- Email: [email protected]
The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study
JGO-00631; No. of pages: 6; 4C: Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Journal of Geriatric Oncology
The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study☆ Linda M. Pak a,b,⁎, Tingsong Yang a, Jiping Wang a,b a b
Division of Surgical Oncology, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States Center for Surgery and Public Health, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States
a r t i c l e
i n f o
Article history: Received 12 June 2018 Received in revised form 6 September 2018 Accepted 17 October 2018 Available online xxxx Keywords: Gastrectomy Gastric cancer Octogenarians
a b s t r a c t Objective: The safety of minimally-invasive (MIG) and open gastrectomy (OG) in the older patients has been demonstrated in several international studies but has not been evaluated in the context of a large, Western population. The objective of this study was to evaluate the safety of gastrectomy by these two approaches among octogenarians in the United States. Materials and methods: The National Cancer Database (2010–2014) was queried for patients with gastric adenocarcinoma who underwent MIG or OG. Disease and treatment characteristics and outcomes were compared across age groups (b65, 65–79, ≥80 years). Multivariable regression analysis was used to identify factors associated with 90-day mortality. Results: 13,845 patients were identified who met study inclusion criteria, of which 2140 patients (16%) were aged ≥80 years. Among octogenarians, MIG was associated with slightly decreased length of stay (11.2 vs 12.7 days, p b .001) compared to OG, with no difference in the rate of margin-positive resections, adequate lymph node sampling, or readmission (p = .30–0.77). 90-day mortality for both OG (17%) and MIG (17.2%) was significantly higher among octogenarians compared to younger patients (p b .001). Treatment at an academic facility was an independent predictor of decreased 90-day mortality among octogenarians (OR 0.534, 95%CI 0.322–0.886, p = .015). Conclusion: In this Western population, we report comparable oncologic and post-operative outcomes between MIG and OG. However, the overall post-operative mortality rate among octogenarians remains unacceptably high. Better patient selection criteria for surgery and efforts to refer these patients to higher volume, academic facilities should be considered to improve patient outcomes. © 2018 Elsevier Ltd. All rights reserved.
1. Introduction Gastric cancer is the fifth most common cause of cancer and the third leading cause of cancer-related deaths worldwide [1,2]. With a peak incidence in the seventh and eighth decades of life, gastric cancer has demonstrated a rising incidence in the older patients as life expectancies have increased [3–6]. The mainstay of treatment is resection, typically involving a partial or total gastrectomy with D2 lymphadenectomy to ensure all microscopic disease has been surgically cleared. However, many older patients with gastric cancer also present with multiple comorbidities and decreased functional reserve which can hinder their ability to recover from a large abdominal operation. Several ☆ This study was presented as a poster presentation at the Society for Surgical Oncology 71st Annual Cancer Symposium, March 21-24, 2018, in Chicago, IL. ⁎ Corresponding author at: 75 Francis St, Boston, MA 02215, United States E-mail address: [email protected] (L.M. Pak).
studies have previously shown older patients to have an increased risk of post-operative morbidity and mortality after open gastrectomy (OG) [7–9]. The introduction of minimally-invasive gastrectomy (MIG) has garnered significant interest as a less invasive and traumatic approach to resection, especially for these older patients. Prior studies evaluating the physiologic effects of pneumoperitoneum have found laparoscopic gastrectomy to be associated with decreased impairment of pulmonary function post-operatively when compared to open gastrectomy, as well minimal adverse cardiac effects [10,11]. Subsequent international studies, primarily consisting of Asian patients, have demonstrated similar or improved post-operative outcomes, morbidity, and mortality in MIG compared to OG among older patients [12–18]. However, the safety and efficacy of either technique of gastrectomy among older patients has not been studied within the context of the Western population.
https://doi.org/10.1016/j.jgo.2018.10.012 1879-4068/© 2018 Elsevier Ltd. All rights reserved.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
2
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
In this context, the objective of this study was to evaluate the safety of both OG and MIG among octogenarians in the United States using population-based data from the National Cancer Database (NCDB).
2. Methods This was a retrospective, population-based study using data provided by the esophagogastric participant use file of the NCDB. The NCDB is a collaborative program between the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The NCDB currently captures 70% of newly diagnosed cancer cases in the United States and collects data from over 1500 CoCaccredited cancer hospitals19. The NCDB was queried for all patients ≥18 years of age with invasive gastric adenocarcinoma with potentially resectable disease (defined as American Joint Committee on Cancer AJCC Stage Ib-IIIc) who underwent elective partial or total gastrectomy with curative intent. This data query was limited to the years 2010 through 2014 as the data on operative approach (open, laparoscopic, or robotic) only became available since 2010. Patients were excluded from analysis if they had less than a partial gastrectomy performed, a palliative resection, or an emergent resection. A resection was defined as emergent if the number of days between diagnosis and surgery was recorded as b1 day. Disease staging was performed in accordance with the AJCC Staging System, 7th edition20. Sufficient lymph node sampling was defined as ≥15 lymph nodes examined in accordance with the National Comprehensive Cancer Network (NCCN) guidelines21. MIG included patients who had undergone either laparoscopic or robotic gastrectomy. Patient demographic, disease and treatment characteristics, and post-operative and long-term outcomes were abstracted from NCDB for patients meeting inclusion criteria.
Patients were categorized into three age groups for analysis (b65 years of age, 65–79 years, and ≥ 80 years). While there is no standard definition, many groups have used a cut-off of 65 years and older to define the older patient population. However, this definition is rapidly shifting to an older age parameter as the average life expectancy in many countries often exceeds 80 years of age; thus, we elected to stratify patients into the age groups listed above and define “older patients” as those age ≥ 80 years [22,23]. Clinical and treatment characteristics and patient outcomes were compared among these three age groups using the chi-squared test for categorical variables and the KruskalWallis test for continuous variables. The primary outcome evaluated in this study was the safety of gastrectomy as defined by the 90-day post-operative mortality in octogenarians. Multivariable logistic regression tests were used to identify patient characteristics and outcome measures associated with 90-day mortality. Factors significant on univariable analysis were included in the multivariable analysis. The results of the multivariable analysis were presented as odds ratios (OR) with the associated 95% confidence intervals (CI) and p-values. A p-value b.05 and a 95% CI exclusive of 1 were considered significant. Post hoc subgroup analysis was performed comparing outcome measures between community and academic facilities, as this was identified in the multivariable analysis as a significant predictor of 90-day mortality. All statistical analysis was performed using STATA software (v14.1, StataCorp, College Station, TX). This study was approved by the Partners Health Research Committee institutional review board. 3. Results Between 2010 and 2014, 13,845 patients were identified who metstudy inclusion criteria (Fig. 1). Overall, 42.2% of patients were b 65 years of age (n = 5843), 42.3% were 65–79 years of age (n =
Patients in NCDB ≥18 years with gastric cancer 2010-2014 n = 84,414 Excluded: • 4,719 neuroendocrine tumors • 8,494 sarcomas • 530 histology unknown Patients with gastric carcinoma n = 70,671
