a prospective study on primary gastric stump cancer...

[CANCER RESEARCH 44, 2208-2212, May 1984]

A Prospective Study on Primary Gastric Stump Cancer following PartialGastrectomy for Benign Gastroduodenal Diseases1

Shinkan Tokudome,2 Summon Kono, Masato Ikeda, Masanori Kuratsune, Chiaki Sano, Kiyoshi Inokuchi,

Yoshifumi Kodama, Hitoshi Ichimiya, Fumio Nakayama, Nobuaki Kaibara, Shigemasa Koga, HiroyasuYamada, Taiji Ikejiri, Naokata Oka, and Hironaga Tsurumaru

Department of Community Health Science, Saga Medical School, Saga 840-01 [S. T.]; Department of Public Health [S. Kono, M. I., M. K.Â¡,The Second Department of

Surgery [C. S., K. I.], The First Department of Surgery [H. I., F. N.], Faculty of Medicine, Kyushu University, Fukuoka 812; The Second Department of Surgery, School ofMedicine, Fukuoka University, Fukuoka 814 [Y. K.]; The First Department of Surgery, Faculty of Medicine, Tottori University, Yonago 683 [N. K., S. Koga]; Department ofSurgery, National Fukuoka Central Hospital, Fukuoka 810 [H. Y., T. I.]; and Department of Surgery, Saga Municipal Koseikan Hospital, Saga 840 [N. O., H. T.], Japan

ABSTRACT

A prospective study was made on 3827 Japanese patientswho had undergone partial gastrectomy for benign gastroduo-

denal diseases to examine whether they are at a high risk ofmortality from primary gastric stump cancer (PGSC) and whetherthe risk is determined by the surgical procedure. The patientswere followed up from the time of surgery (from 1948 to 1970)to June 30, 1981. Of 3,701 patients (96.7%), the vital status atthe end of observation was determined, the total person-years

at risk being 62,286.33. The observed deaths were comparedwith the expected deaths calculated from the mortality rates ofJapan. An elapsed time of 10 years from operation to death wasset not only to exclude possible recurrent, remaining, or multiplecancers but also to allow a certain latency period for the development of PGSC. The observed and expected deaths fromPGSC were 11 and 52.85, respectively, the ratio being 0.21 (p< 0.01 ). The ratios were uniformly <1 for both sexes and acrossthree operative groups: Billroth I, Billroth II with Braun's anastomosis; or Billroth II without Braun's anastomosis. No difference

was observed between the death rates from PGSC by operationtype. The possible role of the postoperative nonphysiological(pathological) environment or duodenogastric reflux in gastricstump carcinogenesis was not detected in the present study.

INTRODUCTION

Low acidity (achlorhydria), lack of vagai and gastrin stimulation,duodenogastric reflux, and changes of bacterial flora are seen inthe gastric stump in patients partially gastrectomized for benigngastroduodenal diseases (6,18,22,42). In addition, pathologicalfeatures such as chronic atrophie gastritis, hyperplastic or polypoid changes, and intestinal metaplasia or dysplasia are observed in the stump (6, 8, 19, 22, 29, 31, 39, 40, 42). Highconcentrations of nitrite and carcinogenic A/-nitroso compounds

are detected in the gastric juice (33, 35, 38). Furthermore, inanimal experiments, the incidence of PGSC3 is higher in the

group subjected to Billroth II procedures with obligatory duodenogastric reflux than in the group treated with Bl procedures

1Presented in part at the 38th Meeting of the Japanese Research Society for

Gastric Cancer, Tokyo, 1982, and at the 21st General Assembly of the JapanMedical Congress, Osaka, 1983. Supported in part by a Grant-in-Aid for CancerResearch (55-3) from the Ministry of Health and Welfare and a Grant-in-Aid forCancer Research (57010068) from the Ministry of Education, Japan.

