Reprinted from

ADVANCES IN PEDIATRICS, VOL. VI, S. Z. LEVINE, EDITOR
   (CHICAGO: YEAR BOOK PUBLISHERS, INC., 1953 )
COPYFWHT, 1953, BY THE YEAR BOOK Pun~~srnms, INC.

Intestinal Obstruction in the
   Neonatal Period

C. EVERETT KOOP

School of Medicine and Graduate School of Medicine, University of
  Pennq,dvaniu; The Children's Hospital, Philadelphia, Pa.

GENERAL CONSIDERATIONS

INTESTINAL obstruction in the newborn infant may seem at fist con-
sideration to represent a rather small segment of neonatal pathology.
Yet it is responsible for most of the surgery in the newborn period
and for the most discouraging mortality figures of any of the neonatal
surgical entities within reach of therapy. The purpose of this discus-
sion is to provide that perspective which could lead to an aggressive
suspicion regarding intestinal obstruction in the neonatal period.

Almost all of the congenital anomalies incompatible with life yet
amenable to surgical correction pose problems in intestinal obstruc-
tion or produce symptoms of it. Atresia of the esophagus, diaphrag-
matic hernia, atresia of the small and large bowel, imperforate anus,
and extrinsic intestinal obstructions due to a variety of lesions, are
all without question among these.

In addition to the obstructive lesions which are uniformly fatal
when not corrected by surgery, a surprisingly large group of con-
genital defects may produce intestinal obstruction in the neonatal
period. Such defects, to be sure, can exist without ever producing
signs or symptoms or may become associated with intestinal obstruo
tion in adult life, but a sufficient number cause trouble in the new-
born period to warrant their enumeration at least.
Except for procedures for an occasional omphalocele or, rarely,
for the relief of respiratory obstruction or urinary retention, most
neonatal surgery in the pediatrician's experience is for the relief of

                      03


64                                    C. EVERElT KOOP
intestinal obstruction. Although this condition in `neonatal life has
long been recognized as a challenge to the surgeon, only in recent
years has therapeutic progress been made. Unfortunately, the increas-
ing success obtained by some surgeons has not been widely reported,
and a 100 per cent mortality for certain lesions is still considered by
some as a reasonable prognosis. Surgeons continue to quote statistics
on the pioneer work (26) with neonatal intestinal obstruction, little
realizing that these figures were accumulated 20 years ago and appar-
ently unaware that increasing opportunities in the surgery of infancy
have led to repeated success in the management of all varieties of
intestinal obstruction in the newborn.
Intestinal obstruction in the neonatal period is, to be sure, a
many-sided problem. In the first place, among those who care for
the newborn, many do not appreciate the frequency of neonatal in-
testinal obstruction. Second, surgeons are rarely invited to observe
the variations in physiology in the evacuation of the meconium in the
newborn; they are presented with intestinal obstruction in advanced
stages, and frequently treat such patients with little or no previous
experience and with the expectation of defeat. Third, single success-
ful case reports appear in the literature, while series of patients
treated by surgeons with broad pediatric opportunities are accumu-
lated until a definitive work can be written. The resulting criteria of
success and standards of therapy are therefore low.
Because of the small number of lesions producing obstruction in
the alimentary tract above the pylorus and because of the similarity
of symptoms when they do occur, this discussion will include ob-
structions from the esophagus to the anus.
The mortality in neonatal obstruction is roughly directly propor-
tional to the patient's age. Some infants, of course, cannot be saved,
but the best opportunities lie in the first 2 days of life; thereafter,
the mortality rises sharply, with survival in cases of obstruction for a
week being rather uncommon. Most surgeons with experience in the
management of intestinal obstruction prefer to operate as soon as the
diagnosis has been made. But in advanced cases, a delay of several
hours, during which fluids can be administered, plasma given, and
the infant kept in an atmosphere of high oxygen concentration, may
alter the surgical risk sufficiently to make a satisfactory outcome at
least a possibility. However, if such preparation lasts more than a
few hours, a point of diminishing return is reached at which more


,@T'pJAL OBSTRUCTION IN THE NEONATAL PERIOD       65
.
benefit can be expected from the surgical procedure than from con-
tirluation of supportive measures.
In the literature, there are frequent references to patients con-
sidered too ill for operative procedures. The question arises whether
the judicious use of supportive measures might not have altered
the picture. The operative risk in all infants with intestinal obstruction
is great but in the untreated patients with obstruction, the mortality
is 100 per cent. Increasing experience in neonatal surgery indicates
bat the preoperative condition of a given patient must always be
@midered in reference to the best result obtainable in that infant,
rather than in relation to the ideal result. The temptation to continue
measures which seem to improve the infant's preoperative condition
is strong. Although the use of several hours for this type of support
is to be earnestly recommended, what cannot be accomplished in
that time is usually unobtainable and the continued delay of sur-
gery diminishes the chance for success. One learns to be content
with any preoperative improvement that can be achieved, and pro-
ceeds with operation.
An occasional delay in surgery may be due to the frequency with
which multiple congenital anomalies are found in infants with in-
testinal obstruction; there is a tendency to wait for another lesion
incompatible with life to make itself known. ObviousIy, anticipating
the unknown rather than treating the obvious, can only result in the
death of many infants who might have been saved. The not too in-
frequent association of neonatal intestinal obstruction with mongol-
ism is another deterrent to surgical effort. Here again I believe that
sufficient error is made in the diagnosis of mental deficiency in the
newborn to warrant proceeding with necessary surgery. Mongolian
idiots, as a rule, are not good operative risks, as compared with other
children; much of the decision is thereby taken out of the surgeon's
hands. The surgeon who has seen the successful management of mul-
tiple defects leave a child as normal as he might have been without
them is naturally less timid in attempting to correct intestinal ob-
struction in the infant with other anomalies.
Fortunately, the decision concerning the propriety of surgery in
these circumstances is only infrequently necessary. .Our surgical staif
has shared in the heartaches of many parents whose children have
died, but only rarely has a child survived whose death after neonatal
Sugery would, in retrospect, have been a blessing.


86                                  c. EvERFIT KOOP

EFFECTS OF -AL OBSI-RUCITON

Because infants lack the reserve found in adults, many of the physi-
ologic disturbances which result from mechanical intestinal obstruc-
tion become serious earlier in infants. Not only does the untreated
obstruction result in distention of the bowel and delayed return of
motor function, thereby preventing the normal nutritional processes
from being carried on, but far-reaching and progressive changes in
the respiratory and circulatory systems also take place.
With the increased distention and the elevation of the diaphragm,
which then becomes limited in its excursion, the thoracic cavity
available for breathing space is diminished and dyspnea ensues; the
resulting fatigue may lead to exhaustion in a newborn infant. Such
a situation also encourages atelectasis and hypoxia. One of the most
disturbing problems in infant surgery is that of hypoxia under anes-
thesia. Surgeons and anesthetists alike tend to evaluate their results
by the survival of their patients. In my opinion, cerebral damage,
which cannot be measured because of the nonexistence of a base line,
must inevitably result from prolonged hypoxia under anesthesia. This
in turn alters respiratory physiology.

