Journal of Cancer Therapy, 2013, 4, 80-91
http://dx.doi.org/10.4236/jct.2013.41A012 Published Online January 2013 (http://www.scirp.org/journal/jct)
A Population-Based Outcomes Analysis of the Impact of
Age on Morbidity and Mortality Following Gastrectomy:
An Analysis of 13,799 Patients from the Nationwide
Inpatient Sample Database
Ami Karkar1, Sachin Patil1, Ronald S. Chamberlain1,2,3*
1Department of Surgery, Saint Barnabas Medical Center, Livingston, USA; 2Saint George’s University School of Medicine, St.
George, Grenada; 3Department of Sur g ery, Univer si ty of Medicine and Den tistry of Ne w Jersey, Newark, USA.
Received August 29th, 2012; revised September 30th, 2012; accepted October 12th, 2012
ABSTRACT
Introduction: Gastric cancer is the 4th most common malignancy and second leading cause of cancer-related death
worldwide, both its incidence and mortality have decreased over the past 70 years. Advancing life expectancy, as well
as subtle change in the type and location of gastric cancer in the US, has resulted in an increased number of elderly pa-
tients requiring gastric surgery. By 2050, the population older than 85 years is projected to reach 20.9 million, and as a
result, the need to assess the operative outcomes and mortality following gastrectomy in this group is imperative. This
study sought to assess age-related clinical outcomes following gastric cancer surgery across all age groups to provide
more precise data for pre-operative surgical risk stratification. Methods: Discharge data on 40,276,240 patients was
collected from Nationwide Inpatient Sample Database (NIS) (2004-2008). Data on patients undergoing gastrectomy as
the primary procedure was analyzed including age, gender, elective/non-elective admission, pre-operative length of stay
(LOS), total LOS, and mortality. Eight age groups were compared for two procedures: to tal gastrectomy (TG) and par-
tial gastrectomy (PG). Categorical data was compared using the Chi square test an d continuous data usin g the Student’s
t test. Univariate analysis and multivariate regr ession analysis were performed to assess independent variables. Results:
13,799 patients underwent gastrectomy surgery with 23.7% having TG and 76.3% PG. Gastric carcinoma was the most
common indication for TG, while benign gastric disease was more common for PG, especially in years 51 - 70 (p <
0.001).The mean ag e for TG and PG groups were 63 ± 12.8 and 64 ± 15 years respectively. Males underw ent twice the
number of TGs (p < 0.001), whereas equal number of males and females underwent PG (p < 0.001). The number of
TGs increased over the 5-year study period, with the highest % change noted in those 41 - 50 years (1500%). PGs per-
formed decreased overall, especially in patients <60 years, however PGs increased in patients >81 years with the great-
est % change in the oldest patients >9 1 years (13%). Non-elective admissions were more common for PGs (N = 4844,
41%) than TGs (N = 695, 21.2%). Mean pre-operative LOS and total LOS increased with advancing age for both TG
and PG (p < 0.001). HTN (45%), electrolyte imbalances (28%) and chronic pulmonary disease (18%) were the most
prevalent co-morbidities and significantly affected mortality on univariate analysis (p = 0.001). Respiratory (18%) and
GI complications (11%) were the most common post-operative complications following TG, while GI (9.6%) and bile
duct fistulas (7.2%) were most common after PG. Overall TG and PG mortality rates were 7.6% and 6.4% respectively.
Mortality increased with advanc ing age in both groups (p < 0.001). Multivariate analysis id entified HTN, valvular dis-
ease, anemia, malignancy and non-elective admissions as independent predictors of mortality (p = 0.001). Conclusions:
Advancing age is not an independent predictor of mortality following gastric surgery. Gastrectomy for gastric cancer is
associated with a higher mortality than for benign gastric diseases. Non-elective admission, and pre-existing hyperten-
sion, valvular disease and anemia independently predicted increased morbidity and mortality following gastric surgery
and should be carefully considered in surgical planning and counseling. Gastric carcinoma is the most common indica-
tion for TG, while benign gastric disease is a more common indication for PG. The number of TGs performed is in-
creasing, especially in the males and younger patients, and may reflect an increased incidence of body and cardiac le-
sions.
Keywords: Gastric Cancer; Gastrectomy; Partial Gastrectomy; Total Gastrectomy
*Corresponding a uthor.

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
81
1. Introduction
Gastric cancer is the fourth most common malignancy
and second leading cause of cancer-related death world-
wide. Gastric cancer primarily presents in advanced
stages especially among the elderly (>80 years) and is
associated with a poor outcomes unless diagnosed early
[1]. Hazanaki et al. observed advanced gastric cancer in
56% of patients >80 years compared to 27.9% of patients
<80 years old (p < 0.01). Overall in-hospital mortality
(2% versus 0%), and 5-year overall survival rates (46.1%
versus 71.1% (p < 0.01)) were substantially higher
among those >80 years compare those <80 years [2].
Precise explanations for the higher incidence of advanced
disease and worse outcomes in the elderly remains
largely unknown but is likely related to life expectancy,
pre-existing co-morbidities, as well as age-related differ-
ences in tumor location and histology [3,4]. Complete
surgical resection with a R0 margin provides the highest
survival benefit for gastric cancer, however this is achi-
evable in only a select group of patients (approximately
50%). Moreover, gastrectomy is associated with signifi-
cant morbidity and mortality, particularly in the elderly,
and in-hospital mortality rates as high as 16.6% after
total gastrectomy (TG) and 15.3% after partial gastrec-
tomy (PG) have been reported [5]. The percent of pa-
tients over 80 years is the fastest growing segment of the
US population, and by 2050, those over 85 years is pro-
jected to reach 20.9 million. As a result, there is a grow-
ing and important need to have precise information re-
lated to the operative risks, outcomes and mortality fol-
lowing gastrectomy in elderly patients in order to im-
prove surgical decision making and patient counseling.
This study sought to compare population-based outcomes
following gastrectomy across a wide range of age groups
to determine whether advancing age, or other factors,
were independently associated with increased morbidity
and mortality.
2. Methods
Discharge data for 40,276,240 patients from the 2004-
2008 Nationwide Inpatient Sample Database (NIS), a
part of the Healthcare Cost and Utilization Project
(HCUP) of the Agency for Healthcare Research and
Quality (AHRQ) was analyzed. The NIS involved no
identifying patient information and was thus exempt
from Institutional Review Board approval at our institu-
tion. Information on patients who had undergone total
gastrectomy (ICD 9 code: 439.1) and partial gastrectomy
(ICD 9 codes: 438.1 and 438.9) as the primary procedure
was abstracted. Eight age groups (in increments of 10
years starting at 21 years) were compared for two proce-
dures: total gastrectomy (TG) and partial gastrectomy
(PG). The number of patients, gender ratio, most com-
mon primary diagnosis, non-elective nature of admission,
co-morbidities, pre-operative length of stay (LOS), total
length of stay (LOS), and morbidity and mortality rates
were abstracted separately for both TG and PG. Informa-
tion on complication rates was extracted using ICD-9
codes. The type of complications and their broad group-
ing is detailed in Table 1 . In addition, the total number of
procedures performed (both TG and PG) over the five
year period, as well as the percentage change in the num-
ber of procedures for each type of gastrectomy (Figures
1(a) and (b)), were calculated by dividing the number of
gastrectomies performed in 2008 by the number gastric-
tomies performed in 2004.