2To whom requests for reprints should be addressed.3The abbreviations used are: PGSC, primary gastric stump cancer; Bl, Billroth

I; Bll, B(-i-), Billroth II with Braun's anastomosis; Bll, B(-), Billroth II without Braun's

anastomosis; O/E ratio, observed versus expected ratio.Received April 7, 1983; accepted February 3, 1984.

without reflux (6, 22).A follow-up study was done on patients who had undergone

partial gastrectomy for benign gastroduodenal diseases to examine whether they are at high risk for PGSC mortality andwhether the risk is related to type of surgery.

MATERIALS AND METHODS

The medical records of the patients who had undergone partial gastrectomy for benign gastroduodenal diseases from 1948 to 1970 werereviewed at 5 departments of surgery in 4 institutions: The First andSecond Departments of Surgery, Kyushu University; The First Department of Surgery, Tottori University; Department of Surgery, NationalFukuoka Central Hospital; and Department of Surgery, Saga MunicipalKoseikan Hospital. Information on name, sex, date of birth, presentaddress, Honseki (permanent address), name of the head of household,date of operation, operative procedure, diagnosis after operation, andhistological diagnosis of the disease was obtained. When the patientswere known to have died at particular institutions, information on dateof death, diagnosis at autopsy, and causes of death was also obtained.The vital status was determined by referring to the city (town, village)office of the patient's Honseki. For the deceased, death certificates were

requested from the District Legal Affairs Bureau.The patients were divided into 3 study groups according to operative

procedure: Bl; Bll, B(+); and Bll, B(-).The patient population at risk was obtained by the person-years

method (24). The patients were observed from the date of operation tothe date of death or the end of observation (June 30, 1981). Somepatients lost to follow-up were observed up to the date last known alive,

the others were assumed to be alive up to the end of observation whenthe date last known alive was unknown.

According to the eighth revision of the ICD (45), underlying causes ofdeath were coded by S. Tokudome and S. Kono independently. Somecases coded discordantly were referred to the nosologist at the Healthand Welfare Statistics and Information Department, Minister's Secretar

iat, Ministry of Health and Welfare, Japan. The observed number ofdeaths from stomach cancer was compared with the expected numberof deaths, which was calculated by multiplying the patients' calendar-,

operative procedure-, sex-, and age-specific person-years at risk by sex-and age-specific mortality rates from stomach cancer in 1955, 1960,

1965,1970, and 1975 for Japan (26, 27). The O/E ratio was calculated,and the difference was tested statistically by Poisson distribution (16).

RESULTS

The number of patients by medical institution and operationtype is shown in Table 1. The total number of patients was 3827.Of these, 1488 (38.9%), 309 (8.1%), and 2030 (53.0%) wereoperated on according to Bl; Bll, B(+); and Bll, (B(-), respec

tively. The number of patients by sex and vital status as of June30, 1981, is shown in Table 2. There were 3159 (82.5%) males

2208 CANCER RESEARCH VOL. 44

Research. on February 8, 2020. © 1984 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Primary Gastric Stump Cancer

and 668 (17.5%) females. There were 2617 (68.4%) living patients as of the end of observation and 1084 (28.3%) patientshad died. There were 126 (3.3%) cases of unknown status, andthe follow-up rate was 96.7%. In Table 3, the number of patientsand person-years at risk by sex and operation type are shown.The total person-years at risk was 62,286.33 at a mean obser

vation period of 16.3 years. In this connection, the mean age atoperation was 46.4 years. The number of patients by benigngastroduodenal disease is shown in Table 4. There were 2169(56.7%) patients with gastric ulcer (including gastric and duodenal ulcers), 1040 (27.2%) patients with duodenal ulcer, 303(7.9%) patients with benign tumor (mostly stomach polyp), 239(6.2%) patients with gastritis, and 76 (2.0%) patients with otherdiseases. HistolÃ³gica! examinations of the resected stomachwere done for 2412 (63.0%) patients.

The observed and expected deaths from stomach cancer withtheir ratios by operation type, irrespective of postoperative timeinterval, are shown in Table 5. All 0/E ratios were <1, except inthe female Bll, B(+) group. The ratios for males and both sexesin Bl; Bll, B(-); and total operative groups were <1 (p < 0.01),2-tailed). The numbers of cases with time intervals from operation

to death <1 year, 1 to 4 years, 5 to 9 years, and 10 years ormore were 6, 9, 8, and 11, respectively.

