THE GROJTTII IK SURFACE ARE,% O F THE HUhlAX GASTRIC AICCOSA1 THREE FIGURES INTRODUCTION The demand f o r ail accurate knowledge of the structure of the stomach during the growth period is evidenced 1)y tlie publication of many papers on the subject. At the present tjme tlie qualitative clia'nges in the various layers of the stomach wall are moderately well described. There are considerable published data available on the ponderal and capacity growth in this viscus, aiicl its position and anatomic relations, at all ages, have been described in the literature. The dimensional and numerical growth of the crypts and glands has heen considered t o a certain degree. That the growth changes in surface area of the mucous membrane have been neglected is not astonishing. The many difficulties which attend the pursuance of such an investigation are obvious. Yet such knowledge is of considerable importance in estahlishing the anatomic norm, at various ages, for this organ. In connection with a study of the growth of the crypts and glands of the human stomach (Scott, '23) it was necessary to arrive at an estimate of the surface area of the gastric mucosa. The observations made in this investigation and in a later quantitative study of the developmental changes in the wall of the liuman stomach (Scott, '29) hronght to 'This study was aided by a grant f r o m the Fenger Meriiorial Association. 131 132 GORDON H. SCOTT light points of such interest that it was decided to make furtlier observations upon the growth changes in the area of tlie gastric mucous membrane. I n addition, it was deemed advisable to attempt to influence experimentally, if at all possihle, this surface of the stomach. The technical difficulties met with in such an investigation are many. First and probably tlie greatest of these is procuring material which has iiot become macerated. Unless one Iias a(*cess to such a mass of material that the most rigid selection may be exercised ~vithoutfear of completely disrupting the continuity of the series, it hecomes difficult, to obtain suitable specimeiis for examination. I n the arailable literature there is recorded a relatively small number of measurements of the gastric mucosal area. Toldt ('80) tletermined the area of the mucous membrane of the stomach in a series of fire individuals ranging in age from the eighth fctal month to fifteen years. Dargein ('99) made a comparison of the mucosal area of the stomach and duodenum in twenty-two adults and of seven infants in the first pear of life. Lissenko ('99) calculated the surface area of the gastric miicosa in a series of seven humans of different ages. With the generous cooperation of the Department of Pathology of the University of Minnesota and of the Children *s Bureau of the Department of Labor, Washington, 1.. C'., under the direction of Dr. F. L. Adair, it was possible to have access to a large iiumber of stomachs taken from iiecropsied infants and children. This abundance of material made it possible to collect a sizable series of suitable organs in the course of three years' time. Another technical difficulty that was encountered was the devising of a method for the determination of the mucosal area which was a t the same time workable and productive of reasonably uniform results. The method used by Toldt ( 'SO) consisted of spreading the fixed and hardened stomach flat upon a glass plate and outlining the area covered. The area in square centimeters was then determined with the Amsler planimeter. This method does not take into account the GROWTH IX SURFACE AREA OF GASTRlC MUCOSA 133 very small folds and surface irregnlnrities of tbc miieosa. Dargeiii ('99) nsed a. wreak solution of acid to macerate the stomach and then removed the mucosa in pieces. Afterward, the bits of mucous membrane were placed on squared paper and the approximate integration method made use of in ascertaining the area covered. Tlic errors introduced by these methods of determining irregular areas in morphological studies have been pointed out by Scammon and Scott ( '27). The results thus obtained hy the several authors are at considerable variance. This is probably due in part to the very divergent tecl~niqiiesemployed. The differences of the mucosal area of the newborn hnmaii stomach, as determined by the rarious workers, map be taken as a fair example of these cliscrepancics. These figures range from 12.75 sq.cm. (Dargein, '99) to 52.7 sq.cm. (Toldf, '80). These data ham heen summarized and depicted graphically by Scammoii ( '19) a11d by Scott .