Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006), pp. 268–273 ( C© 2006)DOI: 10.1007/s10620-006-3123-1

Evaluation of Residual Stomach Motility After

Proximal Gastrectomy for Gastric Cancer by

Electrogastrography

TOMOHIKO HAYASHI, MD,* SHINICHI KINAMI, MD, PhD,* SACHIO FUSHIDA, MD, PhD,*TAKASHI FUJIMURA, MD, PhD,* KOICHI MIWA, MD, PhD,* and KATUMI INOUE, BS†

The relationship between the motility and the size of the residual stomach after proximal gastrectomywas evaluated using electrogastrography (EGG). Based on fast Fourier transformation, recorded slowwaves could be analyzed to obtain the following parameters: dominant frequency (DF), percentagenormal frequency (% 3 cycles per minute [cpm]), and power ratio (PR). EGG parameters, the lengthof the greater curvature of the residual stomach (LGC), were recorded in 18 gastrectomized patients.Compared to 12 healthy controls, the gastrectomized patients had abdominal EGG parameters includ-ing lower %3cpm (43 ± 21% vs 83 ± 7%; P < 0.05), DF (2.2 ± 0.4 vs 3.0 ± 0.2 cpm; P < 0.05),and PR (1.5 ± 0.8 vs 2.5 ± 0.8; P < 0.05). In relation to LGC and parameters, there was no dif-ference between the patients whose LGC was >20 cm and controls in PR (2.3 ± 0.9 vs 2.5 ± 0.8;n.s.). In conclusion, the motility of the residual stomach would be equal to that of the nonresectedstomach as if the volume of the residual stomach was more than half.

KEY WORDS: electrogastrography; gastric motility; proximal gastrectomy; gastric cancer.

Total gastrectomy has been the standard treatment formalignant gastric tumors located in the upper part ofthe stomach. In such cases, however, there is remarkabledeterioration in the quality of life after surgery. There-fore, proximal gastrectomy has been assessed as a limitedsurgery for early gastric cancer located in the upper thirdof the stomach. It has been expected that the postopera-tive nutritional status of the patients could be improvedby preserving the antrum. Detailed analysis of clinicalcases and studies of lymphatic flow in the stomach haveshown that a radical cure can be achieved by proximalgastrectomy if it is performed strictly according to thecorrect indications (1). Though not as serious as those

Manuscript received January 10, 2005; accepted May 17, 2005.From the *Department of Gastroenterological Surgery, Graduate

School of Medical Science, and †Department of Basic Science, Schoolof Health Science, Kanazawa University, Kanazawa, Japan.

Address for reprint requests: Tomohiko Hayashi, MD, Departmentof Surgery, Kamiichi General Hospital, 51, Hohonji Kamiichi-machiNakaniikawa-gun, Toyama, Japan 930-0391; toide@nifty.com.

caused by total gastrectomy, proximal gastrectomy alsocauses some postgastrectomy symptoms, such as refluxesophagitis and delayed gastric emptying (2). Thus, inorder to reduce these postgastrectomy symptoms, it is es-sential not only to select the proper reconstructive proce-dure but also to decide on a proper resectable line derivedfrom the status of the physiological function of the resid-ual stomach. However, the physiological function of theresidual stomach after proximal gastrectomy has not beenclearly evaluated in humans. As a possible cause of the de-layed gastric emptying for residual stomach, some authorsmentioned the influences of truncal vagotomy or transec-tion of the stomach in the era when gastrectomy was donefor peptic ulcer (3, 4). Moreover, there have been no re-ports on whether spontaneous contractions of the residualstomach occur, whether there is a relationship betweengastric removal volume and spontaneous contractions, orwhether there is a relationship between residual gastricvolume and electrical activity in patients after proximalgastrectomy.

268 Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006)0163-2116/06/0200-0268/0 C© 2006 Springer Science+Business Media, Inc.