Intestinal obstruction also affects the circulatory system, not only
by the loss of plasma as a result of the obstruction, but also by caus-
ing hypotension with a resultant decrease in pulse pressure and a
less efficient blood flow through the tissues. This further complicates
the problem of hypoxia.

The combination of respiratory and circulatory difBculties caused
by the obstruction, coupled with the loss of fluid by vomiting and by
accumulation in the intestine, is undoubtedly responsible for the high
mortality in this group of patients, despite the excellence of surgical
technics and the increased knowledge of postoperative fluid and
electrolyte problems.

SYMFTOMATOLOCY

Vomiting, absence of stool, and distention are the classic signs of
advanced intestinal obstruction in any age group. In the newbom,
however, it is frequently possible to make the diagnosis early. Wait-
ing for the appearance of the symptom triad results in a loss of
valuable time, and the delay in diagnosis is reflected in a high mop
tality. If it is possible to make the diagnosis of intestinal obstructica
early, why are so many infants treated expectantly for 4 or 5 days?


,5.1.ES~~~AL OBSTRUCTION IN THE NEONATAL PERIOD        67
The answer to this question is not a simple one, but certainly
,l,o,,ld m&de the fact that the signs which lead to a diagnosis of
   obstruction are also physiologic signs in many newborn
;[,t(&inal
ti,f,lrlts. Inaccuracies in charting by nursery attendants, especially in
Izfrrence to the passage of meconium and regurgitation, contribute
              >
to the pediatricians dilemma and to the delay in diagnosis.
\rornTmc.-Vomiting may be physiologic, usually due to swallowed
U,rliotic fluid and vemix or to air swallowed in crying. The latter
is ,nore common in those understaffed nurseries where feedings are
Ilot &ntained on schedule. Such regurgitation during the first day
of life may be relieved by aspirating the stomach of its contents and
`$\.inz the baby a fresh start. Should regurgitation be encountered
after aspiration of the stomach or for the first time after the first day
of life, a roentgenogram of the abdomen should be made without
del;lV.
Since vomiting is the cardinal symptom of mtestinal obstruction
i,, the neonatal period, the type of vomiting and character of the
,.ornitus provide important information. Esophagc~al vomiting is really
due to an impediment to swallowing and high esophageal obstruction
bf the complete variety produces a type of regurgitation which can-
lnot be confused readily with any other. When such vomiting is spon-
taneous, the vomitus is a sticky material, composed largely of mucus
and saliva.
Vomiting produced by obstruction at the pylorus or some point
`ia the duodenum is forceful in character, revealing the action of a
bypertrophic gastric musculature. The vomitus is colorless if the ob-
struction is above the ampulla, but otherwise contains bile. There is
puch mucus in this vomitus and more often than not, in our experi-
price, particularly in neglected cases, there is changed blood, produc-
b, a coffee-grounds appearance.
Vomiting caused by obstructions of the small or large bowel may
be quite varied. Even distal obstructions of the colon may have a
bomiting pattern not very different from that seen with pyloric or
fuodenal obstruction. Its onset is usually somewhat later than in the
Pther varieties of vomiting, it is accompanied by distention, and the
vomitus may contain meconium. Vomitus without bile may be an
early sign of intestinal obstruction, and the suspicion of such obstruc-
tion need not wait for the appearance of bile in the vomitus.
It should be remembered that vomiting can also result from Ie-
lions other than those of the gastrointestinal tract. Notable among


68                           c. - KO0p
these are central nervous system lesions and obstructive lesions of
the urinary tract; occasionally, vomiting may be associated wi&
cardiorespiratory dif&zulties.
VARIATIONS IN PASSAGE OF bmcoNruM.-The physiologic passage of)
meconium is a variable phenomenon and therefore readily lends itsel
to confusion with pathologic states associated with intestinal obstmc.
tion. In maintaining an aggressive suspicion concerning intestinal oh'
struction, the meconium must be evaluated both qualitatively and
quantitatively.
The infant who passes wet meconium shortly after birth in large
quantities is almost never the victim of an intrinsic obstructive lesion,
Rarely, a baby with an extrinsic lesion causing complete or nearly
complete obstruction will pass a normal amount of meconium during
the first several hours of life and then stop.
The complete absence of meconium is, of course, definitely patho
logic, but a diminished amount may also be so. Even in complete in
trinsic obstruction of the small or large bowel the diaper may be
stained or something that looks like meconium may be passed. It
seems needless to suggest that in the presence of distention or vomit.
ing, some inquiry be made into the volume of meconium, which on
the patient's chart may be indicated by a check mark. Occasionally,
an infant is seen whose referring physician explains the delay in diag
nosis to be the attendant's report that meconium stools were passed
when actually these were mere stains.
When meconium does not pass early, a rectal examination or per.
haps an irrigation is in order. If the presence of an imperforate anu
has been ruled out at the time of delivery, a digital examination may
reveal a rectal valve, or, more rarely, an atresia of the colon just above
the sphincter.
In the newborn, the caliber of the colon is quite small and thf
rectal ampulla is usually somewhat larger. In an infant with obstruc
tion higher than the descending colon, in whom the meconium ha
never passed through the anus, the caliber of that portion of tbc
bowel is often even smaller than normal. A finger inserted in such a
rectum can seldom get beyond the ampulla and almost never beyond
the peritoneal shelf. The sensation is that of reaching a blind end 01
bowel; this physiologic state may lead the examiner to believe thal
the site of obstruction is at a low level, which is quite unusual.
If the meconium is inspissated, a digital stimulus may initiate it
passage. If no meconium is encountered, irrigation of the lower c&~


I,~T~NAL OBSTRUCTION IN THJ3 NEONATAL PERIOD       69
,,,itl, saline by means of a fine rubber catheter attached to a syringe
ellly be rewarding. Indeed, a meconium cast of the lower colon may
ti evacuated, followed by normal meconium stools. If no meconium
is produced by either stimulus, roentgen examination is mandatory.
811,-&l a small bit of apparent meconium or intestinal contents be
ohtaiaed, it is then available for chemical and microscopic study.
The small meconium stool which may or may not indicate an
obs@Wd.iOn can be studied microscopically, as described by Farber
(3 ), for epithelial cells swallowed in utm. The gelatin f&n test for
the presence of trypsin is also helpful on some occasions, if one can
dord the time necessary for incubation of the specimen (39). The
sbsence of cells indicates an atresia of the bowel, but their presence
does not rule out an incomplete intrinsic lesion or an extrinsic ob-
strt&on.
DrsrzWrxoN.-Distention in the newborn infant may be physio-
lo@c until the meconium plug is passed. However, one should not
wait too long for the appearance of meconium in a distended infant
before thinking of obstruction. When distention is present on the
second day of life, or even on the Grst day following the passage of
meconium, a roentgenogram of the abdomen should be made without
delay.
Further evaluation of the infant is necessary, particularly with
reference to a decision concerning an immediate operation as opposed
to a matter of several hours' preparation. But once the suspicion of
obstruction has led to a roentgen examination of the abdomen and
the presence of obstruction has been established, there is little else
to do.

