SOMEOBSERVATIONS UMBILICAL IN INFANTS

SOME OBSERVATIONS ON UMBILICAL HERNIAIN INFANTS

BY

GRACE E. WOODSFrom the Department of Child Health, UniversitY ofBristol

(RECEPVrE FO PUBLICATION JULY 20, 1953)

The common condition of infantile umbilicalhernia may cause considerable concern amongmothers, and there still seems to be some diversity ofopinion concerning appropriate treatment. It isprobable that the condition is present in one baby insix.

In this paper I have considered 283 infants withumbilical hernia. The time of the first appearanceof the hernia has been noted and particular attentionhas been given to the study of treatment and generalprognosis.

HiorcThe condition of the umbilical cord at birth and

the umbilicus during infancy has aroused the interestof medical authors down the centuries. Celsus (Istcentury A.D.) describes an operation by 'ligature'for umbilical hernia at 7 to 14 years. Soranus (A.D.98-117) suggests that 'it is better to double the cordover and roll it in a little bit of wool and then lay itgently against the middle of the navel, for perchanceby the pressure of its weight the part will be mouldedto a better shaped depression'.Widal de Cassis (1848) first described the

umbilical fascia as immediately superficial to theparietal peritoneum and posterior to the linea albaand derived from the transversalis fascia.

Richet (1856) considered the fibrous make-up ofthe umbilical ring and demonstrated that theumbilical ring cannot be closed by direct muscularaction but in early infancy it is reinforced by elasticfibres from the obliterated umbilical arteries andtakes on the nature of a sphincter. Richet alsodescribed an umbilical canal, and he postulatesthree main types of umbilical hernia-the congenital,the infantile and the adult hernia.Levadoux (1907) (as quoted at length by Allen)

noted the variations in the umbilical fascia and in itsdisposition, and of the remnants of the umbilical

arteries and veins and of the urachus. He concludedthat the local fibrous tissue does not reach its fulldevelopment until after puberty.More recently Gorelow (1934), on the evidence of

300 dissections, agrees with Richet concerning theidentity of an umbilical fascia and umbilical canal.No case of strangulation of an infantile umbilical

hernia has been recorded, and Erichsen (1884)declares that 'these small umbilical herniae are neverstrangulated and consequently never cause death,yet it is very rare to see one in a child of ten years ofage, though in infants we meet them by the score'.Nearly all writers have advocated strapping or atruss or some other restrictive appliance. Opinion onoperation has varied if the hernia has not made aspontaneous cure by three to six months of age(Herzfeld, 1938) or to wait as late as five to sevenyears (Fevre and Barcat, 1947).

Embr-yolo-Wyburn (1939) has shown that the mature

umbilical cord contains vestiges of the allantois. theomphalomesenteric duct (yolk sac) and the extra-embryonic coelom, and is mainly composed ofconnective tissue formed from the primitive meso-derm. The abdominal parietes, i.e., the rectus sheath,the linea alba and the fascia of the anteriorabdominal wall, are formed from intra-embryonicmesoderm which has its origin as the middle layer ofcells in the embryonic plate. Fusion of these twodistinct forms of mesoderm is at the embryonic rimwhich is the future umbilical orifice. Proliferation oflateral connective tissue plates passes from the um-bilical cord at this point and is responsible forclosure of the umbilical ring. A failure in the normalgrowth of the dense connective tissue which shouldultimately occupy the umbilical ring may contributeto local congenital abnormality. In the greater dis-

450

Protected by copyright.

on October 2, 2021 by guest.

http://adc.bmj.com

/A

rch Dis C

hild: first published as 10.1136/adc.28.142.450 on 1 Decem

ber 1953. Dow

nloaded from

http://adc.bmj.com/

LIMBILICAL HERNIA IN INFANTS

order, exomphalos, the supra- and infra-umbilicalportions of the abdominal wall have formed, but theintervening area is covered only by amnion andthere is no proper umbilical ring.

'Herniae into the umbilical cord' may be con-sidered to be the contrary type of failure. At birth itis evident that a portion of gut has failed, duringembryonic life, to return normally into theabdominal cavity through a narrowed neck. Theintestinal wall is usually firmly adherent to the upperrim of the umbilical orifice and the 'hernia' cannotbe automiatically reduced into the peritoneal cavity.The commonest type of abnormality is when the

umbilical ring does not close. A true hernia presentsthrough the ring within a peritoneal sac. Theultimate closure of the umbilical ring dependsupon the proliferation of encircling mesoderm.When this obliterative proliferation is incomplete apatent umbilical ring is the result and herniae throughthis ring during infancy are under particular con-sideration in this paper.

AnatomyFrom anatomical descriptions and evidence

obtained by dissections on 20 neonates it is possibleto give a simple picture of the anatomy. The umbili-cal region is bounded laterally by the rectus musclesenclosed in their sheaths, posteriorly by the parietalperitoneum and anteriorly by the skin. At theumbilical orifice in the linea alba the fibrous rem-nants of the umbilical arteries are firmly attached tothe lower edge of the orifice. The venous remnantsare loosely attached to the superior edge of theorifice but are more firmly attached to the inferiorand lateral edges (Fig. I a). This was noted by

A. Cfett fissete.a

t6riaI'anneamchew unhewresaisn

-8 A. La Hi1

blanc3-

S. La v

bilica

C. L

F ombi

D. L'our

E. L'anntiqueter a

F. Le pecolli.

FiG. la.-Richet's onrginal drawing (4rch. gen. Afed., 1857, 1. 62)loose attachment of the umbilical vein to the walls of the orifice and the

attachment of the arteries and urachus.

Richet (1856). A large amount of embryonic con-nective tissue fills the lower portion of the umbilicuswith a firm layer of tissue which helps to act as anefficient barrier against the development of a hernia(Moschcowitz, 1915).

