A Rapid Screening Test for Gastric Achlorhydria CHARLESB. BEAt., *I.D., and JAMEs E. BROWN, M.D.

N aN Evvorcr to find a more satisfactory screening te~t [or the ab~nce of acid

in the stomach, the authors observed that a thread could be partially Ienclosed in an ordinary pharmaceutical capsule which, if swallowed, would easily deliver one end of tile thread to the stomach while the other end was being held in the mouth. Upon retrieval of the thread, an attached pH indicator or other sampling device could also be retrieved. Our first purpose was to develop the method; our second, to determine its acceptability and freedom from undesired effects; and our third, to test its reliability in detecting the absence of gastric acidity. MATERIALS AND METHODS Description o[ the Device

The device (Fig. 1A) consists of a gelatin pharmaceutical capsule into which is coiled a length of white cotton thread partially impregnated with Congo-red dye (sodium diphenyldiazo-bis-~-naphthylamine-sulfonate). One end of the thread protrudes through a hole in the capsule. This end is held while tile capsule is swallowed with a little tepid water; the thread unwinds as the capsule travels down into the stomach. The patient reclines on the examining table for 1 min., by which time thecapsule has softened and fallen away. The thread then can be withdrawn from the stomach with minimal discomfort. The impregnated thread is crimson at pH 7 and dark red at pH 5; below this pH, the color changes to purple and then to black (Fig. 1B). Thus, when the thread is retrieved tile portion that was in the esophagus remains red, while that which was within the stomach normally darkens (Fig. IC). The color change is usually not uniform. The thread frequently shows black spotting, presumably were it touched acid-secreting areas of tile stomach wall or the gastric pool. Since it appeared that varying degrees of acidity could exist in different parts of the stomach at any one time, we did not attempt to assign a specific pH to a thread showing a variable color change. Instead, we used From the Department of Preventive Medicine, Stanford Medical School, Palo Alto, Calif. Supported in part by a grant from the Wm. S. Merrell Co.. Cincinnati, Ohio, and by a student fellowship supplied by the Northern California Division of the Americatl Cancer Society. We are grateful to the many students and staff of the Palo Alto-Stanford I-fospital Center who cooperated to make these studies possible, particularly Mi~ Marie Ha~neros and Mrs. Jean Lynch of the Employee Health Service and Dr. Keith B. Taylor. Professor of Medicine, and his staff in the Gastrointestinal Laboratory.

Neqv Series, Vol. 13, No. 2, 1968

113

B.a1 & Brown

the color change from red to purple (pH 4---5) on any part of the thread as an indication of the presence of acid in tile stomach. Correlation with gastric analysis was on this basis.

Ez,aluation of the Technic T o determine the accuracy and acceptability of this technic, tests were administered to 500 medical students, patients, and Palo Alto-Stanford Medical Center employees. Student Group. T h i r t y healthy, young, fasting medical students (25 men and 5 women) were tested in their class laboratory; they measured their own secretions by gastric aspiration or by azuresin-blue studies. The thread test was administered to each student at the beginning of the experiment. Some then performed a gastric aspiration while others collected their urine for colorimetr3/after taking the azuresin-blue dye. Patient Groilp. One hundred patients, either hospitalized or outpatients, were given the test capsule just before intubation for gastric analysis. After the tube was removed, the test was again administered, except in those individuals in whom gastric washing was performed for cytologic studies. T h e mean age of this group of 54 men and 46 women was 51 years. Employee Group. In an effort to determine whether the test could readily be administered to large groups of apparently healthy individuals, 370 employees were given the thread test at the termination of their routine physical examinations, irrespective of prior food eaten. The employee's physical tolerance to the test was observed, as was the color change of the thread. If the thread remained red, the employee was given a cup of black coffee (which has a neutral pH) and the test was repeated in S0 min. Some were tested after being given a S00-mg. capsule of caffeine. Six submitted to gastric analysis. Most of the employees were middle-aged, the mean being 44 years, although the range was from 18 to 72 years. There were 61 men and 309 women in the group. RESULTS STUDENT GROUP T h e acidity range for all the students was normal, as tested by standard technics. The thread test was in a~eement with the determined pH in all but 1 student, in whom the thread showed no color change although the aspirate was acid (Tables 1 and 2). In 1 of 2 subsequent trials on this person, the thread did show a definite change to black. This individual had a very sensitive gag reflex and during these tests produced great amounts of mucus, which possibly interfered with access of acid to the thread. Furthermore, this test was performed with the student in the sitting position. (This was before we learned that results are often more definite if the patient lies down after swallowing the capsule.) ||4

