Patients' Characteristics

The ratio of males to females was 2.3:1 in the elderly group and 2.1:1 in the control group (Table 1). There was no significant difference in the distribution of TNM stages between the two groups, although the control group included more stage I patients.

The resection rate and curative resection rate were similar in the two groups. The frequencies of total gastrectomy and adjacent structure resection were similar. However, the extent of nodal dissection was significantly different between the two groups. Among curative cases, 43.5% underwent D0,1 operation in the elderly group, while 86.9% underwent extended dissection in the control group.

Morbidity and Mortality

Table 2 summarizes pre-existing morbidity in the two groups. The rate of preoperative morbidity was higher in the elderly group than in the control group (76.8 vs 53.1, p < 0.001). Cardiovascular impairment was the predominant preoperative risk factor in both groups. There were significant differences in the incidence of cardiovascular impairment, renal impairment and anemia between the two groups.

Table 3 shows postoperative complications. The overall frequency of postoperative complications was similar in the two groups (32.1 vs 27.9%). Only respiratory impairment was significantly more common in the elderly group (p < 0.05).

The operative mortality rates were 2.7% in the elderly group and 0.9% in the control group (p = 0.17). The 90-day mortality rates were 10.7 and 3.9%, respectively (p < 0.01).

Table 4 summarizes the patients who underwent resection. The operative mortality rates were 3.0% in the elderly group and 0.8% in the control group overall (p < 0.1). The 90-day mortality rates were 8.0 and 2.5% overall, respectively (p < 0.01). In the patients without pre-existing morbidity, the frequencies of death during operation, death within 3 months of the operation and postoperative complication were similar between the groups.

Table 2. Pre-existing morbidity

	Age (years)		pa
	[ge]80	50-69
Respiratory	15 (13.4%)	262 (9.8%)	N.S.
Cardiovascular	34 (30.4%)	410 (15.4%)	0.001
Hepatic	2 (1.8%)	126 (4.7%)	N.S.
Renal	12 (10.7%)	77 (2.9%)	0.001
Diabetes mellitus	14 (12.5%)	293 (11.0%)	N.S.
Anemia	24 (21.4%)	293 (11.0%)	0.01
Overall	86 (76.8%)	1410 (53.1%)	0.001

^aChi-squared test.

Table 3. Incidence of postoperative complications

	Age (years)		pa
	[ge]80	50-69
Non-surgical
Respiratory	9 (8.0%)	96 (3.6%)	0.05
Cardiovascular	4 (3.6%)	46 (1.7%)	N.S.
Hepatic	4 (3.6%)	113 (4.2%)	N.S.
Renal	3 (2.7%)	23 (0.9%)	N.S.
Diabetes mellitus	2 (1.8%)	25 (0.9%)	N.S.
Surgical
Stenosis, obstruction	5 (4.5%)	94 (3.5%)	N.S.
Hemorrhage	4 (3.6%)	46 (1.7%)	N.S.
Anastomotic leakage	6 (5.4%)	154 (5.8%)	N.S.
Intra-abdominal infection	8 (7.1%)	288 (10.8%)	N.S.
Overall	36 (32.1%)	742 (27.9%)	N.S.

^aChi-squared test.

Table 4. Subgroup analysis of patients who underwent resection

	Number of patients		30-day mortality (%)		90-day mortality (%)		Complication (%)
	[ge]80	50-69	[ge]80	50-69	[ge]80	50-69	[ge]80	50-69
All resected cases
Overall	100	2513	3.0a	0.8a	8.0b	2.5b	35.0	28.4
Preoperative risk
Yes	76	1309	3.9	1.3	9.2c	3.5c	39.5	31.2
No	24	1204	0	0.2	4.2	1.5	20.8	25.2
Resection type
Total gastrectomy
Overall	32	644	9.4d	1.4d	18.8e	4.4e	62.5	47.0
Preoperative risk
Yes	25	344	12.0f	2.3f	24.0g	5.8g	72.0h	48.0h
No	7	300	0	0.3	0	2.7	28.6	46.0
Distal gastrectomy
Overall	64	1744	0	0.5	3.1	1.7	23.4	20.5
Preoperative risk
Yes	49	894	0	0.9	2.0	2.3	24.5	23.7
No	15	850	0	0	6.7	0.9	20.0	17.1
Nodal dissection
D0,1
Overall	48	514	2.1	1.4	8.3	6.4	37.5	27.0
Preoperative risk
Yes	37	314	2.7	2.2	10.8	8.6	40.5	28.3
No	11	200	0	0	0	3.0	27.3	25.0
D2
Overall	51	1933	3.9	0.7	7.8i	1.6i	33.3	27.6
Preoperative risk
Yes	39	960	5.1	1.0	7.7j	2.1j	38.5	31.2
No	12	973	0	0.3	8.3	1.0	16.7	23.9
Combined resection
Spleen
Overall	14	397	7.1	2.3	14.3	5.8	71.4	54.7
Preoperative risk
Yes	10	215	10.0	2.8	20.0	6.5	90.0	57.2
No	4	182	0	1.6	0	4.9	25.0	51.6
Distal pancreas
Overall	7	208	0	1.9	0	5.3	57.1	67.3
Preoperative risk
Yes	5	115	0	1.7	0	4.3	80.0	70.4
No	2	93	0	2.2	0	6.5	0	63.4

^{a, j} p < 0.1. ^{c, f, h} p < 0.05. ^{b, d, e, g, i} p < 0.01.

In the subgroup undergoing total gastrectomy, both the death rate during operation and the 90-day mortality rate were significantly higher in the elderly group than in the control group. The 90-day mortality rate in the elderly group was also significantly higher in the subgroup undergoing D2 dissection than for the younger patients having D2 dissection.

Survival

The overall 5-year survival rate of the elderly group was significantly lower than that of the control group [32.6% (42.4-22.9%) vs 61.1% (63.0-59.1%), p < 0.05]. When curative surgery was performed, the 5-year survival rate of elderly patients was again significantly lower than that of the controls [44.4% (52.3-28.5%) vs 74.0% (75.9-72.0%), p < 0.05]. This significant difference, however, was not observed when non-cancer deaths were excluded [62.5% (78.8-46.2%) vs 79.9% (81.8-78.0%)].

The elderly group had a higher non-cancer death rate than the younger group (Table 5). Twenty-eight elderly patients (~30%) died from causes other than primary cancer within 5 years of the curative operation.

Preoperative risk factors were observed in 25 patients (89.3%) of those elderly patients who died from causes other than primary cancer, whereas it was observed in 71 patients (43.1%) in the control group. Moreover, nine elderly patients, all of whom had preoperative risk factors, died within 1 year.

The recurrence rates were similar between the groups.