RESIDUAL STOMACH MOTILITY AFTER GASTRECTOMY FOR GASTRIC CANCER

On the other hand, percutaneous electrogastrography(EGG) is steadily being improved with commerciallyavailable software/hardware packages. EGG is useful forinvestigating gastric myoelectrical abnormalities in pa-tients with nonulcer dyspepsia and autonomic neuropa-thy (5). Though some reports have analyzed percutaneousEGG after gastrectomy, there have been no reports afterproximal gastrectomy (6).

In this study, we evaluate the motility of the residualstomach after proximal gastrectomy from an electrophysi-ological viewpoint using percutaneous EGG, assess the re-lationship between the motility and the size of the residualstomach, and discuss the recommendable volume of theresidual distal stomach following proximal gastrectomy.

MATERIALS AND METHODS

Patients and Surgical Procedure. A total of 18 patientswho were diagnosed with gastric tumor (16 gastric cancers, 1leiomyosarcoma, and 1 lymphangioma) localized to the upperpart of the stomach underwent proximal gastrectomy with dou-ble tract reconstruction (PG group) at our department betweenJanuary 1995 and December 2001. They consisted of 14 menand 4 women aged 37 to 79 years (average, 59 years). The con-trol group (N group) consisted of 12 healthy volunteers (7 men,5 women) without any gastrointestinal symptoms or digestivediseases, aged 31 to 70 years (average, 57 years). The researchwas approved by the ethical committee of our hospital and writ-ten informed consent was obtained from all patients before thebeginning of the study.

All patients underwent truncal vagotomy with regional lymphnode dissection, and the sizes of the residual stomach dependedon the resectable lines decided on according to the size and lo-cation of the tumor (Figure 1). Following proximal gastrectomy,a selected jejunal portion with its stem was elevated at the retro-colic portion, and an esophagojejunostomy was carried out usingan automatic anastomosis device (Premium Plus CEEA25; AutoSuture Japan, Tokyo). Gastrojejunostomy was performed usinglayer-to-layer anastomosis with a continuous suture of 4-0 ab-sorbable thread and intermittent sutures of 4-0 silk about 10 cmdistal from this anastomosis. Subsequently, the jejunal segmentswere joined by side-to-end anastomosis and the reconstructionwas completed. To estimate the residual stomach size, we mea-sured the lengths of the lesser and greater curvatures from thegastrojejunostomy to the pylorus during the surgery.

Electrogastrography Recording. The EGG was recorded atleast 1 year after surgery. The study was commenced at about0800 hr, after an overnight fast. All medication that could af-fect gastric function was discontinued at least 48 hr before therecording. In order to diminish the impedance between elec-trodes (silver-silver chloride) (Magnerode; Fukuda Denshi Co.,Tokyo), upper abdominal skin was cleaned and gently scrubbeduntil the skin reddened. An active electrode and a reference elec-trode coated with electrode jelly were placed on the sites of theresidual gastric pylorus and anastomosis of gastrojejunostomy,respectively. For correct recording of the EGG, it is essential toascertain the location of the residual stomach. The locations ofthe anastomosis of the gastrojejunostomy and the pylorus of theresidual stomach were identified using a barium meal study, and

Fig 1. Schematic presentation of double tract reconstruction after prox-

imal gastrectomy. After proximal gastrectomy, esophagojejunostomy is

carried out by end-to-end anastomosis using an automatic anastomosis

device and gstrojejunostomy is performed by side-to-end anastomosis.

The distance between the two anastomosis is 10-cm.

their positions were marked on the abdominal skin on the daybefore EGG. The electrodes were connected to a portable EGGrecorder (NIPRO EG; NIPRO, Osaka, Japan). Fasting activitywas recorded for 30 min with the patient in a sitting position.Then the patient consumed the entire test meal (400 kcal/250 ml)(Terumeal mini; Terumo, Tokyo) within 5 min. After taking themeal, the patient resumed the same position, and continuousEEG recording was carried out for 1 hr. The recording was madein a quiet environment, and the subject was asked not to talkand to remain as still as possible during the recording to avoidmotion artifacts.