In selecting PGSC cases, the postoperative time interval wastaken into account (Table 6). An elapsed time of 10 years, asproposed previously (4,7,9,18,41 ), was set not only to excluderecurrent, remaining, or multiple cancers which had been initiallymisdiagnosed or overlooked at operation but also to allow somelatency period, which is indispensable under the assumption that

Table 1

Number of patients by medical institution and operation type

InstitutionF"STNKTotalYrofoperation1950-19701950-19701948-19701963-19701957-1970Bl11351479252171488Bll,B(+)4110784707309Bll,B(-)3164645022265222030Total470108513783485463827

F, The First Department of Surgery, Faculty of Medicine, Kyushu University;S, The Second Department of Surgery, Faculty of Medicine, Kyushu University; T,The First Department of Surgery, Faculty of Medicine, Tottori University; N, Department of Surgery, National Fukuoka Central Hospital; K, Department of Surgery,Saga Municipal Koseikan Hospital.

Table 2

Number of patients by sex and vital statusVital status3 Male Female Total

AliveDeadUnknown

Total

212493897

3159

49314629

668

26171084

126

3827

a nonphysiological (pathological) environment in the stump playssome role in carcinogenesis. For comparison, the expectednumber was based on the population at risk who survived 10years or more after the operation. Although the resected stomachs of 2 cases were not histologically verified as benign, asshown in Table 9, they were included in the figure because theyhad long elapsed times (19.6 and 14.0 years). The total observedand expected deaths from PGSC were 11 and 52.85, respectively. The 0/E ratios were <1 in all sexes and operation types.The ratios for males and both sexes in Bl; Bll, B(-); and total

operative groups were <1 (p < 0.01).As shown in Table 6, PGSC mortality was not increased in the

Bll groups, regardless of Braun's anastomosis. In addition, direct

(internal) comparison of mortality rates from PGSC by operationtype and postoperative time interval was made for the patientswho died 10 years or more after the operation (Table 7). Nostatistically significant difference was noted between the deathrates from PGSC by operation type as well as by postoperativetime interval (25).

In Table 8, PGSC mortality was compared among the groupsby benign gastroduodenal disease. All O/E ratios were <1,except in females with stomach polyp. The ratios for males andboth sexes in the groups with gastric ulcer as well as duodenalulcer were <1 (p < 0.05 or 0.01). Therefore, no difference inPGSC mortalities was attributable to benign gastroduodenaldiseases that had been treated surgically.

In Table 9, details of 11 PGSC cases (9 males, 2 females)were listed. Their original diseases were 5 gastric ulcers, 3duodenal ulcers, and 3 stomach polyps. The number of casesoperated on according to Bl; Bll, B(+); and Bll, B(-) were 5, 2,

and 4, respectively. The mean age at operation was 52.2 years.Underlying causes of death in 8 cases were described simply asstomach cancer on the death certificates, whereas those of theremaining 3 cases were described as PGSC, stomach cancer inthe stump, and gastric cancer at the site of the gastrointestinalanastomosis, respectively. Autopsy was done for only one case.

DISCUSSION

In order to examine the incidence of (or mortality from) PGSCamong patients partially gastrectomized for benign gastroduodenal diseases, 2 types of epidemiological studies (retrospective

Table 4

Number of patients by benign gastroduodenal diseaseGastric ulcer8

Duodenal ulcerPolypGastritisOthers

Total

21691040303239

76

38273 As of June 30, 1981. '' Includes gastric and duodenal ulcers.

Table 3Number of patients and person-years at risk by sex and operation type

MaleOperation

typeBl

Bll, B(+)Bll,B(-)TotalNo.