( '25). MATEXIAL AND NETHODS The material nsed hi this study consisted of ninety-six stomachs from individuals ranging in age from six fetal months to fifty years. The stomachs were obtained as soon after death as was possible. In no case was t.he elapsed time between death and fixation greater than four and one-half hours, and in a number of cases it was one hour or less. Specimens which showed marked contraction or dilatation were rejected from the series. The weights of the stomachs were compared with the data of Scammon ( '19). Those which showed any marked variation (more than 100 per cent) from the mean weight for the stated age were discarded. Every attempt was made to secure only those organs which were normal. Full recognition was given to the fact that the stomach is a highly variable organ in both weight and capacity. The estimates of the ages of these specimens were made with the aid of the empirical formulas for determining fetal age developed by Scammon and Calkins ( '23). 134 GOltDOK H. SCOTT ‘L’he stomachs were fixed as soon after removal from the lmclj- as practicable. In erery cast1 tlic stomachs were filled with the fixing fluid (Zenker-acetic) and immersed in a dish containing tlie same mixture. After fixation f o r twenty-four hoars, they were washed in riinniiig water for a n equal length of time, transferred to 50 per cent alcohol for three hours, il1lcl placed in 70 per cent alcohol. , W e r the specimens had liardened in alcohol f o r three days, the? were ~ ~ e m o v e dsplit , in halves by incisions along the g w a t e i - and Icsspr curvatures, and a plaster cast made of the miicmal surface. This was done by pouring a 20 per cent soliltion of glycerin over the stomach and then pouring into tlie \-eiituicular cavity a paste of quick-setting, non-shrinking 1)lastt.r of Paris. The cast mas removed from the stomach aiicl allowed to dry. As woii as tlie plaster of P a r i s was sufficiently dry, thiii tissiich lxtper. was moistened a i d fitted carefully over all of tlic hurfaee irregularities of the cast. This paper mas allowed to clry in position, carefully removed, and spread out upon standard-weight paper (Resolute Ledger no. 3 2 ) . The outlines of the tissue paper were marked and the covered area cut out witli sharply pointed scissors. This paper was then weighed. The weight of the outlined paper being determined iIlld the weight per square centimeter being known, it was a simple matter to rcduce it to terms of square centiineters and a r r i r e thus indirectly at a n estimate of the area of the gastric miicosa in terms of the same unit of measurement. T o test the accuracy of this method, the area of one stomach was determined twenty times. The area of this stomach, that of a nev-born infant, was 32.13 sq.cm. The probable error of tlic twenty observations was i- 0.38 sq.cm. The standard dcviation was 1.37, with a probable error of 4 0.12, and tlie coefficient of variability was 1.02 per cent. It will be seen from this set of observations on a single stomach that the method is proclnctive of reasonably accurate results. GROWTH I N SURFACE AREA OF GASTRIC MUCOSA 135 FINDINGS Prenatal I n a fetus of six moiiths, calculated age, the surface area of the gastric mucosa averages 9.65 sqxm., with a probable error of fr: 0.21. The variability at this age is small, as indicated by the standard deviation of 0.435 sq.cm. and a coefficient of variability of 4.51 per cent. By the seventh fetal month this value has increased approximately twofold and is now 18.16 sqxm., with a probable error of + 0.10. The rariability is even less than in the preceding lunar month. I n the following two fetal months, the eighth and ninth, the