- Email: [email protected]
The negative effect of splenectomy on the prognosis of gastric cancer
The Negative Effect of Splenectomy on the Prognosis of Gastric Cancer
Shlnlchi Suahlro, MD, Hiroshima, Japan Naofuml Nagasue, MD, Hiroshima, Japan Yulchho Dgawa, MD, Hiroshima, Japan Yuklharu Sasakl, MD, Hiroshima, Japan Soro HIrose, MD, Hiroshima, Japan Hlrofuml Yukaya, MD, Hiroshima, Japan
It is generally accepted that surgical treatment, not only for gastric cancer but also for other malignant neoplasms, should be as radical as possible. For the past two decades, the development of medicine has enabled us to safely perform combined resection of such organs as the spleen, pancreas, colon, mesocolon, and liver in the treatment of gastric cancer. Such aggressive resections have contributed to improving the long-term survival rate of patients with gastric cancer. Splenectomy has been most commonly performed simultaneously with total gastrectomy. Since the fmt report of severe infectious disease in children after splenectomy was published [I], this complication has become well known as one of the common and fatal complications after splenectomy, not only in young children but aleo in adults [2,3]. Although the immunologic role of the spleen is not fully understood, some reports suggest that the spleen has a negative effect on tumor immunity [4-6]. There has been much discussion regarding the propriety of splenectomy in terms of surgical treatment of gastric cancer. At present, there are very few randomized studies, and a uniform conclusion has not been gained as yet. In the present study, the prognostic significance of combined resection of the spleen in the treatment of gastric cancer was elucidated by comparing the long-term survival rate, early surgical complications, and the incidence of late infectious disease in patients with and without splenectomy. Fan
the Oeptrtmnt of Sugary, l+lmsMm Red Cross k@tal.
Japan.
Hkoshhna,
Reqwfi4s for reprintsshould be a&Mssed to Shlnichi Suehlro. MD. DeparbmrltOf surgay.Hlroshlma Red crossHospital, sendemachl l-9-6,
Hiroshima
vduma
730. Japan.
148, Novombu
1004
Material and Methods During the 11 year period from 1970 through 1980, at Hiroshima Red Cross Hospital, 671 patients underwent surgical operation for cure, palliation, or exploration of gastric cancer. During that period, total gastrectomy was performed in 103 patienta. It was performed in conjunction with splenectomy in 53 of the patients and without it in 50 patients. Splenectomy was not carried out in the latter patients hecause the tumor was located in the upper or entire stomach. Two patients had previously undergone distal resection of the stomach for benign ulcer. Resection of the remnant stomach was performed in these patients, one of whom underwent splenectomy simultaneously. The surgical procedures undertaken in the 103 patients are listed in Table I. End-to-aide esophagojejunostomy was performed in 82 patients (79.6 percent). It was performed more often than the other methods, probably due to the chief surgeon’s preference. Splenectomy was carried out in 53 of 103 patients because of either intraoperative accidental injury of the spleen or cancer involvement of the splenic hilus. Except for 53 splenectomies and 2 cholecystectomies for gallbladder stones, combined resections of the other organs were performed in 21 patients. In all of the 21 patients, cancer involvement of the organs was suspected or confiied by operative findings. None of the patients received a large enough doeage of chemotherapy or radiotherapy to cause severe bone marrow suppression or immunosuppression throughout the hospital stay and follow-up days. The mean ages of the splenectomized group of patients (35 men and 18 women) and the nonsplenectomized group (25 men and 25 women) were 57.3 and 53.7 years, respectively. The cure rates of the splenectomized and nonsplenectomized groups were 58.5 percent and 56 percent, respectively. The numbers of patients in each stage of disease are described in Table II. There were no significant differences between the groups. The stage of disease and
645
Suhiro et al
TABLE I
Analysts of Surgkal Procedures Performed In 103 Patlents Wlth Gastrk Cancer No. of Patients
Procedures
101 2
Total gastrectomy Resection of residual stomach
l
TABLE II
Reconstructive procedures Esophagojejunostomy, end-to-side Esophagojejunostomy, Roux-Y Wayama procedure’ Jejunal Interposition
82 7 1 12
The organs resected simultaneously Spleen Pancreas Colon, mesocofon ovary Liver Gallbladdar
53 16 a 2 1 2
Classltlcatkn by Stage of Disease In the Splenectomlzed and Nonsplenectomlzed Groups Splenectomized QouP n (n=53)
Disease stage I II Ill IV
TABLE IV
[a?].