Ofpatients1,179

2591,7213,159Person-yr

at

risk18,794.47

3,906.6628,927.6651,628.79FemaleNo.

ofpatients309

50309668Person-yr

atrisk4,729.89

881.635,046.0210,657.54TotalNo.

ofpatients1,488

3092,0303,827Person-yr

atrisk23,524.36

4,788.2933,973.6862,286.33

MAY 1984 2209



S. Tokudome et al.

TablesObservedand expected number of deaths from stomach cancer with their ratios by operation type

OperationtypeBlBll,

B(+)Bll,B(-)TotalObserved

no.ofdeaths11

51127MaleExpected

no. ofdeaths36.79

6.0546.7589.590/E

ratio0.30"

0.830.24a0.30aObserved

no. ofdeaths2

147FemaleExpected

no. ofdeaths4.99

0.735.3211.040/E

ratio0.40

1.370.750.63Observed

no. ofdeaths13

61534TotalExpected

no. ofdeaths41.78

6.7852.07100.630/E

ratio0.31a

0.880.29a0.34"

" Significantat p < 0.01 (2-tailed).

Table6Observedand expected number of deaths from primary gastric stump cancer with their ratios by operationtype for

patients who died 10 years or more after partial gastrectomy

OperationtypeBl

Bll, B(+)Bll, B(-)Observed

no. ofdeaths324MaleExpected

no. ofdeaths18.80

3.1325.540/E

ratio0.16a

0.640.16aObserved

no. ofdeaths2

00FemaleExpected

no. ofdeaths2.15

0.422.810/E

ratio0.93

00Observed

no. ofdeaths524TotalExpected

no. ofdeaths20.95

3.5528.350/E

ratio0.24a

0.560.14a

Total 47.47 0.19a 5.38 0.37 11 52.85 0.21a

a Significant at p < 0.01 (2-taited).

Table 7Deaf/i rates from PGSCby operation type and postoperative time interval for

patients who died 10 years or more after partial gastrectomy

BlPostopera-five time in- No. of Person-yrterval (yr) deaths at risk

Bll, B(+) or Bll, B(-)

Death No. ofrate/105 deaths

Person-yrat risk

Deathrate/105

10-19

Total (a1

8,297.841.138.50

60.260

13,714.203,424.01

21.8887.62

9,436.34 52.99 17,138.21 35.01

and prospective) (24) were carried out. In the retrospective (case-

control) study, stomach cancer cases were compared with controls in terms of previous partial gastrectomy. Stalsberg andTaksdal (42) reported that the frequency of partial gastrectomyamong autopsied gastric cancer patients was higher than thatof controls and concluded that the statistically significant difference was not explained by biased study groups. On the otherhand, Kivilaakso ef al. (17) reported no difference in the pasthistory of partial gastrectomy between autopsied cases andcontrols. Since selection bias (24, 36) is unavoidable in autopsydata, it is recommended that retrospective study using controlsand unbiased incident stomach cancer cases should be done toexamine the alleged association.

In the prospective study, patients partially gastrectomized forbenign gastroduodenal diseases were followed up, and theirincidence (or mortality) was examined. Some studies (7,12, 20)have reported increased risk of PGSC among the patients;however, others (11, 13, 14, 23, 34, 44) observed lowered (oreven) risk of PGSC. We reported that PGSC mortality was lowamong the partially gastrectomized (p < 0.01). This means thatnot only relative risk but also attributable risk (24) (the proportionof PGSC in the whole stomach cancer) is low in the patients. Nocarcinogenic effect of nonphysiological (pathological) environment in the gastric stump on PGSC was detected in the presentstudy. This evidence supports the idea that the risk of PGSC is

naturally low because the distal two-thirds of the stomach, the

most common location of the cancer, is resected. This alsocoincides with the fact that case reports of PGSC are limited inJapan (8, 41), although, as is well known, the incidence ofstomach cancer in Japan is the highest in the world.