  ROENTGENOGRWIY IN DIAGNOSIS OF I NTESTINAL OBSI-RUCXION
There is little doubt that plain or flat films of the abdomen in the
supine and erect positions are the most valuable diagnostic aids when
an obstructive lesion of the bowel is suspected. They are quickly ob-
tained, atraumatic, and the examination is easy to perform. The con-
trast produced by the gas contained in the bowel is as informative in
most cases as that produced by a medium administered for contrast,
and the risk of aspiration of some radiopaque material which may be
vomited is obviated, as well as the hazard of increasing the obstruc-
`t&n with a substance such as barium.
A knowledge of the normal gas pattern is important for a correct
whtion of the pathologic findings. Many observers (7, 38) have


70                                 C. EVERETT KOop
recorded the appearance of gas in the small bowel in the first 5
hours of life and the passage of flatus by rectum as early as the sixth
to eighth hour. Most infants show a predominance of small-bowel
gas patterns in the first 12 hours of life, with the colon showing some
distention thereafter until the bulk of the meconium has been passed.
Afterward, the small-bowel gas patterns are predominant and not
well delineated.
The best information is obtained from an upright fihn. The ab
sence of gas in the abdomen indicates an atresia of the esophagus
without Cstula communicating with the tracheobronchial tree. When
the latter is present, the abdominal fihn does not vary from the
normal.
In a diagnosis of obstruction of the duodenum and upper jejunum,
the flat film is at its best. There is distention above the point of ob
struction and an absence of gas below it. A similar picture may be
obtained with midgut volvulus, in association with mahotation of the
colon, or a few bubbles beyond the point of maximal obstruction
may be seen.
The diagnosis of obstruction is quite obvious on the flat films of
lesions of the more distal part of the bowel, but the exact level of
the obstruction is obscure. In an adult, it is usually easy to tell the
difference between small and large bowel shadows, but in infants
the plica circulares of the small bowel and the haustrations of the
large bowel are not as well developed as in adults, and the differ-
entiation is always difficult and sometimes impossible. A common
error is to assume that the obstruction is lower than it actually is be-
cause of the tremendous distention and apparently long segments
of dilated bowel.
Starch et al. (40) have pointed out two progressive signs that in-
dicate small bowel obstruction in infants: "Squaring off where the
several gas-filled loops have roughly square lines of demarcation,
and "continuity" where the squaring off is disappearing and the dis.
tended loops are long, broken only by the appearance of septums
incompletely interrupting the gas-filled lumen. Fluid levels in normal
loops of small bowel in infants do not necessarily indicate intestinal
obstruction, but the presence of fluid levels in dilated loops is diag
nostic.
In addition to the gas pattern, the shadows of tumor masses, me
conium, or calcifications can be sought on roentgenograms. Meconi*


-TIN& OBSTRUCTION IN THE NEONATAL PERIOD       71
rrrrr with the appearance characteristic of adult fecal shadows is
s&ongly suggestive of meconium ileus, whereas calciiication outside
the bowel or attached to the bowel wall is diagnostic of meconium
peritonitis. However, meconium ileus due to pancreatic achylia may
e..st without this roentgenographic sign,

All these signs can be seen on the flat ti alone. Occasionally, a
barium enema helps in diagnosing megacolon or in ascertaining the
position of the cecum in suspected malrotation of the colon. In gen-
eral, however, unless such a study will be of extraordinary help to
the surgeon, the better plan is to operate with the diagnosis of in-
testinal obstruction, exact cause unknown. Particularly in hospitals
where congenital bowel lesions are uncommon, there is a tendency
to accumulate a series of films for teaching purposes. Since infants
are easily exhausted, and particularly so if they have abdominal dis-
tention with the accompanying strain on cardiorespiratory reserve,
the quickest, simplest method of getting information is preferable.

The small patient with partial obstruction or an intermittently
obstructing lesion is frequently given an opaque meal. In itself, this
may be indicated under special circumstances; but the infant is fluo-
roscoped repeatedly and kept in the department of radiology for hours
at a time while awaiting the completion of a study. Such an examina-
tion can be more difhcult for the patient than an operative proce-
dure, because in the latter adequate supportive measures are taken
while in the former these are completely neglected.

Repeat films have a special place in the problems of imperforate
anus, or when a meconium obstruction is suspected and a meconium
stool is then passed.

TREATMENT: GENERAL ASPECTS
   A

Anesthesia in neonatal surgery is a highly specialized art, but one
must use the agents and technics which prove best in a particular
institution with its own' personnel. Anesthesia is more than the ad-
ministration of an agent, however, and it is the physiologic approach
to the problem of infant anesthesia that has been largely responsible
for the success of pediatric surgery.O Preanesthetic medication is
*The author gratefully acknowledges the personal skill of Dr. Margery Dem-
kg, and her concepts of neonatal anesthesia which are embodied here.


  72                                  C. - KOOP
omitted in the infant weighing less than 5 lb. In patients above that
weight, l/666 gr. of scopalamine is used. Morphine is not given to
infants not yet fed by mouth because it may easily depress to a
critical degree their low internal heat production and their basal
metabolism. When infants weigh over 7 lb., and have been fed, l/480
gr. morphine is given, primarily to control tachypnea. The effect of
the morphine is to decrease the respiratory rate and increase the tidal
exchange. Efficient respiratory rate in a full term infant under anes-
thesia is 60 per minute or less.
  Preinduction care should include aspiration of the stomach, since
.  regurgitation may occur without evidence of vomiting. Even though
a Levin tube or catheter has been used for suction drainage, aspira-
tion should be carried out because the tube may be blocked and the
  stomach full.
  We prefer endotracheal technics and we use intubation for a num-
ber of reasons. Adequate assistance to respiration can be given, pul-
monary aspiration may be prevented, a patent airway is maintained
at all times, and laryngeal spasm due to irritation by the anesthetic
and secretions is avoided. The assistance to respiration is particularly
helpful because the tidal exchange is decreased and may be inade-
quate when the abdomen is distended. Closure of the abdomen is also
facilitated when respirations are controlled.
  Cyclopropane, oxygen, and ether is the combination of agents
most frequently used, the last mentioned to prevent the bradycardia
which so often accompanies the use of cyclopropane. Curare prepara-
tions and muscle relaxants are hazardous, for they cause a fall in
blood pressure, delay return of consciousness, and really do not help
as much as one would expect. Muscle tone is poor in the newborn
and a distended bowel makes wound closure difbcult. Succinylcholine
is probably the safest of relaxants because of its short action. The
dosage, however, is quite variable in the newborn.
  Following anesthesia one should make certain that the stomach is
empty. The infant should be placed in an oxygen-enriched atmos-
phere if distended or if the respiratory rate remains rapid.

