another treatment modality, such as neoadjuvant chemotherapy. [27] or continuous hyperthermic peritoneal perfusion [28], should be used to improve the ...
@
World J. Surg. 19, 707-713, 1995
WORLD Journal of
SURGERY 9 1995 by the Soci&6 Internationale de Chirurgie
Survival after Resection of Gastric Cancer and Prognostic Relevance of Systematic Lymph Node Dissection: Twenty Years Experience in Taiwan Wei-Jei Lee, M.D., 1 W e n - C h u n g Lee, M.D., PhD., 2 Shyh-Jinn H o u n g , M.D., 1 Chia-Tung Shun, M.D., 3 R e n - L o n g H o u n g , M.D., 4 P o - H u n g Lee, M.D., 1 King-Jen Chang, M.D., I T a - C h e n g Wei, M.D., 1 Kai-Mo Chen, M.D. 1 ~Department of Surgery, National Taiwan University Hospital, National Taiwan University, No. 7, Chung Shan South Road, Taipei, Taiwan, Republic of China 2Department of Public Health, National Taiwan University Hospital, National Taiwan University, No. 7, Chung Shan South Road, Taipei, Taiwan, Republic of China 3Department of Pathology, National Taiwan University Hospital, National Taiwan University, No. 7, Chung Shan South Road, Taipei, Taiwan, Republic of China 4Cancer Research Center, National Taiwan University Hospital, National Taiwan University, No. 7, Chung Shan South Road, Taipei, Taiwan, Republic of China
Abstract. A retrospective study of 954 resectable gastric cancers in a single institute of Taiwan from 1971 to 1990 was performed to evaluate improvements in gastric cancer surgery. The patients were divided into four time periods representing an overall experience of progressive implementation of aggressive resection and increased extent of systematic lymph node dissection. The clinicopathologic data and survival rates were statistically compared and the significance of the extent of resection on survival analyzed. A significant increase in the proportion of upper one-third tumors (from 14.8% to 20.4%) and a decrease in the incidence of intestinal type (73.6% to 41.5%) was found within the overall period. The proportion of patients with early gastric cancer increased from 11.5% to 19.4%. Patients who underwent total gastrectomy and combined visceral resection increased from 13.7% to 27.4% and 19.8% to 41.1%, respectively. An increase of both total dissected lymph node number and the incidence of detected lymph node metastases in early gastric cancer were associated with more extensive lymphadenectomy. An improved 5-year survival rate following aggressive resection was found for all stages except stage IV and T4 lesions, and the surgical mortality decreased from 5.5% to 2.0%. Patients with earlier stage lesions benefited more from radical resection, especially those with stage II and T2 lesions. Systematic lymph node dissection increased the 5-year survival of patients by about 10% for stage III or T3 lesions but not for patients with stage IV or T4 lesions. Multivariate analysis confirmed the significance of the improved technique of lymphadenectomy on the prognosis of gastric cancer following resection in Taiwan. In conclusion, a changing pattern of gastric cancer epidemiology in Taiwan is documented. A decrease in surgical mortality indicates that curative resection with extensive lymph node dissection can be safely performed. Although potential for increasing survival by upstaging remains, radical gastrectomy with extended lymphadenectomy could be adopted for gastric cancer resection with possibly more adequate control of loeoregional disease.
Although gastric cancer is a major malignancy that occurs worldwide, its incidence varies geographically [1]. In Taiwan, gastric cancer remains the third leading cause of cancer death with Correspondence to: K-M. Chen, M.D.
annual death rates in 1990 of 15.4 and 7.2 deaths per 100,000 for males and females, respectively [2]. Surgery remains the primary therapeutic modality for gastric cancer. Although radical gastrectomy with systematic lymph node dissection has been advocated by both Japanese surgeons [3-6] and others [7, 8] as standard procedure for gastric cancer treatment, its status has not been characterized by prospective randomized trials [9, 10]. In our hospital, an aggressive approach to lymph node dissection for gastric cancer has been used since the end of 1970. Because of an encouraging initial experience, this approach was subsequently adopted as routine gastric cancer resection. It is not only interesting but of clinical importance to evaluate the overall result of gastric cancer surgery during the past 20 years when this aggressive program has been used. In this study, we focus on the survival rate following curative resection of gastric cancer for the period since the aggressive approach toward gastric cancer resection and lymphadenectomy was progressively implemented.