When the resected stomach is diagnosed as benign histolog-

ically, cancer in the remaining stomach is proposed to be PGSC,irrespective of time interval from operation to death (5). On thecontrary, however, when the postoperative time interval is notlong enough and whether the resected stomach is verified asfree of cancer or not, cancer in the residual stomach does notseem to be PGSC but recurrent, remaining, or multiple (4, 7, 9,18, 41). This is partly because the state of the residual stomach,as usual, is not thoroughly verified histologically at operation. Inthis respect, time interval (elapsed time, latency period) as wellas histological confirmation always should be taken into accountin the analysis of PGSC.

It was hard to evaluate the risk of PGSC among partiallygastrectomized patients in some prospective studies (1-4, 10,21, 30-32) since the observation period was not taken into

account and the number of patients was used as denominatorwhen calculating the risk of PGSC. By using person-years at risk

as denominator, DommellÃ¶f and Janunger (7), Helsingen andHillestad (12), and Krause (20) from Scandinavian countries havereported significantly elevated incidence of (or mortality from)PGSC in partially gastrectomized patients. Hirohata (13) and Ihreef a/. (14) reported that the 0/E ratios were also >1; however,like other authors (11, 23, 34, 44), they concluded that the riskwas not increased among patients. Some possible factors toexplain the discrepancy between the reports from Scandinaviancountries and ours, such as width of resection, length of observation, and dietary factors, should be investigated in the future.

Since the expected numbers in most prospective studies werenot based on mortality rates of the remnant stomach [cardiacregion, upper part of the stomach, C region (15)] but on those




Primary Gastric Stump Cancer

TablesObserved and expected number of deaths from PGSCwith their ratios by benign gastroduodenaldisease for patients

who died W years or more after partial gastrectomy

BenigndiseaseGastriculcer8

DuodenalulcerPolypObserved

no. ofdeaths5

31MaleExpected

no. ofdeaths32.03

10.481.850/E

ratio0.166

0.29C

0.54Observed

no. ofdeaths0

02FemaleExpected

no. ofdeaths2.66

0.721.340/E

ratio0

01.49Observed

no. ofdeaths5

33TotalExpected

no. ofdeaths34.69

11.203.190/E

ratio0.146

0.27"

0.94"' Includes gastric and duodenal ulcers.6 Significant at p < 0.01 (2-tailed).c Significant at p < 0.05 (2-taited).

Table 9Details of 11 deaths from PGSC

PatientInstitution1234567891011"S,|-~-,.

.14-.S"

(0523)*S

(0536)S

(0540)S

(0713)S

(0883)T

(1554)T

(1658)T

(2216)T

(2300)T(2584)T

(2681)SexMaleMaleMaleMaleMaleMaleMaleMaleFemaleMaleFemaleBenign

diseaseGastric

ulcer(PV)Duodenal

ulcer(PV)Duodenal

ulcer(PV)Duodenal

ulcer(PUV)Polyp

(PUV)Gastric

ulcer(PV)Gastric

ulcer(PV)Gastric

ulcer(PV)Polyp

(PV)Gastriculcer(PV)Polyp

(PV)Opera

tiontypeBll,B(-)Bll,

B(-)Bll,

B(-t-)Bll,

B(-)BlBll,

B(+)Bll,

B(-)BlBlBlBlThe

Second Department of Surgery, Faculty of Medicine,Age

atoperation(yr)50.837.848.652.471.544.548.065.352.162.040.7Ageat

death(yr)66.665.162.772.085.571.973.081.969.576.053.4Kyushu

University;T,Time

interval be

tween op-eration

anddeath(yr)15.827.314.119.614.027.425.016.617.414.012.7CauseofdeathStomach

cancer0Gastric

stumpcancer*Stomach

cancer0Stomach

cancercGastric

stumpcancer0Stomach

cancer0Gastric

stumpcancer0Stomach

cancer0Stomach

cancer0Stomachcancer0Stomach

cancer0The

First Department of Surgery,Faculty of Medicine,Tottori University;PV, pathologicallyverified; PUV,pathologicallyunverified.

" Numbers in parentheses,hospital identificationnumber.0Autopsy not done.