increase in surface area of the gastric mucosa is steady and relatively slower. By the eighth prenatal month it has attained 23.75 sq.cm. and in the succeeding month has reached 27.20 sq.cm. I n the tenth fetal month there is an average increase of practically one-third of its former value, so that by this time the mucosal area is 36.93 sq.cm. The variability has increased in proportion t o the rate of growth and its coefficient at this age is 20.92 per cent. At birth the area of the mucosa in square centimeters is 39.17, with a probable error of t- 1.36. The coefficient of variability is at this time 24.68 per cent and represents the highest variability of any group of the series. This may be explained by the fact that a part of this class were stillborn infants and a part had lived f o r some varying lengths of time. I n some instances the infants had lived f o r several days and had undoubtedly taken food. It has been shown by Scammon ('19) that the stomach increases in capacity very markedly in the first weeks of life. That this is an actual growth rather than an adjustment in muscle tonicity is shown by the members of this series which had lived and ingested food. Invariably, these specimeiis exhibited greater mucosal areas than the stillborn infants. This fact is also emphasized by the increase in stomach weight in early postnatal life. The frequency curve given in figure 1 illustrates the distribution of the observations on the neonatal series. 136 G0I:L)ON H. SCOTT Although the selectioii of material for this group was made d o i g very rigid liiics, oiie could iiot jnstifiahly iiicliitle 011ly sti1ll)orn infants. The maiiy factors wliicli may iiifiueiice stomach growth make themselves apparent in the iieoiiatal material presented in this series. The tu-eiity-three cases giveii in table 1were selected from a group of over 150 speci- Fig. 1 A poilit-to-point freyuoicy c u r r e of the tlistrilmtion of the 01~scrr:itions tlie surf:ice aw:i of tlicx g i s t r i c iiiucosa ill tlie newborn infant. A l d s s a s : nui1iIicr of c:ises in r:rcli class. 0rtliii:ites : c*l:iss groupings i n terms of square IW t iiucters. 011 were collected and examined. Tlie most of these \\-ei*e rejected because of postmoi*tern cliaiigcs o r tlcfiiiite pntliological findings. nic~iiswliicali I'OSt ?I ntnl The postiiatal material was groupetl in table 1 awordiiig t o tlici number of specimcus collected for a given time leiigtli. I t i v a s o1)vionslj- impossible t o secure an eveii tlistribntioii GROWTH IN SURFACE AREA OF GASTRIC MUCOSA 137 of matcrial in a study of this sort. The relatively small number of cases given was selected from considerably over 4-00 stomachs collected. All specimens in which definite or suspected gastro-intestinal lesions were demonstrable were rejected. Stomachs from persons dying of diseases with which gastro-intestinal disturbances arc associated were discarded. TABLE 1 STANDARD DEVIATION AUE 8q.m. 6 fetnl moutlis 7 fetal molltlls 8 fetal mouths 9 fetnl moutlis 10 fctril montlis Birtli 0- 3 montlis 3- 6montlis 6-12 nloiitshs 1- 2years 1- 4 years 4- 7 yell3 7-14 Y W ~ S 20-60 pears 2 2 4 4 10 23 H 7 5 9 10 2 4 6 9.G5 f 02 .1 18.16 0.10 13.75 f 0.91 27.20 f ]..A8 36.93 -t 1.6.5 39.17 2 1.36 1.28.69 2 7.07 169.58 f2.17 174.58 f3.15 248.06 23.73 236.40 2 6.90 263.69 k 2.64 329.31 f2.03 525.88 zk 5.7G * 8q.cni. 0.436 0.205 2.G98 4.678 7.725 9.668 19.987 8.513 10.446 16.582 32.363 5.535 6.027 24.143 2 0.15 f 0.07 2 0.64 Per cent 4.51 84.46 1.13 28.91 22.12 f 1.12 f 1.16 f0.96 11.36 17.20 20.92 24.68 -C 3.37 15.53 f1.53 5.33 5.99 7.97 13.75 2.10 1.83 4.59 2 2.23 f 2.64 f 4.88 f2.G4 i0.14 f 0.47 19.32 14.23 13.42 4.15 3.19 3.01 2.53 2.23 I .n9 1.60 0.00 IIn several of tho age groups the number of ensea is too small to lcud siguificance to the statistical nierisuree employed, but they were applied to the data for thc sake of uniformity. In the group aged from birth to threc months the surface area of the gastric mucosa is 128.69 sq.cm., or over three times its value a t birth. The probable error is 2 '7.07 and the coefficient of variability is 15.53 per