96
Nonsplenectomized Croup n (n=50)
%
3 3 22
5.6 5.6 41.5
3 3 18
6 6 36
25
47.2
26
52
Relatlonshlp Between Anastomotlc Leakage and Reconstructive Procedure
Procedues
No. of Patients
Anastomotic Leakage
Death
82
2
1
7 1 12
0 0 4
0 0 2
Esophagoklunostomy,
end-to-side
TABLE Ill
lncldence of Early Compllcatlon and Late Infectlous Disease In the Splenectomlred and Nonsplenectomlted Groups splenectoflllzed Group
Nonsplenectomired Group
4153 4153
4150 3150
116 l/6
118 210
Early complication Anastomotic leakage Respiratory abnormalities Late infectious disease Pneumonia Common cold
the grade of regional lymph node metastasis were assigned according to the general rules of the gastric cancer study in Japan [7]. The 5 year survival rate was calculated by the cumulative method. Late-onset infectious diseases were investigated in the living patients by mail-in interview.
Results The mortality rate in the splenectomized group was 2 percent (one patient). This patient died from respiratory complications. He was 83 years old and his preoperative lung function was extremely poor. The mortality rate in the nonsplenectomized group was 5.7 percent (three patients). Two of them died from acute peritonitis caused by anastomotic leakage and one from respiratory complications. The overall operative mortality rate was 3.8 percent (Table III). There was no significant difference in the operative mortality rates between the two groups. Early complications included anastomotic leakage, which occurred in four patients in each group and resulted in death in three of the nonsplenectomized patients. The incidence of anastomotic leakage was more correlated with the reconstructive method than
Esphagolelunostomy, Roux-Y Nakayama procedure’ Jejunal interposition l
[ 78].
with whether or not splenectomy had been carried out (Table IV). Respiratory complications, pneumonia, or bronchial obstruction by sputum occurred in four patients in the nonsplenectomized group and in three patients in the splenectomized group. Three patients died from respiratory complications, two in the nonsplenectomized group and one in the splenectomized group. There were no differences in the incidences and mortality rates of early complications between the two groups. However, postoperative fever elevation, the course of which was not determined, tended to last longer in patients in the splenectomized group than in those in the nonsplenectomized group. Six splenectomized and eight nonsplenectomized patients answered our mail-in interviews. The number of patients who had suffered from late pneumonia was only two (one in each group). One patient in the splenectomized group and two in the nonsplenectomized group have suffered from the common cold more often than before operation. There were no differences between the two groups in regard to the incidence of late infectious disease. At last follow-up, 87 patients had died. Four were operative deaths, 74 were cancer-related deaths, and 5 were noncancer-related deaths. The causes of the five noncancer-related deaths were as follows: liver failure due to fulminant viral hepatitis (one patient), airway obstruction by sputum (one patient), suicide (one patient), traffic accident (one patient), and unknown cause (one patient). The 5 year survival
Effect of Splenectomy
\
- - - -
Over all (n=l03)
-.-.-
PaIllatIve
Radical
resectton
on Gastric Cancer
(n=59)
resectlon
(n=44)
\ \ \ \
0
I
2
3
4 postoperatlve
5 years
cwn&the swvtval rata9 h 103 pathds who total gaatmctomy hw gaatrk cancw.
fllowv 1.-Y-f
fW2-W
urrdwmmt
and 50
rate overall was 23.1 percent and that in the curative cases, 37.1 percent (Figure 1). In the nonsplenectomized and splenectomized groups, the 5 year survival rates were 23.1 percent and 21.8 percent, respectively (Figure 2). Splenectomy did not significantly affect the survival rate up to 5 years after operation. comments Total gastrectomy has been performed for gastric cancer located either in the upper stomach or in the entire stomach. Maruyama [8] visualized lymphatic flow from the stomach by injecting oil contrast media (Lipiodol@) into the submucosal layer of the stomach. The lymphograms revealed that the lymph of the left upper region flowed to the splenic hilus and along the splenic artery to the celiac artery. The findings indicated that it is important in the surgical treatment of upper stomach cancer to dissect the lymph nodes of the splenic hilus. In regard to the clinical aspects, the rate of microscopic cancer involvement of the splenic hilus has been reported to be 30 to 40 percent in cases of total gastrectomy with splenectomy (9111. Kanai [12] reported that the rate was higher in splenectomized patients than in nonsplenectomized patients. It is conceivable that total gastrectomy combined with splenectomy may precipitate a more radical operation.