Autopsy done.

of the total stomach, there was a pitfall in the comparison of theobserved and expected numbers. Therefore, it is recommendedthat the expected number be calculated on the basis of the rateslimited to the cardiac region. Because the proportion of mortalityrates of the cardiac region in "the stomach cancer with specifiedparts" was 20%, which was close to the 25% reported by

Saegesser and James (37), in Vital Statistics, 1975, Japan (28),the mean year at death of 11 PGSC cases, the proportion of0.20 was applied to the total expected number of 52.85. Theobtained result of 10.57 was almost equal to that in 11 of theobserved cases, which showed that mortality from stomachcancer was not increased after partial gastrectomy even whenthe region of the stomach was adjusted.

Even if PGSC mortality is low in the partially gastrectomizedas a whole, any difference in the mortality by operative procedurewould be explained by the difference in duodenogastric reflux inthe gastric stump. As mentioned before, there was neitherincreased PGSC mortality in the Bll group in which duodenogastric reflux is obligatory nor a difference in the mortality rates fromPGSC between Bl and Bll operative groups. Accordingly, nocarcinogenic effect of duodenogastric reflux was detected in the

present epidemiological study, possibly because of the limitednumber of PGSC cases. In this regard, the least significantrelative risk of 3.72 was calculated by using the population atrisk and the mortality rate of the nonexposed (Bl) group and thetype I and II error rates [e.g., a = 0.05 (2-tailed) and ÃŸ= 0.20(one-tailed)] (43). We might therefore need further observation in

this issue.Dietary habits, smoking habits, and socioeconomic factors

related to stomach cancer were not studied, and geographicalvariation of stomach cancer mortality was not adjusted in thepresent study. However, the results obtained were not considered to be explained by these factors alone.

ACKNOWLEDGMENTS

The authors wish to thank Dr. M. Nishizumi, K. Funatsu, K. HarÃ¡,Y. Kaizuka,T. Tsuji, N. Machara, and Susan M. Arnold, Saga MedicalSchool; K. ImamuraandK. Irita, Kyushu University; Dr. 0. Sasaki, Fukuoka Dental College; N. Kawasaki,Tottori University; N. Marita and K. Morooka, National Fukuoka Central Hospital;Y. Ide and H. Ohmi, Saga Municipal Koseikan Hospital; the staff of the city (town,village)office and District Legal Affairs Bureau referred; and the staff of the Healthand Welfare Statistics and InformationDepartment, Minister's Secretariat, Ministry

of Healthand Welfare, Japan, for their valuableadvice and/or technical assistance.

MAY 1984 2211



S. Tokudome et al.

REFERENCES

1. Bahr, R., RÃ¶hrte-Lehmann, S., and Geisbe, H. Statistische Untersuchungenzur Ã„tiologie des Magenstumpfkarzinoms. Med. Welt, 31: 123-127, 1980.

2. Bilton, J. L., Magno, M. G., Jr., and Kampani, S. Primary cancer in the gastricstump following gastric resection for benign peptic ulcer. Int. Surg., 56: 95-100. 1971.

3. Blikra, G. Partial gastrectomy with Braun's anastomosis in gastroduodenal

ulcer. A follow-up after 3-29 years. Acta Chir. Scand., 117: 123-136, 1959.4. Clemencon, G., Baumgartner, R., Leuthold, E., Miller, G., and Neiger, A. Das

Karzinom des operierten Magens. Dtsch. Med. Wochenschr., 707: 1015-1020, 1976.

5. CÃ´tÃ©,R., Dockerty, M. B., and Cain, J. C. Cancer of the stomach after gastricresection for peptic ulcer. Surg. Gynecol. Obstet., 707: 200-204,1958.

6. Oahm, K., Eichen, R., and Mitschke, H. Das Krebsrisiko im Resektionsmagen.Zur Bedeutung des duodenogastrischen Refluxes bei verschiedenen gastroen-teralen Anastomosen. Langenbecks Arch. Chir., 344: 71-82, 1977.

7. DomellÃ¶f, L., and Janunger, K-G. The risk for gastric carcinoma after partialgastrectomy. Am. J. Surg., 734: 581-584, 1977.