Patients who underwent partial or total gastrectomy with curative intent n = 22,490
Excluded: • 43,416 no surgery • 2,525 local excision only • 252 unknown surgery • 183 palliative resections • 1,805 emergent procedures
Excluded: • 117 Stage 0 • 4,021 Stage 1a • 1,262 Stage 4 • 3,254 unknown stage Patients with potentially resectable disease (AJCC stage Ib-IIIc) n = 13,836
Patients aged <65 years n = 5,837
Patients aged 65-79 years n = 5,859
Patients aged ≥80 years n = 2,140
Fig. 1. Flowchart of study population selection.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx Table 1 Comparison of patient and treatment characteristics among patients with gastric cancer aged b65 years, 65–79 years, and ≥ 80 years. b65 years n = 5843 Age, in years, median (range) 56 (18–64) Sex, male 3865 (66.2%) Race White 3872 (66.3%) Black 1171 (20.0%) Asian 574 (9.8%) Other/Unknown 226 (3.9%) Charlson-Deyo Score 0 4250 (72.7%) 1 1268 (21.7%) 2 325 (5.6%) Tumor Size ≤2 cm 762 (15.1%) 2-4 cm 1660 (32.9%) 4-6 cm 1282 (25.4%) N6 cm 1345 (26.6%) Tumor Location Proximal 2097 (35.9%) Body 469 (8.0%) Distal 1439 (24.6%) Lesser curve 570 (9.8%) Greater curve 234 (4.0%) Overlapping 516 (8.8%) Unknown 518 (8.9%) Tumor Differentiation Well-differentiated 153 (2.6%) Moderately-differentiated 1245 (21.3%) Poorly-differentiated 3973 (68.0%) Undifferentiated 127 (2.2%) Unknown 345 (5.9%) AJCC T Classification T1 347 (5.9%) T2 1046 (17.9%) T3 2859 (48.9%) T4 1591 (27.2%) AJCC N Classification N0 2002 (34.3%) N1 1288 (22.0%) N2 1117 (19.1%) N3 1436 (24.6%) AJCC Stage Stage Ib 828 (14.2%) Stage II 2421 (41.4%) Stage III 2594 (44.4%) Lymphovascular Invasion Yes 2420 (41.4%) No 2533 (43.4%) Unknown 890 (15.2%) Extent of Resection
65–79 years ≥80 years n = 5862 n = 2140
p-valuea
72 (65–79) 3879 (66.2%)
– b0.001
83 (80–90) 1177 (55.0%)
b0.001 4268 (72.8%) 908 (15.5%)
1603 (74.9%) 274 (12.8%)
546 (9.3%) 140 (2.4%)
223 (10.4%) 40 (1.9%)
3591 (61.3%) 1653 (28.2%) 618 (10.5%)
1280 (59.8%) 578 (27.0%)
3
Table 1 (continued)
Partial gastrectomy Total gastrectomy Multivisceral/en bloc Surgical Approach Open Laparoscopic Robotic Chemotherapy None
b65 years n = 5843
65–79 years ≥80 years n = 5862 n = 2140
3745 (64.1%) 1362 (23.3%) 736 (12.6%)
3950 (67.4%) 1233 (21.0%) 679 (11.6%)
1636 (76.5%) 318 (14.9%)
4678 (80.1%) 941 (16.1%) 224 (3.8%)
4676 (79.8%) 971 (16.6%) 215 (3.7%)
1759 (82.2%) 326 (15.2%) 55 (2.6%)
887 (15.2%)
1804 (30.8%) 1628 (27.8%) 1933 (33.0%) 497 (8.5%)
1519 (71.0%) 298 (13.9%)
Neoadjuvant
282 (13.2%) b0.001
761 (14.5%) 1693 (32.2%) 1377 (26.2%) 1428 (27.2%)
196 (9.7%) 607 (29.9%)
Unknown Radiation therapy None
3571 (60.9%) 1205 (20.6%) 888 (15.2%)
1848 (86.4%) 195 (9.1%)
Adjuvant 597 (29.4%) Neoadjuvant 632 (31.1%) Unknown b0.001
198 (3.4%)
46 (2.2%)
1911 (32.6%) 515 (8.8%) 1521 (26.0%) 609 (10.4%) 263 (4.5%) 493 (8.4%) 550 (9.4%)
281 (13.1%)
209 (3.6%) 1685 (28.7%) 3591 (61.3%) 110 (1.9%) 267 (4.6%)
74 (3.5%) 647 (30.2%)
398 (6.8%) 1191 (20.3%) 2833 (48.3%) 1440 (24.6%)
123 (5.8%) 412 (19.3%)
2132 (36.4%) 1405 (24.0%) 1082 (18.5%) 1243 (21.2%)
715 (33.4%)
984 (16.8%) 2489 (42.5%) 2389 (40.8%)
338 (15.8%) 810 (37.9%)
2565 (43.8%) 2494 (42.6%) 803 (13.7%)
1179 (55.1%) 745 (34.8%)
186 (8.7%) 0.041
b0.001
b0.001 Adjuvant
p-valuea
1845 (31.6%) 2575 (44.1%) 536 (9.2%)
155 (7.2%) 168 (7.85%) b0.001
2903 (49.7%) 1569 (26.9%) 1145 (19.6%) 226 (3.9%)
51 (2.4%)
Cm indicates centimeters; AJCC, American Joint Committee on Cancer. a Bold indicates significant p-values.
230 (10.8%) 848 (39.6%) 239 (11.2%) 131 (6.1%) 198 (9.3%) 213 (10.0%) b0.001
1324 (61.9%) 45 (2.1%) 50 9(2.3%) b0.001
924 (43.2%) 681 (31.8%) b0.001
514 (24.0%) 432 (20.2%) 479 (22.4%) b0.001
992 (46.4%) b0.001
216 (10.1%) b0.001
5862), and 15.5% were ≥ 80 years of age (n = 2140) (Table 1). In comparison to the two younger age groups (b65 years and 65–79 years), older patients aged ≥80 years had more comorbidities and presented with more advanced stage of disease (both p b .001). Histopathologically, older patients had larger, moderately-differentiated tumors located in the distal portion of the stomach associated with the presence of lymphovascular invasion (all p b .001). Octogenarians more often underwent partial gastrectomy rather than total or multi-visceral, en bloc gastrectomy (p b .001). A higher percentage of these older patients underwent OG compared to MIG and only a minority received any neoadjuvant or adjuvant chemoradiation therapy (all p b .001). These disease and treatment details are summarized in Table 1. Comparing short-term post-operative and oncologic outcomes among patients who underwent either OG or MIG, the two older age groups, especially those ged ≥80 years, had consistently poorer outcomes than the group aged b65 years (Table 2). In particular, the 30- and 90-day mortality rates increased dramatically by age group. Patients b65 years at 30- and 90-day mortality rates of 1.8% and 4.4% respectively, compared to 8.4% and 16.9% in patients ≥80 years (p b .001). Older patients also had lower rates of adequate lymph node sampling (50.5% in patients ≥80 years vs. 57.7% in 65–79 years vs. 61.8% in b65 years, p b .001). There was no difference in the rate of marginpositive resections by age group (p = .21). Among octogenarians, 82.2% underwent OG (n = 1759) and 17.8% underwent MIG (n = 381) (Table 3). Octogenarians who underwent MIG were more likely to be treated at an academic comprehensive cancer program compared to those who underwent OG (p b .001), but there were no differences in the extent of gastrectomy performed or the use of neoadjuvant or adjuvant chemoradiation (p = .07–0.45). In evaluating post-operative outcomes between octogenarians who had MIG compared to OG, there was no difference in the rate of marginpositive resections, adequacy of lymph node sampling, 30-day readmissions, or 30- or 90-day mortality (p = .30–0.92) (Fig. 2). Patients who underwent MIG did have a slightly decreased length of stay compared to OG patients (mean 11.2 vs. 12.7 days, p b .001). On univariable analysis, factors associated with increased likelihood of 90-day mortality among octogenarians who underwent either OG or
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
4
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
Table 2 Comparison of post-operative and oncologic outcomes in patients of various age groups who underwent gastrectomy.
LOS, days, mean (standard deviation) 30-day Readmission 30-day Mortality 90-day Mortality Positive Margin Lymph node sampling ≥15 LN
b65 years n = 5843
65–79 years n = 5862
≥80 years n = 2140
p-valuea
11.8 ± 10.1 359 (6.1%) 85 (1.8%) 201 (4.4%) 912 (15.7%) 3594 (61.8%)
12.6 ± 10.6 442 (7.5%) 245 (5.2%) 462 (10.0%) 866 (14.9%) 3365 (57.7%)
12.4 ± 9.6 177 (8.3%) 143 (8.3%) 289 (16.9%) 347 (16.3%) 1079 (50.5%)
b0.001 0.001 b0.001 b0.001 0.21 b0.001
LOS indicates length of stay; LN, lymph nodes. a Bold indicates significant p-values.
MIG were increased tumor size, tumor location diffusely throughout the stomach, and performance of total gastrectomy or multivisceral/en-bloc resection (Table 4). Asian race, treatment at an academic/research facility, and tumor location along the lesser curve of the stomach were associated with decreased likelihood of 90-day mortality. The surgical approach (OG vs. MIG) was not significantly associated with 90-day mortality among octogenarians. On multivariable analysis, increased tumor size and total gastrectomy were independently associated with increased likelihood of 90-day mortality while Asian race, treatment at academic/research facility were independently associated with decreased likelihood of 90-day mortality (Table 4.) On post hoc subgroup analysis comparing post-operative outcomes among octogenarians between community and academic facilities, octogenarians who underwent gastrectomy at academic facilities had lower rates of 90-day mortality (10.7% vs 17.4%, p = .017), lower rates of margin-positive resections (13.1% vs 20.7%, p = .030), and higher rates of adequate lymph node sampling (58.3% vs 40.2%, p b .001) (Supplemental Table 1). 4. Discussion As life expectancies increase globally and more older patients are being diagnosed with gastric cancer, the indications for surgery have Table 3 Comparison of treatment characteristics and post-operative outcomes in octogenarians who underwent open compared to minimally-invasive gastrectomy.