           OPEFWITVE AND POSTOPERATIVE CARE
   Preparation for all possible exigencies should precede any opera-
  tive procedure. In addition to competently administered anesthesia,
an in-dwelling catheter of suitable plastic material or a metal cannula


-TINAL OBSTRUCTION IN THE NEONATAL PERIOD       73
olust be in a vein and fastened securely enough to permit rapid in-
fusion of blood, if needed. In our experience, a wide transverse inci-
sion facilitates exposure and wound healing and the cosmetic appear-
ance of scars is superior to that obtained with longitudinal incisions.
bong the standard postoperative measures are: (1) gastric suc-
uca drainage; (2) use of chemotherapy, antibiotics, and vitamins;
(3) repeated small infusions of blood or plasma; and (4) maintenance
of fluid and electrolyte balances. Perhaps most important of all is
that infants who have undergone surgery are critically ill and must
he seen at repeated short intervals and their management changed
as indicated by the postoperative course.
In general, feeding by mouth is attempted as soon as possible.
Usually, 5 days must pass before an anastomosis, such as a duodeno-
jejmOStOIIlY or enteroenterostomy, permits the passage of sufficient
fluid and air to prove that it is functioning. This is the very practical
reason why search should be made for multiple obstructive lesions at
operation; in the absence of knowledge of their existence the pro-
longed postoperative block to the passage of fluids and gas may lead
to an unnecessary exploration at just the time when the patient is
least able to withstand it.
Although planning fluid therapy according to well-accepted stand-
ards, it is well to lean heavily upon chemical determinations of CO,-
combining power, and determinations of serum chloride, serum sodi-
um, and serum potassium.
In-dwelling tubes passing through anastomoses have been used to
feed infants postoperatively (18). It is safe to avoid such a practice,
if possible, although very fine-caliber plastic tubes (rather than rub-
ber or neoprene catheters of larger caliber) across anastomoses may
prove to be quite safe. Our experience with the use of such tubes
following end-to-end anastomosis of the esophagus seem to indicate
this (B). Only rarely does an infant tolerate feedings placed directly
into his jejunum; tube feeding into the jejunum, although mechan-
ically possible, may therefore prove to be physiologically unsound.
The use of an incubator, such as the %olette,"* seems to have im-
proved the postoperative course of many infants undergoing surgery,
although a number of other, seemingly minor, technical advances
carried out coincidentally may be at least equally responsible for the
lowered morbidity and mortality rates.
*Air-Shields, Inc., Hatboro, Pa.


74

C. EVERFIT KOOP

OBSTRUCTIONS BY SITES AND TYPES

ESOPHAGUS

A-M.-Most infants with atresia of the esophagus are born
cyanotic, with excessive amounts of mucus in de nasophaxynx. Fol-
lowing aspiration and oxygen administration they become pink, and
until accumulation of secretions again produces cyanosis are free of
pathologic signs for some minutes to hours. An astute observer may
suspect esophageal atresia at this time, but in any case the diagnosis
should be obvious at the time of the fist feeding. The second swallow
on its way down meets the forcefully ejected first swallow on its way
back in the nasopharynx; the infant chokes, coughs, and sneezes,
simultaneously with the appearance of the feeding at the nostrils.
Aspiration usually occurs at this time, and aspiration pneumonia, fre-
quently seen first on the right and most commonly in the upper lobe,
has its beginning.

Intranasal passage into the esophagus of a no, 8 soft rubber cath-
eter, which will stop at the site of the atresia, usually high in the
mediastinum, easily co&rrns the suspected diagnosis. Further con-
firmation is not necessary, but can be obtained by injection of several
cubic centimeters of iodized oil into the catheter. Barium, so danger-
ous when aspirated, should not be used.

Atresia of the esophagus is usually accompanied by a tracheo-
esophageal fistula, and any combination of atresia with or without
proximal or distal fist&s is possible. About 80 per cent of patients
with this problem have a proximal atresia of the esophagus with a
distal tracheoesophageal fistula. The symptoms are due to the atresia
and not to the Sstula. The latter may complicate the situation by
permitting vomiting of stomach contents into the tracheobronchial
tree via the fistula. The air which enters the gastrointestinal tract
through the fistula, when seen roentgenographically, assures the sur-
geon that a distal esophagus is present which in other cases may be
nonexistent (Plates 1 and 2). In our experience, this lesion is second
in frequency only to imperforate anus as a cause of alimentary ob-
struction in infants.

A number of surgeons have, in the past decade, achieved remark-
able results in the operative treatment of this lesion, which before
1943 was invariably fatal. Their procedure consists of a retroperitoneal
approach to the defect, with closure of the Ptula and end-to-end


~TINfi OBSTRUCXIONINTHE NEONATALPERIOD       75
anastomosis of the esophagus (15, 42). Even after 10 years, success
does not uniformly reward the effort that many have contributed to
the correction of this anomaly.
We prefer a transpleural approach to the lesion, a simple anasto-
mosis instead of a two-layer one, and postoperative feedings by
mouth or fine polyethylene tubing instead of the usual gastrostomy
(21). These departures from the more common retropleural technics
with gastrostomy have cut operating and anesthesia time by 60 per
cent; have shortened the hospital stay (an economy to the famiIy
and hospital), and have virtually eliminated the need for postopera-
tive dilatation of an esophageal stricture at the site of anastomosis,
which is frequently encountered when feeding by mouth is delayed.
With a team used to managing neonatal problems, a survival rate of
about 80 per cent can be expected, but at this writing there are few
clinics able to achieve this consistently. Such factors as multiple le-
sions, delayed diagnoses, and other defects incompatible with life
will probably prevent the attainment of much better figures in any
long series of cases.
S~F.NOSIS.-This is far less frequent than atresia and is not as difll-
cult to manage because the obstruction is incomplete.
A~HALMLc-T~~~ condition is amenable to medical management
and alterations in position during feeding.
CONGENITAL SHORT xsoPriAous.-The obstruction produced is actu-
ally of the cardiac portion of the stomach as it passes through the
diaphragm. In our experience, this is a very rare lesion, representing
the most unusual variety of obstruction above the pylorus. Large
numbers, however, have been reported from England (48).
DUPLIC~ATI~N OF xsoPHhcvs.-The presenting symptom in these
cases, which are seen occasionally, is inability to swallow during the
neonatal period. The duplications are usually large (Plate 3) and, as
such, displace pulmonary tissue so that respiratory distress or cyano-
sis may lead to a roentgenographic examination of the chest before
the inability to swallow is actually known. The common wall between
the esophagus and duplication usually dictates a procedure such as
marsupialization, although we have more often than not been able
to dispose of the duplication by a one-stage extirpation.

                 STOMACH
Only a few lesions produce obstruction at the level of the stomach
in children, and fewer still cause difhculty in the neonatal period.


76                                     C. EVERETT KOOP
DUPIXATION.-A duplication of the stomach must assume tremen-

dous size or be located at the cardia or pylorus to cause obstruction
(14), so that obstruction of the gastrointestinal tract due to duplica-
tion of the stomach is more of a possibility than a probability in the
neonatal period.

CONGENITAL HYPERTROPHIC PYLOFUC STENOSIS.-This would seem to
be a congenital rarity. Many have been impressed with the marked
hypertrophy of the pylorus at a very early age, and the typical py-
loric tumor has been seen as early as the seventh month of fetal life.
Of interest in this regard is the work of Wallgren (45), who made a
roentgenographic study of 1,000 newborns by means of a contrast
meal. He found that the roentgenographic features of 5 infants in
whom lesions of hypertrophic pyloric stenosis subsequently developed
were indistinguishable from those of 995 infants in whom stenosis
did not develop.