Amer;can dourn41 of D;cjest;v, D;seases

@sstric Achlorhydrla TABLE l. T H R E A D - T E S T AND GASTRIC-ANALYSIS CORRELATIONS

No. tested by both procedures

Group Students Employees

Good correlations

30 •

6

Patients

TOTALS

Questionable correlation o1" ]ailure of correlation

29

1

5

3

I00

95

5

136

127

9

*Seven were given azuresin test only. TABLE 2. ANALYSIS OF FAILURES OF CORREI.ATIONS OF T H R E A D TESTS W I T H GASTRIC ANALYSIS

Group

Individual

Student

B.T.

Thread remained red; pH of gastric aspirate, 1.5.

Nausea, with much mucus production; repeated thread test demonstrated presence of acid

Patient

E.H.,L.K.

Thread minimally sponed; achlorhvdria (lowest pH 6.7 and 7.0).

Undetermined; possibly error of observation of thread

L.L.,M.M., M.B.

Thread remained red; pH measured 2.7, 2.4, and 3.0.

Undetermined; repetition of the thread test would have been appropriate

A.P.,C.C.

Thread minimally spotted; gastric analysis normal.

Coffee inadequate stimulant; thread turned black after capsule containing 300.mg. of caffeine was taken

R.B.

Thread minimally spotted; hyperchlorhydria found on gastric analysis.

Same; however, this result is surprising

Employee

Discrepancy

Probable explanation

PATIENT GROUP

In 95 of the 100 patients on whom gastric analysis was performed, the thread test agreed with the gastric analysis (Tables 1-3). In some instances the thread was black, although tile subsequent basal pH measurement (made during the gastric analysis) revealed a pH above 5. After administration of histamine, however, the aspirate became acid. (See patient V.T. in Table 3.) This result was consklered a corroboration of the thread test rather than a discrepancy. Some difficulty of interpretation did exist in 2 patients, in whose threads slight, black spotting occurred. In these, the measured pH of the aspirate remained in the 6--7 range, and the patients were considered achlorhydric. In Nmv Series, Vol. 13. No. 2, 1968

||5

M

F

M

F

T.A.

M.C.

R.S.

V.T.

39

55

45

65

¢,a.Slric uh'cr

(ia~llic ulcer Blatk

spot I lllg

Black

spols i(ll!

ph)sical

t*~illllil|i|l

I or 2 purple

Red

Thread color belore +maly~is

7,2

3.5

7.6

7,2

phi

UORRI.'LATIONS

t;astric carcinoma Jlcalihy; I'olltinc

eFil+~! of '2 aliquols. tl.~twt-sl rt.adhlg of 4 aliqtlOlS.

Sex

Patient

.+lge (yr.)

i'Jc+~etlti~g +ymlJlom or diagt~osis

'i'AIII.E ~1. R E P R E S E N T A T I V E

190

73

37

27

Vol. (ml.)

Basal

IIETWEEN

0,75

0.77

0.0

0.07

Total (mEq.)