Chemotherapy None Adjuvant Neoadjuvant Unknown Radiation therapy None Adjuvant Neoadjuvant Unknown Surgery Facility Type Community Comprehensive Community Academic Integrated Network Extent of Resection Partial gastrectomy Total gastrectomy Multivisceral/en bloc LOS, days, mean (standard deviation) 30-day Readmission 30-day Mortality 90-day Mortality Positive Margin Lymph node sampling ≥15 LN
OG N = 1759
MIG N = 381
1244 (70.7%) 242 (13.8%) 126 (7.2%) 147 (8.4%)
275 (72.2%) 56 (14.7%) 29 (7.6%) 21 (5.5%)
1520 (86.4%) 164 (9.3%) 38 (2.2%) 37 (2.1%)
328 (86.1%) 31 (8.1%) 13 (3.4%) 9 (2.4%)
165 (9.4%) 751 (42.7%) 658 (37.4%) 185 (10.5%)
19 (5.0%) 142 (37.3%) 183 (48.0%) 37 (9.7%)
1328 (75.5%) 269 (15.3%) 162 (9.2%) 12.7 ± 10.0 141 (8.0%) 117 (8.1%) 242 (16.9%) 292 (16.7%) 889 (50.7%)
308 (80.8%) 49 (12.9%) 23 (6.3%) 11.2 ± 7.9 36 (9.5%) 26 (9.4%) 47 (17.2%) 55 (14.6%) 190 (49.9%)
p-valuea 0.31
0.45
b0.001
0.07
b0.001 0.36 0.47 0.92 0.30 0.77
OG indicates open gastrectomy; MIG, minimally-invasive gastrectomy; LOS, length of stay; LN, lymph nodes. a Bold indicates significant p-values.
continued to expand to accommodate the shifting epidemiology of this disease. There have been several reports in the literature on the safety and efficacy of minimally-invasive gastrectomy for the treatment of gastric cancer in older patients compared to open gastrectomy [12,14,15,24]. However, these studies have emerged predominantly from hospital centers in Asia which tend to encounter and treat a different population demographic compared to Western centers [5,25–27]. In this context, the present study examined the safety and efficacy of both open and minimally-invasive gastrectomy among octogenarians in the United States using the NCDB. While we did not identify any clinically significant differences in short-term oncologic outcomes or post-operative outcomes between octogenarians who underwent MIG compared to OG, octogenarians consistently had significantly worse outcomes than younger age groups. The 30- and 90-day mortality rates for patients ≥80 years was 8.4% and 16.9%, respectively, which was nearly four times higher than that of the youngest age group b65 years. This postoperative mortality rate among octogenarians reported in the NCDB is significantly higher than prior reports in the literature, largely from Asian institutions. In a series of 2422 patients who underwent open gastrectomy at Chang Gung Memorial Hospital in Taiwan, including 164 patients ≥80 years, the 30-day mortality was 3% among octogenarians compared to 2% among non-octogenarians (p = .18) [8]. Overall, patients had the same median age of 83 years, similar distribution in sex, comorbidities, extent of disease, and extent of gastric resection, as compared to the present study. In another series of 2175 patients treated at Nihon University in Japan, which included 80 patients ≥80 years, Eguchi et al. reported a similarly low 30-day mortality rate of 4% among octogenarians, with no difference when compared to younger patients b80 years (3%, p = .84) [28]. The relatively higher post-operative mortality rates among octogenarians reported in the NCDB from over 1500 CoC-accredited hospitals compared to prior reports from Asian institutions indicates a pressing need for quality assessment and improvement measures, as these other institutions have set the example that safe surgery is possible among older patients with gastric cancer. In the present study, we identified treatment at an academic comprehensive cancer program as an independent factor associated with reduced 90-day mortality (OR 0.534, 95% CI 0.322–0.886); we also found that octogenarians treated at academic facilities had lower rates of 90-day mortality (p = .017), lower rates of margin-positive resections (p = .030), and higher rates of adequate of lymph node sampling (p b .001), compared to those treated at community facilities. In addition to having the highest minimum caseload requirement of all CoC facility categorizations (N500 newly diagnosed cancer cases/year), academic facilities also participate in cancerrelated clinical research/trials and in training of resident physicians. The impact of hospital volume on improved outcomes has been well established [29–32]. This has prompted efforts to promote centralization of certain procedures to high-volume centers. The centralization of pancreaticoduodenectomies, one of the most complex abdominal surgical procedures, worldwide to higher-volume centers has led to consistently decreased mortality rates and improved outcomes [33–35]. This hospital volume-outcome relationship may be particularly critical for older patients. In a study of 2155 pancreaticoduodenectomies performed in the Netherlands, de Wilde et al. identified a larger reduction in post-
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
5
Table 4 Univariable and multivariable logistic regression model for 90-day mortality in octogenarians undergoing gastrectomy. Multivariableb
Univariable
Gender (Ref: Male) Race (Ref: White) Black Asian Charlson/Deyo Score (Ref: 0) 1 2 Facility Type (Ref: Community) Comprehensive Community Academic Integrated Network Tumor Size (Ref: ≤2 cm) 2-4 cm 4-6 cm N6 cm Tumor Location (Ref: Proximal) Body Distal Lesser curve Greater curve Overlapping Neoadjuvant chemotherapy Neoadjuvant radiation therapy Resection (Ref: Partial) Total gastrectomy Multivisceral/en bloc Surgical approach (Ref: OG) MIG Adequate lymph node sampling ≥15 LN
OR
95% CI
p-valuea
OR
95% CI
p-value
0.872
0.675–1.125
0.29
–
–
–
0.754 0.316
0.505–1.126 0.173–0.577
0.17 b0.001
0.833 0.354 –
0.537–1.291 0.190–0.659 –
0.41 0.001 –
0.888 1.121
0.658–1.200 0.772–1.627
0.440 0.549
0.831 0.608 1.033
0.539–1.283 0.389–0.953 0.610–1.747
0.40 0.030 0.91
0.776 0.534 0.874
0.477–1.262 0.322–0.886 0.485–1.576
0.31 0.015 0.66
2.887 2.601 4.036
1.456–5.722 1.307–5.177 2.056–7.926
0.002 0.006 b0.001
2.720 2.399 3.075
1.311–5.646 1.151–5.001 1.487–6.359
0.007 0.020 0.002
1.004 1.047 0.539 0.989 1.804 0.632 0.577
0.589–1.711 0.700–1.566 0.297–0.978 0.530–1.846 1.098–2.964 0.355–1.124 0.203–1.641
0.99 0.82 0.042 0.97 0.020 0.12 0.30
0.965 1.167 0.561 0.929 1.695 – –
0.552–1.687 0.747–1.825 0.299–1.052 0.477–1.810 0.984–2.919 – –
0.90 0.50 0.07 0.83 0.06 – –
1.885 1.726
1.367–2.598 1.130–2.636
b0.001 0.012
1.980 1.631 –
1.335–2.936 1.001–2.657 –
0.001 0.05 –
1.018 0.914
0.722–1.435 0.709–1.179
0.92 0.49
OR indicates odds ratio; CI, confidence interval; Ref, reference group; cm, centimeter; OG, open gastrectomy; MIG, minimally-invasive gastrectomy; LN, lymph node. a Bold indicates significant p-values. b Factors significant (p b .05) on univariable analysis were included in the multivariable analysis.