cent, or approximately 10 pcr cent less than at birth. By six moiitlis the mucosal area has increased to 159.58 sq.cm. and by the end of the first year of life it has attaiiied 174.52 sq.cm. I n the first ycar of lifc the surface area of the gastric mucosa has increased fourfold. In the succeeding ycar the rate of growth has become considerably retarded, for at the cnd of twenty-four months the mean value is but 208.06 sq.cm., with a probable error of 3.i3. * 138 GOliDOX €1. SCOTT I11 the iicxt two years tlie growth in mucosal area of ilie .stomach is but slight, f o r a t the fourth pear it was found to 1~ 235.40 sq.cm., with a probable error of r+ 6.90 and a coefficaicnt of variability of 13.75 per cent. From four to seven !.ears represents ailother period of slow growth f o r this surface of the stomach. The mean value of tlie surface area of the miicosn at this timcb is hut 263.69 sq.cm. 111 the succeeditig sevm years therr=is an inertlase of about 6<5sq.cm. for the l\\-rragc of the pei.iod of seren to fourteen >-ears. At this tirnc it is 229.31 sq.cm. k’ig. 2 fi siniple liistograril, tnkrn from the data in table -1, which illustmtt~s the growth i>li:inges in t h e riiucosal a r e a of the stomacli. The figiires given f o r tlic adult s t o m a c h a r c an average of six specimens, four of Tvhicli were accidental deaths and the r~rnaiiiiiigtwo were suicides by mercuric-bichloride poisoning. The age range is large, but no significant difference was found in any one of the group. The range of the observations u-as from 493.26 to 551.12 sq.cm. The mean value f o r this 5.76 group was 525.58 sqxm., with a prohable error of I sqxm. a i d a coeficieilt of rariability of 4.59 per cent. The data from which the above figures were taken a r e summarized and preseiitocl iii table 1 and illustrated graphically by a simple histogram in figiire 2. GI:OWTH N i i i ~ i l w rof I N SURFACE ~ ~ m .OF 4 GASTRIC XUCOSA f o l d iwrfiase it2 surface arca 139 of t h e gastl-ic mzicosa By the simple expedient of dividing the mean area of the mucosa in tlie adult series hy tlie mean d u e a t any selected age, the result is the number of times which the area has increased betwecii a givc~iiage and matnrity. These figures are t o be found in thc last column of table 1. T t will be seen that the surface area of the gastric miicosa increases 54.46 times betwemi tlie sixth fetal month and maturity. By the next fetal moiitli this d u e has dropped to 28.94 times and decreases steadily thereafter iiiitil birth. From birth to maturity the mncosal area increases but 13.42-fold. After the sudden drop in the number of fold increase j i i surface area in the first three moiitlis of life, there is a slow steady decrease until the adult stage is attained. These figures sliosv clearly the periods of most rapid growth in mucosal area of the stomach a s represented iii the preseiit series. Obaio11sly, the most rapid growth rate is bet~veen birth and the third month of life. There i s aiiotlier period of accelerated developmelit between the sixth aiid sereiith fetal months. At other times during development, p r t i c u larly postnatal growth, the i n c ~ e a s ein mucosal area is slow and even. I n the prenatal series tlie same steady course of eiilargement is followed, but it is more rapid tliaii that which i s foiiiid after birth. E f w t s of distPiitiou iipou gust r i c 111 tlir S I . I Y . ~ U ( Oc l r m of t h c iirosa Tt was ilecidetl to test c~xperinieiitdlytlic efiects of marked t~s-pansioiiupon tlie surface area of the mucosa. The technical difficulties n.hic.11 p ~ ~ c s ~ ~tliemsclvcs ft'd obviatccl any attempts at direct measjliwmeiits of the rcwilts. The fact that it u7as impossible to amirately determine tlie mucosal area of a stomach contracted aiid later measnre the samci area of the same stomach when distended made it necessary to draw upoii inferential evideiice to aid iii tlie solution of this proldem. 