Statistical analysis was performed using th e Chi squ are
test for categorical data and the Student’s t test for con-
tinuous data. A value of p < 0.05 was considered statisti-
cally significant. Univariate analysis and multivariate
logistic regression analysis were performed to assess
factors affecting mortality following gastrectomy. Data
analysis was performed using SPSS version 19.0 (SPSS,
Inc). As per NIS database reporting guidelines, values
<10 are not shown in the results section and in the tables.
3. Results
3.1. Demographic Information
13,799 patients identified in the NIS underwent either
total gastrectomy (TG) or partial gastrectomy (PG) be-
tween 2004 and 2008, and formed the study population.
Clinicopathological characteristics are detailed in Table
2. 23.7% (N = 3271) patients underwent a TG, while
76.3% (N = 10,528) underwen t a PG. Th e largest nu mber
of TG were performed in patients age 61 - 70 years (N =
980, 30%), whereas the largest number of PG were per-
formed in those age 71 - 80 years (N = 2436, 23.1%).
The mean age for the TG and PG groups were 63.1 ±
12.8 and 64.1 ± 15 years, respectively. Twice the number
of males compared to females received a TG (p < 0.001 ),
whereas equal numbers received a PG (p < 0.001). The
greatest gender difference was observed among those 61 -
70 years old for the TG group (M:F, 2.9:1), and among
those 21 - 30 for the PG group (M:F, 1:1.6). Figures 1(a)
and (b) detail the change in number and percent of TG
and PG performed during the study period. TGs in-
creased over the 5-year study period, with the highest
percentage change among those 41 - 50 years (1500%).
The number of PGs performed decreased overall, espe-
cially in patients aged <60 years, but it increased among
patients >80 years with the highest percentage change
noted in patients >90 years (13%).
3.2. Pathology and Co-Morbidities
Gastric carcinoma was the most common indication for

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
82
Table 1. Complications studied by ICD code and their broad grouping for analysis (national inpatient sample database
2004-2008).
Broad group ICD 9 code Complication Components
Medical
complications
Respiratory
complications
512.1
415.11
518.5
518.4
997.31
997.39
Iatrogenic pneumothorax
Iatrogenic pulmonary embolic inf a rct
Post traumatic pulmonary insufficiency
Acute lung edema NOS
Ventilator associated pneumonia
Respiratory complications NEC
Cardiac
complications 997.1 Post-operative complications of Heart
Cardiac arrest during or resulting fr om a procedure
Cardiac insufficiency during or resulting from a procedure
Cardiorespiratory failure during or resulting from a procedure
Heart failure during or resulting from a procedure
Post-operative stroke 997.02 Post-operative stroke (Begins 1995)
Urinary tract
complications 997.5 Post-operative complications of urinary tract
Complications of:
External stoma of urinary tract
Internal anastomosis and bypass of urinary tract, including tha
t
involving intestinal tract
Oliguria or anu ria specified as due to procedure
Renal failure (acute) specified as due to procedure
Insufficiency (acute) specified as due to procedure
Tubular necrosis (acute) specified as due to procedure
Sepsis 995.91
Intra-abdominal post-operative
Stitch post-operative
Subphrenic post-operative
Wound post-ope rative
Surgical
complications
Gastrointestinal
complications 997.4 Post-operative complications of GI tract
Complications of: intestinal (internal) anastomosis and bypass,
not elsewhere classified, except that involving urinary tract
Hepatic failure specified as due to a procedure
Hepatorenal syndrome specified as due to a procedure
Intestinal obstruction NOS specified as due to a procedure
Fistula of bile duct 576.4 Fistula of bile duct
Post-operative
infection 998.59 Other post-ope rative infection Abscess: post-operative
Accidental operative
laceration 998.2 Accidental operative laceration Accidental perforation by catheter or other instrument during a
procedure on:
Blood vessel, Nerve, Organ
Post-operative
hemorrhage 998.11 Post-operative hemorrhage
Disruption of
internal operative
wound 998.31 Disruption of internal operative wound
Disruption of
external operative
wound 998.32 Disruption of external operative woun d
Intestinal fistula 569.81 Intestinal fistula
Deep vein
thrombosis
453.9
453.40
453.41
453.42
Venous thrombosis NOS
DVT/Embolism lower ext NOS
DVT/Embolism proximal lower ext
DVT/Embolism distal lower ext

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database 83
0%200%400%600%800%1000% 1200% 1400% 1600%
21-30 years
31-40 years
41-50 years
51-60 years
61-70 years
71-80 years
81-90 years
>90 years
(a)
-50% -40%-30% -20%-10%0%10%20%
21- 30 y ears
31- 40 y ears
41- 50 y ears
51- 60 y ears
61- 70 y ears
71- 80 y ears
81- 90 y ears
>90 year s
(b)
Figure 1. (a) Percentage change in the number of total gastrectomy performed among all age groups between 2004-2008 (na-
tional inpatient sample database); (b) percentage change in the number of partial gastrectomy among all age groups between
2004-2008 (national inpatient sample database).
TG overall (87.5%) and in each age group. Benign dis-
ease was the most common indication for PG overall
(57%) and for all patients under 70 years. Gastric carci-
noma wa s t he mos t commo n indication for PG in patients
>70 years (Table 3(a)) . An increa sing number o f PGs wer e
performed for both malignant and benign disease with ad-
vancing age, while an increasing percentage of TGs com-
pared to PGs were performed in younger patients (pri-
marily in those with malignant disease). ( T abl e 3(a) ) Over
the course of the study, there was an increasing trend in
the number of patients operated upon for gastric cancer,
with a corresponding decrease in the number of patients
operated upon for benign disease in all but the seventh
and tenth decades (Table 3(b)). Gastroesophageal
junction (GEJ) or abdominal esophagus carcinoma was
the most indication for total gastrectomy in all age

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
84
Table 2. Clinicopathological data on 13,799 patients undergoing total and partial gastrectomy (national inpatient sample da-
tabase 2004-2008).