operative mortality in patients ≥70 years (from 10.4% to 4.4%) than compared to the overall population of all ages (from 9.8% to 5.1%) [36]. As facility type was the only factor we identified in the present study to be associated with decreased 90-day mortality among octogenarians, the potential centralization of gastrectomies should be considered. Of note, we did not find any association between the use of neoadjuvant chemotherapy or radiation, surgical approach by MIG vs. OG, or extent of lymph node sampling (b15 vs ≥15LN) and 90-day mortality. In the present study, we also found octogenarians to have decreased rates of adequate lymph node retrieval of ≥15 LNs compared to younger groups (50.5% in patients ≥80 years vs. 57.7% in those 65–79 years vs. 61.8% in b65 years). We did not find LN retrieval of ≥15 to be associated with 90-day mortality (OR 0.914, 95% CI 0.709–1.179, p = .49). The progressively decreasing rate of LN retrieval ≥15 by increasing age group reflects a growing trend towards a limited lymphadenectomy in older patients, particularly as several studies have demonstrated a lack of survival benefit of extended lymphadenectomy in this population [37–40]. The limitations of this study include the potential selection bias inherent to a retrospective review. Secondly, we used proxy measures of oncologic outcomes based on the variables captured in the NCDB, such as utilizing the number of lymph nodes as a surrogate marker for extent of lymph node dissection which is typically measured anatomically by the lymph node stations. Thirdly, we accounted for the potential effect of surgery facility type in the outcomes analysis but were not able to adjust for individual surgeon or provider-level effects. Lastly, the NCDB does not include data on post-operative complications which is another important outcome measure to consider when evaluating the benefits of MIG in the older patient population. 5. Conclusions Among the Western population represented in this study, MIG achieved similar oncologic and post-operative outcomes as OG among
octogenarians. However, the overall post-operative mortality rate among octogenarians remains unacceptably high and is independently associated with the facility type where the surgery was performed. Better patient selection for surgery and efforts to refer older patients with gastric cancer to higher volume, academic facilities should be considered to improve post-operative outcomes. Conflict of interest No authors declare no conflicts of interest. Disclosures The National Cancer Data Base (NCDB) is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The CoC's NCDB and the hospitals participating in the CoC NCDB are the source of the de-identified data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Author contributions LMP, TY, and JW contributed to study conception and design. LMP and TY contributed to data acquisition and data quality control. LMP, TY, and JW contributed to data analysis and interpretation. LMP and JW contributed to statistical analysis and manuscript preparation. LMP, TY, and JW contributed to manuscript editing and manuscript review. Acknowledgements None.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
6
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
References [1] Bray F, Jemal A, Grey N, Ferlay J, Forman D. Global cancer transitions according to the Human Development Index (2008-2030): a population-based study. Lancet Oncol 2012;13(8):790–801. [2] Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136(5):E359–86. [3] Kitamura K, Yamaguchi T, Taniguchi H, et al. Clinicopathological characteristics of gastric cancer in the elderly. Br J Cancer 1996;73(6):798–802. [4] Bittner R, Butters M, Ulrich M, Uppenbrink S, Beger HG. Total gastrectomy. Updated operative mortality and long-term survival with particular reference to patients older than 70 years of age. Ann Surg 1996;224(1):37–42. [5] Crew KD, Neugut AI. Epidemiology of gastric cancer. World J Gastroenterol 2006;12 (3):354–62. [6] Thakkar JP, McCarthy BJ, Villano JL. Age-specific cancer incidence rates increase through the oldest age groups. Am J Med Sci 2014;348(1):65–70. [7] Fujiwara Y, Fukuda S, Tsujie M, et al. Effects of age on survival and morbidity in gastric cancer patients undergoing gastrectomy. World journal of gastrointestinal oncology 2017;9(6):257–62. [8] Hsu JT, Liu MS, Wang F, et al. Standard radical gastrectomy in octogenarians and nonagenarians with gastric cancer: are short-term surgical results and long-term survival substantial? Journal of gastrointestinal surgery 2012;16(4):728–37. [9] Persiani R, Antonacci V, Biondi A, et al. Determinants of surgical morbidity in gastric cancer treatment. J Am Coll Surg 2008;207(1):13–9. [10] Kitano S, Shiraishi N, Fujii K, Yasuda K, Inomata M, Adachi Y. A randomized controlled trial comparing open vs laparoscopy-assisted distal gastrectomy for the treatment of early gastric cancer: an interim report. Surgery 2002;131(1 Suppl): S306–11. [11] Suzuki S, Nakamura T, Imanishi T, et al. Carbon dioxide pneumoperitoneum led to no severe morbidities for the elderly during laparoscopic-assisted distal gastrectomy. Ann Surg Oncol 2015;22(5):1548–54. [12] Kwon IG, Cho I, Guner A, Kim HI, Noh SH, Hyung WJ. Minimally invasive surgery as a treatment option for gastric cancer in the elderly: comparison with open surgery for patients 80 years and older. Surg Endosc 2015;29(8):2321–30. [13] Inokuchi M, Tanioka T, Nakagawa M, Okuno K, Gokita K, Kojima K. Laparoscopic Distal Gastrectomy is Feasible in Very Elderly Patients as Compared with Open Distal Gastrectomy. Journal of investigative surgery 2017:1–7. [14] Yasuda K, Sonoda K, Shiroshita H, Inomata M, Shiraishi N, Kitano S. Laparoscopically assisted distal gastrectomy for early gastric cancer in the elderly. Br J Surg 2004;91 (8):1061–5. [15] Okumura N, Son T, Kim YM, et al. Robotic gastrectomy for elderly gastric cancer patients: comparisons with robotic gastrectomy in younger patients and laparoscopic gastrectomy in the elderly. Gastric cancer 2016;19(4):1125–34. [16] Tokunaga M, Hiki N, Fukunaga T, et al. Does age matter in the indication for laparoscopy-assisted gastrectomy? Journal of gastrointestinal surgery 2008;12(9): 1502–7. [17] Cho GS, Kim W, Kim HH, Ryu SW, Kim MC, Ryu SY. Multicentre study of the safety of laparoscopic subtotal gastrectomy for gastric cancer in the elderly. Br J Surg 2009;96 (12):1437–42. [18] Cai J, Wei D, Gao CF, Zhang CS, Zhang H, Zhao T. A prospective randomized study comparing open versus laparoscopy-assisted D2 radical gastrectomy in advanced gastric cancer. Dig Surg 2011;28(5–6):331–7.
[22] Ramesh HS, Pope D, Gennari R, Audisio RA. Optimising surgical management of elderly cancer patients. World J Surg Oncol 2005;3(1):17. [23] Aging NIo, Health NIo. Why Population Aging Matters: A Global Perspective; 2007https://www.nia.nih.gov/sites/default/files/2017-06/WPAM.pdf, Accessed date: 5 September 2017. [24] Inokuchi M, Nakagawa M, Tanioka T, Okuno K, Gokita K, Kojima K. Long- and shortterm outcomes of laparoscopic gastrectomy versus open gastrectomy in patients with clinically and pathological locally advanced gastric cancer: a propensity-score matching analysis. Surg Endosc 2018;32(32):735–42. [25] Balakrishnan M, George R, Sharma A, Graham DY. Changing Trends in Stomach Cancer Throughout the World. Curr Gastroenterol Rep 2017;19(8):36. [26] Yamamoto S. Stomach cancer incidence in the world. Jpn J Clin Oncol 2001;31(9): 471. [27] Sasako M, Inoue M, Lin JT, Khor C, Yang HK, Ohtsu A. Gastric Cancer Working Group report. Jpn J Clin Oncol 2010;40(Suppl. 1):i28–37. [28] Eguchi T, Fujii M, Takayama T. Mortality for gastric cancer in elderly patients. J Surg Oncol 2003;84(3):132–6. [29] Patel VI, Mukhopadhyay S, Ergul E, et al. Impact of hospital volume and type on outcomes of open and endovascular repair of descending thoracic aneurysms in the United States Medicare population. J Vasc Surg 2013;58(2):346–54. [30] Birkmeyer JD, Finlayson SR, Tosteson AN, Sharp SM, Warshaw AL, Fisher ES. Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery 1999;125(3):250–6. [31] Gouma DJ, van Geenen RC, van Gulik TM, et al. Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000;232(6):786–95. [32] Lidsky ME, Sun Z, Nussbaum DP, Adam MA, Speicher PJ, Blazer DG, III. Going the Extra Mile: Improved Survival for Pancreatic Cancer Patients Traveling to Highvolume Centers. Ann Surg 2016. [33] O'Mahoney PRA, Yeo HL, Sedrakyan A, et al. Centralization of pancreatoduodenectomy a decade later: Impact of the volume-outcome relationship. Surgery 2016;159(6):1528–38. [34] Topal B, Van de Sande S, Fieuws S, Penninckx F. Effect of centralization of pancreaticoduodenectomy on nationwide hospital mortality and length of stay. Br J Surg 2007;94(11):1377–1381. [35] Gooiker GA, Lemmens VE, Besselink MG, et al. Impact of centralization of pancreatic cancer surgery on resection rates and survival. Br J Surg 2014;101(8):1000–5. [36] de Wilde RF, Besselink MG, van der Tweel I, et al. Impact of nationwide centralization of pancreaticoduodenectomy on hospital mortality. Br J Surg 2012;99(3): 404–10. [37] Hartgrink HH, van de Velde CJ, Putter H, et al. Extended lymph node dissection for gastric cancer: who may benefit? Final results of the randomized Dutch gastric cancer group trial. Journal of Clinical Oncology 2004;22(11):2069–77. [38] Rausei S, Ruspi L, Rosa F, et al. Extended lymphadenectomy in elderly and/or highly co-morbid gastric cancer patients: A retrospective multicenter study. Eur J Surgical Oncology 2016;42(12):1881–9. [39] Passot G, Vaudoyer D, Messager M, et al. Is Extended Lymphadenectomy Needed for Elderly Patients With Gastric Adenocarcinoma? Ann Surg Oncol 2016;23(8): 2391–7. [40] Takeshita H, Ichikawa D, Komatsu S, et al. Surgical outcomes of gastrectomy for elderly patients with gastric cancer. World J Surg 2013;37(12):2891–8.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
Contents lists available at ScienceDirect
Journal of Geriatric Oncology
The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study☆ Linda M. Pak a,b,⁎, Tingsong Yang a, Jiping Wang a,b a b
Division of Surgical Oncology, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States Center for Surgery and Public Health, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States
a r t i c l e
i n f o
Article history: Received 12 June 2018 Received in revised form 6 September 2018 Accepted 17 October 2018 Available online xxxx Keywords: Gastrectomy Gastric cancer Octogenarians
a b s t r a c t Objective: The safety of minimally-invasive (MIG) and open gastrectomy (OG) in the older patients has been demonstrated in several international studies but has not been evaluated in the context of a large, Western population. The objective of this study was to evaluate the safety of gastrectomy by these two approaches among octogenarians in the United States. Materials and methods: The National Cancer Database (2010–2014) was queried for patients with gastric adenocarcinoma who underwent MIG or OG. Disease and treatment characteristics and outcomes were compared across age groups (b65, 65–79, ≥80 years). Multivariable regression analysis was used to identify factors associated with 90-day mortality. Results: 13,845 patients were identified who met study inclusion criteria, of which 2140 patients (16%) were aged ≥80 years. Among octogenarians, MIG was associated with slightly decreased length of stay (11.2 vs 12.7 days, p b .001) compared to OG, with no difference in the rate of margin-positive resections, adequate lymph node sampling, or readmission (p = .30–0.77). 90-day mortality for both OG (17%) and MIG (17.2%) was significantly higher among octogenarians compared to younger patients (p b .001). Treatment at an academic facility was an independent predictor of decreased 90-day mortality among octogenarians (OR 0.534, 95%CI 0.322–0.886, p = .015). Conclusion: In this Western population, we report comparable oncologic and post-operative outcomes between MIG and OG. However, the overall post-operative mortality rate among octogenarians remains unacceptably high. Better patient selection criteria for surgery and efforts to refer these patients to higher volume, academic facilities should be considered to improve patient outcomes. © 2018 Elsevier Ltd. All rights reserved.