voluma 148, Novodmr 1994
cwnMthv~alr~hS33
rwmpbmd&edpattent8
otter lotsi gaetrectomy.
Several nonrandomized studies by Japanese investigators have shown that the long-term survival rates of patients with splenectomy are slightly lower than those of patients without splenectomy [13-161. However, in most of their studies, the stage of disease was more advanced in the splenectomized patients than in the nonsplenectomized patients. As a result, there is no consensus of opinion regarding the value of splenectomy; some physicians are in favor of it and some are not. Incidentally, many Western reports have warned of an increased incidence of severe infectious disease after splenectomy. However there are few Japanese reports that infectious disease will be a clinical problem after splenectomy. Our series also showed no substantial difference in the incidence of infectious disease between the two groups. No patient died from late, severe infectious disease in the splenectomized group. It is unclear as yet why the results in Japanese patients are different from those in reports from the West. Possibly it is due to a racial difference. As long as extensive chemotherapy or radiotherapy is not performed postoperatively, Japanese patients do not seem to suffer from severe infectious disease, regardless of the presence or absence of the spleen. In an experimental study, suppressor T cells, which play a role in suppressing cellular immunity against
647
Suhiro et al
cancer, were found in the spleen [4). Robinette [17], in a long-term follow-up study of 740 American veterans splenectomized because of trauma, reported no significant difference in the mortality rate from cancer compared with the rate in nonsplenectomized veterans. Splenectomy does not seem to be deleterious in the treatment of gastric cancer, at least from an immunologic standpoint. Since the procedure of total gastrectomy can be simplified by the addition of splenectomy, it may be performed in poor-risk patients without immunodeficiency. The present study indicates that splenectomy does not induce any detrimental effect in terms of cancer prognosis and postoperative complications in Japanese patients and that combined resection of the spleen should be performed during surgical treatment of patients with gastric cancer when lymph node metastasis is suspected in the splenic hilus. Summary During the 11 year period from 1970 through 1980, 103 patients underwent total gastrectomy for gastric cancer. Splenectomy was performed in 53 of the patients because of perioperative accidental injury or an operative finding of tumor metastasis to the lymph nodes or direct invasion to the splenic hilus. Retrospective comparisons were carried out among the 53 splenectomized and 50 nonsplenectomized patients in terms of the long-term survival rate and early and late complications. The two groups of patients were quite similar with regard to rates of curative resection and stages of disease. The 5 year cumulative survival rates overall and in the splenectomized and nonsplenectomized groups were 23.1,21.8 and 23.8 percent, respectively. Splenectomy did not affect the prognosis of gastric cancer. In addition, there were no differences between the two groups in the incidences of early surgical complications and infectious disease of late onset. Our results suggest that, in the treatment of gastric cancer, splenectomy may not be avoided, either when the spleen is injured perioperatively or when cancer involvement is suspected in the splenic hilus.