Patients’ Age Groups by Decade
Overall 21 - 30 31 - 40 41 - 50 51 - 60 61 - 70 71 - 80 81 - 90 >91 p
value*
Total Gastrectomy
N, (%) 3271 32 (1.0) 110 (3.4)356 (10.9)733 (22.4)980 (30.0)802 (24.5) 250 (7.6) (-)
Male: Female r a t i o 2.1: 1 0.9: 1 1.2: 1 1.7: 1 2.2: 1 2.9: 1 2.1: 1 1.7: 1 1:01 0.001
Non-elective admission, N (%) 695 (21.2) 11 (34.4) 25 (22.7)66 (18.5)152 (20.7)174 (17.8)168 (20.9) 94 (37.6) (-) 0.001
Pre-operativ e LOS, mean ± SD 1.2 ± 3.8 1.8 ± 4.6 0.7 ± 2.61.0 ± 3.11.0 ± 2.81.0 ± 3.51.5 ± 5.0 2.3 ± 4.1 6.4 ± 5.60.001
Total LOS, mean ± SD 16.5 ± 14.7 16.2 ± 16.7 13.5 ± 9.314. 8 ± 1 2 .715.5 ± 13.216.7 ± 15.417.7 ± 15.4 18.7 ± 17.0 17.4 ± 8.50.001
Mortality, N (%) 250 (7.6) (-) (-) 11 (3.1)39 (5.3)60 (6.1)88 (11.0) 45 (18.0) (-) 0.003
Partial Gastrectomy
N, (%) 10,528 153 (1.5) 531 (5.0)1336 (12.7)2158 (20.5)2349 (22. 3)2436 (23.1) 1429 (13.6) 136 (1.3)
Male: Female r a t i o 1:1.1 1:1.6 1:1.5 1:1.3 1:1.1 1:0.9 1:1.0 1:1.3 1:1.4 0.001
Non-elective admission, N (%) 4884 (46.5) 87 (56.9) 246 (46.4)567 (42.5)932 (43.4)1001 (42.7)1142 (46.9) 820 (57.5) 89 (65.4)0.001
Pre-operativ e LOS, mean ± SD 2.1 ± 4.4 2.6 ± 5.1 2.3 ± 8.51.7 ± 3.62.0 ± 4.21.9 ± 3.92.2 ± 4.0 2.9 ± 4.4 3.3 ± 4.30.001
Total LOS, mean ± SD 13.9 ± 13.4 13.7 ± 18.3 11.7 ± 17.812.2 ± 11.912.9 ± 13.313.9 ± 13.214.9 ± 13.6 16.1 ± 12.2 15.8 ± 9.80.001
Mortality, N (%) 670 (6.4) (-) (-) 31 (2.3)61 (2.8)135 (5.7)199 (8.2) 206 (14.4) 27 (19.9) 0.001
Abbreviations: N: number of patients; LOS: length of stay; SD: standard deviation; (-): As per NIS database reporting guidelines, values <10 are not shown in
the table; *p value < 0.05 statistical significance for the range.
Table 3. (a) Distribution of malignant and benign cases in 13,799 patients undergoing total and partial gastrectomy separated
according to patient age groups by decades (national inpatient sample database 2004-2008); (b) Distribution of malignant and
benign cases in 13,799 patients undergoing total and partial gastrectomy separated according to patient age groups by dec-
ades and year of study period (national inpatient sample database 2004-2008).
(a)
Main Group
Age groups Malignant disease, N (%) Benign disease, N (%) p value*
Total gastrectomy Partial gastrectomy Total gastrectomy Partial gastrectomy
21 - 30 19 (50.0) 19 (50.0) 13 (8.8) 135 (91.2) 0.001
31 - 40 79 (40.7) 115 (59.3) 31 (6.9) 420 (93.1) 0.001
41 - 50 280 (43.1) 370 (56.9) 76 (7.3) 970 (92.7) 0.001
51 - 60 634 (47.3) 706 (52.7) 99 (6.3) 1462 (93.7) 0.001
61 - 70 884 (45.1) 1074 (54.9) 96 (7.0) 1284 (93.0) 0.001
71 - 80 736 (35.0) 1367 (65.0) 66 (5.8) 1075 (94.2) 0.001
81 - 90 222 (20.7) 849 (79.3) 28 (4.6) 583 (95.4) 0.001
>90 (-) 74 (90.2) 0 (0) 63 (100.0) 0.001
Abbreviations: N: number of patients; (-): As per NIS database reporting guidelines, values <10 are not shown in the table; *p value < 0.05 statistical signifi-
cance for the range.
(b)
Number of Malignant and Benign cases d uring study period (20 04 -2008)
Age
Groups Malignant disease, N (%) Benign disease, N (%)
2004 2005 2006 2007 2008 2004 2005 2006 2007 2008
21 - 30 (-) 12 (31.6) (-) 7 (18.4) 8 (21.1) 43 (29.1) 28 (18.9) 19 (12.8) 32 (21.6) 26 (17.6)
31 - 40 26 (13.4) 40 (20.6) 41 (21.1) 47 (24.2) 40 (20.6) 130 (28.8)86 (19.1) 74 (16.4) 68 (15.1) 93 (20.6)
41 - 50 76 (11.7) 143 (22.0) 101 (15.5) 162 (24.9)168 (25.8)242 (23.1)215 (20.6)195 (18.6) 184 (17.6)210 (20.1)
51 - 60 154 (11.5) 283 (21.1) 268 (20.0) 331 (24.7)304 (22.7)344 (22.0)295 (18.9)320 (20.5) 268 ( 17.2)334 (21.4)
61 - 70 237 (12.1) 397 (20.3) 433 (22.1) 432 (22.1)459 (23.4)267 (19.3)286 (20.7)269 (19.5) 256 ( 18.6)302 (21.9)
71 - 80 279 (13.3) 474 (22.5) 422 (20.1) 468 (22.3)460 (21.9)258 (22.6)240 (21.0)208 (18.2) 231 ( 20.2)204 (17.9)
81 - 90 158 (14.8) 230 (21.5) 210 (19.6) 219 (20.4)254 (23.7)129 (21.1)140 (22.9)129 (21.1) 103 ( 16.9)110 (18.0)
>90 17 (20.7) 12 (14.6) 18 (22.0) 13 (15.9) 22 (26.8) 13 (20.6) (-) 14 (22.2) 11 (17.5) 15 (23.8)
Abbreviations: N: number of patients ; (-): As per NIS database r eporting guidelines, values <10 are not shown in the table.

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database 85
groups (59.2%), whereas partial gastrectomy for malign-
nancy was most commonly performed for antral or pylo-
ric tumors (47.2%) in all age groups (Table 4). Overall,
15.1% of gastrectomies were performed for GEJ/ab-
dominal esophagus tumors, while 17.5% were performed
for antral tumors. Elective admissions were far more
common in the TG group (N = 2576, 78.8%) compared
to the PG group (N = 5648, 53.6%) with more non-elec-
tive admissions occurring in the both the youngest and
oldest age groups (Table 2). The three most common
co-morbidities overall were hypertension (TG; 43.7 % vs.
PG; 45.5%), fluid and electrolyte imbalances (TG; 27%)
vs. PG; 28.1%), and chronic pulmonary disease (TG;
9.9% vs. PG; 18.2%). (Tables 5(a) and (b)) The per-
centage of patients with hypertension and fluid and elec-
trolyte imbalances was highest in those >80 years for
both the TG and PG groups. Chronic pulmonary disease
occurred most commonly in those aged 71 - 80 years for
both the TG and PG groups, but occurred with high fre-
quency in all patients >70 years (~19% - 26%).