Patients and Methods
From 1 January 1971 through 31 December 1990 there were 954 patients who underwent gastric resections for primary adenocarcinoma of the stomach in the Department of Surgery, National Taiwan University. Patients' medical records were reviewed to ascertain age, sex, tumor location and size, and type of gastric resection. Survival status was determined from medical records and by personal follow-up. A detail histopathologic review of the pathologic slides were performed to determine the histologic type, depth of gastric wall invasion, number of lymph nodes excised, and the extent of lymph node metastases. The classification and operation procedure was performed according to the protocol of the Japanese Research Society for
708
World J. Surg. Vol. 19, No. 5, Sept./Oct. 1995
Table 1. Clinicopathologic characteristics at different periods.
Characteristic
1971-1975 (n = 182)
1976 -1980 (n = 249)
1981-1985 (n = 224)
1986-1990 (n = 299)
p
Age (mean _+ SD) Male/female ratio
59.4 _+ 11.9 2.6
57.3 _+ 12.0 2.3
58.8 _+ 12.0 1.8
59.1 _+ 12.1 2.0
NS NS
Location Upper third Middle third Lower third
27 (14.8) 29 (15.9) 126 (69.2)
36 (14.5) 53 (21.3) 160 (64.2)
46 (20.5) 58 (25.6) 120 (53.6)
61 (20.4) 84 (28.0) 155 (51.6)
Histology Intestinal Diffuse
134 (73.6) 48 (26.4)
180 (72.3) 69 (27.7)
114 (51.1) 109 (48.9)
124 (41.5) 175 (58.5)
Invasion depth T1 T2 T3 T4
21 (11.5) 41 (22.5) 101 (55.5) 19 (10.4)
39 (15.7) 69 (27.7) 128 (51.4) 13 (5.2)
32 (14.3) 66 (29.5) 116 (51.8) 10 (4.5)
58 (19.4) 53 (17.7) 160 (53.5) 28 (9.4)
0.021
< 0.001
< 0.01 NS NS NS
SD: standard deviation; NS: no significance; numbers in parentheses are percents.
Gastric Cancer (JRSGC) [11]. Nodal involvement is classified into n ( - ) , n l ( + ) , n2(+), n3(+), and n4(+) according to the Japanese enumeration. The first three designations are similar to the pN0, pN1, and pN2 of the TNM system [12]. Both n3(+) and n4(+) are defined as metastasis, or pM1 in the TNM system. The extent of lymphadenectomy is classified as R1, R2, R3, and R4 according to the "n" level for which lymph node dissection was performed. For cancer in the upper part, total gastrectomy with splenectomy and distal hemipancreatectomy was routinely performed as an R2 resection for clearance of associated lymph nodes. The histologic pattern was classified into intestinal and diffuse types on the basis of Lauren's classification [13]. The patients were grouped according to four periods: Period I (1971-1975) represented 182 patients on whom R1 resection was routinely performed. Period II (1976-1980) contained 249 patients on whom R1 or R2 resection was routinely performed. Period III (1981-1985) comprised 221 patients in whom R2 resection was routinely performed. Period IV (1986-1990) comprised 299 patients on whom R2 to R3 resection was the principal form of surgery. The depth of cancer invasion was defined according to the TNM system: pT1, the cancer has invaded the mucosa and submucosa; pT2, the tumor invaded the muscularis or the subserosa; pT3, the cancer invaded over the serosa; pT4, the cancer had invaded adjacent structures. The site of the cancer was classified into upper third (C), middle third (M), and lower third (A) of the stomach. The 5-year survival curves for the periods were compared and the results assessed with regard to differences in TNM stage, tumor location, and invasion depth of the cancer. Survival curves were generated using the method of Kaplan and Meier. Statistical differences between survival curves were evaluated using the log-rank test. Differences between patient populations were assessed using the chi-square test, Kruskal-Wallis test, or ANOVA, as appropriate. Multivariate analysis with Cox's regression model was performed only for the most recent three periods because the data of the first period were not sufficiently detailed. Statistical analysis was done using the SAS package.