Dissections have been made on stillborn infantswhose umbilical cord was still present. Wharton'sjelly was first dissected away. The still patent arteriesand vein then lay loose in the dissection and theircourse through the umbilical orifice to their destina-tions could be traced. It was then found that thearteries were firmly attached to the lower edge of theumbilical orifice and the corresponding region of theanterior abdominal wall. The umbilical vein had nosuch attachment and could easily be pulled in and outof the orifice. Whether or not the obliterated veinlater became attached to the inferior edge of theumbilical orifice could not be verified by dissection.Thus it is suggested that the umbilicus is weakest atthe upper edge of the orifice. Usually hernia presentsat the upper edge of the umbilicus.The umbilical orifice is thus essentially a gap in the

linea alba. The size of the gap varies considerablyand the recti play no part in its direct forma-tion. By dissection it was found that at the levelof the umbilicus the edges of the two recti wereseparated by 1 to 1-7 cm. The recti in infants aremore divaricant supra-umbilically than infra-umbilically and 1 to 2 cm. below the orifice therecti muscles are in close apposition at all ages andthe linea alba is only a fibrous raphe. This factor,coupled with the additional strength of the inferioredge of the umbilical orifice, creates an effectiveanatomical barrier against infra-umbilical herniation.Some divarication of the recti is common in infants

and it is possible for wide divarication tobe present without a hernia. Alter-

'repr~ natively, widely divaricant recti are

re de compatible with a spontaneous dis-It 4 apparance of an umbilical hernia. The

apr" in tonicity of these muscles is however, ofconsiderable importance. The rectusmuscles are attached to the posteriorsurface of their sheaths and thus to the

ie os linea alba by fibrous insertions. Manyle. anatomical studies show that one set of

I_art6r these insertions is always at the level ofthe umbilicus. Hamilton has noted that'tendinous fibres develop in fascia in

ou sphinc- direct response to mechanical stress'.wnbilicai. Thus the tone and strength of the rectusoine- muscles undoubtedly affect the

strength of the linea alba and theshow.ing the fascial covering of the anterior abdo-much firmerminal wall.

451

- D)



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

ARCHIVES OF DISEASE IN CHILDHOOD

The umbilical fascia is a direct extension of thetransversalis fascia. Gorelow (1934) observed thatit was represented by a small number of obliquelyrunning fibres, varying from 0 5 cm. to 6-5 cm. inheight. It varied in strength and in position, some-times lying entirely above the umbilical orifice,sometimes over the orifice, only rarely below theorifice, and always firmly adherent to the peritoneum.My dissections have confirned that at the level

of the orifice the peritoneum becomes thickened andsomewhat toughened and appears to be reinforcedby fibrous tissue. Richet considered that this fasciadevelops in strength during the early months andstrengthening the superior edge of the umbilicalorifice forms a protection against hernia.Where there is no appreciable umbilical fascia

the peritoneum is not supported and a directumbilical hernia may form through the umbilicalorifice. It is probably this type of hernia whichoccurs in infants, particularly in premature infants,and spontaneously disappears in the first fewmonths.

Various workers have postulated the existence ofan umbilical canal, formed anteriorly by the lineaalba, posteriorly by-the umbilical fascia and laterallyby the medial edges of the rectus sheaths with anentrance up to 3 6 cm. above the site of the obliter-ated umbilical veins, while in the canal the vein

A. L'ouraque etles art6res om%-bilicales sou-d6es A la cica-

-tr-ice omibilicale

B. La veine ombili-Cale.

. ~~~~~~~~C C, C. Ouvertrepar iaquelle lavein. ombili-

Da- ottiere ou m-raobilical..

pants doatraersdibes ferarnestdoti.fsci om-bilicalis.

perioneuhasbeenremo.ed e~eang te ubilical cna.

is loosely surrounded by adipose tissue. Richet'soriginal drawings are shown in Figs. Ib and 1 c.The umbilical canal, like the inguimal canal,

traverses obliquely the abdominal parietes from theperitoneum to the skin surfaces and both have asimilar posterior surface of transversalis fascia andperitoneum. It is suggested that herniation downthis canal leads to the adult type of umbilicalhernia.

It appears that there are two main types ofumbilical hernia. A direct umbilical hernia can formthrough the umbilical orifice when there is an absenceor weakness of fascial strengthening. The herniatends to 'point' at the superior edge of the umbilicalorifice as this area is less strengthened by fibroustissue. This may be the common type of umbilicalhernia seen in infants. It is not affected by thepresence or absence of divarication of the recti.The hernia tends to be cured spontaneously whenthe fascia of the abdominal wall has increased instrength. A fairly wide umbilical orifice does notnecessarily mean the presence of a hernia, and ahernia may disappear before the orifice has com-pletely closed.An oblique umbilical hernia may take place

through the umbilical canal in later infancy, child-hood or adult life. A portion of omentum or bowelpasses down the umbilical canal and presents at the

A. Cicatrice ombili- Gcale ou fusion de F G E F'ouraque, dest6res etdela veineornbilicaJes. B

caleC. Terminaison de la

Souttiere ombili-cale, au fond delaquelle se vokt1'espace circncipar la demi-rconf4rence superioure

D, D. Art6res ombili- ACcafes.E, E. La goutti4mre om-

bikicalee; graa *t6 en ke,laveine isole, et lesfibres apFvotiques qui fermenFl. fond de la gout-tiere diss_quees.

F, F, F. F. Dbis de -*fl'insertion dui.fscia

G, G, G. Vaissea

mex.H. H. Les faisceaux

fibreux qui formentpari anterieure deD Dla goutti4re.

FiG. Ic.-Richet's original drawing (Arch. gen. Med 1856, 1. 650)showing a similar dissection after the remov.al of the

umbilical fascia.

A

C

452



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

UMBILICAL HERNIA IN INFANTS

upper edge of the umbilicus. This type is morelikely to be persistent and surgical complicationsmight occur.No essential advances in these aspects of the

anatomy of umbilical herniation have been madesince Richet's (1856) original publication. Hisobservations on the different anatomical arrange-ment of infantile and adult hernia closely approxi-mate to the present clinical findings.