3.3. Complications
Complications in the TG and PG groups are detailed in
Tables 6(a) and (b). The overall complication rate was
54% in the TG group and 36% in the PG group. The
most common complications in the TG group were res-
piratory (18%) and gastrointestinal (GI) (11.4%), with
the highest percent of complications occurring in those
>51 years. An increasing number of respiratory compli-
cations were seen with advancing age, with those age 21 -
30 years having those lowest incidence (9.4%) and those
>91 having the highest (25%). The age-related trend in
GI complications (intestinal obstruction and anastomotic
and hepatic complications) noted. Among PG patients,
GI complications (9.6%) and bile duct fistulas (7.2%)
were most common peaking incidence among those 61 -
70 years. Similar to the TG group, no appre- ciable
age-related trend in GI complications was noted in the
PG group. There was an increase age-related incidence
in bile duct fistulas in the PG group ranging from 2% in
those 21 - 30 years to 13.2% in those >91 years. Cardiac
events occurred more commonly in the TG groups with
the highest incidence in those 61 - 70 (7.7%) and 71 - 80
years old (8.1%) (p < 0.001). The overall cardiac compli-
cation rate in the PG group (3.2%) was half of that ob-
served in the TG group (6.1% ) (p < 0.00 1).
3.4. Clinicopathological Data and Mortality
Clinical outcomes from the two surgery groups are de-
tailed in Table 2. 59.5% of all admissions for gastric
surgery were elective, including 78.8% of TGs and
53.6% of PGs (p = 0.001). Elective admission status did
not different for the PG groups, but was highest in those
>81 years old. Elective admission status was far more
common in those <80 years old undergoing TG com-
pared to those >81 years old (p = 0.001). The mean
pre-operative and total hospital LOS generally increased
with advancing age for both the TG and PG with the ex-
ception of the youngest age group (age 21 - 30 years)
whose LOS equaled those >61 years (p < 0.001). The
overall mortality rate for all gastrectomy patients in the
study groups was 6.7% (Table 7). The mortality rate for
the TG group was 7.6%, while the mortality rate for the
PG group was 6.4% (p = 0.01) (Table 2). Mortality rates
increased in a linear fashion with advancing age also for
both TG and PG beginning with those >30 years of age
(Table 2). However, it is notable that the youngest
group had a mortality rate higher than (TG group) or
similar to (PG group) those aged 51 - 60 years (p < 0.001)
(Table 7). The highest mortality rate of 20% was
Table 4. Anatomic site of malignancy in 13,799 patients undergoing total and partial gastrectomyseparated according to pa-
tient age groups by decades (national inpatient sample database 2004-2008).
Total Gastrectomy, N (%) Partial Gastrectomy, N (%)
Site of
Malignancy Overall 21 - 30 31 - 40 41 - 50 51 - 60 61 - 70 71 - 8081 - 90>91Overall21 - 3031 - 4041 - 5051 - 60 61 - 70 71 - 80 81 - 90>91
Lower 1/3rd
and
Abdominal
Esophagus
422
(15.6) 0 11 (13.6) 41 (16) 100 (16.8) 150 (18.2) 94 (13.7) 26 (11.7) 059 (1.3) 0 (-) (-)18 (2.5) 22 (2.1) (-) (-) 0
GE
Junction 1176 (43.6) (-) 30 (37) 106 (41.4) 279 (46.9) 388 (47.1) 296 (43.1) 70 (31.5) (-) 428 (9.4) (-) (-) 41 (11.2) 69 (9.8) 99 (9.2) 143 (10.4) 63 (7.4) (-)
Gastric
Body 167 (6.2) (-) (-) 18 (7.0) 21 (3.5) 40 (4.9) 59 (8.6) 20 (9) (-) 483 (10.6) (-) 15 (13.3) 38 (10.4) 76 (10.7) 101 (9.4) 168 (12.2) 73 (8.6) (-)
Antrum/
Pyloric/
Pre-pyloric
Tumor
261
(9.7) (-) (-) 28 (10.9) 64 (10.8) 75 (9.1) 53 (7.7) 27 (12.2) (-) 2159 (47.2) (-) 53 (46.9) 158 (43.3) 299 (42.3) 526 (49.1) 662 (48.1) 414 (48.7) 41 (55.4)
Abbreviations: N: number of patients; GE: gastroesophageal; NEC: not elsewhere classified; NOS: not otherwise specified; (-): As per NIS database reporting
guidelines, values <10 are not shown in the table.

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
86
Table 5. (a) Distribution of co-morbidities in 3271 patients undergoing total gastrectomy separated according to patient age
groups by decades (national inpatient sample database 2004-2008); (b) distribution of co-morbidities in 10,528 patients un-
dergoing partial gastrectomy separated according to patient age groups by decades (national inpatient sample database
2004-2008).
(a)
Patients’ Age Groups by Decades, N (%)
Co-morbidities Overall
21 - 30 31 - 4041 - 5051 - 60 61 - 70 71 - 80 81 - 90 >91 p value*
Hypertension 1429 (43.7) (-) (-)
72
(20.2) 291 (39.7) 473 (48.3) 438 (54.6) 138 (55.2) (-) 0.001
Fluid and electrolyte
imbalances 882 (27.0) (-) 27 (24.5) 74 (20.8) 173 (23.6) 284 (29.0) 224 (27.9) 87 (34.8) (-) 0.001
Chronic pulmonary
disease 651 (19.9) (-) (-) 52 (14.6) 114 (15.6) 211 (21.5) 212 (26.4) 48 (19.2) (-) 0.001
Weight loss 480 (14.7) (-) 15
(13.6) 48 (13.5) 92 (12.6) 134 (13.7) 132 (16.5) 53 (21.2) (-) 0.043
Obesity 186 (5.7) 0 (-) 26 (7.3)65 (8.9)58 (5.9)28 (3.5)5 (2.0) 0 0.001
Coagulopathy 143 (4.4) (-) (-) 10 (2.8)25 (3.4)41 (4.2)44 (5.5)19 (7.6) (-) 0.017
Chronic blood loss
anemia 143 (4.4) (-) (-) (-) 30 (4.1)34 (3.5)52 (6.5)18 (7.2) 0 0.001
Renal failure 118 (3.6) (-) (-) (-) 19 (2.6)33 (3.4)41 (5.1)18 (7.2) 0 0.001
Alcohol abuse 104 (3.2) (-) (-) 16 (4.5)37 (5.0)31 (3.2)17 (2.1)0 0 0.002
Liver disease 75 (2.3) 0 (-) 11 (3.1)17 (2.3)18 (1.8)22 (2.7)(-) 0 0.76
Diabetes 47 (1.4) (-) (-) (-) 15 (2.0)13 (1.3)12 (1.5)(-) (-) 0.83
Abbreviations: N: number of patients; (-): As per NIS database reporting guidelines, values <10 are not shown in the table; *p value < 0.05 statistical signifi-
cance for the range.