8. Feldman, F., and Seaman, W. B. Primary gastric stump cancer. Am. J.Roentgenol, 775: 257-267.1972.

9. Goldenkranz, R., and Thorbjamarson, B. Carcinoma of stomach. Followingprevious peptic ulcer surgery. NY. State J. Med., 78:733-735,1978.

10. Gnesser. G., and Schmidt, H. Statistische Erhebungen Ã¼berdie HÃ¤ufigkeitdesKarzinoms nach Magenoperation wegen eines GeschwÃ¼rsleidens. Med. Welt,35: 1836-1840,1964.

11. Hakkiluoto, A. Long-term follow-up study of patients operated on for benignpeptic ulcer. Ann. Chir. Gynaecol., 65: 361-368,1976.

12. Helsingen, N., and Hiltestad, L. Cancer development in the gastric stump afterpartial gastrectomy for ulcer. Ann. Surg., 743: 173-179, 1956.

13. Hirohata, T. Mortality from gastric cancer and other causes after medical orsurgical treatment for gastric ulcer. J. Nati. Cancer Inst., 47: 895-908,1968.

14. Ihre, B. J. E., Barr, H., and Havermark, G. Ulcer-cancer of the stomach. Afollow-up study of 473 cases of gastric ulcer. Gastroenterologia, 702: 78-91,1964.

15. Japanese Research Society for Gastric Cancer. The General Rules for theGastric Cancer Study in Surgery and Pathology, Ed. 10, pp. 2-3. Tokyo:Kanehara Shuppan, 1979.

16. Kitagawa, T. Tables of Poisson Distribution, pp. 3-148. Tokyo: Baifukan,1951.

17. Kivilaakso, E., Hakkiluoto, A., Kalima, T. V., and Sipponen, P. Relative risk ofstump cancer following partial gastrectomy. Br. J. Surg., 64: 336-338,1977.

18. Kobayashi, S., Frolla, J. C., and Kirsner, J. B. Late gastric carcinoma developing after surgery for benign conditions. Endoscopie and histologie studiesof the anastomosis and diagnostic problems. Digest. Dis., 75: 905-912,1970.

19. Koga, S., Watanabe, H., and Enjoji, M. Stornai polypoid hypertrophie gastritis.A polypoid gastric lesion at gastroenterostomy site. Cancer (Phila.), 43: 647-657, 1979.

20. Krause, U. Late prognosis after partial gastrectomy for ulcer. A follow-up studyof 361 patients operated upon from 1905 to 1933. Acta Chir. Scand., 774:341-354, 1958.

21. Kurz, L. Development of carcinoma in the gastric stump following resectionfor ulcer. Dan. Med. Bull., 6: 277-280, 1959.

22. Langhans, P., Heger, R. A., Hohenstein, J., Schlake, W., and Bunte, H.Operation-sequel carcinoma of the stomach. Experimental studies of surgicaltechniques with or without resection. World J. Surg., 5: 595-605,1981.

23. Liavaag, K. Cancer development in gastric stump after partial gastrectomy for

peptic ulcer. Ann. Surg., 755. 103-106,1962.24. Lilienfeld, A. M. Foundations of Epidemiology, pp. 164-214. New York: Oxford

University Press, 1976.25. Mantel, N. Evaluation of survival data and two new rank order statistics arising

in its consideration. Cancer Chemother. Rep., 50:163-170,1966.26. Ministry of Health and Welfare, Japan. Death Statistics of Malignant Neo

plasms: A Special Report of Vital Statistics, Japan, pp. 34-35. Tokyo: Healthand Welfare Statistics Association, 1961.

27. Ministry of Health and Welfare, Japan. Death Statistics of Malignant Neoplasms: A Special Report of Vital Statistics, Japan, p. 31. Tokyo: Health andWelfare Statistics Association, 1973.

28. Ministry of Health and Welfare, Japan. Vital Statistics, 1975, Japan, Vol. 2, pp.76-79. Tokyo: Ministry of Health and Welfare, Japan, 1977.

