
49 (37) 16 (12) 5 (4) 0 (0) 0 (0) 0 (0) 44 (33) 20 (15)
44 (61) 27 (38) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0)
9 (23) 15 (38) 2 (5) 1 (3) 0 (0) 0 (0) 7 (18) 6 (15)
4 (13) 10 (33) 1 (3) 0 (0) 1 (3) 0 (0) 12 (40) 2 (7)
Charlson comorbidity index, n (%) 0 1 2 3+
227 (56) 141 (35) 28 (7) 10 (2)
73 (56) 42 (32) 12 (9) 3 (2)
69 (51) 50 (37) 11 (8) 4 (3)
47 (65) 21 (29) 1 (1) 3 (4)
21 (53) 15 (38) 4 (10) 0 (0)
17 (57) 13 (43) 0 (0) 0 (0)
ASA class, n (%) II III IV
71 (17) 312 (77) 23 (6)
12 (9) 107 (82) 11 (8)
25 (19) 99 (74) 10 (7)
20 (28) 51 (71) 1 (1)
10 (25) 29 (73) 1 (3)
4 (13) 26 (87) 0 (0)
M AN U
SC
<0.001
TE D
Smoking status, n (%) Former Current Never
RI PT
0.79
0.34
0.01
0.13
235 (60) 96 (25) 60 (16)
78 (63) 31 (25) 15 (12)
79 (60) 39 (30) 14 (11)
42 (63) 11 (16) 14 (21)
20 (53) 8 (21) 10 (26)
16 (53) 7 (23) 7 (23)
83 (69-96)
81 (66-95)
80 (66-94)
84 (75-100)
87 (73-100)
84 (68-103)
0.16
64 (52-77)
64 (55-75)
63 (49-80)
66 (51-78)
66 (62-83)
60 (50-70)
0.31
281 (69) 54 (13) 33 (8) 20 (5) 15 (4) 1 (<1) 2 (<1)
74 (57) 15 (12) 17 (13) 10 (8) 11 (8) 1 (1) 2 (2)
110 (82) 11 (8) 6 (5) 6 (5) 1 (1) 0 (0) 0 (0)
54 (75) 15 (21) 2 (3) 0 (0) 1 (1) 0 (0) 0 (0)
22 (55) 8 (20) 4 (10) 4 (10) 2 (5) 0 (0) 0 (0)
21 (70) 5 (17) 4 (13) 0 (0) 0 (0) 0 (0) 0 (0)
Preoperative PET scan, n (%)
371 (92)
125 (96)
120 (90)
64 (89)
38 (97)
24 (83)
Extent of Resection, n (%) Wedge Segmentectomy Lobectomy Sleeve lobectomy Lobectomy & wedge Bilobectomy Pneumonectomy
35 (9) 7 (2) 272 (67) 2 (<1) 66 (16) 14 (3) 10 (2)
19 (15) 3 (2) 71 (55) 1 (1) 19 (15) 9 (7) 8 (6)
9 (7) 2 (1) 102 (76) 0 (0) 18 (13) 1 (1) 2 (1)
5 (7) 0 (0) 46 (64) 0 (0) 17 (24) 4 (6) 0 (0)
1 (3) 2 (5) 28 (70) 1 (3) 8 (20) 0 (0) 0 (0)
1 (3) 0 (0) 25 (83) 0 (0) 4 (13) 0 (0) 0 (0)
MIS Approach (%)
292 (72)
84 (65)
129 (96)
41 (57)
33 (83)
5 (17)
<0.001
Right sided resection (%)
247 (61)
81 (62)
81 (60)
39 (54)
28 (70)
18 (60)
0.75
% Predicted FEV1, median (IQR)
AC C
Clinical stage, n (%) IA IB IIA IIB IIIA IIIB IV
EP
% Predicted DLCO, median (IQR)
<0.001
0.049 0.001
ACCEPTED MANUSCRIPT
Pathologic Stage, n (%) 0 IA IB IIA IIB IIIA IV
5 (1) 201 (50) 84 (21) 38 (9) 32 (8) 41 (10) 5 (1)
2 (2) 51 (39) 24 (18) 15 (12) 15 (12) 22 (17) 1 (1)
1 (1) 79 (59) 28 (21) 11 (8) 8 (6) 5 (4) 2 (1)
1 (1) 40 (56) 17 (24) 5 (7) 2 (3) 7 (10) 0 (0)
Pathologic Nodal Stage, n (%) pN0 pN1 pN2
317 (79) 47 (12) 36 (9)
93 (73) 18 (14) 17 (13)
110 (83) 17 (13) 5 (4)
55 (79) 7 (10) 8 (11)
0.008 0 (0) 20 (50) 6 (15) 1 (3) 5 (13) 7 (18) 1 (3)
1 (3) 11 (37) 9 (30) 6 (20) 2 (7) 0 (0) 1 (3)
RI PT
0.053
32 (80) 2 (5) 6 (15)
27 (90) 3 (10) 0 (0)
AC C
EP
TE D
M AN U
SC