Umbilical Henia in Infancv

Celsus defines a true hernia as a 'protrusion ofloop or knuckle of an organ or tissue through anabnormal opening'. I have accepted as the criterionof diagnosis the presence of a saccular swelling whichprotrudes on coughing or crying and which can bereduced by simple pressure through a gap in theumbilicus. By contrast, the absence of any saccularprotrusion despite the palpability of a gap at theumbilical orifice would indicate the absence of atrue hernia. It should be noted that in some infants atbirth there is an unusual length of skin on theumbilical cord which when the cord separates mayfalsely suggest a hernia (Fig. 2).My clinical study of umbilical hernia comprises

283 infants in whom the hernia appeared before theage of 21 years. Clinical details of the birth, thebirth weight, the length of the cord, the day ofseparation of the cord, the time of appearance of thehernia and the possible related causes have beenobtained. Of these infants, 140 were seen at an infantwelfare clinic and the remaining 143 were observedin a hospital surgical out-patient department.

Infants who first attended under the age of 3months at the same infant welfare clinic were takenas controls. With the assistance of health visitors acomplete follow-up has been possible, and nocontrol infants have shown any tendency to umbilicalhernia.

FIG. 2.-Baby aged 34 weeks (10 lb. 12 om birth weight). The umbilical protuuionis due to an excessive amount of skin only. No umbilical orifice was palpable. Theumbilicus was still moist. Strapping was not indicated and would have aggravated

the septic condition of the umbiicus.

FrequencyOut of a total of 573 infants who were brought

to the infant welfare clinic between December,1948, and July, 1951, 106 developed an umbilicalhemia before the age of 6 months. This is anincidence of 1 in 5 -4.

Tvpe of Birth

Table I compares the type of birth in 276 infantswith umbilical hernia with 453 controls who werebom between December, 1948, and July, 1951, inthe same area as the clinic cases, on the basis ofinformation obtainable from hospital and midwiferecords.

TABLE 1COMPARISON OF TYPE OF BIRTH IN AFFECTED INFAN-TS

AND CONTROLS

Umbilical Herniae ControlsType of Birth

No. 0 of Total No. 0 of Total

Normal birth .. 244 88-9 439 96 9Forceps .. .. 5 1 9 8 1-7Breech .. . 7 2-5 3 07Caesarean .. .. 5 1-9 3 0-7Twins . .. 13 4-8 0 0-0

274 100 453 100

These figures show significantly that hernia occursmore frequently in twin births. In fact, all the twinsseen in the survey developed umbilical herniae.There is a possible tendency for a higher incidencein breech births.The birth weights of infants developing umbilical

hernia and of controls are shown in Table 2 andFig. 3.

It will be noted that 2- 80 of the controls wereunder 54 lb. at birth. 9 0% of the affected childrenwere under 51 lb. at birth, 5-30% of the controls

were over 9 lb. at birth and 9-00% ofthe affected children were over 9 lb. atbirth. There is, therefore, evidence thatumbilical hernia is more common inpremature babies (i.e., 54 lb. or less atbirth) and a suggestion that the same istnre of babies above 9 lb. at birth. Theincidence in large babies may be due toa larger cord, which may be particularlynoticeable in some large babies.

Lenth of Cord

The length of the cord at birth was

453



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

ARCHIVES OF DJSEASE IN CHILDHOODTABLE 2

BIRTH WEIGHTS IN AFFECTED INFANTS AND CONTROLS

Herniae Controls in Same AreaBirth Weight (lb.)

No. 0 No. 0

Under 3 0 0-0 0 0-03-34 0 0-0 2 0-434-4 0 0-0 1 0-244k 4 1-5 0 0-044-5 6 2-2 2 045-5k 11 4-1 8 1-754-6 24 9-0 22 4 76-64i 16 6-0 51 10-964-7 40 14-8 62 13-27-71 51 19-0 77 16- 471-8 49 18-3 86 18-384X8 29 10-7 90 19-184-9 15 5-6 44 9-49-94 8 3-0 12 2-69i-10 7 2-5 8 1-710-104 7 2-5 3 0-6104-11 0 0-0 1 0-21 1-1 14 1 0-4 1 0-21 1-12 0 0-0 0 0-012-12i 1 0-4 0 0-0

268 100 470 100

obtained wherever possible and compared with thecontrols (Table 3).

I am indebted to Dr. G. Herdan for the followingcomment on these figures:

TABLE 3

Length of Cord in ControlsAffected Infants (in.)

0-6 (0)6-12 (1)12-18 (7) 2018-24 (20) 3624-30 (12! 6Over 30 (0) 2

40 64

'Although num-bers in Table 3 arenot suflicient forstatistical testing ofany difference thatmay exist in the twodistributions, yetmere inspectionshows distributionsso strikingly differentthat the influence ofthe length of thecord upon the in-vestigated pheno-menon seems at leastlikely.'

I have no evidence toexplain a possible rela-tionship between lengthof the cord and the

4.

00-i 21X

u4d

to

E_ 50a

0La

E

z

-5

a

4'.

0

0

a

A.

occurrence of umbilical hernia, nor have I evidencethat traction on the cord during birth contributes toa hernia, although a possibility is suggested by thesomewhat higher incidence in breech births.

Day of Separation of CordTABLE 4

DAY OF SEPARATION OF THE CORD

Dav Affected Children Controls

3rd 1 (0- 70o)4th 4 (3 00,) 12 (5-0°o)5th 20 (14-9-0) 30 (12-50,)6th 37 (27-60o) 74(30 7°o0)7th 26 (19s5O,) 53 (22O00o)8th 23 (17-20o) 32(13 3°O)9th 9 (6-70,) 24 (10-00o)10th 9 (6-7-,) 9 (3 70)11th 2 (1-50,) 3 (1-20,)12th 1 (0 70o) 3 (1-2°o)13th 1 (0-40,)14th 2 (1-5SO)15th

134 (1000o) 241 (100%o)

No significant difference is noted between the herniacases and the controls and no evidence is obtainedthat the type of treatment of the cord after birth hasan influence on the later development of a hernia.Of the various modes of dressing the cord, it was

noted that of 43 infants treated by a method in whichno dressing or binder is applied but the cord ismerely covered with collodion and allowed to dropoff, 10 developed herniae, a similar incidence to thatin the more orthodox method of treating the cord.