1,8

I,fi

5,1

6.9

~!12

I(i3

50

73

8.~,2

1+1.8

0,23

0.16

Port,lira fdaticm ....................... IPol. Tolal tJHt (ml.) (mEq.)

l|la(k

i|la~ k

Notdone

Not d o n e

Thread color after atJaly~is

"I'ES+!" ANI) ( ; A S T R I C ANALYSIS

G+,~tric tmalysis

TIIREAD

atidily

gi|M t i(: at hltty Nol,nal

Notm~d

(Initl kt.d)

Achlof hydria Ilypc)chhJrhydria

Comment

B

oN 7.0. 5.0 4.4

L

1.5

Fig. 1. A. Gelatin capsule containing Congo-red-stained thread, with unstained end protruding from top. B. Approximate colors of same thread at various pH measurements. C. Drawing indicates appearance of thread used to test person with normal gastric acidity. Portion of thread retained in m o u t h is not impregnated; portion from esophagus remains bright red, while that from stomach is ~potted.

G,stri¢ Achlorhydr;a

3 other patien.ts, the thread remained red even .though the aspirate was pH 3 or less, These 3 patients were examined early in our study, a n d were not retested after gastric stimulation. EMPLOYEE GROUP

In tile 370 employees tested, no untoward effects, other titan occasional gagging and a single episode of vomiting, were observed. If the thread remained in place longer than 1 min., the capsule sometimes softcned too mucll and did not separate from the thread. In such cases, g a g g i n g u s u a l l y o c c u r r c d when the thread was pulled up. Of this group, 55 persons demonstrated no color change or only a minimal one when their threads were examined (Table 4). T h e color change was considered minimal-or questionable when it consisted of only 1 or 2 black spots a millimeter or so in length on an otherwise totally red thread. Midway in our study it became evident that minimal black spotting of the thread could be consistent with hypochlorhydria, even though the color of the small spot indicated a low pH. Until we realized this fact, wc did not consider it necessary to retest after stimulation of patients showing this reaction; thus, 33 of the employees demonstrating minimal spotting had no further studies. Twenty-two were retested after drinking black coffee, and a definite color change was seen in 15. T h e remaining 7, who showed little or no color change after the coffee, were judged appropriate candidates for gastric analysis, which was performed in 6 of them (Tables 1 and 2). Of these 6, 3 were achlorhydric or nearly so, 2 were normal, and 1 was hyperchlorhydric. In the la,tter 3 a definite black coloration was seen when a 300-mg. capsule of caffeine and sodium benzoate was given, instead of coffee, before repetition of the thread test. Thus, after adequate caffeine stimialation, 3 persons still showed no evidence oE acid on the thread test. In a~'l 3, marked hypochlorhydria or achlorhydria was demonstrated by gastric analysis. It is evident that in about 85c7o of the employees the presence of gastric acid TABLE 4. T H R E A D TESTS OF EMI'LOYEES D t ' R I N G R O U T I N E PHYSICAL E X A M I N A T I O N

Groups A B

C D E

F G

Totals No. of employees tested without regard to food eaten No. from Group A showing definite black coloration of thread No. from Group A showi,~g unchanged or only minimally spotted thread No. from Group C who were retested after drinking black coffee No. from Group D who still showecl unchanged or minimally changed thread after drinking coffee No. from Group E having gastric analysis with histamine stimulation No. from Group F showing severe hypochlorhvdria or achlorhvdria