(b)
Patients’ Age Groups by Decades, N (%)
Co-morbidities Overall
21 - 30 31 - 4041 - 50 51 - 6061 - 70 71 - 80 81 - 90 >91 p value*
Hypertension 4875 (45.5)
24
(15.6) 103 (19.1) 344 (25.2) 888 (40.5) 1224 (51.0) 1415 (57.2) 792 (54.9) 85 (62.0) 0.001
Fluid and electrolyte
imbalances 3011 (28.1) 31 (20.1) 106 (19.7) 302 (22.1) 536 (24.4) 642 (26.8) 789 (31.9) 539 (37.4) 66 (48.2) 0.001
Chronic pulmonary
disease 1949 (18.2) 12 (7.8) 55 (10.2) 184 (13.5) 364 (16.6) 487 (20.3) 549 (22.2) 270 (18.7) 28 (20.4) 0.001
Weight loss 1430 (13.4 ) 19
(12.3) 67 (12.4) 163 (11.9) 243 (11.1) 307 (12.8) 352 (14.2) 257 (17.8) 22 (16.1) 0.001
Chronic blood loss
anemia 1016 (9.5) 12 (7.8) 47 (8.7) 99 (7.3)175 (8.0) 223 (9.3)259 (10.5) 186 (12.9) 15 (10.9) 0.001
Obesity 548 (5.1) (-)
43
(8.0) 102 (7.5) 159 (7.2) 131 (5.5)71 (2.9)30 (2.1) (-) 0.001
Renal failure 543 (5.1) (-) 12
(2.2) 37 (2.7)63 (2.9)107 (4.5)173 (7.0) 130 (9.0) 16 (11.7) 0.001
Coagulopathy 471 (4.4) (-)
13
(2.4) 41 (3.0)79 (3.6)95 (4.0)139 (5.6) 89 (6.2) (-) 0.001
Liver disease 396 (3.7) (-) 32
(5.9) 65 (4.8)115 (5.2) 87 (3.6)69 (2.8)19 (1.3) (-) 0.001
Alcohol abuse 379 (3.5) (-) 24
(4.5) 93 (6.8)102 (4.6) 89 (3.7)57 (2.3)(-) (-) 0.001
Diabetes 197 (1.8) (-) (-) (-) 38 (1.7)59 (2.5)59 (2.4)26 (1.8) (-) 0.001
Abbreviations: N: number of patients; (-): As per NIS database reporting guidelines, values <10 are not shown in the table; *p value < 0.05 statistical signifi-
cance for the range.

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database 87
Table 6. (a) Distribution of post-operative complications in 3271 patients undergoing total gastrectomy separated according
to patient age groups by decades (national inpatient sample database 2004-2008); (b) Distribution of post-operative complica-
tions in 10,528 patients undergoing partial gastrectomy separated according to patient age groups by decades (national inpa-
tient sample database 2004-2008).
(a)
Patients’ Age Groups by Decade, N (%)
Complications Overall 21 - 3031 - 4041 - 5051 - 60 61 - 70 71 - 80 81 - 90>91
Medical Complications
Respiratory complications 590 (18.0) (-) 12 (10.9)44 (12.4)117 (16. 0 )207 (21.1) 158 (19.7) 47 (18.8)(-)
Cardiac complications 201 ( 6.1 ) (-) (-) 15 (4.2)29 (4.0)75 (7.7) 65 (8.1) 12 (4.8)0
Urinary complications 47 (1.4) 0 1 (0.9) (-) (-) 13 (1.3) 16 (2.0) (-) (-)
Post-operative stroke 5 (0.2) 0 0 (-) 0 (-) (-) 0 0
Sepsis 3 (0.09) 0 0 0 0 (-) 0 0 0
Surgical Complications
Gastrointestinal complications 374 (11.4) (-) 12 (10.9)42 (11.8)85 (11.6)120 (12.2) 85 (10.6) 28 (11.2)(-)
Other post-operative infection 208 (6.4) (-) (-) 25 (7.0)45 (6.1)74 (7.6) 44 (5.5) 13 (5.2)0
Accidental laceration during a procedure 112 (3.4) 0 (-) 14 (3.9)29 (4.0)36 (3.7) 19 (2.4) 9 (3.6) (-)
Post-operative Hemorrhage 83 (2.5) (-) (-) 12 (3.4)17 (2.3)19 (1.9) 23 (2.9) 9 (3.6) 0
Venous thrombosis 53 (1.6) 0 (-) (-) (-) 20 (2.0) 15 (1.9) (-) 0
Disruption of external operation wound 39 (1.2) 0 (-) (-) (-) 20 (2.0) - 0 0
Disruption of internal operation wound 35 (1.1) 0 0 (-) (-) (-) 12 (1.5) (-) 0
Intestinal fistula 3 (0.1) 0 0 0 (-) 0 - 0 0
Abbreviations: N: number of patients ; (-): As per NIS database r eporting guidelines, values <10 are not shown in the table.
(b)
Patients’ Age Groups by Decades, N (%)
Complications Overall 21 - 3031 - 4041 - 5051 - 6061 - 70 71 - 80 81 - 90>91
Medical Complications
Cardiac Complications 333 (3.2) (-) (-) 15 (1.1)40 (1.9)61(2.6) 105 (4.3) 91(6.4)12 (8.8)
Respiratory Complications 160 (1.5) (-) ( -) 21 (1.6)27 (1.3)46 (2) 31 (1.3) 24 (1.7)(-)
Urinary Tract Complications 97 (0.9) 0 (-) (-) 18 (0.8)23 (1.0) 27 (1.1) 20 (1.4)(-)
Post-operative Stroke 22 (0.21) 0 0 (-) (-) (-) (-) (-) 0
Sepsis 18 (0.17) 0 0 0 (-) (-) (-) (-) 0
Surgical Complications
Gastrointestinal Complications 1013 (9.6) 10 (6.5)44 (8.3)128 (9.6)199 (9.2)230 (9.8) 234 (9.6) 15 8 (11.1)10 (7.4)
Fistula of bile duct 763 (7.2) (-) 19 (3.6)61 (4.6)99 (4.6)190 (8.1) 209 (8.6) 164 (11.5)18 (13.2)
Post-operative infecti o n s 568 (5.4) 10 (6.5)21 (4.0)71 (5.3)123 (5.7)137 (5.8) 119 (4.9) 76 (5.3)11 (8.1)
Accidental operative laceration 266 (2.5) 0 15 (2.8)31 (2.3)50 (2.3)63 (2.7) 67 (2.8) 34 (2.4)(-)
Post-operative hemorrhage 220 (2.1) (-) 12 (2.3)26 (1.9)35 (1.6)61 (2.6) 47 (1.9) 34 (2.4)(-)
Disruption of external operation wound 103 (1.0) (-) (-) 11 (0.8)23 (1.1)24 (1.0) 22 (0.9) 1 5 ( 1.0)(-)
Disruption of internal operation wound 69 (0.7) (-) (-) 11 (0.8)18 (0.8)18 (0.8) 12 (0.5) (-) (-)
Intestinal fistula 29 (0.28) 0 (-) (-) (-) (-) (-) (-) 0
Deep vein thrombosis 98 (0.9) 0 (-) (-) 11 (0.5)20 (0.9) 3 5 (1.4) 23 (1.6)(-)
Abbreviations: N: number of patients ; (-): As per NIS database r eporting guidelines, values <10 are not shown in the table.

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
88
Table 7. Univariate analysis of in-hospital mortality for
13,799 patients undergoing total and partial gastrectomy
(national inpatient sample database 2004-2008).