There are, occasionally, infants who vomit at birth and slowly, but
early, develop the typical vomiting pattern of pyloric stenosis and, in-
deed, have the lesion. Just when these patients pass into the clinical
stage of true pyloric stenosis is not known but the tumor has been
found at operation as early as the fourth day of life. Most pyloric
tumors are palpable, and if positively felt, no roentgenography is nec-
essary. In the neonatal period, however, the diagnosis is unusual
enough, so that the possibility is not considered, as a rule, and roent-
gen examinations are commonly done without any effort to palpate
the pylorus. Certainly, the presence of continued vomiting without
distention should lead one to consider the diagnosis and co&m it
by means most reliable in his hands.

There is no question that a Fredet-Ramstedt type of pyloroplasty
is the procedure of choice, through the smallest incision the surgeon
can safely use. Some prefer medical management with antispasmod-
its and atropinelike drugs. In view of the safety of surgery for pyloric
stenosis, as well as the certainty of results, nonsurgical management
of this condition at the present stage of knowledge, in my opinion
fails to make the best use of available treatment.

DIAPHRAGM

Defects in the diaphragm may cause obstruction of the gastro-
intestinal tract at the level of the esophagus, stomach, small intestine,
or splenic flexure of the colon. Not infrequently, diaphragmatic hernia
on the left is associated with mahotation of the colon, so that either


INTESmfi OB~~UCTION IN THE NEONATAL PEXIOD       77
a volvulus of the small bowel or bands obstructing the duodenum
may be the cause of obstructive symptoms rather than the compres-
sion of bowel by the edges of the diaphragmatic defect itself. Or,
the two phenomena may occur concomitantly.
In our experience, when diaphragmatic hernia is diagnosed in the
first few hours or days of life, the signs and symptoms have usually
been those of cardiorespiratory embarrassment. Signs of obstruction
in the absence of respiratory distress lead to a diagnosis of a dia-
phragmatic lesion later in the neonatal period. The cardiorespiratory
signs are usually a function of the volume of abdominal viscera dis-
placed into the thorax which, in turn is somewhat a measure of the
size of the hernia.
Assuming that the diagnosis of diaphragmatic hernia has not been
made in the Srst few hours of life on the basis of cardiorespiratory
signs, obstructive phenomena may make their appearance when gas
fills the intestine, the viscera expand, and the loops of large and
small bowel are forced into the pleural cavity. Distention, depend-
ing on the level of obstruction, usually is of secondary importance to
vomiting as a sign which leads to roentgenography of the abdomen,
and in turn to a diagnosis.
Fortunately, most abdominal roentgenograms of infants reveaI
enough of the chest to give a clear view of the diaphragm (Plate 4).
The clariflcation of diagnostic problems concerned with diaphrag-
matic hernia is not always as easy by roentgen examination as might
be surmised. Eventration of the diaphragm, which is rarely, if ever,
a cause of obstruction, may be easily accompanied by vomiting, and
can be confused with a true diaphragmatic hernia or a diaphragmatic
defect permitting the passage of intestine into the thorax.
Occasionally, supposed roentgenographic evidence of gas-filled
small bowel in the thorax has proved on operation to have been the
shadows of a multiloculated empyema.
Small diaphragmatic defects, as of the foramen of Morgagni, can
be the site of incarceration of a loop of bowel which is difficult to
prove roentgenographically, particularly if the situation is grave and
the use of a radiopaque medium is precluded. Some diaphragmatic
defects are asymptomatic throughout life and without neonatal signs.
But in those which do produce symptoms in the neonatal period and
remain uncorrected, the mortality rises to about 90 per cent by the
end of the third month.
Once the suspicion of a diaphragmatic hernia is confhmed by


78                                    C. EVERElT KOOP
x-ray, operative correction should be carried out forthwith. Because
some diaphragmatic hernias are known to be asymptomatic, there is
a tendency to postpone surgery until the infant is older. Procrastina-
tion usually results in death, or the necessity of an emergency opera-
tion with even greater risk than at the time of first diagnosis.
The aim of all operative procedures is to reposition the abdominal
viscera in the peritoneal cavity and to close the diaphragmatic defect.
This may be accomplished by the transabdominal or transthoracic
routes. Judging by the literature, the former seems to be preferred;
but we routinely use the latter, feeling that it presents advantages
over the abdominal procedure. The mortality rate, in our experience,
is 13 per cent (24). The thoracic approach permits rapid emptying
of abdominal organs from the chest in the distressed infant, and the
exposure is superior to that achieved with an abdominal procedure.
We encounter no greater diihculty in putting the viscera in their
proper place through the aperture in the diaphragm than through the
abdominal incision. Neither method is easy. In 2 patients with con-
genital absence of the diaphragm it has been possible, using the
transthoracic route to construct a diaphragm out of the abdominal
parietes; this probably could not have been done from below. We
believe, furthermore, that the abdominal wall adjusts to the over-
crowding of viscera better if there has been no abdominal incision.

               V~LWLUS OF Mroowr
Fundamental to the understanding of the etiology in midgut vol-
vulus is the knowledge of the embryologic rotation of the colon and
its variations. At the risk of oversimplifying a complex series of events
it can be said that failure of the cecum to descend into a fixed posi-
tion in the right lower quadrant can permit the ileocecal valve to lie
in close proximity to the duodenojejunal junction with a relatively
long, and at times poorly attached, mesentery. Or, failure of attach-
ment of the ascending colon to the abdominal parietes may permit
the proximal colon, which is then freely moveable on its blood supply,
to behave as though it were an additional length of small intestine.
With two fixed points, the duodenojejunal junction and ileocecal
valve or duodenojejunal junction and hepatic flexure lying side by
side instead of being widely separated, as in the normal anatomic
Position, the small bowel is free to twist on itself in either direction
and to pull the ascending part of the colon with it when the anatomic
situation permits.


-TIN& OBSTBVCTION IN THE NEONATAL PERIOD       79
Several hundred cases of this lesion have been reported in the
literature. About half of them presented symptoms the first day or
~0 of life, and most of the remainder in the iirst two weeks. ~rob-
ably, many have a compromised blood supply even in utero. Roughly,
a puter of the ~ohdized loops were of the whole small bowel,
while the remainder included the right colon (6, 12, 27).
Persistent bilious vomiting is the cardinal sign of volvulus of the
midgut. The infants do not seem to be in pain but seem hard hit by
the situation. Distention is not a problem, especially at ilrst. Blood is
sometimes seen in the stools but usually after the passage of normal
meconium. An abdominal mass is sometimes palpable; .in our ex-
perience this is due to collapsed bowel and mesentery twisted upon
itself.
Other lesions of embryologic variation are not uncommonly as-
sociated with malrotation of the colon. Bands crossing the duodenwn
from the malrotated colon to some other fixed point are perhaps the
most usual, but other adventitious (probably not inflammatory)
bands, internal hernias, volvulus of the duodenum, and situs inversus,
either partial or complete, are also found. Roentgenognaphic examina-
tion of the abdomen confirms the diagnosis of high intestinal ob-
struction and may lead to an accurate preoperative diagnosis (Plate 5).
The diagnosis of high intestinal obstruction, even without exact
knowledge of its etiology, makes surgical exploration mandatory. At
operation, the surgical procedure will be indicated by the exact nature
of the lesion found, but the operation will not be complete until the
malrotation is corrected by the Ladd procedure. In this operation
the malrotated colon is detached from its abnormal fixation, leaving
it free to fall into the lower abdomen where a greater distance will
be maintained between the ileocecal valve and the duodenojejunal
junction.
               h'lTW3USCEPTION
`Ihis lesion is very rare in the newborn. Intussusception in the very
young is usually of the idiopathic variety rather than due to demon-
strable mechanical factors, and usually is ileocolic. It may be ileoca
cal and be externally visible through a patent omphalomesenteric duct
(Plate 6).
Mussusception has been reported in a 7 month premature infant
only a few days old. JeErey (20) reported on 17 cases in the neo-
natal period collected from the literature, with only 1 survival. His