New Series, Vol. 13, No. 2, 1968

370 315 55 22 7 6 3

119

B**I & Brown

was demonstrated by the thread test, regardless of prior preparation of die stomach. Fifteen per cent needed additional stimulation, black coffee usually being sufficient. Probably, less than 5 ~,o require a stronger stimulus to respond. Two per cent or less do not respond, as confirmed by gastric analysis. COMMENT The thread test indicates whether an)" of the material with which it comes in contact within the stomach has a pH of 5 or less. If the thread is spotted red and black upon retrieval, material o f yawing de~ees of acidity has come in contact with it, and there can be no precise agxeement between the color of the thread and the pH of the aspirated gastric pool. However, on the basis that the thread indicated whether acid can be produced by the stomach, correlation with standard tests is good. If we eliminate from consideration those patients who did not have repeat tests after fnitial doubtful results, agreement seems excellent. T h e normal fasting stomach produces very little acid; thus adequate stimulation of secretion is often necessary before hypochlorhydria or achlorhydria can be ruled out. When the thread test is used, the administration of black coffee is usually sufficient. A 300- to 500-mg. capsule of caffeine is necessa~" in some cases, and perhaps should be given on a routine basis to minimize the occurrences of doubtful results. We believe there is no contraindication to the performance of this test. Congo-red dye has been widely used in medicine and often given intravenously: it has no known carcinogenic effect and is nontoxic except in very high doses. 1 T h e only persons on whom t h e t h r e a d test could not be accomplished were a very few who stated that the,v were unable to swallow any type of capsule. The thread itself produced surprisingly little irrkation or nausea. In all of our tests, only 1 patient vomited. We have considered using substances other than Congo red. An indicator with a broader range might have some advantage. However, the criteria for both hyperacidity and hypoacidity depend not only on pH but also on the total amount of acid produced over a period of time.2 Thus, we feel that a single measurement, no matter how accurate the indicator, migltt not give a reliable indication of h,vperchlorhydria or hypochlorhydria. On the other hand, if after appropriate stimulation the thread is unchanged or only minimally changed, the patient is likely to be found achlorhydric or hypochlorhydric by standard gastric analysis. In determining the indications for the u ~ of such a screening test, both the pathologic significance and the approximate distribution of achlorhydria in the population should be appreciated. Ahnost all patients with pernicious anemia, and many individuals with gastric malignancies, are achlorhydric. Apparently the factor common to both of these conditions is atrophic gastritis, s'-s It was once thought that achlorhydria was present in 15 to 20% of |20

AmoHcan Journ,! of D;qesfive D;sease$

Glstrlc Achlorhydria

the U.S. population over 40 years of age. ~ Based on a modern definition of achlorhydria (failure of gastric pH to fall below 6 after a given stimulus, 7 the incidence is undoubtedly much lower. Probably less than 2% of our employee group over 40 years of age was truly achlorhydric. Tile fact that achlorhydria is now seen to be limited to a small percentage of the population increases its significance as a finding. It is this group whose stomachs merit repeated careful examination, even when the individuals are asymptomatic, particularly since early diagnosis of gastric malignancy may favorably atlect the salvage rate. s The well-known "tubeless" gastric analysis" can demonstrate hyperchlorhydria, whereas the Congo-red thread test cannot. However, since neither test can clearly differentiate hypochlorhydria from true achlorhydria, both are useful primarily as screening tests. T h e fornter requires laboratory analysis, whereas the thread test does not. T h e method in which a tiny radio transmitted is swallowed *° is said to give a quantitative measure of gastric acklity, b u t both cost and time requirements have restricted its use somewhat. A normal stomach, with proper stimulation, should be able to produce enough acid to lower the pH of its contents to 3.5 or less. A hypersecreting stomach produces not only a low pH but also an abnormally large volume of acid, best measured by repeated aspiration of gastric contents over a period of ,time. On the other hand, suspicion of an inadequately secreting stomach may be aroused by a single p H measurement. Such a determination is very simply made by means of the "thread test." Final confirmation, of course, is dependent on a properly performed gastric analysis. SUMMARY A device was developed for delivering a pH indicator to the stomach and retrieving it without discomfort to the patient. T h r e a d impregnated with Congo-red d y e was encased in a gelatin capsule, with one end protruding through a hole. While the end of the thread is held the capsule is swallowed; as the capsule descends the thread unwinds. T h e thread turns purple or black at a pH below 5; above ,this p H the thread turns various shades of red. After 1 rain. the thread can be withdrawn. T h e technic was tested in 500 subjects, In 127 of 136 subjects in whom gastric analysis was also done the agreement between the thread test and the determined pH was good; in the other 9, either results could not be correlated or correlation was questionable. Kaiser Foundation International JO0 Lakeside Dr. Oakland, Calif. 94604