646
References 1. King H, Shumaker HB. Susceptibility to infection after sple nectomy performed in infancy. Ann Surg 1952;136:23942. 2. Singer DB. Post-splenectcmy sepsis. Perspectives in pediatric pathology. Chicago: Year Book Medical, 19783284-307. 3. Standage BA, Gross JC. Cutccme and sepsis after splenectomy in adults. Am J Surg 1982;143:545-8. 4. Gerson RK. Lancer EM, Kondo K. Immuno-regulatwy role of spleen localizing thymocytes. J lmmunol 1974;122:54654. 5. Meyer JD, Argyris BF, Meyer JA. Splenectcmy, suppressoc cell activity and survival in tumor bearing rats. J Surg Res 1980;29:527-32. 8. Takahashi M, Fujimoto S. Clinical study on the conelation between immunologic status and total gasfrectomy combined with splenectomy. Jpn J Surg 1980;10:100-4. 7. Japanese Research Society for Gastric Cancer. The general rules for the gastric cancer study In Surgery. Jpn J Surg 1979;3:61. 8. Maruyama K. Lymphatic flow of gastric cardia. Stom lntest 1978;13:1535-42. 9. Fly OA, Dockerty MB, Waugh JM. Metastasis to the regional nodes of the splenic hilus from carcinoma of the stomach. Surg Gynecol Obstet 1956;102:279-86. 10. Omori Y. Some problems on extent of gastric surgery in relation to clinicopathological flndings. J Clin Surg 1971;26: 1855-62. 11. Ohashl I, Takagi K, Dta H, Kamiya J, Nakagoe T, Kajltani T. Pancreaticosplenectcmy for advanced gastric carcinoma. Jpn J Gastroenterol Surg 1979;12:993-9. 12. Kanai H. Significance of pancreaticosplenectomy in gastric resection for gastric cancer. Nippon Gan Chiryo Gakkai Shi 1967;2:328-38. 13. Sugimachi K, lnokuchi K. Critical evaluation of prophylactic splenectomy in total gastrectomy for the stomach cancer. Gann 1980;71:704-9. 14. Koga S. Kaibara N, Kimwa 0. Nishkfori H, Kishimoto H. Prognostic significance of combined splenectomy of pancreaticosplenectomy in total and proximal gastrectcmy for gastric cancer. Am J Surg 1981;142:546-50. 15. Yoshino K, Haruyama K, Nakamura S, et al. Evaluatlon of splenectomy for gastric cancer. Nippon Gan Chiryo Gakkal Shi 1979;12:944-9. 16. Miwa H, Kojima K, Drita K. et al. Effect of splenectomy and immunotfierapy on advanced gastric cancer associated with total gastrectomy. Jpn J Surg;13:20-4. 17. Robinette CD. Splenectomy and subsequent mortality In veterans of the 1939-45 war. Lancet 1977;2:127-9. 18. Nakayame K. New reconstructive method after excision of lower esophagus and cardiac portion of stomach: beta anastomosis combined with antrectomy. Surgery 1963; 54:281.
lha American
Journal d Surfpry
Shlnlchi Suahlro, MD, Hiroshima, Japan Naofuml Nagasue, MD, Hiroshima, Japan Yulchho Dgawa, MD, Hiroshima, Japan Yuklharu Sasakl, MD, Hiroshima, Japan Soro HIrose, MD, Hiroshima, Japan Hlrofuml Yukaya, MD, Hiroshima, Japan
It is generally accepted that surgical treatment, not only for gastric cancer but also for other malignant neoplasms, should be as radical as possible. For the past two decades, the development of medicine has enabled us to safely perform combined resection of such organs as the spleen, pancreas, colon, mesocolon, and liver in the treatment of gastric cancer. Such aggressive resections have contributed to improving the long-term survival rate of patients with gastric cancer. Splenectomy has been most commonly performed simultaneously with total gastrectomy. Since the fmt report of severe infectious disease in children after splenectomy was published [I], this complication has become well known as one of the common and fatal complications after splenectomy, not only in young children but aleo in adults [2,3]. Although the immunologic role of the spleen is not fully understood, some reports suggest that the spleen has a negative effect on tumor immunity [4-6]. There has been much discussion regarding the propriety of splenectomy in terms of surgical treatment of gastric cancer. At present, there are very few randomized studies, and a uniform conclusion has not been gained as yet. In the present study, the prognostic significance of combined resection of the spleen in the treatment of gastric cancer was elucidated by comparing the long-term survival rate, early surgical complications, and the incidence of late infectious disease in patients with and without splenectomy. Fan
the Oeptrtmnt of Sugary, l+lmsMm Red Cross k@tal.
Japan.