=hernia0 = control

-* , -o

_M

-L Z.~IT-L -. i -1.r S" -6.0 7 S8 =I^ t r. -Z0 O aO = M

Birth Weight (lb.)FsG 3.-Percentage of total number of cases at each birth weight.

454



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

UMBILICAL HERNIA IN INFANTSCondition of Umbilicus after Separation of CordA common belief is that sepsis of the umbilicus

after separation of the cord tends to induce a hernia.The condition of the umbilicus after separation ofthe cord was carefully recorded in all children wholater attended the infant welfare clinic both in thosewho ultimately developed hernia and in the controls.In the majority of cases the umbilicus became drywithin a few days of separation of the cord and nofurther dressing was required. In some cases minorlocal bleeding recurred for several weeks. In othersthe umbilical area became frankly septic with apurulent discharge, the folds of skin around theumbilicus becoming crusted and adherent with pus.In a small number of cases this septic condition ledto the formation of a local soft granuloma, whichcould readily be distinguished from a cord stumpwhich was firm and tough.

Table 5 gives the incidence of five commonconditions in the affected children and in the controls.

TABLE 5INCIDENCE OF COMMON CONDMONS IN THE AFFECTED

CHILDREN AND CONTROLS

Condition of Umbilicus Affected Children Controls

Healthy .184 (76 3°o) 364 (77-90)Bleeding .. .. .. 15 (6-2-o) 25 (5 4° )Mildsepsis .37(15.40o) 46 (9-9°°)Granuloma .0 (0-00) 30 (6 40o)Stumps .5 (2 1Oo) 2 (0 40o)

241 ( I00) 467 (I00%o)

These figures suggest a greater tendency for ahernia to develop after mild sepsis. A significantfact is that no case of umbilical hernia occurred aftergranulation. A direct follow-up of a further 28 casesof umbilical granulation personally treated hasrevealed no example of herniation. It is presumedthat the umbilical orifice becomes firmly closed byscar tissue.

Size of the Umbilical OrificeRoutinely at the baby's first attendance at the

clinic the size of the gap in the linea alba at theumbilicus was recorded. It was noted in many casesthat the orifice through which the umbilical arteriesand veins had passed into the cord appeared to beclosed. One was, therefore, tempted to believe thata hernia would not develop. Sometimes there was alarge gap, but such a gap did not always indicate afuture hernia. It is interesting to note that in fiveneonates with a particularly large gap, sufficient toengage the little finger, no hemia subsequentlydeveloped, even though one infant had an attack ofwhooping cough. The possible anatomical explana-tion for this has been noted (vide supra). Theincidence of gaps was recorded in the last 310 seen

at the infant welfare clinic. (1) Quite a definite gapwithout a hernia was noted in 19 cases. None of thesecases developed a hernia within the first six months oflife. (2) A very slight pinpoint gap was noted in50 cases, but none developed hernia. (3) The orificeappeared closed and there was no sign of an orificeto the examining finger in 167 cases and none ofthese developed hernia. (4) The orifice appearedclosed as in (3) and yet a hernia developed later inseven cases. (5) A gap and a hernia were noted in 54cases. (6) A granulation which required treatmentwas noted in 13 cases and thus a gap could not bepalpated. (7) In three there was no gap but markeddivarication of recti was noted. It is probable, there-fore, that the palpable size of a gap in the lineaalba at the umbilicus in early infancy is not anessential and determining factor as to whether ahernia will develop or not.

Family HistoryIn 264 cases where a family history could be

obtained 32 gave a definite history of a hernia in abrother, sister or parent. But many may have for-gotten about a small hernia in another child. It is acommon observation at the clinic that if one infanthas a marked hernia frequently the next baby in thefamily has one also.

Associated Inguil HerniaOf these 283 cases, nine children had an inguinal

hernia in addition to an umbilical hernia. Althoughit is known that both inguinal herniae and umbilicalherniae have a higher incidence in premature babiesonly three of these nine cases were, in fact, prematurebabies. Two other cases of a umbilical hernia had theallied condition ofhydrocoele of the hernial sac. Thusit was noted that 11 cases of inguinal hernia, allrequiring operation, occurred in 283 cases ofumbilical hernia, whereas in 213 controls in theyear December, 1948, to December, 1949, there wereonly two cases of inguinal hernia.

Asociated Supra-Umbilical HerniaeAs suggested above, when firm fascial coverings of

the abdominal wall at birth are not formed, both aweakness of the umbilical orifice and a weakness inthe linea alba may result. A small supra-umbilicalhernia may therefore co-exist with an umbilicalhernia, and this was noted on seven occasions. In oneof these cases the gap through which the umbilicalhernia protruded was an indefinite oblong shape,possibly emphasizing the existence of weak fibroustissue. It is possible for a supra-umbilical hernia tooccur in the absence of true umbilical hernia and

455P

rotected by copyright. on O

ctober 2, 2021 by guest.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.28.142.450 on 1 D

ecember 1953. D

ownloaded from

http://adc.bmj.com/


this was so of the only case mentioned in the birthrecords. These supra-umbilical hemiae tend to becured spontaneously in a similar manner to anumbilical hernia, presumably due to the develop-ment of the umbilical fascia in later infancy.

Umbilical Heria in Coloued ChildrenThere is a widespread opinion that umbilical

hernia is an almost universal condition in colouredchildren. I have been interested to note, for example,that English fathers who have served abroad haveremarked on the size of umbilical hernia in nativechildren and in consequence have been unable tounderstand their wife's concern over the compara-tively small size of the hernia in their own babies.The condition in coloured babies has been ascribedto malaria or other diseases causing a large liver orspleen. There have been four full negro babies at myinfant welfare clinic in the last two years. One had asevere umbilical granulation and no herniadeveloped. The other three children developedhernia. Two cleared up in a few months but the other,which was the largest hernia seen in this series,ultimately reached operation (Fig. 4). These fourwere healthy babies born in England, and had noevidence of malaria.