80                                  C. EVEUEZT KOOP
own case, which he believed to be the youngest at that time, was 28
hours old. Few neonatal cases have been reported since. In one of
our cases, the infant passed the typical stool of blood and mucus im-
mediately after the passage of the meconium plug in the delivery
room. Since it normally takes several hours at least after intussuscep-
tion begins for blood to be passed by rectum, this must have begun in
utero. Vomiting began almost immediately, and a mass was palpable
before laparotomy was performed a few hours later. The infant gave
no evidence of discomfort. Recovery was uneventful.
The use of a barium enema to relieve intussusception is permis-
sible only if the surgeon is on hand to take over the management of
the patient if barium reduction is unsuccessful and if certain criteria
of safety regarding height of the column of barium and abdominal
palpation are rigidly enforced. The mortality in our cases of intus-
susception compares favorably with that of published series in which
barium was tried first. The author's preference is for operative treat-
ment, which prevents delay, picks up organic lesions when present,
and makes reduction absolutely certain. In the newborn particularly,
in whom the diagnosis may not be clear-cut, and in whom delay may
be tragic, exploration of the abdomen should probably be carried out
without recourse to preliminary barium enema. Abnormalities in
blood coagulation should also be searched for in newborn infants
who bleed in quantity from the rectum. No bowel resection for
neonatal intussusception has been necessary in our clinic, but, if it
were, direct anastomosis would be preferred unless the patient's con-
dition permitted no more than the Mikulicz type of external anas-
tomosis.

              ABNORMAL MECONIUM
Intestinal obstruction in the newborn due to meconium is an ex-
ample of a disturbed physiology resulting in an organic obstruction.
Pathologic obstruction resulting from abnormal meconium must be
differentiated from the delayed passage of a physiologic meconiurn
plug. This plug may be spontaneously evacuated after some intestinal
distention has been built up behind it, or it may require the stimula-
tion of a digital examination of the rectum or, occasionally, the irri-
gation of the lower colon with saline. Subsequent studies in patients
who pass dry meconium plugs have nof in our experience, contrib-
uted any positive knowledge concerning the etiology of this condi-
tion (Plate 7, A).


retinal OBSTRVCTION IN TJ3E NEONATAL PERIOD       81
The pathologic group may be divided into four varieties: (1) ob-
struction caused by chemically abnormal meconium due to pancreatic
&ylia; (2) altered meconium in conjunction with an apparently
associated anatomic defect, such as atresia of the bowel; (3) atypical
meconium and a concomitant lesion, such as volvulus of the small
bowel; and (4) complications of the meconium obstruction, such as
Perforation of the bowel with abscess formation or a plastic variety of
peritonitis.
The clinical picture varies in accordance with the site of the
cbstruction; it also depends on associated findings. These infants do
not, as a rule, pass meconium by rectum, although occasionally there
are reports of diapers being stained green-black. In addition to the
vomiting, distention, and absence of stools, the careful observer may
note excessively tenacious nasopharyngeal mucus. The roentgenogram
is usually diagnostic, presenting "fecal shadows" and other phenom-
ena of intestinal obstruction as demonstrated by gas patterns and
fluid levels (Plate 7, B). The characteristic calcification of meconium,
in the bowel or extruded into the peritoneal cavity, is diagnostic of
atypical meconium. The history of a similar condition in a sibling
is helpful, for the familial tendency in pancreatic fibrosis and me-
conium ileus seems to be well established.
Almost all agree that the absence of pancreatic secretions in the
intestines is etiologic in meconium obstruction of the bowel. Attention
was fbst called to this association by Landsteiner in 1905. The pan-
creatic enzymes may be prevented from reaching the bowel either
as a result of pancreatic fibrosis or of atresia of the pancreatic ducts,
as Farber described (9, 10). Hinden ( 17) has reported meconium
ileus without pancreatic fibrosis. His patient, however, had a vol-
vulus associated with it at birth and later died of intestinal obstruc-
tion, so that there may be some question concerning the inspissated
meconium.
The patient with pancreatic insticiency may have meconium of
green-black color, thick, sticky, and mucilaginous, or it may resem-
ble dried putty and leave no stain on one's examining fingers. Atten-
tion has been called to the chemical differences between normal
meconium and that found with pancreatic achylia, but, unfortunately,
seldom is there enough of a specimen in the obstructed infant to aid
the preoperative diagnosis. The gelatin Elm test for trypsin is simple
enough to help the surgeon occasionally before his hand is forced by
the patient's condition (39).


82                                 C. EL'EREXT KOOP
Meconium obtained at operation may be studied and is of some
prognostic significance. Buchanan and Rapoport (5, 88) have noted
that normal meconium contains larger amounts of total reducing
sugar, as compared with that of patients with pancreatic achylia.
Variations were also reported in the nitrogen partition and in pepsin
and trypsin activity.
Why only a small percentage of patients with pancreatic fibrosis
present the syndrome of meconium ileus is not known. Hurwitt and
Amheim ( 19) have suggested that the pancreatic lesion is one of early
intrauterine life and that the time of its beginning might be roughly
estimated on the basis of the distance the meconium has traveled.
(Meconium is in the cecum by the fourth month and is in the rectum
by the fifth month of fetal life. )
Statements in pediatric literature that meconium ileus is invariably
fatal, coupled with the ever-present hope that an obstruction due to
meconium will relieve itseIf, lead not only to delay in the treatment
of meconium ileus but encourage resort to ineffective measures in an
effort to avoid a laparotomy which might prove unnecessary.
It is safe to say that the laparotomy, unnecessary in retrospect,
seldom has an adverse effect on the patient, but that early laparotomy
in the case of meconium ileus may result in a successful outcome
that would be precluded by delay.
Until quite recently, meconium ileus was uniformly fatal. Since
the report of Hiatt and Wilson (16) in 1948, a number of surgeons
have successfully treated the obstructive manifestations of this entity
in the neonatal period. After the successful management of the ob-
struction, the infant still has the future of a child with pancreatic
fibrosis but we do not believe that this in itself is s&cient reason to
withhold surgery from the infant.
At operation, the typ ical lesion of meconium ileus consists of a
segment of bowel tightly packed with meconium of the inspissated
variety, the bowel distal to it collapsed, and the proximal intestine
dilated with gas and liquid contents. The wall of the bowel containing
the meconium is thickened, showing the effects of peristaltic effort
to move the meconium distally. Several surgical procedures are pas-
sible. The one most commonly used is that of opening the bowel, re-
moving the meconium by mechanical means or irrigation, and then
closing the bowel after patency of the distal bowel has been estab-
lished. Some surgeons use pancreatin suspensions for the irrigation and
leave some behind in the bowel before suturing.