Hkoshhna,
Reqwfi4s for reprintsshould be a&Mssed to Shlnichi Suehlro. MD. DeparbmrltOf surgay.Hlroshlma Red crossHospital, sendemachl l-9-6,
Hiroshima
vduma
730. Japan.
148, Novombu
1004
Material and Methods During the 11 year period from 1970 through 1980, at Hiroshima Red Cross Hospital, 671 patients underwent surgical operation for cure, palliation, or exploration of gastric cancer. During that period, total gastrectomy was performed in 103 patienta. It was performed in conjunction with splenectomy in 53 of the patients and without it in 50 patients. Splenectomy was not carried out in the latter patients hecause the tumor was located in the upper or entire stomach. Two patients had previously undergone distal resection of the stomach for benign ulcer. Resection of the remnant stomach was performed in these patients, one of whom underwent splenectomy simultaneously. The surgical procedures undertaken in the 103 patients are listed in Table I. End-to-aide esophagojejunostomy was performed in 82 patients (79.6 percent). It was performed more often than the other methods, probably due to the chief surgeon’s preference. Splenectomy was carried out in 53 of 103 patients because of either intraoperative accidental injury of the spleen or cancer involvement of the splenic hilus. Except for 53 splenectomies and 2 cholecystectomies for gallbladder stones, combined resections of the other organs were performed in 21 patients. In all of the 21 patients, cancer involvement of the organs was suspected or confiied by operative findings. None of the patients received a large enough doeage of chemotherapy or radiotherapy to cause severe bone marrow suppression or immunosuppression throughout the hospital stay and follow-up days. The mean ages of the splenectomized group of patients (35 men and 18 women) and the nonsplenectomized group (25 men and 25 women) were 57.3 and 53.7 years, respectively. The cure rates of the splenectomized and nonsplenectomized groups were 58.5 percent and 56 percent, respectively. The numbers of patients in each stage of disease are described in Table II. There were no significant differences between the groups. The stage of disease and
645
Suhiro et al
TABLE I
Analysts of Surgkal Procedures Performed In 103 Patlents Wlth Gastrk Cancer No. of Patients
Procedures
101 2
Total gastrectomy Resection of residual stomach
l
TABLE II
Reconstructive procedures Esophagojejunostomy, end-to-side Esophagojejunostomy, Roux-Y Wayama procedure’ Jejunal Interposition
82 7 1 12
The organs resected simultaneously Spleen Pancreas Colon, mesocofon ovary Liver Gallbladdar
53 16 a 2 1 2
Classltlcatkn by Stage of Disease In the Splenectomlzed and Nonsplenectomlzed Groups Splenectomized QouP n (n=53)
Disease stage I II Ill IV
TABLE IV
[a?].
96
Nonsplenectomized Croup n (n=50)
%
3 3 22
5.6 5.6 41.5
3 3 18
6 6 36
25
47.2
26
52
Relatlonshlp Between Anastomotlc Leakage and Reconstructive Procedure
Procedues
No. of Patients
Anastomotic Leakage
Death
82
2
1
7 1 12
0 0 4
0 0 2
Esophagoklunostomy,
end-to-side
TABLE Ill
lncldence of Early Compllcatlon and Late Infectlous Disease In the Splenectomlred and Nonsplenectomlted Groups splenectoflllzed Group
Nonsplenectomired Group
4153 4153
4150 3150
116 l/6
118 210
Early complication Anastomotic leakage Respiratory abnormalities Late infectious disease Pneumonia Common cold
the grade of regional lymph node metastasis were assigned according to the general rules of the gastric cancer study in Japan [7]. The 5 year survival rate was calculated by the cumulative method. Late-onset infectious diseases were investigated in the living patients by mail-in interview.
Results The mortality rate in the splenectomized group was 2 percent (one patient). This patient died from respiratory complications. He was 83 years old and his preoperative lung function was extremely poor. The mortality rate in the nonsplenectomized group was 5.7 percent (three patients). Two of them died from acute peritonitis caused by anastomotic leakage and one from respiratory complications. The overall operative mortality rate was 3.8 percent (Table III). There was no significant difference in the operative mortality rates between the two groups. Early complications included anastomotic leakage, which occurred in four patients in each group and resulted in death in three of the nonsplenectomized patients. The incidence of anastomotic leakage was more correlated with the reconstructive method than
Esphagolelunostomy, Roux-Y Nakayama procedure’ Jejunal interposition l
[ 78].