Cullen (1916) considered the higher incidenceamong coloured babies to be due to a thicker cordand a larger umbilicus. Mori (1938) observed aseasonal occurrence of hernia correspondingapproximately to the curve for craniotabes andrickets, i.e., the incidence being highest amongbabies born in January. This finding may, I think,have some relationship to respiratory infection.Jones (1941) was unconvinced that the high incidenceof umbilical hernia in negro babies was related either

FiG. 4.-A negro baby aged 14 months was first seen when the umbisize and there swas little umbilical protrusion. After five months continuthe umbilical orifice was 12 mm. The umbilical area was continuc

condition was as shown abose, she had no symptc

to malnutrition, rickets or a large cord. Informationreceived from Nigeria indicates that umbilicalhernia is very common among babies there, butthey nearly always disappear spontaneously and asenior medical officer states that he has neverheard of a case of hernial strangulation. Probably thehigher incidence in coloured babies represents aninherited trait.

Immediate Causes for the Development of UmbilicalHeria

The preceding observations give a picture of thepossible inherent natal and post-natal factors thatmay influence the later development of an umbilicalhernia. However, in many cases there appeared to bean immediate physical cause for the occurrence ofthe hernia, such as coughing, crying or vomiting.Mothers frequently stated that the hernia was noted

directly after the child's birth (Table 6). This appearedto be erroneous information, since in only one casewas an umbilical hernia mentioned in the birthrecords and that particular hernia was, in fact,supra-umbilical. It would, therefore, be moreaccurate to regard these hernias as developingin the early weeks of life, possibly soon after theseparation of the cord. The initial appearance of aninfantile umbilical hernia after the age of 6 months iscomparatively rare. Only one case was noted aslate as 21 years of age. Fig. 5 shows the time of thefirst appearance of the hernia in the out-patient andcases seen at the infant welfare clinic. The latterwere seen personally and may be more accurate.Though a hernia may develop in a healthy, full-

term, breast-fed baby, some immediate cause for thehernia was often apparent, as Table 7 shows.

Crying. Crying is often blamed by the motheras the cause of the hernia. Inone case the parent gave a

>M:- definite day and hour, andL r>!_ stated that the child cried all

day and the hernia made itsfirst appearance in the evening.Nine hospital cases appearedto be of this type and five ofthese had an associated in-guinal hernia.

Coughing. In 16 hospitalcases the hernia was reputedto have occurred during oneparticular attack of bronchialtrouble. In a few of the cliniccases where the mother blamed

lical orifice was 3 mm. in a cough, a definite orifice hadIOUs strapping the size of previously been felt in the lineaouslv alba, andpreviooms. ~~alba, and it was considered

456P

rotected by copyright. on O

ctober 2, 2021 by guest.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.28.142.450 on 1 D

ecember 1953. D

ownloaded from

http://adc.bmj.com/

UMBILICAL HERNIA IN INFANTSTABLE 6

AGE OF FIRST APPEARANCE OF HERNIA

No. of No. ofAge (Months Hospital Cases* Clinic Cases+

0-1 54 131-2 37 722-3 6 293-4 7 104-5 2 15-66-7 57-8 2 38-9 19-12 912-15 2 115-18 118-21 521-24 224- 1

* According to mother's history. + Personaoly seen.

TABLE 7IMMEDIATE CAUSE FOR DEVELOPMENT OF HERNIA

No. of No. of InfantCause Out-Patient Cases Welfare Clinic Cases

Persistent crning .. 9 5Cough .. .. 18 20Pertussis .. .. 16 6Constipation .. 3 0Flatukence .. .. 0 1Marked vomiting .. I IBronchopneumonia .. 1 2Gastro-enteritis I.. ,

that a slight rise in intra-abdominal pressure wouldreadily lead to umbilical protrusion. In a few casesthe child had an attack of pneumonia or severebronchiolitis in the first month of life and the herniafollowed immuediately after the illness. Fig. 5emphasizes an increased tendency for a hernia 4to appear during the winter months.

Pertussis. This was a frequent cause. Thesevere cough and straining can be expectedto raise the intra-abdominal pressure 34sufficiently to cause a hernia. Infective illnessin a small child also lowers the muscle tone.In three cases a small hernia had beenpreviously noted and had apparently under- 2gone a clinical cure, but the attack of Upertussis caused its reappearance. o

Feeding Diffiulties. Fe-ding difficulties and o 14intestinal disturbance appeared occasionally Zto be the cause of a hernia. Crying due tounderfeeding and aerophagy was sometimesevident, and once there was a history offrequent vomiting. Constipation may be eithera cause or a symptom of the umbilical herniaand in each of four cases the hernia persistedafter 2 years of age. Possibly a general stateof sub-nutrition and hypotonia were the cause FS.both of the constipation and the hernia. Two

infants had severe gastro-enteritis necessitatinghospital treatment during the first few weeks oflife, and a hernia was noted immediately afterwards.Thus illness causing a rise in intra-abdominal

pressure may initiate herniation through a baby'srelatively insecure umbilical orifice, and an illnessor infection, which lowers the general physiquecausing hypotonicity, may create a backgroundfavourable to the development of an umbilicalhernia.

Conditions Associated with Umbilical HeriaCertain definite conditions are associated with

umbilical hernia.Mongolism. Umbilical hernia is commonly

present in mongolism, and it may be an associatedsign related to the hypotonicity. Over a period ofyears about 50 mongols have come under mypersonal supervision and many, but not all, havehad umbilical hernia. In each instance the hernia hasdisappeared spontaneously as in normal children.None required operation or persisted.

Cretnism. Similarly in cretinism an umbilicalhernia is frequently seen and may again be related tosevere hypotonia. Of 11 cretins personally seen, eighthad had umbilical hernia. Six of these umbilicalhernias had disappeared under thyroid treatment.and a case is shown in Fig. 6. Three cretins now over7 years of age have never had an umbilical hernia.

Other Conditions. Umbilical hernia may bepresent in other conditions leading to a poor

)

3

3

3

gE 6 0.1 £6 CI0 0;oa , X

= i o

Months of the Year

-The month of vear at which hernia first noticed in 225 cases s-herethe parents statement was known to be accurate.