REFERENCES 1. SW~LmOFF,I. J., and BEL'~SaXlN,I. J. Systemic reaction to the intravenous administration of Congo red. Quart Bull Sea View Hosp 8:!31, 1946. New Series, Vol. 13, No. 2, 1968

|2|

Seal & Brown

2- R o ~ r . ~ , R. A. Progress in gastroenterology. Gastric analysis. Gastroenterology 45.'90, 1965. 3. Sit'itALY, M.. Er-~xtA.~. E.. and N~BFRG, W. Pernicious anemia and atrophic gastritis. Acta Med Scand 166:213, 1960. 4. SIi;~,~L~, M., and SEt'PaL~, K. Atrophic gastritis as a possible precursor of gastric carcinoma and pernicious anemia. Acta Med Scand 166:455, 1960. 5. Sit.'L,,t.~, M.. Vt.'OgLn~n, Y., and SePPXt.A, K. Follow-up studies of patients with atrophic gastritis. ,4cta .~led Scand 170:131, 1961. 6. CostroRr, M. W. Gastric acidity before and after development of gastric cancer: Its etiologic, diagnostic and prognostic significance. Ann l~ttern Med 3t:1331, 1951. 7. C_,~I~,W. I.. and Strcus. W. "Auaciditv." In Modern Trends in Gastroenterology (Series 2), Jones, F. A., Ed. Hoeber, New York, 19'38, p. 177. 8. SH_~l,,on, D. B, and Wa.nC.I.:',STeEn. O. H. Early diagnosis of cancer in the G.I. tract. Postgrad Med 27:806, 1960. 9. Seaxt., H. L., Mtu_er, L. L.. and PLUXm,E. J. Tubeless gastric analysis with an azure A ion-exchange compound. Ga~stroenterology 28:4~, 1955. 10. STortEa, E. H., DODD, D. T., SxsoeR. P. A., and EDt~t.EMEX,C. O. A telemetering capsule for ga.~trointestinal pH measurement. J~IMA 178:830, 1961.

International Symposium on Parenteral Nutrition An I n t e r n a t i o n a l S~ml~osittm on Parenteral N u t r i t i o n will be held at V a n d e r b i l t University, Nashville, T e n n . , on April 4-6, 1968. T h e r e will be half-day sessions on I n d i c a t i o n s for P a r e n t e r a l N u t r i t i o n , Carbohydrates, A m i n o Acids, Fat Emulsions, a n d Studies of C o m p l e t e P a r e n t e r a I N u t r i t i o n . T h e co-chairmen of the S y m p o s i u m are H. C. Meng, M.D., Ph.D., and D. H. Law, M.D. Guest speakers include: Dr. H. XV. Bansi, H a m b u r g , G e r m a n y ; Dr. K. H. B~issler, Mainz, West G e r m a n y ; Dr. G. Berg, Erlangen, West G e r m a n y ; Dr. G. F. Cahill, Boston, Massachusetts; Col. J. E. C a n h a m , Denver, Col.; Dr. D. A. Coats, M e l b o u r n e , Australia; Dr. XV. J. Darby, Nashville, T e n n . ; Dr. A. C. Frazer, London, England; Dr. R. P. Geyer, Boston, Mass.; Dr. A. E. H a r p e r , Madison, XVis.; Dr. K. Lang, Mainz, West G e r m a n y ; Dr. R. Levine, N e w York, N. Y.; Dr. H. Mehnert, Munich, G e r m a n y ; Col. J. A. Moncrief, San Antonio, T e x . ; Dr. J. F. Mueller, Brooklyn, N. Y.; Dr. H. N. Monro, Camt)ritlge, .Mass.; Dr. M. J. T . Peaston, Liverpool, England; Dr. H. T . Randall, Providence, R. I.; Dr. J. E. Rhoads, Philadelphia, Pa.; Dr. O. Schuberth, Stockholm, Sweden; Dr. R. O. Scow, Bethesda, Md.; and I)r. A. Wretlind, Stockholm, Sweden. Advance registration is required. For information, contact symposium secretary, Mrs. C o s O. T . Bat.t., Station 17, V a n d e r b i l t University, Nashville, T e n n . 37203.

122

Amer;c~,n Jou.rnal of O;gest;ve D;seases