1. Introduction Gastric cancer is the fifth most common cause of cancer and the third leading cause of cancer-related deaths worldwide [1,2]. With a peak incidence in the seventh and eighth decades of life, gastric cancer has demonstrated a rising incidence in the older patients as life expectancies have increased [3–6]. The mainstay of treatment is resection, typically involving a partial or total gastrectomy with D2 lymphadenectomy to ensure all microscopic disease has been surgically cleared. However, many older patients with gastric cancer also present with multiple comorbidities and decreased functional reserve which can hinder their ability to recover from a large abdominal operation. Several ☆ This study was presented as a poster presentation at the Society for Surgical Oncology 71st Annual Cancer Symposium, March 21-24, 2018, in Chicago, IL. ⁎ Corresponding author at: 75 Francis St, Boston, MA 02215, United States E-mail address: [email protected] (L.M. Pak).
studies have previously shown older patients to have an increased risk of post-operative morbidity and mortality after open gastrectomy (OG) [7–9]. The introduction of minimally-invasive gastrectomy (MIG) has garnered significant interest as a less invasive and traumatic approach to resection, especially for these older patients. Prior studies evaluating the physiologic effects of pneumoperitoneum have found laparoscopic gastrectomy to be associated with decreased impairment of pulmonary function post-operatively when compared to open gastrectomy, as well minimal adverse cardiac effects [10,11]. Subsequent international studies, primarily consisting of Asian patients, have demonstrated similar or improved post-operative outcomes, morbidity, and mortality in MIG compared to OG among older patients [12–18]. However, the safety and efficacy of either technique of gastrectomy among older patients has not been studied within the context of the Western population.
https://doi.org/10.1016/j.jgo.2018.10.012 1879-4068/© 2018 Elsevier Ltd. All rights reserved.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
2
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
In this context, the objective of this study was to evaluate the safety of both OG and MIG among octogenarians in the United States using population-based data from the National Cancer Database (NCDB).
2. Methods This was a retrospective, population-based study using data provided by the esophagogastric participant use file of the NCDB. The NCDB is a collaborative program between the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The NCDB currently captures 70% of newly diagnosed cancer cases in the United States and collects data from over 1500 CoCaccredited cancer hospitals19. The NCDB was queried for all patients ≥18 years of age with invasive gastric adenocarcinoma with potentially resectable disease (defined as American Joint Committee on Cancer AJCC Stage Ib-IIIc) who underwent elective partial or total gastrectomy with curative intent. This data query was limited to the years 2010 through 2014 as the data on operative approach (open, laparoscopic, or robotic) only became available since 2010. Patients were excluded from analysis if they had less than a partial gastrectomy performed, a palliative resection, or an emergent resection. A resection was defined as emergent if the number of days between diagnosis and surgery was recorded as b1 day. Disease staging was performed in accordance with the AJCC Staging System, 7th edition20. Sufficient lymph node sampling was defined as ≥15 lymph nodes examined in accordance with the National Comprehensive Cancer Network (NCCN) guidelines21. MIG included patients who had undergone either laparoscopic or robotic gastrectomy. Patient demographic, disease and treatment characteristics, and post-operative and long-term outcomes were abstracted from NCDB for patients meeting inclusion criteria.
Patients were categorized into three age groups for analysis (b65 years of age, 65–79 years, and ≥ 80 years). While there is no standard definition, many groups have used a cut-off of 65 years and older to define the older patient population. However, this definition is rapidly shifting to an older age parameter as the average life expectancy in many countries often exceeds 80 years of age; thus, we elected to stratify patients into the age groups listed above and define “older patients” as those age ≥ 80 years [22,23]. Clinical and treatment characteristics and patient outcomes were compared among these three age groups using the chi-squared test for categorical variables and the KruskalWallis test for continuous variables. The primary outcome evaluated in this study was the safety of gastrectomy as defined by the 90-day post-operative mortality in octogenarians. Multivariable logistic regression tests were used to identify patient characteristics and outcome measures associated with 90-day mortality. Factors significant on univariable analysis were included in the multivariable analysis. The results of the multivariable analysis were presented as odds ratios (OR) with the associated 95% confidence intervals (CI) and p-values. A p-value b.05 and a 95% CI exclusive of 1 were considered significant. Post hoc subgroup analysis was performed comparing outcome measures between community and academic facilities, as this was identified in the multivariable analysis as a significant predictor of 90-day mortality. All statistical analysis was performed using STATA software (v14.1, StataCorp, College Station, TX). This study was approved by the Partners Health Research Committee institutional review board. 3. Results Between 2010 and 2014, 13,845 patients were identified who metstudy inclusion criteria (Fig. 1). Overall, 42.2% of patients were b 65 years of age (n = 5843), 42.3% were 65–79 years of age (n =
Patients in NCDB ≥18 years with gastric cancer 2010-2014 n = 84,414 Excluded: • 4,719 neuroendocrine tumors • 8,494 sarcomas • 530 histology unknown Patients with gastric carcinoma n = 70,671
Patients who underwent partial or total gastrectomy with curative intent n = 22,490
Excluded: • 43,416 no surgery • 2,525 local excision only • 252 unknown surgery • 183 palliative resections • 1,805 emergent procedures
Excluded: • 117 Stage 0 • 4,021 Stage 1a • 1,262 Stage 4 • 3,254 unknown stage Patients with potentially resectable disease (AJCC stage Ib-IIIc) n = 13,836
Patients aged <65 years n = 5,837
Patients aged 65-79 years n = 5,859
Patients aged ≥80 years n = 2,140
Fig. 1. Flowchart of study population selection.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx Table 1 Comparison of patient and treatment characteristics among patients with gastric cancer aged b65 years, 65–79 years, and ≥ 80 years. b65 years n = 5843 Age, in years, median (range) 56 (18–64) Sex, male 3865 (66.2%) Race White 3872 (66.3%) Black 1171 (20.0%) Asian 574 (9.8%) Other/Unknown 226 (3.9%) Charlson-Deyo Score 0 4250 (72.7%) 1 1268 (21.7%) 2 325 (5.6%) Tumor Size ≤2 cm 762 (15.1%) 2-4 cm 1660 (32.9%) 4-6 cm 1282 (25.4%) N6 cm 1345 (26.6%) Tumor Location Proximal 2097 (35.9%) Body 469 (8.0%) Distal 1439 (24.6%) Lesser curve 570 (9.8%) Greater curve 234 (4.0%) Overlapping 516 (8.8%) Unknown 518 (8.9%) Tumor Differentiation Well-differentiated 153 (2.6%) Moderately-differentiated 1245 (21.3%) Poorly-differentiated 3973 (68.0%) Undifferentiated 127 (2.2%) Unknown 345 (5.9%) AJCC T Classification T1 347 (5.9%) T2 1046 (17.9%) T3 2859 (48.9%) T4 1591 (27.2%) AJCC N Classification N0 2002 (34.3%) N1 1288 (22.0%) N2 1117 (19.1%) N3 1436 (24.6%) AJCC Stage Stage Ib 828 (14.2%) Stage II 2421 (41.4%) Stage III 2594 (44.4%) Lymphovascular Invasion Yes 2420 (41.4%) No 2533 (43.4%) Unknown 890 (15.2%) Extent of Resection