457



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/


(a) (b) (c) (d)FiG. 6.-A cretin. shossing how the umbilical condition cleared under thvroid treatment with no strapping.

physique and may be associated with congenitaldefects. Recently observed examples include umbili-cal hernia associated with congenital cataract, withcongenital heart disease, with hypospadias, withspina bifida and myelomeningocoele and withgargoylism.

SymptomsIn this series of 283 cases of umbilical hernia

there was no apparent disability in the baby. Allobservers were in agreement with the mothers thatthe hernia was causing no feeding difficulty, loss ofweight, intestinal disturbance or pain. The childrenwere in good health and full activity. Even the twovery large hernias, one of which is shown in Fig. 4gave rise to no symptoms.Among 143 cases referred to hospital for consul-

tant opinion symptoms were absent in all but five.Four of these mothers stated that the child com-plained of pain around the navel, but one motherthought it was due to constipation. One motherblamed a tiny hernia for severe vomiting andabdominal pain which occurred periodically, andthis child was operated on without delay, but with-out relief of the symptoms. Henry (1941) states thatsevere abdominal catastrophe may cause painapparently referred to a small umbilical hernia andso give the impression that the umbilical hernia iscausing the symptoms. But in the few cases wherethe umbilical hernia was blamed for abdominalsymptoms the cause was almost invariably elsewhere.

In one child aged 21 years a protrusion appearedsuddenly through the umbilical orifice four daysbefore the first attendance at hospital. The protru-

sion had caused pain and vomiting, and on examina-tion there was a tender mass, presumably a smallportion of omentum nipped in the orifice. Operationcleared the symptoms and signs. Anatomicallythis appeared to be of the so-called adult type ofhernia in which the hernial protrusion was directeddown the umbilical canal and may be an exampleof the 'semi-umbilical' hernia mentioned by Browne(1952).No example of strangulation was seen in any of

these 283 cases. There is no record of a case ofstrangulated infantile umbilical hernia in Bristolhospital records.A child may play with and apply traction upon its

own umbilical hernia, and if the hernia is markedlyprotuberant other children in the family mayremark about it. Fig. 7 emphasizes this possibility.For this reason surgical treatment may be demandedby the parents.

TreatmentCareful consideration must be given to the

treatment of a condition which is causing no

symptoms and rarely gives rise to complications orfuture trouble. Strapping with adhesive plaster isfrequently advocated but it is not without its dis-advantages. If the umbilicus has not completelyhealed sepsis can occur under the strapping and an

angry purulent condition of the umbilicus may befound. Two cases were seen. Impetigo developsin the damaged skin occasionally. The strapping is

undoubtedly irritating to some babies and may bethe cause of frequent crving.

'e)

458



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

UMBILICAL HER.VIA IN INFANTS

FIG. 7-A.B., aged 11 months, a poor physical specimen, who hadcongenital absence of nght external auditory meatus. Umbilicus w.asstrapped from birth to 6 months of age. Defect in the linea alba swas

fairly small. Operated on at II months.

Trusses are designed to prevent the protrusion ofthe bowel into the hernial sac, supposedly by thepressure of an inflated rubber cushion.On anatomical grounds these methods can hardly

be defended. Forcible apposition to the two rectusmuscles will not alter the actual size of the umbilicalorifice and cannot accelerate a cure which is notalready taking place naturally. Richet made thisobservation in 1856. It is highly important to notethat a restriction on the normal activity of the

abdominal muscles tends to cause loss of tone and ageneral ballooning of the abdominal wall. In fact,some children after a long period of strapping or theuse of a truss have an anterior abdominal wallresembling in shape and tone the abdominal wall ofa multipara (Fig. 8). Strapping may, therefore, havea very significantly deleterious effect. Trusses act inexactly the opposite way, the rubber cushion separ-ating the internal edges of the rectus muscles andtending, contrary to common hypothesis, to enlargethe umbilical orifice and make an automatic cureless likely.The common absence of symptoms or sequelae

and the common finding of spontaneous cure evenup to 5 or 6 years of age make the call for surgicaltreatment rather problematic. In some cases of long-protuberant hernia, with which the child caninterfere and play. it may be justifiable to operate.The only cases of long-protuberant hernia in thisseries have been in children who had persistentumbilical strapping applied. I believe that if thecondition was left and no strapping applied at anytime this type of umbilical hernia in infants wouldnot occur and certainlv the child's attention wouldnot be drawn to it.

For the past three years my policy has been toleave these umbilical hernias alone and wait for aspontaneous cure. A planned experiment to treatalternate cases with strapping and without strappingproved impossible to carry out. A mother whosebaby was strapped finding another baby beingcured spontaneously without strapping would soon

(b (c)FIG. 8a.-A premature baby, breast fed, aged 11 months. This umbilical protrusion had been continuously strapped since I month of age. The

hernia protrudes through the superior edge of the umbilical orifice. The condition improsed after strapping w-as stopped.FIG. 8b.-Child aged 3 years. Hernia well on the way to automatic cure, and umbilical orifice now barely palpable. Abdominal muscles lax and

flabby. No symptoms. Baby brother had a large umbilical hernia, which was cured in seven months swith no strapping.FiG. 8c.-Lateral and anterior view in a child aged 3 years.

459



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/


d

a b

FIG. 9a and b.-Healthv breast-fed baby aged 5 weeks wsith umbilicalhernia.

remove the strapping. On the other hand, at thebeginning of the investigation some mothers couldnot be persuaded not to apply strapping. In some

cases strapping had to be discontinued because ofthe condition of the skin. For the last two years

all members of my infant welfare clinics, healthvisitors, clinic nurses and mothers have been so

convinced that strapping is not necessary that themethod has been quite abandoned.The last 122 umbilical hernias seen at the infant

welfare clinic have not been treated in any way. Allbut three have made an automatic cure with com-

plete absence of a peritoneal sac or pulsation on

coughing. The three outstanding cases are well on

the way to a natural cure.