140 (;oI:rJos 73. SCOTT Frcsli ncw-boni liuinaii stomaclis wercl distended with fixing Auid at cliff creiit pressures. The pressure in each stoniacli W ~ Skcpt constant foi- a period of ten hours and recorded mcaiirvhile with a mei*cury manometer. After the stomach had tlioronglily hardcnecl, pieces n % r ~cut from each of tlic sevei*alparts and prepared f o r sectioning. The sections ~ ~ e r c stained witli liaematoxylin and cosiii. Whcn compared with swtioiis of corrqmii(1iiig ])arts of normal and undistendcd stomachs, they rcv-calecl but little cliarige in the thicbiiess 01 1 lie mncosa and g e n c i d histologic: appearance. 7’he musclc laycrs m’re tliiiiiietl somewhat and the submucosa was flattcwctl. TV4ieii the pi-c~ssurc~ was over 200 mni. of mercury, tlicrci was a cliange iii tlic mncoc;al layer. Tlie crypts WC~I’C shortened, and wh~11tlic pressnre amounted to as much as 250 to 3OU mm. of Iyi(’i*ciiry,cr.acl;s appeared in tlie epithelium. of tlisf(~iitioiiupon thcl newFigurc 3 illustrattbs tlir e Imrn infant’s stomach. In first figure, u,tlie mucosa is indicated by solid 1)lacB. This figure exliihits tlic typical histologic appearaiice of the normal nen-born in sttrmavli. 7’he secoittl figure, 0, is a stomach from il specim ’ i i of the same cJalculatcc1 age, wliich was distmdecl at ii uro of 120 rnin. of mrrcwry. Tllere is n o cvideiicc of Iiangc i n the miwosal layer aiid but very slight d t e r a tioii i n thc sihmucosul aricl niusclc layers. In tlie last figure, I , tlic stomac.li pictirrcd js tlint ttf a iiex1)oi~n iiifaiit. The 1)rcssure of disteiitiou in this instance was 2%) mm. of n i c ~ r (ry. ~ i The ~~pitlic~liurn s1iou-s 110 evitlence of flatteiiing, the rypt h s o m e d 1 a t short enecl, and t 11e g.1and s a IF likewise K ~ O U ’I)lunt tlian tliow ovdiiiarily seeii in the untlisteudecl b i ornacli. The miisc~le1ayei.s arc considerably flattened aiitl 11i(k hi~iiclles more scparatotl. ‘L’liis is also true of the (d ~lll)Irlllcos;L. Fig. 3 A wi ies of figuivs ill i i trlig the effects of distrrition iipoii t h t . ~ i c ~ w l ) o i iiif:uit ii ’s stoni:rc*li. All figure3 weie i1r:iwii with the Edinger app:iratur ,it a stant1:irtl niaynificatioii. The epithelial parts of tlie inucosa are rcprcseiitml 111 soh(l blavk. u, :I nnimil uiidistciided stoiiiacli; 11, stomacli disteiidrd a t 1% in[ii. rrf riiei( [iry pressure: c, :I stoni:icli (listended a t :I picssuie of ?t‘O mrii. of mriwry. G R O W T H IN SURFACE AREA OF GASTRIC b C Figure 3 MUCOSA 141 &Ill figures were drawn at a standard magnification and from sections taken from corresponding parts of the st omaclis -ill this case, the corpus from the midventral portioii-ant1 at a~qxosimatelythe same location. Tlic same experimtiiits \verc tried wit 11 the white-rat, rahhit, vat, and clog stomaclis. These were distended in sitn in ;incstlietized aiiimals. Iii the c a s e of the rat stomachs, the miicosa (lid iiot show evidence of thinning until pr near to t h e h r s t i n g poiiit were used. The mean h \troilgtli f o r the adult white-rat stomach was 127 mm. of mcrcurj-. T l i c ~ e n’as 110 mwi*kcd x-ariatioii in himtiiig streligtli, foi. the total range of ohsermtions was from 121 t o 13% mm. of mercury. I n the rat the tear als\-nys took place iii tliat portion of the stomac.li wliich was lined with strntificd hcpamou s epitbeliurn. The rabbits’ stomachs gave esseiitially the same resnlts a h ~ w r eol)taiaecl for the lien-horn Iiuman stomachs. The 1)iirstiiig strength varietl from ICiO to 210 mm. of mercury. ‘i’lierci \vas little eriticiice of miicosal (*liaiige either iii sliortof’ c ~ y p t s01’ of thimiiig of tlie mwosal I a p r . Whc11 iires of approximatel:- 130 mm. of mercury ~vercircaclirtl, t l w glnii(1s Iw+aine more compac+tly arranged than i n tlie iioi*mally c~spandeclstomach. \\-lieii vath’ stomaelis wci*e subjected to distention at 1)rws u i w var~-ingfrom 90 I(J150 mrn. of mercnry, a somewliaf tljffcreiit cJfYcct was prodi~ced. Pressures 111) to 110 mm. of rnc’rcury gave no cff’wt ul)on t l i c iniieosn. The glaiids rctxiiietf ii c d . IT11a i hi pli e r 1)r e s 8111 w-ere attained, the glands took the appearance of Iiaving been coiled upon tliemselrt+. 