Variable Mortality
rate, % p value*
Overall 6.7
Age groups 0.001
21 - 30 years 3.8
31 - 40 years 1.4
41 - 50 years 2.5
51 - 60 years 3.4
61 - 70 years 5.9
71 - 80 years 8.9
81 - 90 years 15.0
>90 years 20.0
Gender 0.002
Male 7.3
Female 6.0
Race 0.015
Native American 10.2
White 7.3
Hispanic 6.3
Black 5.7
Asian or Pacific Islander 5.1
Others 3.7
Type of gastrectomy 0.011
Total gastrectomy 7.6
Partial gastrectomy 6.4
Indications for gastrectomy 0.001
Benign disease 8.0
Malignant disease 5.6
Type of admi s s i o n 0.001
Non-elective admission 11.4
Elective admission 3.5
Co-morbidities
Coagulopathy 25.2 0.001
Renal failure 19.3 0.001
Congestive heart failure 19.1 0.001
Fluid and electroly te disorders 13.0 0.001
Liver disease 12.4 0.001
Peripheral vascular disorders 12.0 0.001
Weight loss 11.6 0.001
Diabetes with chronic complications 9.9 0.042
Alcohol abuse 9.9 0.005
Chronic blood loss anemia 9.1 0.001
Valvular disease 8.5 0.059
Chronic pulmonary disease 8.4 0. 0 0 1
Diabetes, uncomp licated 5.9 0.129
Hypertension (combine uncomplicated
and complicated) 5.5 0.001
*p value < 0.05 statistical significance for the range.
seen in those >91 years old (p < 0.001).
3.5. Univariate and Multivariate Analysis
Univariate analysis identified several patient factors
which impacted mortality rates. (Table 7) Specifically,
mortality rates were significantly higher among patients
>90 years (20%) (p = 0.001), male patients (7.3%) (p =
0.002), Native Americans (10.2%) (p = 0.015), those
having a gastrectomy for benign disease (8%) (p = 0.001),
and patients with non-elective admission status prior to
gastrectomy (11.4%) (p = 0.001). In addition, the pre-
sence of several co-morbidities also increased mortality
rates including coagulopathy (25.2%), renal failure
(19.3%), congestive heart failure (19.1%) and fluid and
electrolyte imbalances (13%). On multivariate analysis
(Table 8), age did not independently impact mortality
and only non-elective admission (p = 0.001), hyperten-
sion (p = 0.001), valvular heart disease (p = 0.001), ane-
mia (p = 0.001), and g astric cancer (p = 0.001) incr eased
in-hospital mortality in p a tients undergoing gastrectomy.
4. Discussion
Today approximately 21,000 Americans are diagnosed
annually with gastric cancer and half will die due to the
lethality of the disease. It has been predicted that the
proportion of the US population over 80 years old is ex-
pected to double in the next two decades going from 5%
(1990) to 10% (2030) [6]. In addition, Americans 65 to
79 years who con stituted 9.3% of the population in 2000,
are projected to increase to 12.5% by 20 50 and 12.9% by
2070 [7]. Given current life table analysis, which predicts
that patients who survive to the age of 80 - 85 years old
will live another 8 years on average, the number of eld-
erly patients (>80 years)—and as a result the number of
patients overall who will require gastric surgery—is in-
creasing and will increase exponentially in the future
[8,9]. Despite these facts, little is known about the clini-
cal outcomes of gastrectomy for either benign or malig-
nant conditions in elderly patients, as they are largely
excluded from clinical trials given their presumed “high
risk” status. More precisely, increased severity and fre-
quency of co-morbidities, malnutrition and diminished
organ reserve in the elderly and their demonstrable im-
port on post-operative morbidity and mortality, has led to
significant surgical bias towards performing less radical
procedures or failure to consider the elderly for gastric
surgery at all. Recently identified epidemiological shifts,
from distal intestinal type to proximal diffuse type ade-
nocarcinoma of the gastric cardia, most notably in the
elderly patients and Western countries, may further com-
plicate this picture [10,11]. At present, the best available
data on the outcomes of gastric surgery in the elderly are

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database 89
Table 8. Multi-variate analysis of in-hospital mortality for 13,799 patients undergoing total and partial gastrectomy (national
inpatient sample database 2004-2008).
Overall Total gastrectomy Partial gastrectomy
Variable OR (95% C.I.) p value OR (95% C.I.)p value OR (95% C.I.) p value*
Hypertension 1.7 (1.4 - 2.0) 0.001 1.7 (1.2 - 2.4) 0.001 1.7 (1.4 - 2.1) 0.001
Valvular heart disease 1.6 (1.1 - 2.3) 0.001 1.5 (1.0 - 2.3) 0.00 1
Anemia 1.5 (1.2 - 1.8) 0.001 1.5 (1.2 - 1.9) 0.001
Non-elective admission 2.4 (2.0 - 2.9) 0.001 2.7 (1.9 - 3.7) 0.001 2.4 (1.9 - 3.1) 0.001
Gastric cancer 1.6 (1.3 - 2.0) 0. 00 1 2.0 (1.3 - 3.2) 0.001 1.6 (1.2 - 2.0) 0.001
Abbreviations: OR: odds ratio; C.I.: confidence interval; *p value < 0.05 statistical signif icance.
conflicting, lack inh eren t unifo rmity, and suffer pr imarily
from the problem of small sample sizes and single insti-
tutional an alysis [2,10,12-18].
Pisanu et al. evaluated 135 patients over 75 years old
treated over an 11-year period, and determined that ad-
vanced age was not a contraindication to gastrectomy,
nor was it predictive of increased morbidity and mortal-
ity (REF). These authors noted that the increased rate of
co-morbidities and malnutrition were the primary factor
contributing to increased mortality in the elderly rather
than age alone. (REF) They also concluded that PG as
opposed to TG was associated with improved quality of
life (QOL) for this group, without sacrificing cure. Simi-
lar to these results Katai et al. noted significantly in-
creased 30-day and 90-day mortality rates of 9.4% and
18.8% in patients >80 years undergoing TG compared to
those 50 - 69 years (1.4% and 4.4%, p < 0.001) [11].
Among patients >80 years old undergoing PG, mortality
rates were 0% and 3.7% compared to 0.5% and 1.7% in
patients 50 - 69 year s of ag e (p = N.S). [14] Most notably,
PG was also associated with an improved long-term sur-
vival compared to TG [19,20]. Sánchez-Bueno et al. also
reported improved five-year survival among patients un-
dergoing PG (48.1%) compared to those undergoing TG
(18%) (p = 0.001) [20]. However despite these encourage
results, there remains no clear consensus on whether age
alone independently affects morbidity and mortality after
gastric surgery (and for that matter many surgeries). This
is primarily due to the fact that most studies, in which
increased mortality rates among the elderly have been
reported, have failed to perform multivariate analysis
allowing independent assessment of the contribution of
advanced age [14]. In one of the few studies to perform
multivariate analysis, Pisanu et al. reported a signifi-
cantly higher mortality rate 17.2% following all types of
gastrectomy in patients >75 years (17.2%) compared to
patients <75 years (5.6%). However, they noted that sev-
eral co-morbidities rather than age were the strongest
predictors of increased post-operative mortality. Similar
to the report by Pisanu et al., the current study also re-
ports higher mortality rates with advancing age, but age
by itself was not an independent predictor of mortality
alone. Rather, our data supports the notion that being
“old” is not nearly as important as being “sick and old”
when it comes to mortality following gastric surgery.