Of the 132 cases of umbilical hernia over the ageof 1 year sent for consultant opinion, 124 had beenstrapped for a considerable period-some as long assix years. Of the eight that had not been strapped,four became cured spontaneously while underobservation. In the remaining four the hernia did notappear until the child was over a year old and may,based on anatomical considerations, have been theadult type of hernia.

In comparison with 122 untreated cases ofumbilical hernia, 26 cases from infant welfareclinics were strapped for periods of one month or

over. Fifteen of these 26 cases became cured auto-matically fairly quickly, but five spontaneously onlyafter three and a half to six years. Three childrenaged between 31 and 4 years have not yet becomenormal and three have received surgical treatment.

Possibly comparison should be made betweensevere and less severe cases to compare the effect ofstrapping. However, when first seen at the infantwelfare clinic soon after birth, no case could be trulycalled severe. The gap in the orifice felt by pressurevaried in size from about 2 mm. to the size of thethumb. The size of the sac and the amount ofprotrusion were not proportional to the size of theorifice. There might be considerable protrusion

FIG. 9 c and d.-Same bab%- four months later, healths, still breastfed. No treatment swhatsoever was gi'.en for umbilhcal condition.

through a small orifice and very little protrusionthrough a large orifice. Forty-nine of the cases thatwere not strapped had a palpable umbilical orificeas large as the tip of a little finger and all becamecured spontaneously. Two cases are illustratedin Figs. 9 and 10. Of 14 cases where the orifice was ofsimilar size and which were strapped, five herniashave not completely disappeared in spite of the factthat the children have reached 3 years of age.The negro baby with the very large umbilical

hernia (Fig. 4) was initially seen at a few weeks of ageand a record was made that the orifice was 3 mm. insize and the protrusion was small. The baby wasregularly strapped for a year and the size of theorifice and the hernia increased rapidly. Exactlysimilar events took place in the case of a large herniain a white baby, who eventually required operation.None of the hernias that were not strapped reachedany large proportions and this was also true of thehernias in the other two negro babies. The oppor-tunity arose to strap one and leave one of two

460

c



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

UMBILICAL HERNIA IN INFANTS

FiG. I la.-D.P. and S.P., identical twins aged 2 months.FIG. I lc.-S.P. was never strapped. At 2i years of age cure is nearly

complete.FiG. 1 lb.-D.P. wias strapped or trussed for eight months. Nosw aged

21 years cure of umbilical hernia is not complete.

identical twins. This procedure could only becontinued for about eight months as the child whowas strapped developed impetigo in the area ofstrapping. It can be seen from the Fig. 11 that(although both children are of poor physical status)the child whose umbilical hernia was not strappedreveals more advanced healing.

In 17 cases from infant welfare clinics that had notcleared by one year of age, 10 had been strapped.Three of those children who had not been strappedhad had a severe attack of pertussis whichapparently only delayed the automatic cure. Threeothers who came from poor homes had had severalattacks of bronchitis, but despite this, and in theabsence of treatment for the hernia, a natural cure isproceeding. One healthy little boy made a perfectlysatisfactory cure by 20 months. The clinical emphasisis strongly against strapping.

In many cases where the parents expressed a desirefor operation the baby's name was placed on thewaiting list. Owing to shortage of beds there wasfrequently an interval of several months before thebaby was admitted to hospital and in this interval,and in the absence of any form of treatment, spon-taneous cure had frequently occurred. Duringthe course of these investigations long-term observa-tions were made on a series of nine cases. Four caseswere followed up and found to be cured by 5 yearsof age, two were cured by 6 years, two by 7 yearsand one by 8 years of age.

As already mentioned, the incidence among 573healthy babies seen at infant welfare clinics was1: 5-4. The high incidence of umbilical hernia ininfants might suggest that nature would design a cure.An opportunity arose to note the incidence in

nursery school children. Ten very small herniae, all

in children aged 2 years, were found among 105children in one school. All had cleared before thechildren left school at 5 years. This relatively highincidence may have been related to the practice ofrecruiting children for nursery schools from homeswhere maternal care and circumstances are un-favourable. The healthier life in a nursery markedlyimproved the muscular tone of these children andmay have been responsible for an accelerated cure.In another nursery school, among 300 children agedbetween 3 and 5 years no hemia was found.Among 250 school children aged 5 to 11 years

personally examined, five hernias only were noted.Four were very small and one was large and thishad been strapped for six years. Two half-castechildren had been operated on for umbilical herniain infancy. In another survey of approximately3,000 school children aged between 5 and 18 yearsonly three umbilical hernias were found, and one ofthese became cured spontaneously in the followingfew months. These very few hernias in older childrenwere causing no morbidity. Routine examinationof 800 university students showed no evidence ofumbilical hernia. This suggests that the ultimateresiduum of infantile umbilical weakness or abnor-mality in the population is negligible.

DiscssionFrom the evidence available in this investigation

and from consideration of the anatomy, it wouldappear that a patent orifice at the umbilicus is presentfor some time after birth and can be regarded asnormal. The size of the orifice may vary from aminimum of 1 mm. to the dimensions of an adultthumb. The observation that the fascial strengthen-ing of the umbilical region has not fully developedin the early months of life offers a simple explanationfor the occurrence of a small hemia through theumbilical orifice even in a perfectly healthy infant.

*_ Y..

LW 'I,

461



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

462 ARCHIVES OF DISEASE IN CHILDHOOD

The anatomical state can be regarded as no morethan a variation within the normal.The occurrence of an umbilical hernia in a parti-

cular infant might be related to difference in embryo-logical, anatomical and inherited make-up, and withthe normal growth of the baby the umbilical fasciadevelops to an extent which will bring about thespontaneous disappearance of the hemia.