We used a filter setting with a high cutoff of 8.6 cycles perminute (cpm) and a time constant of 1.0 sec in order to filterout unnecessary noises such as heartbeat, respiration, and bodymovement artifacts. The EGG data stored on the recorder weredownloaded to a computer and analyzed with exclusive com-puter software (EGS 1; Gram Co., Saitama, Japan). The datawere obtained by running spectral analysis. Using a fast Fouriertransformation (FFT) algorithm of a 512-sec “window” of theraw data, power spectra of overlapping stretches for 384 seg-ments between the subsequent stretches of electric signal werecomputed and displayed as a functional time, yielding frequencyand amplitude information over the course of study. The follow-ing parameters, derived from the computer analysis of the EGGsignal, were determined for each EGG.

1. Dominant slow wave frequency (DF) in both the fastingand the postprandial periods: This was calculated by aver-aging the FFT results (power as a function of frequency)for running spectra during the fasting or postprandial pe-riod, respectively. The DF represents the frequency withthe highest power.

Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006) 269

HAYASHI ET AL.

TABLE 1. COMPARISON OF ELECTROGASTROGRAPHY (EGG) PARAMETERS IN

CONTROL AND DOUBLE TRACT RECONSTRUCTION AFTER PROXIMAL

GASTRECTOMY

EGG parameter PG group (n = 18) N group (n = 12)

%3cpmFasting 30 ± 20.48* 76.25 ± 10.23Postprandial 42.78 ± 20.96* 83.33 ± 6.87

Dominant frequency (DF; cpm)Fasting 1.94 ± 0.46* 2.91 ± 0.19Postprandial 2.16 ± 0.4* 3.04 ± 0.21

Frequency ratio (FR) 1.14 ± 0.21 1.08 ± 0.1Power ratio (PR) 1.53 ± 0.83* 2.47 ± 0.79

Note. N group, control; PG group, proximal gastrectomy with double tract recon-struction. Results are mean ± SD.*P < 0.05.

2. Frequency ratio (FR), i.e., the ratio between the postpran-dial and the fasting DFs.

3. Percentage of normal slow wave (%3cpm) in both the fast-ing and the postprandial periods: The normal frequencywas in the range of 2.4–3.6 cpm.

4. Power ratio (PR), the ratio of postprandial to fasting pow-ers of the DF: The absolute value of the electrical power(amplitude) is influenced by many factors (e.g., skin resis-tance, electrode distance from the stomach wall), and thusonly the PR was analyzed.

Statistical Analysis. Results are expressed as the mean ±SD. Numerical data were analyzed using the Mann-Whitney test.Linear regression was used to study the correlation coefficient oftwo variables. Patients were divided into two groups at the lowestP value in the Mann-Whitney test. Both groups were testedby regression analysis. Values of P < 0.05 were considered toindicate statistical significance.

RESULTS

According to the criteria of the Japanese Research Soci-ety for Gastric Cancer of 16 gastric cancer patients, therewere 8 with stage IA, 5 with stage IB, 1 with stage II,1 with stage IIIA , and 1 with stage IV disease. With

TABLE 2. THE RELATIONSHIP BETWEEN THE LENGTHS OF THE LESSER AND GREATER

CURVATURES AND THE ELECTROGASTROGRAPHY (EGG) PARAMETERS IN DOUBLE TRACT

RECONSTRUCTION AFTER PROXIMAL GASTRECTOMY

Greater curvature Lesser curvature

Lowest LowestCorrelation P-value Correlation P-value

EGG parameter coefficient point (cm) coefficient point (cm)

%3cpmFasting 0.772* 13 0.722* 7Postprandial 0.723* 10 0.615* 11

Dominant frequency (DF; cpm)Fasting 0.675* 18 0.698* 8Postprandial 0.752* 13 0.663* 11

Frequency ratio (FR) No correlation No correlationPower ratio (PR) 0.664* 20 0.607* 11

*P < 0.05.

respect to pathological types, there were nine with well-differentiated, five with moderately differentiated, and twowith poorly differentiated cancers (7). None of the patientsshowed evidence of recurrence.

Comparison of EGG parameters between the PG groupand the N group is shown in Table 1. DF revealed a sig-nificant difference between the PG group and the N groupin both the fasting period and the postprandial period. The%3cpm in the PG group was significantly lower than thatin the N group in both the fasting state and the postpran-dial state. PR was significantly lower in the PG group thanin the N group. There was no significant difference in FRbetween the two groups.