Much like a machine that continues to function well de-
spite age, if giv en the p roper mainten ance (on or off war-
ranty), the old patient with neglected health or chronic
conditions is at greatest risk of failure (or death) when
stressed. While the current study is certainly limited to
drawing conclusions b ased on the variables assessed, it is
noteworthy that the presence of gastric cancer, and pre-
existing hypertension, valvular disease, and anemia were
all strongly associated with increased mortality. Of note,
hypertension was the most prevalent co-morbid condition
in those >50 years; while fluid and electrolyte i mbalance
was the most common co-morbidity in younger groups.
These results are similar to those of Pisanu et al., who
identified hypertension and stroke as the two most im-
portant factors affecting mortality following gastrectomy,
as well as a report by Roviello et al. who noted that the
two most common pre-existing conditions among pa-
tients undergoing surgery for gastric cancer were car-
diovascular (57.7%) and anemia/hypoproteinemia (41%)
[21,22]. The congruence of these findings across all three
studies may be interpreted as implying that, whereas eld-
erly patients are far more likely to have cardiovascular
disease and suboptimal nutrition, particular attention to
these co-morbidities, and optimization or remediation of
these when possible, may result in decreased in-hospital
mortality.
Several authors have reported that emergency surgery
or non-elective gastric surgery (in addition to many other
types of surgery) is associated with both higher morbid-
ity and mortality rates [21,23,24]. In agreement with this
literature, the current study also found that the elective or
non-elective admission for gastrectomy correlated well
with survival. Interestingly, the number of non-elective
admissions for both TG and PG peaked at both ends of
the age spectrum (youngest and oldest), with the in-

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
90
creased number of non-elective admissions in the PG
group compared to the TG group perhaps indicative of
surgical bias and teaching, which urges performing the
least radical procedure necessary in the sick patient.
Published reports of increased medical and surgical
complications post gastrectomy among the elderly are
numerous, and unifor mly comment on decreased compli-
cations when performing PG compared to TG—albeit
this is not just in the elderly population [22,25,26]. Gas-
trointestinal (GI) complications (11.3%) were the most
common surgical complication noted in the current study
(ICD 997.4). Medical complications were noted in 16%
of patients, and were far more of a concern in TG pa-
tients (23.6%) than PG patients (13.6%). In the TG group,
the most common complication overall was respiratory
issues (18%) which is similar to the findings of Bittner et
al. [10]. While Bitner et al. did report on variations in
complications between gastrectomy patients <70 and >70
years, the current report is the first study to analyze com-
plications by decade. It is quite clear from the current
results that the increased morbidity in elderly patients
comes primarily in the form of medical complications
(namely respiratory and cardiac events) and not surgical
complications. (Tables 6(a) and (b)) Furthermore, car-
diac events occurred at nearly twice the rate in the TG
(6.1%) compare that PG group (3.2%). Taken on the
who le, these results support the notion that when feasible,
PG is the preferred surgical therapy in elderly patients
(particularly in those with pre-existing co-morbidities
that may increase the likelihood of post-operative com-
plications).
The limitations of the study are that it is primarily ret-
rospective in nature, as well as those factors inherent in
working with large administrative databases. Complica-
tions that may have occurred post-discharge or on re-
admission within 30 days of procedure were not neces-
sarily captured by the data set. That said, the underesti-
mation of the actual complication rates should apply
across all patients equally and thus not alter the findings
of this report. In addition, the current study is limited by
errors in cod ing that may have occurred at the time of the
hospitalization. Under-reporting of co-morbidities in
elderly patients is another potential limitation due to not
including confounding factors like frailty, in which three
or more of these following criteria are present: uninten-
tional weight loss, exhaustion, weakness, slow walking
speed, and low physical activity [27]. Finally, this study
did not differentiate between stages of disease or extent
of lymphadenectomy performed which could affect out-
comes.
In summary, age alone is not an independent factor
predictive of increased mortality following gastrectomy,
while non-elective admissions, gastric cancer, hyperten-
sion, valvular disease and anemia are. That said, whereas
the elderly are more likely to be afflicted with these (and
other co-morbidities), as well as to suffer from increased
rates of post-operative morbidity, these results imply that
surgical risk stratification of patients considered from
gastrectomy, should focus primarily on the overall pre-
operative status of the patient’s health rather than their
chronologic age. Can gastrectomy be safely performed in
elderly patients with acceptable clinical outcomes? The
answer is yes. That said, all efforts should be made to
optimize the patients remediable pre-operative co-mor-
bidities prior to elective procedures, and provide appro-
priate counseling regarding the risks and expected post-
operative course. Which type of gastrectomy is safer or
preferred in the elderly patient remains a more difficult
question to answer. While all surgeons have as their
mantra “do no harm” and should select the least radical
operation with similar outcomes and lowest anticipated
morbidity/mortality, these goals must be undertaken aga-
inst a backdrop of following sound surgical oncologic
principles, by achieving a negative margin and adequate
information to permit appropriate staging.
REFERENCES
[1] R. F. Saidi, J. L. Bell and P. S. Dudrick, “Surgical Resec-
tion for Gastric Cancer in Elderly Patients: Is There a Dif-
ference in Outcome? ” Journal of Surgical Research, Vol.
118, No. 1, 2004, pp. 15-20.
doi:10.1016/S0022-4804(03)00353-6
[2] K. Hanazaki, M. Wakabayashi, H. Sodeyama, et al., “Sur-
gery for Gastric Cancer in Patients Older than 80 Years of
Age,” Hepatogastroenterology, Vol. 45, No. 19, 1998, pp.
268-275.
[3] W. C. Meyers, R. J. Damiano Jr., F. S. Rotolo and R. W.
Postlethwait, “Adenocarcinoma of the Stomach. Chang-
ing Patterns over the Last 4 Decades,” Annals of Surgery,
Vol. 205, No. 1, 1987, pp. 1-8.
doi:10.1097/00000658-198701000-00001
[4] D. M. Parkin, P. Pisani and J. Ferlay, “Estimates of the
Worldwide Incidence of Eighteen Major Ca ncers in 1985,”
International Journal of Cancer, Vol. 54, No. 4, 1993, pp.
594-606. doi:10.1002/ijc.2910540413
[5] B. Martella, C. Militello, S. Spirch, et al., “Palliative Sur-
gery for Gastric Cancer in Elderly Patients,” Acta Bio-
Medica, Vol. 76, Suppl. 1, 2005, pp. 49-51.
[6] B. Cartier, “Carotid Surgery in Octogenarians: Why Not?”
Annals of Vascular Surgery, Vol. 16, No. 6, 2002, pp.