29. Morgenstern, L., Yamakawa, T., and Seltzer, D. Carcinoma of the gastricstump. Am. J. Surg., 725: 29-38,1973.

30. Nicholls, J. C. Carcinoma of the stomach following partial gastrectomy forbenign gastroduodenal lesions. Br. J. Surg., 67: 244-249,1974.

31. Peitsch, W., and Becker, H.-D. Frequency and prognosis of gastric stumpcancer. Front. Gastrointest. Res., 5:170-177,1979.

32. Poli, M. Die HÃ¤ufigkeit des Auftretens von Neoplasmen nach Gastroduoden-alresektionen bei Ukuskranken. Wien. Med. Wochenschr., 703: 964-966,1953.

33. Puju, S., Shuker, D. E. G., Bishop, W. W., Falchuk, K. R., Tannenbaum, S. R.,and Thiiiy. W. G. Mutagenicity of A/-nitroso bile acid conjugates in Salmonellatyphimurium and diploid human lymphoblasts. Cancer Res., 42: 2601-2604,1982.

34. Ross, A. H. M., Smith, M. A., Anderson, J. R., and Small, W.P. Late mortalityafter surgery for peptic ulcer. N. Engl. J. Med., 307: 519-522, 1982.

35. Ruddell, W. S. J., Bone, E. S., Hill, M. J., and Walters, C. L. Pathogenesis ofgastric cancer in pernicious anaemia. Lancet, 7: 521-523,1978.

36. Sackett, D. L. Bias in analytic research. J. Chronic Dis., 32: 51-63, 1979.37. Saegesser, F., and James, D. Cancer of the gastric stump after partial

gastrectomy (Billroth II principle) for ulcer. Cancer (Phila.), 29: 1150-1159,

1972.38. Schlag, P., BÃ¶ckler, R., Urich, H., Peter, M., Merkle, P., and Herfarth, C. Are

nitrite and W-nitroso compounds in gastric juice risk factors for carcinoma inthe operated stomach? Lancet, 7: 727-729,1980.

39. SchÃ¶nleben, K., Langhans, P., Schlake, W., Kautz, G., and Bunte, H. Gastricstump carcinomaâ€”carcinogenic factors and possible preventive measures.Acta Hepatogastroenterol., 26: 239-247, 1979.

40. Schrumpf, E., Serck-Hanssen, A., Stadaas, J., Aune, S., Myren, J., and Osnes,M. Mucosal changes in the gastric stump 20-25 years after partial gastrectomy. Lancet, 2: 467-469, 1977.

41. Shimazu, H., Kobori, O., Hosaka, S., Nomura, K., Tani, M., Tomiyama, J.,Kusama, S., and Miyakawa, S. A study on carcinoma of the stomach aftergastric surgery for benign diseases (in Japanese). Jpn. J. Gastroenterol. Surg.,72:713-723,1979.

42. Stalsberg, H., and Taksdal, S. Stomach cancer following gastric surgery forbenign conditions. Lancet, 2:1175-1177, 1971.

43. Walter, S. D. Determination of significant relative risks and optimal samplingprocedures in prospective and retrospective comparative studies of varioussizes. Am. J. Epidemiol., 705: 387-397,1977.

44. Westlund, K. Mortality of peptic ulcer patients. Acta Med. Scand., 774 (Suppl.402V 1-110, 1963.

45. World Health Organization. Manual of the International Statistical Classificationof Diseases, Injuries, and Causes of Death, Eighth Revision, 1965, Vol. 1, pp.41-400. Geneva: WHO, 1967.




1984;44:2208-2212. Cancer Res Shinkan Tokudome, Suminori Kono, Masato Ikeda, et al. Diseasesfollowing Partial Gastrectomy for Benign Gastroduodenal A Prospective Study on Primary Gastric Stump Cancer

Updated version

http://cancerres.aacrjournals.org/content/44/5/2208

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/44/5/2208To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



a prospective study on primary gastric stump cancer...

Documents