Histology (%) 0.95 Adenocarcinoma 270 (67) 82 (63) 87 (65) 49 (68) 30 (75) 22 (73) AIS 10 (2) 5 (4) 3 (2) 1 (1) 0 (0) 1 (3) Squamous cell carcinoma 86 (21) 30 (23) 30 (22) 12 (17) 9 (23) 5 (17) Large cell carcinoma 7 (2) 3 (2) 2 (1) 2 (3) 0 (0) 0 (0) NSCLC NOS 22 (5) 6 (5) 9 (7) 6 (8) 0 (0) 1 (3) Other 11 (3) 4 (3) 3 (2) 2 (3) 1 (3) 1 (3) Adenocarcinoma in situ (AIS); American Society of Anesthesiologists (ASA); Forced expiratory volume in one second (FEV1); Diffusion capacity of carbon monoxide (DLCO); Minimally-invasive surgery (video-assisted thoracoscopic surgery or robotic ); Nonsmall cell lung cancer (NSCLC); Not otherwise specified (NOS); Length-of-stay (LOS); Interquartile range (IQR), Positron Emission Tomography (PET)
ACCEPTED MANUSCRIPT
Any invasive mediastinal staging, n (%)
268 (66)
119 (92)
100 (75)
16 (22)
Timing of Invasive Staging*, n (%) Before Resection At the time of resection Both before and at the time of resection
97 (36) 147 (55) 24 (9)
3 (3) 95 (80) 21 (18)
87 (87) 12 (12) 1 (1)
4 (25) 10 (63) 2 (13)
Hospital E n = 30
p-value
<0.001 27 (90) 3 (10) 0 (0) 0 (0) 0 (0) 0 (0)
RI PT
Table 2: Patterns of invasive cancer staging for patients undergoing resection of lung cancer across hospitals Hospital Hospital Hospital Hospital All A B C D n = 406 n = 128 n = 132 n = 70 n = 40 Invasive mediastinal staging, n (%) None 138 (34) 11 (8) 34 (25) 56 (78) 10 (25) Mediastinoscopy/Mediastinotomy only 228 (56) 96 (74) 87 (65) 13 (18) 29 (73) EBUS only 9 (2) 2 (2) 5 (4) 1 (1) 1 (3) EUS only 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) EBUS & Mediastinoscopy/Mediastinotomy 30 (7) 21 (16) 7 (5) 2 (3) 0 (0) EBUS & EUS & 1 (<1) 0 (0) 1 (1) 0 (0) 0 (0) Mediastinoscopy/Mediastinotomy 30 (75)
3 (10)
1 (3) 29 (97) 0 (0)
2 (67) 1 (33) 0 (0)
<0.001
SC
<0.001
AC C
EP
TE D
M AN U
Number of nodal stations sampled* Mean (SD) 3.0 (1.2) 3.2 (1.0) 2.7 (1.4) 2.9 (1.3) 3.2 (1.4) 2.7 (0.6) 0.05 Median (range) 3 (0-7) 3 (0-7) 3 (0-5) 3 (0-5) 3 (0-5) 3 (2-3) 0.34 No nodes sampled, n (%) 18 (7) 1 (1) 14 (14) 1 (6) 2 (6) 0 (0) 0.004 Endobronchial ultrasound guided nodal aspirate (EBUS); Esophageal ultrasound guided nodal aspirate (EUS); Standard deviation (SD); Interquartile range (IQR); Collaborative to Improve Native Cancer Outcomes (CINCO) *Calculated among patients who underwent invasive mediastinal staging
ACCEPTED MANUSCRIPT
Table 3: Characteristics of patients receiving and not receiving invasive staging. All Invasive staging n = 406 n=265 Mean age, years [SD] 68.2 (10.0) 68.0 (10.4)
No invasive staging n=137 68.8 (9.3)
p-value
0.95
Women, n (%)
228 (57)
150 (56)
78 (57)
Race, n (%) White Black Asian Native American Pacific Islander Other
354 (88) 7 (2) 26 (6) 4 (1) 3 (1) 8 (2)
231 (87) 6 (2) 17 (6) 4 (2) 3 (1) 4 (2)
123 (90) 1 (1) 9 (7) 0 (0) 0 (0) 4 (3)
Insurance, n (%) Commercial Medicare Medicaid Tricare (military) Other Government Uninsured Government & Commercial Mixed Government
145 (36) 77 (19) 13 (3) 2 (<1) 3 (1) 2 (<1) 126 (31) 38 (9)
89 (33) 35 (13) 11 (4) 2 (1) 3 (1) 1 (<1) 97 (36) 30 (11)