l,.m~NAL OBSTRUCTION IN THE NEONATAL PERIOD       88
Mother technic is to resect the area of inspissation and join by
L,,,,stcmosis the proximal dilated and the distal collapsed segments.
This technic is limited by the length of the loop containing the firm
,,,econium, and adds to the problems of management of the patient
tl,ose connected with an anastomosis. If the obstructing meconium
,s low enough in the bowel, an enterostomy can be performed just
i,&nal to that area, and postoperatively the distal loop irrigated
L,rltil clear. This technic is especially valuable in the infant who is
Jairrg poorly on the operating table and when a short procedure is
Ilesired. The limiting factor, of course, is water and electrolyte loss
bough the enterostomy. We have used all these technics with suc-
cess, and feel that each has its place. In the majority of cases, if the
first method is feasible, it perhaps provides the greatest safety.
If, at laparotomy, the atypical meconium is associated with other
lesions, then both, of course, must be treated. The volvulus which is
,ccasionally seen with meconium ileus is apparently caused by the
&rt of hyperactive peristalsis to dislodge the meconium. The treat-
[Ilent of the meconium obstruction removes the cause of the volvulus
which should not recur. Obviously, if the blood supply is compro-
mised, the gangrenous lesion will dictate therapy (Plate 7, C and 0).
Occasionally, an atresia of the bowel is associated with an appar-
ent meconium ileus or meconium peritonitis. The two phenomena
can occur concomitantly, but some of the apparent atresias are asso-
ciated with evidence of peritonitis so that two possibilities may be
considered: (1) perforation proximal to an organic obstruction, or
(2) apparent organic obstruction as a sequel to obstructive healing
after perforation.
Meconium peritonitis (Plate 8) seriously complicates the various
problems associated with abnormal meconium. It need not be asso-
ciated with pancreatic achylia, although it frequently is. Meconium
does not appear in the bowel until the end of the third month of
gestation. Advanced plastic peritonitis has been seen as early as the
sixth month of fetal life (31).
Meconium is considered to be sterile for several days after birth
aad most meconium peritonitis is assumed to be sterile. It should be
remembered, however, that bacteria from the infant's environment
can enter the peritoneal cavity via the gastrointestinal tract and cause
iu perforation in a matter of hours.
Neuhauser (19), in 1944, described the roentgenographic appear-
ance of meconium peritonitis, noting the areas of calciflc density


84                                  C. EVERElT KOOp
within the peritoneal cavity-within the bowel lumen, within the
bowel wall, or lying loosely adherent to the peritoneal surface of the
bowel parietes. Neuhauser cited others as believing that the calciflca-
tions may begin as early as 24 hours after the onset of peritonitis,
Brunkow et al. (4) followed the fate of such calcified areas and re-
ported that in 1 case they had completely disappeared by 20 months, 3
Our own experience confirms this.
Before 1950 there were no reports of survival in meconium peri-
tonitis, and even now survivals are not common. If the bowel leak
has sealed itself and there is no obstruction, drainage of the abscess
cavity is atraumatic and presents the best opportunity for recovery.
The surgical management of obstruction in the presence of plastic
peritonitis presents as diflicult a problem as the surgeon faces, and
accounts for the high mortality.
Postoperatively, in addition to the surgical measures of decom-
pression and support, these infants should be given pancreatin daily '
by mouth. We have also found that essence of caroid, acting as a
mucolytic enzyme, is helpful.

                 UMBILICUS
The omphalomesenteric duct may persist as an open limb of the
bowel draining meconium to the skin in the region of the umbilicus
or as a band extending from the navel to the ileum. In the latter in-
stance, there may be an associated Meckel's diverticulum on the
ileum, a cyst within the all-but-obliterated duct, or a mucous sac
lying just below the umbilicus and draining onto the skin.
Obstructions caused by the persistent remnant of the omphak
mesenteric duct are unusual; when they occur, the mechanism is
almost always that of small bowel volvulus around the fixed rem
nant extending from ileum to navel. Rarely, an intussusception of
small bowel through the open fistula is responsible for the obstruc
tion (Plate 6).
Definitive diagnosis is seldom possible; rather, one must be c@'
tent with a diagnosis of small bowel obstruction, either complete @r
partial. Occasionally, the presence of drainage of either meconium 0:
mucus at the umbilicus indicates an embryologic vestige and a pre
sumptive diagnosis may be entertained. Exploration of the abdomek
is mandatory, and the therapy is dictated by the situation whid
is found.
Omphaloceles do not produce obstructive symptoms, and tLQ


IeTINAL OBSTRUCTION IN TJXE NEONATAL PERIOD       85
.
i,,cnceration of bowel in congenital umbilical hernias is rare indeed.
It has been reported that an umbilical cord clamp has crushed small
holvel containing an unreoognized omphalocele (la).

           ATRES~A OF SMALL BOWEL
Historically, it is interesting to note that single survivals of patients
,d& atresia of th e small bowel were recorded in 1911, 1926, and
1927. From the n until 1950, there were only 22 others, during which
period about 700 atiesias of the small bowel were reported. More
recently, the mortality rate has improved considerably in many clinks.

                 DUODENUM
Obstructions at the level of the duodenum are caused by a variety
cf lesions, present signs of high intestinal obstruction early, and re-
quire prompt diagnosis and skillful management if the infant is to
survive. In general, the prognosis in these cases is better than in
lower small intestinal lesions, probably due to earlier diagnosis be-
came of earlier signs as well as to fewer physiologic changes caused
by the obstruction.
The signs of duodenal obstruction are early vomiting and disten-
tion limited at first to the upper abdomen. Infants with obstructions
lower than the duodenum may vomit on the first day of life also, so
this sign is not pathognomonic, as sometimes stated. The vomitus
may or may not contain bile and may have flecks of meconinm in it.
In neglected cases, changed blood is frequently seen.
A plain film of the abdomen is usually diagnostic, and the use of a
contrast medium is unnecessary. The aspiration of gas from the stom-
ach by catheter, followed by a second fihn, may delineate the duo-
denal lesion due to removal of the overlying gastric shadow (Plate 9).
BANDs.-These are the simplest obstructions, either adventitious
or in association with malrotation of the colon. They oross the duo-
denum and cause partial or complete obstruction. Their surgical di-
vision is readily accomplished; malrotation, if present, should be oor-
rected as noted previously.
ANNULAR PANCREAS.-A rare but formidable problem in duodenal
obstruction is caused by an annular pancreas. Gross and Chishohn
(13) reported a successfully treated case in 1944, and cited only 2
infant cases, with 1 survival, in previous literature. The diagnosis in
2 of our 4 cases was suggested preoperatively by Dr. John Hope on
the basis of the "double bubble" of gas seen on the abdominal film