751-755. doi:10.1007/s10016-001-0243-0
[7] United States Census Bureau, 2012.
http://www.census.gov/population/www/projections/natd
et-d1a.html
[8] I. Matsushita, H. Hanai, M. Kajimura, et al., “Should
Gastric Cancer Patients more than 80 Years of Age Un-
dergo Surgery? Comparison with Patients Not Treated

A Population-Based Outcomes Analysis of the Impact of Age on Morbidity and Mortality Following
Gastrectomy: An Analysis of 13,799 Patients from the Nationwide Inpatient Sample Database
91
Surgically Concerning Prognosis and Quality of Life,”
Journal of Clinical Gastroenterology, Vol. 35, No. 1,
2002, pp. 29-34.
doi:10.1097/00004836-200207000-00008
[9] R. J. McKenna Sr., “Clinical Aspects of Cancer in the Eld-
erly. Treatment Decisions, Treatment Choices, and Follow-
Up,” Cancer, Vol. 74, Su ppl. 7, 1994 , pp. 21 07-2117.
doi:10.1002/1097-0142(19941001)74:7+<2107::AID-CN
CR2820741719>3.0.CO;2-1
[10] R. Bittner, M. Butters, M. Ulrich, S. Uppenbrink and H.
G. Beger, “Total Gastrectomy. Updated Operative Mor-
tality and Long-Term Survival with Particular Reference
to Patients Older than 70 Years of Age,” Annals of Sur-
gery, Vol. 224, No. 1, 1996, pp. 37-42.
doi:10.1097/00000658-199607000-00006
[11] P. Lauren, “The Two Histological Main Types of Gastric
Carcinoma: Diffuse and So-Called Intestinal-Type Car-
cinoma. An Attempt at a Histo-Clinical Classification,”
Acta Pathologica et Microbiologica Scandinavica, Vol.
64, No. 1, 1965, pp. 31-49.
[12] T. Bandoh, T. Isoyama and H. Toyoshima, “Total Gas-
trectomy for Gastric Cancer in the Elderly,” Surgery, Vol.
109, No. 2, 1991, pp. 136-142.
[13] S. Ishigami, S. Natsugoe, T. Saihara, et al., “Clinical and
Pathologic Features of Early Gastric Cancer in Elderly
Patients,” Hepatogastroenterolo gy, Vol. 44, No. 16, 1997,
pp. 1164-1168.
[14] H. Katai, M. Sasako, T. Sano and K. Maruyama, “The
Outcome of Surgical Treatment for Gastric Carcinoma in
the Elderly,” Japanese Journal of Clinical Oncology, Vol.
28, No. 2, 1998, pp. 112-115. doi:10.1093/jjco/28.2.112
[15] K. Kitamura, T. Yamaguchi, H. Taniguchi et al., “Clini-
copathological Characteristics of Gastric Cancer in the
Elderly,” British Journal of Cancer, Vol. 73, No. 6, 1996,
pp. 798-802. doi:10.1038/bjc.1996.139
[16] H. Kubota, T. Kotoh, D. K. Dhar, et al., “Gastric Resec-
tion in the Aged (> or = 80 Years) with Gastric Carci-
noma: A Multivariate Analysis of Prognostic Factors,”
Australian and New Zealand Journal of Surgery, Vol. 70,
No. 4, 2000, pp. 254-257.
doi:10.1046/j.1440-1622.2000.01802.x
[17] Y. Maehara, T. Oshiro, H. Oiwa, et al., “Gastric Carci-
noma in Patients over 70 Years of Age,” British Journal
of Surgery, Vol. 82, No. 1, 1995, pp. 102-105.
doi:10.1002/bjs.1800820134
[18] Y. Maehara, Y. Emi, S. Tomisaki, et al., “Age-Related
Cha r a c teristics of Ga stric Carcinoma in Youn g and Elderly
Patients,” Cance r, Vol. 77, No. 9, 1996, pp. 1774-1780.
doi:10.1002/(SICI)1097-0142(19960501)77:9<1774::AID
-CNCR3>3.0.CO;2-C
[19] H. J. Meyer, J. Jahne, H. Wilke and R. Pichlmayr, “Sur-
gical Treatment of Gastric Cancer: Retrospective Survey
of 1704 Operated Cases with Special Reference to Total
Gastrectomy as the Operation of Choice,” Seminars in
Surgical Oncology, Vol. 7, No. 6, 1991, pp. 356-364.
doi:10.1002/ssu.2980070607
[20] F. Sanchez-Bueno, J. A. Garcia-Marcilla, D. Perez-Flores,
et al., “Prognostic Factors in a Series of 297 Patients with
Gastric Adenocarcinoma Undergoing Surgical Resec-
tion,” British Journal of Surgery, Vol. 85, No. 2, 1998, pp.
255-260. doi:10.1046/j.1365-2168.1998.00558.x
[21] F. Roviello, D. Marrelli, S. A. De, A. Messano, E. Pinto
and A. Carli, “Complications after Surgery for Gastric
Cancer in Patients Aged 80 Years and over,” Japanese
Journal of Clinical Oncology, Vol. 28, No. 2, 1998, pp.
116-122. doi:10.1093/jjco/28.2.116
[22] A. Pisanu, A. Montisci, S. Piu and A. Uccheddu, “Cura-
tive Surgery for Gastric Cancer in the Elderly: Treatment
Decisions, Surgical Morbidity, Mortality, Prognosis and
Quality of Life,” Tumori, Vol. 93, No. 5, 2007, pp. 478-
484.
[23] A. G. Greenburg, R. P. Saik, J. J. Coyle and G. W. Peskin,
“Mortality and Gastrointestinal Surgery in the Aged: Elec-
tive vs. Emergency Procedures,” Archives of Surgery, Vol.
116, No. 6, 1981, pp. 788-791.
doi:10.1001/archsurg.1981.01380180046009
[24] C. W. Wu, S. S. Lo, K. H. Shen, M. C. Hsieh, W. Y. Lui
and F. K. P’eng, “Surgical Mortality, Survival, and Qual-
ity of Life after Resection for Gastric Cancer in the Eld-
erly,” World Journal of Surgery, Vol. 24, No. 4, 2000, pp.
465-472. doi:10.1007/s002689910074
[25] F. Bozzetti, “Total versus Subtotal Gastrectomy in Cancer
of the Distal Stomach: Facts and Fantasy,” European
Journal of Surgical Oncology, Vol. 18, No. 6, 1992, pp.
572-579.
[26] J. Davies, D. Johnston, H. Sue-Ling, et al., “Total or Sub-
total Gastrectomy for Gastric Carcinoma? A Study of
Quality of Life,” World Journal of Surgery, Vol. 22, No.
10, 1998, pp. 1048-1055. doi:10.1007/s002689900515
[27] J. A. Sosa, P. J. Mehta, T. S. Wang, L. Boudourakis and S.
A. Roman, “A Population-Based Study of Outcomes from
Thyroidectomy in Aging Americans: At What Cost?”
Journal of the American College of Surgeons, Vol. 206,
No. 3, 2008, pp. 1097-1105.
doi:10.1016/j.jamcollsurg.2007.11.023