1’1it~ crypts almost tlisappearcd. ,Tohnson ( ’13) tlescrilwtl ;t sirnilai. c+fect in this animal and also in the giiiiica-l)ig. H e poiiitcd out tliut tliclrc is a sliorteiiiiig of both crFpt+ a i i t l giaiids, wliic4i a t tlie same timc m w widelied a i d spwatl a1)art. ;\Inller ( ’ l 5 ) , in a stiidy of tlie lmrstiiig strenptli of t h e alirnviitai-y t r w l of the cat, states that tlie mucosa of t l i c stomach which is ordiiiai*ily tlirow-n iiito rriai~y rugacl, G R O W T H I S SURF.4CE AREA OF GASTKIC MUCOSA 143 indicating EL considerable reserve for dilatation, reaches its elastic limit very soon and tears. He found the mucosa to be the layer which ruptured in advance of the submucosa and muscle. The stomach of the dog is similar to that of the cat in its reaction to hyperdistention. The glands become coiled and the crypts very much shortened when a pressure of 150 mm. of mercury is reached. The muscle layers a r e very much thinned and the individual bundles a r e torn apart. Cracks a r e present in the epithelium and in most cases the mucosa is completely ruptured in one or more places. It would seem from the experimental results obtained with fresh newborn infants' stomachs and with the various aiiimals tried that normal distention of the stomach influences but slightly, if a t all, the mucosal surface area. Until a very marked degree of distention is experimentally produced, there is no new surface presented by a shortening of the crypts and glands. This is contrary to the findings of Johnson ('13) in the intestine of the cat and guinea-pig. H e found flattening of the epithelial cells and a spreading out of the villi coupled with shortening of the iiitestinaI glands. SlXCh a happening would obviously present additional surf ace to the lumen of the gut. This seems, however, not to be the case in the stomach under normal conditions and, iii many instances, under experimental pressures near to the bursting strength. SIJMXARY The surface area of the gastric mucosa in man may be measured with a reasonable degree of accuracy by means of the method described. The mucosal area increases 54.46-fold between the sixth fetal month and maturity and 13.42-fold from birth to the adult state. The most rapid rate of growth takes place in the first three months of postnatal life, during which time it increases over 100 per cent per month. 144 OORDOK H. SCOTT Sormal distention has little or no effcct upon the amount of surface presented to the gastric cavity. Distention under pressures iicm to tlie bursting strength results in an estension of the area of the mucosa which is evidently brought about by shortening the crypts and glands. No change in shape takes place in the epithelial cells themselves. IdTERATUHE CITED J)AWF.IN, 1’. 18!)!) Surface ct voliimc eoniptirk de I’estomac et du duodenuin. Bibliog. Aiiat., vol. 7, p. 207. JOJINSON. F. P. 1913 The effrcts of distention of the jntcstine upon the. shape of villi and glands. An]. Jour. Anat., vol. 14, no. 2, pp. 23&250. LJSSESKO,W. 1899 The growt.11 and structure of the stomach in ehildhood. Dissort., St. l’otcrsburg. (Quoted from review in Jahresb. f. Anrt., Bd. 3, 1900, and from Gundobin, A. P., 1912, Die Besonderheiten dcs Kindcsalters, 8. 264. Berlin.) YTJJ.W,R, II. R. 1915 Kottrr on tlie biirsting strength of the alimentarF tract of the cat. Antit,. Hee., vol. 10, no. 2, pp. 5345. SVAMMON? It. E. 1919 Some graphs and tables illustrating the grort.11 of tlic liumnn stomach. Am. Jour. Dis. Child., vol. 17, p. 395. Sr.4nmox, R. E., AND CA~KINS,I,. A. 192.3 Simple empirical formulae for expressing the lineal growth of the human fetus. Proc. Soc. Exp. Biol. :md Md.,vol. 20, pp. 35.3456. f+’AMJrON, R..E.,AND S c m , ti. 14. 1927 The technique of determining irregular areas in morphological studies. Anrt. Ree., vol. 35, no. 3, pp. 269-277. &-om, G. H. 1925 Growth of crypts and glands of the human stomach. Am. Jour. Dis. Child., vol. 30, pp. 147-175. 1929 A quantitative study of the fetal growth changes of the parts of the huinwn stomach wall. Am. .Tour. Anat., vol. 44, pp. 1-46. TOLIW, C. 1880 Die Entwickelung untl Ausbildung der Driisen (1- Mngens. Sitz. d. k. Ahid. der Wisscnsch. Wien., Bd. 82, S. G.7-128.