with whether or not splenectomy had been carried out (Table IV). Respiratory complications, pneumonia, or bronchial obstruction by sputum occurred in four patients in the nonsplenectomized group and in three patients in the splenectomized group. Three patients died from respiratory complications, two in the nonsplenectomized group and one in the splenectomized group. There were no differences in the incidences and mortality rates of early complications between the two groups. However, postoperative fever elevation, the course of which was not determined, tended to last longer in patients in the splenectomized group than in those in the nonsplenectomized group. Six splenectomized and eight nonsplenectomized patients answered our mail-in interviews. The number of patients who had suffered from late pneumonia was only two (one in each group). One patient in the splenectomized group and two in the nonsplenectomized group have suffered from the common cold more often than before operation. There were no differences between the two groups in regard to the incidence of late infectious disease. At last follow-up, 87 patients had died. Four were operative deaths, 74 were cancer-related deaths, and 5 were noncancer-related deaths. The causes of the five noncancer-related deaths were as follows: liver failure due to fulminant viral hepatitis (one patient), airway obstruction by sputum (one patient), suicide (one patient), traffic accident (one patient), and unknown cause (one patient). The 5 year survival
Effect of Splenectomy
\
- - - -
Over all (n=l03)
-.-.-
PaIllatIve
Radical
resectton
on Gastric Cancer
(n=59)
resectlon
(n=44)
\ \ \ \
0
I
2
3
4 postoperatlve
5 years
cwn&the swvtval rata9 h 103 pathds who total gaatmctomy hw gaatrk cancw.
fllowv 1.-Y-f
fW2-W
urrdwmmt
and 50
rate overall was 23.1 percent and that in the curative cases, 37.1 percent (Figure 1). In the nonsplenectomized and splenectomized groups, the 5 year survival rates were 23.1 percent and 21.8 percent, respectively (Figure 2). Splenectomy did not significantly affect the survival rate up to 5 years after operation. comments Total gastrectomy has been performed for gastric cancer located either in the upper stomach or in the entire stomach. Maruyama [8] visualized lymphatic flow from the stomach by injecting oil contrast media (Lipiodol@) into the submucosal layer of the stomach. The lymphograms revealed that the lymph of the left upper region flowed to the splenic hilus and along the splenic artery to the celiac artery. The findings indicated that it is important in the surgical treatment of upper stomach cancer to dissect the lymph nodes of the splenic hilus. In regard to the clinical aspects, the rate of microscopic cancer involvement of the splenic hilus has been reported to be 30 to 40 percent in cases of total gastrectomy with splenectomy (9111. Kanai [12] reported that the rate was higher in splenectomized patients than in nonsplenectomized patients. It is conceivable that total gastrectomy combined with splenectomy may precipitate a more radical operation.
voluma 148, Novodmr 1994
cwnMthv~alr~hS33
rwmpbmd&edpattent8
otter lotsi gaetrectomy.
Several nonrandomized studies by Japanese investigators have shown that the long-term survival rates of patients with splenectomy are slightly lower than those of patients without splenectomy [13-161. However, in most of their studies, the stage of disease was more advanced in the splenectomized patients than in the nonsplenectomized patients. As a result, there is no consensus of opinion regarding the value of splenectomy; some physicians are in favor of it and some are not. Incidentally, many Western reports have warned of an increased incidence of severe infectious disease after splenectomy. However there are few Japanese reports that infectious disease will be a clinical problem after splenectomy. Our series also showed no substantial difference in the incidence of infectious disease between the two groups. No patient died from late, severe infectious disease in the splenectomized group. It is unclear as yet why the results in Japanese patients are different from those in reports from the West. Possibly it is due to a racial difference. As long as extensive chemotherapy or radiotherapy is not performed postoperatively, Japanese patients do not seem to suffer from severe infectious disease, regardless of the presence or absence of the spleen. In an experimental study, suppressor T cells, which play a role in suppressing cellular immunity against
647
Suhiro et al
cancer, were found in the spleen [4). Robinette [17], in a long-term follow-up study of 740 American veterans splenectomized because of trauma, reported no significant difference in the mortality rate from cancer compared with the rate in nonsplenectomized veterans. Splenectomy does not seem to be deleterious in the treatment of gastric cancer, at least from an immunologic standpoint. Since the procedure of total gastrectomy can be simplified by the addition of splenectomy, it may be performed in poor-risk patients without immunodeficiency. The present study indicates that splenectomy does not induce any detrimental effect in terms of cancer prognosis and postoperative complications in Japanese patients and that combined resection of the spleen should be performed during surgical treatment of patients with gastric cancer when lymph node metastasis is suspected in the splenic hilus. Summary During the 11 year period from 1970 through 1980, 103 patients underwent total gastrectomy for gastric cancer. Splenectomy was performed in 53 of the patients because of perioperative accidental injury or an operative finding of tumor metastasis to the lymph nodes or direct invasion to the splenic hilus. Retrospective comparisons were carried out among the 53 splenectomized and 50 nonsplenectomized patients in terms of the long-term survival rate and early and late complications. The two groups of patients were quite similar with regard to rates of curative resection and stages of disease. The 5 year cumulative survival rates overall and in the splenectomized and nonsplenectomized groups were 23.1,21.8 and 23.8 percent, respectively. Splenectomy did not affect the prognosis of gastric cancer. In addition, there were no differences between the two groups in the incidences of early surgical complications and infectious disease of late onset. Our results suggest that, in the treatment of gastric cancer, splenectomy may not be avoided, either when the spleen is injured perioperatively or when cancer involvement is suspected in the splenic hilus.