65–79 years ≥80 years n = 5862 n = 2140
p-valuea
72 (65–79) 3879 (66.2%)
– b0.001
83 (80–90) 1177 (55.0%)
b0.001 4268 (72.8%) 908 (15.5%)
1603 (74.9%) 274 (12.8%)
546 (9.3%) 140 (2.4%)
223 (10.4%) 40 (1.9%)
3591 (61.3%) 1653 (28.2%) 618 (10.5%)
1280 (59.8%) 578 (27.0%)
3
Table 1 (continued)
Partial gastrectomy Total gastrectomy Multivisceral/en bloc Surgical Approach Open Laparoscopic Robotic Chemotherapy None
b65 years n = 5843
65–79 years ≥80 years n = 5862 n = 2140
3745 (64.1%) 1362 (23.3%) 736 (12.6%)
3950 (67.4%) 1233 (21.0%) 679 (11.6%)
1636 (76.5%) 318 (14.9%)
4678 (80.1%) 941 (16.1%) 224 (3.8%)
4676 (79.8%) 971 (16.6%) 215 (3.7%)
1759 (82.2%) 326 (15.2%) 55 (2.6%)
887 (15.2%)
1804 (30.8%) 1628 (27.8%) 1933 (33.0%) 497 (8.5%)
1519 (71.0%) 298 (13.9%)
Neoadjuvant
282 (13.2%) b0.001
761 (14.5%) 1693 (32.2%) 1377 (26.2%) 1428 (27.2%)
196 (9.7%) 607 (29.9%)
Unknown Radiation therapy None
3571 (60.9%) 1205 (20.6%) 888 (15.2%)
1848 (86.4%) 195 (9.1%)
Adjuvant 597 (29.4%) Neoadjuvant 632 (31.1%) Unknown b0.001
198 (3.4%)
46 (2.2%)
1911 (32.6%) 515 (8.8%) 1521 (26.0%) 609 (10.4%) 263 (4.5%) 493 (8.4%) 550 (9.4%)
281 (13.1%)
209 (3.6%) 1685 (28.7%) 3591 (61.3%) 110 (1.9%) 267 (4.6%)
74 (3.5%) 647 (30.2%)
398 (6.8%) 1191 (20.3%) 2833 (48.3%) 1440 (24.6%)
123 (5.8%) 412 (19.3%)
2132 (36.4%) 1405 (24.0%) 1082 (18.5%) 1243 (21.2%)
715 (33.4%)
984 (16.8%) 2489 (42.5%) 2389 (40.8%)
338 (15.8%) 810 (37.9%)
2565 (43.8%) 2494 (42.6%) 803 (13.7%)
1179 (55.1%) 745 (34.8%)
186 (8.7%) 0.041
b0.001
b0.001 Adjuvant
p-valuea
1845 (31.6%) 2575 (44.1%) 536 (9.2%)
155 (7.2%) 168 (7.85%) b0.001
2903 (49.7%) 1569 (26.9%) 1145 (19.6%) 226 (3.9%)
51 (2.4%)
Cm indicates centimeters; AJCC, American Joint Committee on Cancer. a Bold indicates significant p-values.
230 (10.8%) 848 (39.6%) 239 (11.2%) 131 (6.1%) 198 (9.3%) 213 (10.0%) b0.001
1324 (61.9%) 45 (2.1%) 50 9(2.3%) b0.001
924 (43.2%) 681 (31.8%) b0.001
514 (24.0%) 432 (20.2%) 479 (22.4%) b0.001
992 (46.4%) b0.001
216 (10.1%) b0.001
5862), and 15.5% were ≥ 80 years of age (n = 2140) (Table 1). In comparison to the two younger age groups (b65 years and 65–79 years), older patients aged ≥80 years had more comorbidities and presented with more advanced stage of disease (both p b .001). Histopathologically, older patients had larger, moderately-differentiated tumors located in the distal portion of the stomach associated with the presence of lymphovascular invasion (all p b .001). Octogenarians more often underwent partial gastrectomy rather than total or multi-visceral, en bloc gastrectomy (p b .001). A higher percentage of these older patients underwent OG compared to MIG and only a minority received any neoadjuvant or adjuvant chemoradiation therapy (all p b .001). These disease and treatment details are summarized in Table 1. Comparing short-term post-operative and oncologic outcomes among patients who underwent either OG or MIG, the two older age groups, especially those ged ≥80 years, had consistently poorer outcomes than the group aged b65 years (Table 2). In particular, the 30- and 90-day mortality rates increased dramatically by age group. Patients b65 years at 30- and 90-day mortality rates of 1.8% and 4.4% respectively, compared to 8.4% and 16.9% in patients ≥80 years (p b .001). Older patients also had lower rates of adequate lymph node sampling (50.5% in patients ≥80 years vs. 57.7% in 65–79 years vs. 61.8% in b65 years, p b .001). There was no difference in the rate of marginpositive resections by age group (p = .21). Among octogenarians, 82.2% underwent OG (n = 1759) and 17.8% underwent MIG (n = 381) (Table 3). Octogenarians who underwent MIG were more likely to be treated at an academic comprehensive cancer program compared to those who underwent OG (p b .001), but there were no differences in the extent of gastrectomy performed or the use of neoadjuvant or adjuvant chemoradiation (p = .07–0.45). In evaluating post-operative outcomes between octogenarians who had MIG compared to OG, there was no difference in the rate of marginpositive resections, adequacy of lymph node sampling, 30-day readmissions, or 30- or 90-day mortality (p = .30–0.92) (Fig. 2). Patients who underwent MIG did have a slightly decreased length of stay compared to OG patients (mean 11.2 vs. 12.7 days, p b .001). On univariable analysis, factors associated with increased likelihood of 90-day mortality among octogenarians who underwent either OG or
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
4
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
Table 2 Comparison of post-operative and oncologic outcomes in patients of various age groups who underwent gastrectomy.
LOS, days, mean (standard deviation) 30-day Readmission 30-day Mortality 90-day Mortality Positive Margin Lymph node sampling ≥15 LN
b65 years n = 5843
65–79 years n = 5862
≥80 years n = 2140
p-valuea
11.8 ± 10.1 359 (6.1%) 85 (1.8%) 201 (4.4%) 912 (15.7%) 3594 (61.8%)
12.6 ± 10.6 442 (7.5%) 245 (5.2%) 462 (10.0%) 866 (14.9%) 3365 (57.7%)
12.4 ± 9.6 177 (8.3%) 143 (8.3%) 289 (16.9%) 347 (16.3%) 1079 (50.5%)
b0.001 0.001 b0.001 b0.001 0.21 b0.001
LOS indicates length of stay; LN, lymph nodes. a Bold indicates significant p-values.
MIG were increased tumor size, tumor location diffusely throughout the stomach, and performance of total gastrectomy or multivisceral/en-bloc resection (Table 4). Asian race, treatment at an academic/research facility, and tumor location along the lesser curve of the stomach were associated with decreased likelihood of 90-day mortality. The surgical approach (OG vs. MIG) was not significantly associated with 90-day mortality among octogenarians. On multivariable analysis, increased tumor size and total gastrectomy were independently associated with increased likelihood of 90-day mortality while Asian race, treatment at academic/research facility were independently associated with decreased likelihood of 90-day mortality (Table 4.) On post hoc subgroup analysis comparing post-operative outcomes among octogenarians between community and academic facilities, octogenarians who underwent gastrectomy at academic facilities had lower rates of 90-day mortality (10.7% vs 17.4%, p = .017), lower rates of margin-positive resections (13.1% vs 20.7%, p = .030), and higher rates of adequate lymph node sampling (58.3% vs 40.2%, p b .001) (Supplemental Table 1). 4. Discussion As life expectancies increase globally and more older patients are being diagnosed with gastric cancer, the indications for surgery have Table 3 Comparison of treatment characteristics and post-operative outcomes in octogenarians who underwent open compared to minimally-invasive gastrectomy.