The relationship between the lengths of the lesser andgreater curvatures and the EGG parameters were also in-vestigated (Table 2). First, we analyzed the correlations ofthe length of the greater curvature from the pylorus to theanastomosis of the gastrojejunostomy and EGG parame-ters. Significant positive correlations were found betweenthe length of the greater curvature and the EGG parame-ters except for FR (Table 2). These results indicated thatthe size of the residual stomach was positively correlated

270 Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006)

RESIDUAL STOMACH MOTILITY AFTER GASTRECTOMY FOR GASTRIC CANCER

with the stomach motility. To evaluate the relationship be-tween the length of the greater curvature and the EGGparameters, the lowest P-value point was determined, atwhich the data were divided into two groups, a large resid-ual stomach and a small residual stomach group. For DF,the point was 18 cm (<18 cm, small group; ≥18 cm,large group) in the fasting state and 13 cm (<13 cm, smallgroup; ≥13 cm, large group) in the postprandial state. For%3cpm, it was 13 cm (<13 cm, small group; ≥13 cm,large group) in the fasting state and 10 cm (<10 cm, smallgroup; ≥10 cm, large group) in the postprandial state.For PR, it was 20 cm (<20 cm, small group; ≥20 cm,large group). Testing differences in the regression linesbetween the two groups divided at the lowest P-valuepoint, we found a significant difference (P < 0.05) onlyin the relationship between the length of the greater curva-ture and the PR. The regression lines of the <20-cm groupand the ≥20-cm group were y = 0.0078x + 0.9579, andy = −0.1876x + 6.4232, respectively (Figure 2).

In a similar way, the correlations between the length ofthe lesser curvature from the pylorus to the anastomosis ofthe gastrojejunostomy and each EGG parameter were sta-tistically analyzed. Significant positive correlations werefound between the length of the lesser curvature and theEGG parameters except for FR (Table 2). However, wefound no significant difference in the relationship between

Fig 2. The relationship between the power ratio (PR) and the length

of the residual gastric greater curvature from the pylorus. A significant

positive correlation was found between the length of the greater curvature

and the PR in the fasting state (r = 0.664, P < 0.05). The lowest P-

value point was 20 cm of the length of the gastric greater curvature. The

regression lines of the <20-cm group and the ≥20-cm group were y =0.0078x + 0.9579 (a), and y = −0.1876x + 6.4232 (b), respectively.

The regression lines of the two groups were significantly different (P <

0.05), indicating that the two groups were regarded as the statistically

different groups.

the length of the lesser curvature and any of the parametersby testing differences in the regression lines between thetwo groups divided at the lowest P-value point.

Therefore, we thought the most effective point of di-vision was 20 cm for the greater curvature of the lengthfrom the pylorus to the anastomosis, since a significantdifference was found only in the relationship between thelength of the greater curvature and the PR by comparingthe regression lines between the two groups divided atthe lowest P-value point. Thus, each EGG parameter wascompared between the <20-cm group and the N group.The results showed that the values of all parameters weresignificantly lower in the <20-cm group than in the Ngroup, except for FR (P = 0.147). When each parame-ter was compared between the ≥20-cm group and the Ngroup in a similar way, there was no significant differencebetween the two groups in PR (P = 0.554).

DISCUSSION

Recently, EGG has been applied to the assessment ofmotility in the residual stomach after distal gastrectomy(6), and it has been one of the most appropriate examina-tions for evaluating the motility of the residual stomach(8). In the present study, we investigated the motility ofthe residual stomach after proximal gastrectomy from anelectrophysiological standpoint using EGG.