Results
Clinicopathologic Background Clinicopathologic data were compared for patients of different time periods (Table 1). There were no differences in age distribution or sex ratio. Significant correlation was found between the location of tumors, distribution of invasion depth, and histologic type. A shift of gastric cancer from the distal third to upper third was observed during this period. Tumors of the upper third increased from 14.8% to 20.4%, whereas the distal third localization decreased from 69.2% to 51.6%. A shift in the intestinal type of gastric cancer was also observed to decrease from 73.6% to 41.5%. Early gastric cancer increased from 11.5% to 19.4%.
Surgery There was a tendency to use the more extensive operation for resection of gastric cancer as time progressed (Table 2). More total gastrectomies and combined visceral resections were performed during the most recent period (1986-1990). Within this period, 27.4% of the patients underwent total gastrectomy and 41.1% underwent combined resection of other viscera. The average number of total dissected lymph nodes per patient increased throughout the study as well. During the first period (1971-1975) the average number of lymph nodes removed was 6.2 +- 7.4 compared to 32.3 -+ 16.7 during the most recent period. This difference reflects the more aggressive approach taken with regard to lymphadenectomy during the latter portion of the study. Although more radical operations were performed within the most recent period, operative mortality decreased from 5.5% to 2.0%. There was no increased incidence in the detection of lymph node metastases or in the number of metastases per patient over the various time periods. However, increased detection of nodal metastases was found for early cancer (T1) when stratified by the factor of invasion depth.
Lee et al.: Progress of Gastric Cancer Surgery
709
Table 2. Pathologic characteristics at different periods.
Characteristic
1971-1975 (n = 182)
1976 -1980 (n = 249)
1981-1985 (n = 224)
1986 -1990 (n = 299)
p
Curative resection
131 (72.0)
221 (88.8)
192 (86.8)
239 (79.9)
NS
Extent of surgery Total gastrectomy Total LN no. (mean -+ SD) Combined resection
25 (13.7) 6.2 _+ 7.4 36 (19.8)
32 (12.8) 11.5 _+ 10.2 68 (27.3)
46 (20.5) 22.9 + 12.3 81 (36.2)
82 (27.4) 32.3 _+ 16.7 123 (41.1)
< 0.001 < 0.001 < 0.01
Surgical mortality
10 (5.5)
11 (5.4)
7 (3.1)
6 (2.0)
119 (65.4) 3.4 __+6.2
140 (56.2) 3.6 -+ 5.8
139 (62.0) 4.8 _+7.8
182 (60.8) 3.7 -+ 5.3
2 (4.7) 20 (48.8) 80 (79.2) 17 (89.5)
3 (7.7) 25 (36.2) 102 (79.7) 10 (76.9)
4 (12.5) 33 (50.0) 94 (81.0) 8 (80.0)
9 (15.5) 25 (47.2) 121 (75.6) 27 (96.4)
LN metastases All Positive number Related to depth T1 T2 T3 T4
NS NS NS
LN: lymph node; SD: standard deviation; NS: no significance; numbers in parentheses are percents.
100 Statistical Comparisons All Four.Curves 80 n
>
,m
P
60
u) 1986-1990 (n=299)
4
52.4 11.6 9.2
58.7 12.5 6.6
72.9 21.8 8.8
75.0 46.2 16.7
< 0.05 < 0.001 < 0.05
Tumor location Upper Middle Lower
13.3 23.1 27.8
5.7 36.7 34.6
19.5 45.4 39.4
20.6 46.8 53.4
< 0.001 NS < 0.05
LN: lymph node. All values are percents.
=- IA (n=50)
100
\
=
IB (n=36)
80 m
.> P u) c r k.
II (n=5t)
60
IliA (n=47) 40
n
20
IIIB (n=49) IV