46 (41) 42 (30) 2 (1) 0 (0) 0 (0) 1 (1) 29 (21) 8 (6)
0.48
RI PT
0.33
ASA class, n (%) II III IV
71 (17) 312 (77) 23 (6)
Smoking status, n (%) Former Current Never
235 (60) 96 (25) 60 (16)
M AN U
227 (56) 141 (35) 28 (7) 10 (2)
148 (55) 90 (34) 22 (8) 8 (3)
79 (57) 51 (37) 6 (4) 2 (1)
34 (13) 217 (81) 17 (6)
37 (27) 95 (69) 6 (4)
TE D
Charlson comorbidity index, n (%) 0 1 2 3+
SC
<0.001
0.36
0.002
0.002
159 (62) 70 (27) 28 (11)
76 (57) 26 (19) 32 (24)
80 (69-94)
85 (72-100)
0.03
63 (51-76)
66 (56-80)
0.15
281 (69) 54 (13) 33 (8) 20 (5) 15 (4) 1 (<1) 2 (<1)
170 (63) 36 (13) 25 (9) 20 (7) 14 (5) 1 (<1) 2 (1)
111 (80) 18 (13) 8 (6) 0 (0) 1 (1) 0 (0) 0 (0)
Preoperative PET scan, n (%)
371 (92)
255 (96)
116 (85)
Extent of Resection, n (%) Wedge Segmentectomy Lobectomy Sleeve lobectomy Lobectomy & wedge Bilobectomy Pneumonectomy
35 (9) 7 (2) 272 (67) 2 (<1) 66 (16) 14 (3) 10 (2)
20 (7) 6 (2) 180 (67) 0 (0) 40 (15) 12 (4) 10 (4)
15 (11) 1 (1) 92 (67) 2 (1) 26 (19) 2 (1) 0 (0)
MIS Approach (%)
292 (72)
201 (75)
91 (66)
83 (69-96)
% Predicted DLCO, median (IQR)
64 (52-77)
AC C
Clinical stage, n (%) IA IB IIA IIB IIIA IIIB IV
EP
% Predicted FEV1, median (IQR)
0.001
<0.001 0.02
0.05
ACCEPTED MANUSCRIPT
Right sided resection (%)
247 (61)
163 (61)
84 (61)
Pathologic Stage, n (%) 0 IA IB IIA IIB IIIA IV
5 (1) 201 (50) 84 (21) 38 (9) 32 (8) 41 (10) 5 (1)
1 (<1) 121 (45) 56 (21) 26 (10) 26 (10) 34 (13) 4 (1)
4 (3) 80 (58) 28 (20) 12 (9) 6 (4) 7 (5) 1 (1)
0.01
RI PT
Pathologic Nodal Stage, n (%) pN0 pN1 pN2
0.77
0.003
317 (79) 47 (12) 36 (9)
198 (74) 37 (14) 31 (12)
119 (89) 10 (7) 5 (4)
AC C
EP
TE D
M AN U
SC
Histology (%) 0.049 Adenocarcinoma 270 (67) 174 (65) 96 (70) AIS 10 (2) 5 (2) 5 (4) Squamous cell carcinoma 86 (21) 67 (25) 19 (14) Large cell carcinoma 7 (2) 5 (2) 2 (1) NSCLC NOS 22 (5) 10 (4) 12 (9) Other 11 (3) 7 (3) 4 (3) Adenocarcinoma in situ (AIS); American Society of Anesthesiologists (ASA); Forced expiratory volume in one second (FEV1); Diffusion capacity of carbon monoxide (DLCO); Minimally-invasive surgery (video-assisted thoracoscopic surgery or robotic ); Non-small cell lung cancer (NSCLC); Not otherwise specified (NOS); Length-of-stay (LOS); Interquartile range (IQR), Positron Emission Tomography (PET)
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1
SUPPLEMENTAL MATERIAL:
2
Appendix A—Participating Hospitals
3
MultiCare Allenmore Hospital
4
MultiCare Good Samaritan Hospital
5
Harborview Medical Center
6
Peace Health St. Joseph Medical Center
7
Providence Regional Medical Center Everett
8
St. Joseph Medical Center
9
St. Francis Hospital
M AN U
St. Anthony Hospital
11
St. Clare Hospital
12
St. Elizabeth Hospital
13
MultiCare Tacoma General Hospital
14
University of Washington Medical Center
15
Virginia Mason Medical Center
AC C
EP
TE D
10
16
SC
RI PT
1