The frequency of 3 cpm disappears in patients whohave undergone a total gastrectomy, and the meanpostoperative-to-preoperative PR of the 3-cpm peak hasbeen shown to decrease significantly, to that of the preoper-ative control (6, 9). The postoperative-to-preoperative PRfor the 3-cpm peak was significantly reduced followingsubtotal gastrectomy in patients in the postprandial state(6). In a study of the relationship among EGG, electro-gastromyogram, and mechanical activity in canine mod-els, the regular frequency about 3 cpm disappeared, and asignificant correlation among the amplitude of EGG, theamplitude of the electrogastromyogram, and the mechan-ical activity was seen after subtotal gastrectomy (10). Inshort, the frequency of 3 cpm showed gastric electrical ac-tivity in the EGG, and the amplitude of the EGG showedmechanical activity of the gastric antrum in the electrogas-tromyogram. For accurate estimation of the slow wave byEGG, we selected patients with at least 1 year since prox-imal gastrectomy who showed a small influence of theoperation (11, 12). The electrodes should be placed alongthe longitudinal axis of the residual stomach in order tomeasure the difference in electrical potential. Therefore,we placed the electrodes using a barium meal test, be-cause the optimum electrode positions varied accordingto residual gastric size and shape (13, 14).

Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006) 271

HAYASHI ET AL.

In proximal gastrectomy as a curative procedure againstcancer, truncal vagotomy with regional lymph node dis-section and removal of the pacemaker zone located onthe upper one-third of the greater curvature were con-ducted. As a result, it is doubtful whether spontaneouscontractions of the residual stomach after proximal gas-trectomy occurred. To answer this question, two experi-mental studies were performed on the appearance of thepacemaker in dogs. One study, in which the stomach afterthe transection-anastomosis with truncal vagotomy wasevaluated by electromyogram, showed that a new pace-maker zone appeared at the most oral region of transectionand its discharge rate was slower than normal (3). In theother study, Hoshikawa et al. showed that, after proximalgastrectomy, the amplitude of action potentials was gen-erally low, resting phases appeared occasionally, and thecontrol activity-like waves appeared preprandially, but theinterval and duration of these waves were irregular and nosynchronized electrical activity was transmitted betweenthe residual stomach and the duodenum (11). According toboth studies, electrophysiological activity would exist inthe residual stomach without the original pacemaker zoneafter proximal gastrectomy. The present study is the firstto assess this question in humans. Our results revealedthat slow wave signals in both the fasting and the post-prandial period existed in the residual distal stomach ofhumans, though a truncal vagotomy was conducted andthe strongest and most regular pacemaker zone was re-moved by proximal gastrectomy. However, most EGG pa-rameters were weak compared with those of the normalsubjects.

On the other hand, the volume of the residual distalstomach was previously thought to play an important rolein motility of the residual stomach based on empiricalresults. Taihei et al. observed that residual gastric motil-ity was closely related to the volume of resection ratherthan vagotomy in an animal model. In their experiment,food discharge to the duodenum was promoted in a largeremnant stomach (15). In the present study, significantpositive correlations were observed between the lengthof the lesser or greater curvature and many EGG param-eters, such as DF, %3cpm, and PR. We demonstratedthat the EGG parameters of the residual stomach corre-lated with the size of the residual stomach in humans,and that patients with a larger volume of residual distalstomach following proximal gastrectomy showed a betterdischarge ability. What, then, is the recommended vol-ume of the residual distal stomach following proximalgastrectomy? Among the EGG parameters, we thoughtPR was the most important, since the amplitude of thePR on EGG was shown to represent the mechanical ac-tivity of the gastric antrum (10). We were able to divide

our gastrectomized patients into two groups according tothe relationship between the length of the residual greatercurvature and the PR. As the lowest P-value point was20 cm of the length of the greater curvature, the twogroups were divided at this point. Moreover, the regressionlines of the <20-cm group and the ≥20-cm group werey = 0.0078x + 0.9579, and y = −0.1876x + 6.4232, re-spectively. The regression lines of the two groups weresignificantly different (P < 0.05), indicating that the twogroups were regarded as the statistically different groups.Moreover, the value of PR in the ≥20-cm group was equalto that of the normal controls. Thus, in such cases the con-tractivity was expected to be equal to that of the normalstomach from an electrophysiological standpoint, and therecommended volume of the residual stomach would belonger than 20 cm of the length of the greater curvature.In adult Japanese, the normal lengths of the greater cur-vature and the lesser curvature are about 45 and 15 cm,respectively (16). If the length of the greater curvature waspreserved longer than 20 cm, there was a possibility thatthe residual distal stomach was more than half.