Conditions which contribute to under-develop-ment or late maturation of the fascia of theabdominal wall, or a particular weakness of thisfascia, or a marked hypotonicity of the abdominalmuscles, will increase the incidence of umbilicalhemia, or aggravate the condition when present anddelay a spontaneous cure. Thus deficiency in thelocal fascia is particularly demonstrable in prematureinfants and in these children umbilical hemiae occurwith greater frequency. Umbilical sepsis has beenobserved as a contributory factor in the develop-ment of hemia. In upper respiratory infection or anygeneral infection umbilical hernia is more likely todevelop or to show delay in natural cure. Conditionssuch as mongolism and cretinism are classicallyrelated to umbilical hemia and indicate that umbili-cal hernia is more likely to appear in disease ofwhich muscular hypotonicity is characteristic.From the observations recorded in this paper there

appears to be little doubt that in the healthy babyumbilical hernia has a very strong potentiality forspontaneous disappearance. In the present survey93 0 of children became cured automatically in thefirst year of life in the absence of any strapping ortruss. There is -considerable evidence that the localapplication of restrictive appliances may actuallydelay the disappearance of the hernia or even increaseits severity, perhaps because the limitation of spon-taneous and free movements of the abdominalmusculature inhibits stimulus to local anatomicalmaturation.

ConclusionsTwo hundred and eighty-three infants with

umbilical hernia have been studied. In this series afrequency of an umbilical hernia of one in sixinfants has been recorded. The first appearance ofthe hernia is usually within the first few months oflife and rarely after six months.The incidence of herniation after premature or twin

births is significantly higher than after a normal birth.Breech births, babies of high birth weight and

babies whose cords are of unusual length appearsomewhat more predisposed to umbilical hernia.Mild sepsis of the umbilicus after separation of thecord is related to an increased tendency to herniation.

Granulation tissue forming at the umbilicusprevents herniation.

Umbilical hernia is compatible with normalhealthy growth and its occurrence is, in part,dependent on the child's inherited status. Con-stitutional conditions causing either hypotonicityof the abdominal musculature or a rise in the intra-abdominal pressure favour the development of ahernia and aggravate the condition if present.

In infants symptoms directly attributable toumbilical hernia are rare and no case of strangula-tion has been recorded.

Spontaneous cure is a very common event.The use of restrictive appliances such as strappingand trusses frequently aggravate the condition anddelay a spontaneous cure.The ultimate incidence in school children and

early adult life is very small.In a prolonged investigation of this type. I have

received help from many sources. I would particularlywish to thank Professor A. V. Neale, at whose suggestionthis investigation was carried out, for detailed assistancein the preparation of this paper, and Mr. W. A. Jackman,who allowed me to obtain the details from his out-patient cases over a period of years and Professor R. H.Parry for permission to use the facilities of the infantwelfare clinics. My thanks are also due to Dr. G. Herdanfor statistical help, to Dr. A. A. Craig and Dr. S. W.Terry for obtaining the figures from school children andDr. D. E. Oliff for the figures from university students.I wish to thank the general practitioners who allowedme to pay domiciliary visits to their patients and thehealth visitors who gave full and willing cooperation.

BUM-1OGRAPHYApkey, J. (1948). Postgrad. med. J., 24, 588.Barrington-Ward, L. (1947). Practitioner. 159. 376.Beadnell, L. M. M. (1942). B'it. med. J., 1, 565.Belki, A. (1936). Lattante, 7, 359.Bennett-Jones, M. J. (1944). Brit. med. J., 1. 78.Browne, D. (1952). Ibid., 2. 1144.Chamberlain, J. W. (1936). J. Pediat., 9, 215.Cohen, R. (1948). .Amer. J. Dis. Child.. 76, 44.Cullen. T. S. (1916). The Umbilicus and its Diseases. Philadelphia.Erichsen, J. E. (1884). The Science and Art of Surgery, 8th ed-. 'ol. 2,

pp. 822-823. London.Fe-re, M. and Barcat, J. R. (1947). Presse mcd.. 54, 500.Fraser, J. (1926). Surgery of Childhood, Vol. 2. pp. 774-776. London.Frazer, J. E. (1931). A Mfanual of Embryology. London.Freshman. E. (1933). Lancer, 2. 701.Frtedjung, J. K. (1935). Arch. Kinderheilk.. 104. 168.Gorelow. M. A. (1934). Arch. klin. Chir., 181. 3Q5.Gyllenswdrd, C. (1937). .4cta paediat.. L'ppsala. 20. 237.Hamilton. W. J., Bovd, J. D. and Mossm2n, H. W. (1952) Hionan

Enbr-ology, p. 368. Cambridge.Henrv. A. K. (1941). St. Bart's Hosp. Rep.. 2. '34.Herzfeld, G. (1938). Amer. J. Surg., 39, 422.Jones, J. W. (1941). Arch. Pediat.. 58. 294.Kuntznann, J. (1947). Pediatrie, 36, 307.Magnus. P. Von (19421. Nord. med. (HospitalstId). 13. 930.Mahorner, H. (1940). Ann. Surg.. 111. 979.Mori, S. (1938). Orient. J. Dis. Infants, 24. 10.Moschcowitz, A. V. (1915). .4nn. Surg.e 61, 570.O'Leary. C. M. and Clvmer, C. E. (1941). Amer. J. Sure.. 52. '-MParsons, H. H. (1940). Mfilir. Surg., 87, 298.Reilly, W. A. (1941), J. clin. Endocr.. 1. 532.Richet, A. (1856). Arch. gen. fed.. ser. 5, 8, 641.

(1857). Ibid., Ser. 5. 9. 59.Spivack. M. (1946). .4mer. J. Obst. Gvnec.. 52. 387.Still, G. F. (1931). The Historv of Paediatrics. London.Takayama, Y. (1939). Fukuoka .4cta med.. (Autoref. Sect.). 32. 115.Trimingham, H. L. and McDonald. J. R. (1945). Surg. Gynec. Oh,Ier.,

80, 152.Widal de Cassis, A. T. ( 1848). Des hernies ombicales et &pigastr.ques.

Paris.Woods, G. E. (1952). Brit. med. J.. 2. 1148.Wvburn, G. M. (I1939). 1. Anat., 73. 289.



http://adc.bmj.com

/A

rch Dis C


ber 1953. Dow

nloaded from

http://adc.bmj.com/

Documents

SOMEOBSERVATIONS UMBILICAL IN INFANTS