Chemotherapy None Adjuvant Neoadjuvant Unknown Radiation therapy None Adjuvant Neoadjuvant Unknown Surgery Facility Type Community Comprehensive Community Academic Integrated Network Extent of Resection Partial gastrectomy Total gastrectomy Multivisceral/en bloc LOS, days, mean (standard deviation) 30-day Readmission 30-day Mortality 90-day Mortality Positive Margin Lymph node sampling ≥15 LN
OG N = 1759
MIG N = 381
1244 (70.7%) 242 (13.8%) 126 (7.2%) 147 (8.4%)
275 (72.2%) 56 (14.7%) 29 (7.6%) 21 (5.5%)
1520 (86.4%) 164 (9.3%) 38 (2.2%) 37 (2.1%)
328 (86.1%) 31 (8.1%) 13 (3.4%) 9 (2.4%)
165 (9.4%) 751 (42.7%) 658 (37.4%) 185 (10.5%)
19 (5.0%) 142 (37.3%) 183 (48.0%) 37 (9.7%)
1328 (75.5%) 269 (15.3%) 162 (9.2%) 12.7 ± 10.0 141 (8.0%) 117 (8.1%) 242 (16.9%) 292 (16.7%) 889 (50.7%)
308 (80.8%) 49 (12.9%) 23 (6.3%) 11.2 ± 7.9 36 (9.5%) 26 (9.4%) 47 (17.2%) 55 (14.6%) 190 (49.9%)
p-valuea 0.31
0.45
b0.001
0.07
b0.001 0.36 0.47 0.92 0.30 0.77
OG indicates open gastrectomy; MIG, minimally-invasive gastrectomy; LOS, length of stay; LN, lymph nodes. a Bold indicates significant p-values.
continued to expand to accommodate the shifting epidemiology of this disease. There have been several reports in the literature on the safety and efficacy of minimally-invasive gastrectomy for the treatment of gastric cancer in older patients compared to open gastrectomy [12,14,15,24]. However, these studies have emerged predominantly from hospital centers in Asia which tend to encounter and treat a different population demographic compared to Western centers [5,25–27]. In this context, the present study examined the safety and efficacy of both open and minimally-invasive gastrectomy among octogenarians in the United States using the NCDB. While we did not identify any clinically significant differences in short-term oncologic outcomes or post-operative outcomes between octogenarians who underwent MIG compared to OG, octogenarians consistently had significantly worse outcomes than younger age groups. The 30- and 90-day mortality rates for patients ≥80 years was 8.4% and 16.9%, respectively, which was nearly four times higher than that of the youngest age group b65 years. This postoperative mortality rate among octogenarians reported in the NCDB is significantly higher than prior reports in the literature, largely from Asian institutions. In a series of 2422 patients who underwent open gastrectomy at Chang Gung Memorial Hospital in Taiwan, including 164 patients ≥80 years, the 30-day mortality was 3% among octogenarians compared to 2% among non-octogenarians (p = .18) [8]. Overall, patients had the same median age of 83 years, similar distribution in sex, comorbidities, extent of disease, and extent of gastric resection, as compared to the present study. In another series of 2175 patients treated at Nihon University in Japan, which included 80 patients ≥80 years, Eguchi et al. reported a similarly low 30-day mortality rate of 4% among octogenarians, with no difference when compared to younger patients b80 years (3%, p = .84) [28]. The relatively higher post-operative mortality rates among octogenarians reported in the NCDB from over 1500 CoC-accredited hospitals compared to prior reports from Asian institutions indicates a pressing need for quality assessment and improvement measures, as these other institutions have set the example that safe surgery is possible among older patients with gastric cancer. In the present study, we identified treatment at an academic comprehensive cancer program as an independent factor associated with reduced 90-day mortality (OR 0.534, 95% CI 0.322–0.886); we also found that octogenarians treated at academic facilities had lower rates of 90-day mortality (p = .017), lower rates of margin-positive resections (p = .030), and higher rates of adequate of lymph node sampling (p b .001), compared to those treated at community facilities. In addition to having the highest minimum caseload requirement of all CoC facility categorizations (N500 newly diagnosed cancer cases/year), academic facilities also participate in cancerrelated clinical research/trials and in training of resident physicians. The impact of hospital volume on improved outcomes has been well established [29–32]. This has prompted efforts to promote centralization of certain procedures to high-volume centers. The centralization of pancreaticoduodenectomies, one of the most complex abdominal surgical procedures, worldwide to higher-volume centers has led to consistently decreased mortality rates and improved outcomes [33–35]. This hospital volume-outcome relationship may be particularly critical for older patients. In a study of 2155 pancreaticoduodenectomies performed in the Netherlands, de Wilde et al. identified a larger reduction in post-
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
5
Table 4 Univariable and multivariable logistic regression model for 90-day mortality in octogenarians undergoing gastrectomy. Multivariableb
Univariable
Gender (Ref: Male) Race (Ref: White) Black Asian Charlson/Deyo Score (Ref: 0) 1 2 Facility Type (Ref: Community) Comprehensive Community Academic Integrated Network Tumor Size (Ref: ≤2 cm) 2-4 cm 4-6 cm N6 cm Tumor Location (Ref: Proximal) Body Distal Lesser curve Greater curve Overlapping Neoadjuvant chemotherapy Neoadjuvant radiation therapy Resection (Ref: Partial) Total gastrectomy Multivisceral/en bloc Surgical approach (Ref: OG) MIG Adequate lymph node sampling ≥15 LN
OR
95% CI
p-valuea
OR
95% CI
p-value
0.872
0.675–1.125
0.29
–
–
–
0.754 0.316
0.505–1.126 0.173–0.577
0.17 b0.001
0.833 0.354 –
0.537–1.291 0.190–0.659 –
0.41 0.001 –
0.888 1.121
0.658–1.200 0.772–1.627
0.440 0.549
0.831 0.608 1.033
0.539–1.283 0.389–0.953 0.610–1.747
0.40 0.030 0.91
0.776 0.534 0.874
0.477–1.262 0.322–0.886 0.485–1.576
0.31 0.015 0.66
2.887 2.601 4.036
1.456–5.722 1.307–5.177 2.056–7.926
0.002 0.006 b0.001
2.720 2.399 3.075
1.311–5.646 1.151–5.001 1.487–6.359
0.007 0.020 0.002
1.004 1.047 0.539 0.989 1.804 0.632 0.577
0.589–1.711 0.700–1.566 0.297–0.978 0.530–1.846 1.098–2.964 0.355–1.124 0.203–1.641
0.99 0.82 0.042 0.97 0.020 0.12 0.30
0.965 1.167 0.561 0.929 1.695 – –
0.552–1.687 0.747–1.825 0.299–1.052 0.477–1.810 0.984–2.919 – –
0.90 0.50 0.07 0.83 0.06 – –
1.885 1.726
1.367–2.598 1.130–2.636
b0.001 0.012
1.980 1.631 –
1.335–2.936 1.001–2.657 –
0.001 0.05 –
1.018 0.914
0.722–1.435 0.709–1.179
0.92 0.49
OR indicates odds ratio; CI, confidence interval; Ref, reference group; cm, centimeter; OG, open gastrectomy; MIG, minimally-invasive gastrectomy; LN, lymph node. a Bold indicates significant p-values. b Factors significant (p b .05) on univariable analysis were included in the multivariable analysis.