In conclusion, we have found that the motility of theresidual stomach following proximal gastrectomy has al-most the same power as the nonresected stomach if theresidual distal stomach is more than half. Conversely, inthe case that the size of the residual distal stomach is lessthan half of the original stomach, some drainage proce-dures, such as pyloroplasty or securing of an escape route,may be necessary to avoid the possibility of delayed gas-tric emptying.

REFERENCES

1. Kitamura K, Nishida S, Yamamoto K: Lymph node metastasis in

gastric cancer in upper third of the stomach surgical treatment on the

anatomical distribution of positive node. Hepatogastroenterology

45:281–286, 1988

2. Buhl K, Schlag P, Herfarth C: Quality of life and functional results

following different types of resection for gastric carcinoma. Eur J

Surg Oncol 16:404–409, 1990

3. Shiratori T, Kuroda S, Sugawara K, Hatafuku T: Effect of vago-

tomy or spanchnicectomy on the canine stomach with transection-

anastomosis. Tohoku J Exp Med 93:317–329, 1967

4. Kelly KA, Code CF: Effect of transthoracic vagotomy on canine

gastric electrical activity. Gastroenterology 57:51–58, 1969

5. Camilleri M, Hasler WL, Parkman HP, Quigley EMM, Soffer E:

Measurement of gastrointestinal motility in the GI laboratory. Gas-

troenterology 115:747–762, 1998

6. Homma S, Shimakage N, Yagi M, Hasegawa J, Sato K, Matsuo H,

Tamiya Y, Tanaka O, Muto T, Hatakeyama K: Electrogastrography

prior to and following total gastrectomy, subtotal gastrectomy, and

gastric tube formation. Dig Dis Sci 40:893–900, 1995

7. Japanese Gastric Cancer Association: Japanese Classification of

Gastric Carcinoma, 13th ed. JGCA, 1999

8. Koch KL, Stern RM: Handbook of Electrogastrography. New York,

Oxford University Press, 2004

272 Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006)

RESIDUAL STOMACH MOTILITY AFTER GASTRECTOMY FOR GASTRIC CANCER

9. Pezzolla F, Riezzo M, Maselli A, Giorgio I: Electrical activity

recorded from abdominal surface after gastrectomy or colectomy

in humans. Gastroenterology 97:313–320, 1989

10. Kajimoto T: Relation between electrogastrography and gastric elec-

tromyogram, mechanical activity. J Smooth Muscle Res 31:93–107,

1995 (in Japanese)

11. Hoshikawa T, Denno R, Yamaguchi K, Ura H, Hirata K: Chronic

outcome of proximal gastrectomy with jejunal pouch interposition

in dogs. J Surg Res 112:122–130, 2003

12. Geldof H, Schee VD, Blankenstein MV, Smout AJPM, Akkermans

LMA: Effect of highly selective vagotomy on gastric myoelectrical

activity. Dig Dis Sci 35:969–975, 1990

13. Patterson M, Rintala R, Lloyd D, Abernethy L, Houghton D,

Williams J: Validation of Electrode placement in neonatal electro-

gastrography. Dig Dis Sci 46:2245–2249, 2001

14. Mirizzi N, Scafoglieriu U: Optimum direction of the electo-gastro-

graphic signal in man. Med Biol Eng Comput 21:385–389, 1983

15. Taihei Y, Hanyu N, Furukawa Y, Abe S, Nakata K, Aoki T:

Residual gastric motility and gastrointestinal hormone release af-

ter proximal gastrectomy. J Smooth Muscle Res 23:225–228, 1987

(in Japanese)

16. Sakurai A, Ishizuka T: Statistical research of estimation of the parts

in healthy Japanese stomach. Nippon ihou kaisi 18:1778–1781, 1959

(in Japanese)

Digestive Diseases and Sciences, Vol. 51, No. 2 (February 2006) 273