646
References 1. King H, Shumaker HB. Susceptibility to infection after sple nectomy performed in infancy. Ann Surg 1952;136:23942. 2. Singer DB. Post-splenectcmy sepsis. Perspectives in pediatric pathology. Chicago: Year Book Medical, 19783284-307. 3. Standage BA, Gross JC. Cutccme and sepsis after splenectomy in adults. Am J Surg 1982;143:545-8. 4. Gerson RK. Lancer EM, Kondo K. Immuno-regulatwy role of spleen localizing thymocytes. J lmmunol 1974;122:54654. 5. Meyer JD, Argyris BF, Meyer JA. Splenectcmy, suppressoc cell activity and survival in tumor bearing rats. J Surg Res 1980;29:527-32. 8. Takahashi M, Fujimoto S. Clinical study on the conelation between immunologic status and total gasfrectomy combined with splenectomy. Jpn J Surg 1980;10:100-4. 7. Japanese Research Society for Gastric Cancer. The general rules for the gastric cancer study In Surgery. Jpn J Surg 1979;3:61. 8. Maruyama K. Lymphatic flow of gastric cardia. Stom lntest 1978;13:1535-42. 9. Fly OA, Dockerty MB, Waugh JM. Metastasis to the regional nodes of the splenic hilus from carcinoma of the stomach. Surg Gynecol Obstet 1956;102:279-86. 10. Omori Y. Some problems on extent of gastric surgery in relation to clinicopathological flndings. J Clin Surg 1971;26: 1855-62. 11. Ohashl I, Takagi K, Dta H, Kamiya J, Nakagoe T, Kajltani T. Pancreaticosplenectcmy for advanced gastric carcinoma. Jpn J Gastroenterol Surg 1979;12:993-9. 12. Kanai H. Significance of pancreaticosplenectomy in gastric resection for gastric cancer. Nippon Gan Chiryo Gakkai Shi 1967;2:328-38. 13. Sugimachi K, lnokuchi K. Critical evaluation of prophylactic splenectomy in total gastrectomy for the stomach cancer. Gann 1980;71:704-9. 14. Koga S. Kaibara N, Kimwa 0. Nishkfori H, Kishimoto H. Prognostic significance of combined splenectomy of pancreaticosplenectomy in total and proximal gastrectcmy for gastric cancer. Am J Surg 1981;142:546-50. 15. Yoshino K, Haruyama K, Nakamura S, et al. Evaluatlon of splenectomy for gastric cancer. Nippon Gan Chiryo Gakkal Shi 1979;12:944-9. 16. Miwa H, Kojima K, Drita K. et al. Effect of splenectomy and immunotfierapy on advanced gastric cancer associated with total gastrectomy. Jpn J Surg;13:20-4. 17. Robinette CD. Splenectomy and subsequent mortality In veterans of the 1939-45 war. Lancet 1977;2:127-9. 18. Nakayame K. New reconstructive method after excision of lower esophagus and cardiac portion of stomach: beta anastomosis combined with antrectomy. Surgery 1963; 54:281.
lha American
Journal d Surfpry