operative mortality in patients ≥70 years (from 10.4% to 4.4%) than compared to the overall population of all ages (from 9.8% to 5.1%) [36]. As facility type was the only factor we identified in the present study to be associated with decreased 90-day mortality among octogenarians, the potential centralization of gastrectomies should be considered. Of note, we did not find any association between the use of neoadjuvant chemotherapy or radiation, surgical approach by MIG vs. OG, or extent of lymph node sampling (b15 vs ≥15LN) and 90-day mortality. In the present study, we also found octogenarians to have decreased rates of adequate lymph node retrieval of ≥15 LNs compared to younger groups (50.5% in patients ≥80 years vs. 57.7% in those 65–79 years vs. 61.8% in b65 years). We did not find LN retrieval of ≥15 to be associated with 90-day mortality (OR 0.914, 95% CI 0.709–1.179, p = .49). The progressively decreasing rate of LN retrieval ≥15 by increasing age group reflects a growing trend towards a limited lymphadenectomy in older patients, particularly as several studies have demonstrated a lack of survival benefit of extended lymphadenectomy in this population [37–40]. The limitations of this study include the potential selection bias inherent to a retrospective review. Secondly, we used proxy measures of oncologic outcomes based on the variables captured in the NCDB, such as utilizing the number of lymph nodes as a surrogate marker for extent of lymph node dissection which is typically measured anatomically by the lymph node stations. Thirdly, we accounted for the potential effect of surgery facility type in the outcomes analysis but were not able to adjust for individual surgeon or provider-level effects. Lastly, the NCDB does not include data on post-operative complications which is another important outcome measure to consider when evaluating the benefits of MIG in the older patient population. 5. Conclusions Among the Western population represented in this study, MIG achieved similar oncologic and post-operative outcomes as OG among
octogenarians. However, the overall post-operative mortality rate among octogenarians remains unacceptably high and is independently associated with the facility type where the surgery was performed. Better patient selection for surgery and efforts to refer older patients with gastric cancer to higher volume, academic facilities should be considered to improve post-operative outcomes. Conflict of interest No authors declare no conflicts of interest. Disclosures The National Cancer Data Base (NCDB) is a joint project of the Commission on Cancer (CoC) of the American College of Surgeons and the American Cancer Society. The CoC's NCDB and the hospitals participating in the CoC NCDB are the source of the de-identified data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. Author contributions LMP, TY, and JW contributed to study conception and design. LMP and TY contributed to data acquisition and data quality control. LMP, TY, and JW contributed to data analysis and interpretation. LMP and JW contributed to statistical analysis and manuscript preparation. LMP, TY, and JW contributed to manuscript editing and manuscript review. Acknowledgements None.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012
6
L.M. Pak et al. / Journal of Geriatric Oncology xxx (xxxx) xxx–xxx
References [1] Bray F, Jemal A, Grey N, Ferlay J, Forman D. Global cancer transitions according to the Human Development Index (2008-2030): a population-based study. Lancet Oncol 2012;13(8):790–801. [2] Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136(5):E359–86. [3] Kitamura K, Yamaguchi T, Taniguchi H, et al. Clinicopathological characteristics of gastric cancer in the elderly. Br J Cancer 1996;73(6):798–802. [4] Bittner R, Butters M, Ulrich M, Uppenbrink S, Beger HG. Total gastrectomy. Updated operative mortality and long-term survival with particular reference to patients older than 70 years of age. Ann Surg 1996;224(1):37–42. [5] Crew KD, Neugut AI. Epidemiology of gastric cancer. World J Gastroenterol 2006;12 (3):354–62. [6] Thakkar JP, McCarthy BJ, Villano JL. Age-specific cancer incidence rates increase through the oldest age groups. Am J Med Sci 2014;348(1):65–70. [7] Fujiwara Y, Fukuda S, Tsujie M, et al. Effects of age on survival and morbidity in gastric cancer patients undergoing gastrectomy. World journal of gastrointestinal oncology 2017;9(6):257–62. [8] Hsu JT, Liu MS, Wang F, et al. Standard radical gastrectomy in octogenarians and nonagenarians with gastric cancer: are short-term surgical results and long-term survival substantial? Journal of gastrointestinal surgery 2012;16(4):728–37. [9] Persiani R, Antonacci V, Biondi A, et al. Determinants of surgical morbidity in gastric cancer treatment. J Am Coll Surg 2008;207(1):13–9. [10] Kitano S, Shiraishi N, Fujii K, Yasuda K, Inomata M, Adachi Y. A randomized controlled trial comparing open vs laparoscopy-assisted distal gastrectomy for the treatment of early gastric cancer: an interim report. Surgery 2002;131(1 Suppl): S306–11. [11] Suzuki S, Nakamura T, Imanishi T, et al. Carbon dioxide pneumoperitoneum led to no severe morbidities for the elderly during laparoscopic-assisted distal gastrectomy. Ann Surg Oncol 2015;22(5):1548–54. [12] Kwon IG, Cho I, Guner A, Kim HI, Noh SH, Hyung WJ. Minimally invasive surgery as a treatment option for gastric cancer in the elderly: comparison with open surgery for patients 80 years and older. Surg Endosc 2015;29(8):2321–30. [13] Inokuchi M, Tanioka T, Nakagawa M, Okuno K, Gokita K, Kojima K. Laparoscopic Distal Gastrectomy is Feasible in Very Elderly Patients as Compared with Open Distal Gastrectomy. Journal of investigative surgery 2017:1–7. [14] Yasuda K, Sonoda K, Shiroshita H, Inomata M, Shiraishi N, Kitano S. Laparoscopically assisted distal gastrectomy for early gastric cancer in the elderly. Br J Surg 2004;91 (8):1061–5. [15] Okumura N, Son T, Kim YM, et al. Robotic gastrectomy for elderly gastric cancer patients: comparisons with robotic gastrectomy in younger patients and laparoscopic gastrectomy in the elderly. Gastric cancer 2016;19(4):1125–34. [16] Tokunaga M, Hiki N, Fukunaga T, et al. Does age matter in the indication for laparoscopy-assisted gastrectomy? Journal of gastrointestinal surgery 2008;12(9): 1502–7. [17] Cho GS, Kim W, Kim HH, Ryu SW, Kim MC, Ryu SY. Multicentre study of the safety of laparoscopic subtotal gastrectomy for gastric cancer in the elderly. Br J Surg 2009;96 (12):1437–42. [18] Cai J, Wei D, Gao CF, Zhang CS, Zhang H, Zhao T. A prospective randomized study comparing open versus laparoscopy-assisted D2 radical gastrectomy in advanced gastric cancer. Dig Surg 2011;28(5–6):331–7.
[22] Ramesh HS, Pope D, Gennari R, Audisio RA. Optimising surgical management of elderly cancer patients. World J Surg Oncol 2005;3(1):17. [23] Aging NIo, Health NIo. Why Population Aging Matters: A Global Perspective; 2007https://www.nia.nih.gov/sites/default/files/2017-06/WPAM.pdf, Accessed date: 5 September 2017. [24] Inokuchi M, Nakagawa M, Tanioka T, Okuno K, Gokita K, Kojima K. Long- and shortterm outcomes of laparoscopic gastrectomy versus open gastrectomy in patients with clinically and pathological locally advanced gastric cancer: a propensity-score matching analysis. Surg Endosc 2018;32(32):735–42. [25] Balakrishnan M, George R, Sharma A, Graham DY. Changing Trends in Stomach Cancer Throughout the World. Curr Gastroenterol Rep 2017;19(8):36. [26] Yamamoto S. Stomach cancer incidence in the world. Jpn J Clin Oncol 2001;31(9): 471. [27] Sasako M, Inoue M, Lin JT, Khor C, Yang HK, Ohtsu A. Gastric Cancer Working Group report. Jpn J Clin Oncol 2010;40(Suppl. 1):i28–37. [28] Eguchi T, Fujii M, Takayama T. Mortality for gastric cancer in elderly patients. J Surg Oncol 2003;84(3):132–6. [29] Patel VI, Mukhopadhyay S, Ergul E, et al. Impact of hospital volume and type on outcomes of open and endovascular repair of descending thoracic aneurysms in the United States Medicare population. J Vasc Surg 2013;58(2):346–54. [30] Birkmeyer JD, Finlayson SR, Tosteson AN, Sharp SM, Warshaw AL, Fisher ES. Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery 1999;125(3):250–6. [31] Gouma DJ, van Geenen RC, van Gulik TM, et al. Rates of complications and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000;232(6):786–95. [32] Lidsky ME, Sun Z, Nussbaum DP, Adam MA, Speicher PJ, Blazer DG, III. Going the Extra Mile: Improved Survival for Pancreatic Cancer Patients Traveling to Highvolume Centers. Ann Surg 2016. [33] O'Mahoney PRA, Yeo HL, Sedrakyan A, et al. Centralization of pancreatoduodenectomy a decade later: Impact of the volume-outcome relationship. Surgery 2016;159(6):1528–38. [34] Topal B, Van de Sande S, Fieuws S, Penninckx F. Effect of centralization of pancreaticoduodenectomy on nationwide hospital mortality and length of stay. Br J Surg 2007;94(11):1377–1381. [35] Gooiker GA, Lemmens VE, Besselink MG, et al. Impact of centralization of pancreatic cancer surgery on resection rates and survival. Br J Surg 2014;101(8):1000–5. [36] de Wilde RF, Besselink MG, van der Tweel I, et al. Impact of nationwide centralization of pancreaticoduodenectomy on hospital mortality. Br J Surg 2012;99(3): 404–10. [37] Hartgrink HH, van de Velde CJ, Putter H, et al. Extended lymph node dissection for gastric cancer: who may benefit? Final results of the randomized Dutch gastric cancer group trial. Journal of Clinical Oncology 2004;22(11):2069–77. [38] Rausei S, Ruspi L, Rosa F, et al. Extended lymphadenectomy in elderly and/or highly co-morbid gastric cancer patients: A retrospective multicenter study. Eur J Surgical Oncology 2016;42(12):1881–9. [39] Passot G, Vaudoyer D, Messager M, et al. Is Extended Lymphadenectomy Needed for Elderly Patients With Gastric Adenocarcinoma? Ann Surg Oncol 2016;23(8): 2391–7. [40] Takeshita H, Ichikawa D, Komatsu S, et al. Surgical outcomes of gastrectomy for elderly patients with gastric cancer. World J Surg 2013;37(12):2891–8.
Please cite this article as: Pak LM, et al, The safety and efficacy of gastrectomy for gastric cancer among octogenarians: a western population-based study, J Geriatr Oncol (2018), https://doi.org/10.1016/j.jgo.2018.10.012