86                              C. - KOOp
(Plate 10). In 3 of our cases there was an associated intrinsic obstmc.
tion of the duodenum; for this reason, in addition to the danger to
the ducts, we believe a by-passing operation to be better than one
which attempts division of the pancreatic ring alone. Duodenojejunos.
tomy, if it is feasible, is probably the procedure of choice, but in must
instances gastroenterostomy may be an easier and a safer procedure.
Except for the diagnostic "double bubble" in ammlar pancreas,
great roentgenographic accuracy is not to be expected in separating
one we of small bowel obstruction from another. One must be con-
tent with lmowing whether the obstruction is in the duodenum or
lower, and whether it is complete or partial.   '
Dmrm+cMs.-When a diaphragm obstructs the lumen of the duo-
denum, it should not be perforated. The treatment of choice is resee
tion and anastomosis or anastomosis around the lesion. Such lesions
may be multiple and an effort should be made to rule out the pres-
ence of similar lesions in the lower bowel.
A-.-The management of obstruction depends on the lesion
encountered. The level of the obstruction and the relationship of the
atresia to the bile and pancreatic ducts will dictate the surgical pro-
cedure. Duodenojejunostomy or gastrojejunostomy produce equally
satisfactory results initially. The long-term problems of gastrojejunos-
tomy have not been proved in infants, but can be presumed, judging
from experience in adults, to entail the additional hazard of marginal
peptic ulceration.
          JEJUNUM AND 1~
Generally, lesions obstructing the small intestine more distally an?
increasingly more difkult to manage. The lower the lesion, the later
the onset of vomiting and the greater the distention. These are the
patients referred to the surgeon after the longest delay. Postponement
of diagnosis is reflected in some series, in which half the patients
were considered to be beyond surgical tieaiment. Although this
seems exaggerated, it indicates the poor condition of these infants
when seen by the surgeon. Many infants with intrinsic bowel lesi@
are born prematurely, a factor which increases the risk. Multiple
lesions also seem to be more common in association with atresia of
the ileum than with that of the duodenum and colon, and the COR*
tion of multiple defects adds to the hazard.
The operative procedure in complete obstruction is anastomos&*
This may vary in technic, and the choice of end-to-side or side-to-side


I@TINfi OBSTRUCTION IN THE NEONATAL PERIOD       87
varieties must be dictated by previous experience and the anatomic
situation. Despite traditional belief to the contrary, the delayed anas-
tomosis of the Mikulicx type may be carried out in the lower quarter
of the small bowel provided that the common spur of the two limbs
of bowel is promptly crushed and the time during which the infant
has an iIlteStiM1 fh&lla is not prolonged. Regardless of method, there
is no field of surgery where meticulous atraumatic technic and the
use of fine suture material is as rewarding as it is in an anastomosis
of the small intestine of a newborn infant (18).
SrxNosxs.-Single or multiple stenoses may occur in any portion
of the small intestine. Usually a single stenotic area produces signs
of partial or intermittent obstruction; but with multiple lesions dis-
tention sometimes builds up rapidly and the resultant state must be
considered as complete intestinal obstruction.
DTn?LIcATIoNs.-When these produce obstructive symptoms in the
neonatal period, the lesion is usually a cystlike structure with one
wall in common with the mesenteric wall of the adjacent bowel
(Plate 11). Resection of the lesion with an appropriate anastomosis
has been uniformly successful in our experience.
INTEBNAL -.-Signs of high or low, complete or partial ob-
struction are produced by internal hernias, such as the paraduodenal
variety, those behind the cecum, or those through abnormal mesen-
teric or omental apertures. Exact preoperative diagnoses are rare,
and treatment is usually made mandatory by the obstructive phe-
nomena. Symptoms may exist from birth and require relief in the
neonatal period. More commonly, however, there is either a mild
feeding problem or an asymptomatic interval before the situation
demands correction in the first few months of life.
INGUINAL IXRNIAS.-Particularly in premature infants, such hernias
may produce signs of intestinal obstruction and indeed may incar-
cerate or strangulate a portion of the small bowel. The latter condi-
tion can usually be prevented by elective hemiorrhaphy or by using
a yam truss until surgery can be carried out
In the infant, an incarcerated hernia does not necessarily make
emergency surgery mandatory. Most of these lesions can be reduced
normally. It is noteworthy that an infant bowel suffers no damage
from longer incarceration than is apparently the case in adults.
However, if manual reduction camrot be accomplished early, surgi-
cal reduction should not be postponed.
The small bowel is subject to a great variety of obstructive lesions,


88                                  C. IWKRETT KOOP
among them: polyps; benign and malignant tumors (30) (Plate 12);
abscess (36); volvuhrs around fibrous bands connecting abdominal
viscera with the parieties (23 ); adhesions, perhaps resulting from a
prenatal bowel perforation with recovery; granulomas; and inflam-
matory lesions so old that they undoubtedly originated in utero (25).
In these unusual situations, the preoperative diagnosis of intestinal
obstruction is usually the only possible one. Laparotomy reveals the
exact situation `and treatment is then based upon the findings. Un-
fortunately, in most of these rare problems, the obstruction can be
relieved only by extirpation of the lesion; this frequently introduces
d&s&es which may be insurmountable.

COLON

Except for imperforate anus and agenesis of the myenteric plexus
(megacolon), the obstructive lesions of the colon in the neonatal
period are less common than at any other anatomic site. The lesions
themselves, however, resemble those occurring elsewhere in the gastro-
intestinal tract. Obstructions due to atresia, reduplication, and bands
are infrequently reported,

Abdominal distention is most marked in distal bowel obstruction
but that sign in itself should not be used in diagnosis and indeed is
misleading in late cases of small bowel obstruction. We have seen
mild bleeding by rectum with extrinsic obstruction of the splenic
flexure. One obstructive reduplication of the rectum was visible ex-
ternally over the buttocks.

Treatment is the same as for obstruction of the small bowel, with
the exception of colostomy. A proximal enterostomy in the small
bowel is dangerous because of fluid and electrolyte loss, whereas
distal to the right transverse colon this aspect of a colostomy can be
managed. In general, definitive surgical procedures are to be preferred
if they do not entail too great a risk.

NEUROGENIC OBSTRUCTIONS

The relationship between certain types of intestinal obstruction and
agenesis of ganglion cells in the myenteric plexus of the bowel bar
been made clear by Zuelzer and Wilson (47) and Swenson and Neu-
hauser (4%44). The pathologic physiology of neurogenic intestinal oh-
struction in childhood is therefore better understood now. There are a
number of differences between neonatal obstructions due to disturbed


rnaTINAL OBSTRUCTION IN THE NEONATAL PERIOD       89
+ysiology and those due to organic causes. Generally, operation in
be former discloses no obvious intrinsic or extrinsic lesion responsible
for failure of bowel evacuation nor is there evidence of a work hyper-
uophy proximal to the neurogenic obstruction. These anatomic find-
ings are reflected in the symptomatology and the roentgenograms. (1)
Since a small meconium stool or two is frequently passed by an infant
,~b a neurogenic obstruction and rapidly progressive signs of obstruc-
tion are normally absent, diagnosis is late. (2) The gas pattern on the
roentgenogram indicates in many instances that obstruction is not
,mplete; this further postpones the d&ite diagnosis and need for
treatment. (3) Finally, the signs may vary in intensity from those of
complete obstruction to almost normal meconium evacuation.
However, a fecal impaction in an infant is never normal. It must
always be considered indicative of some disturbance, usually far-reach-
ing, in the enzymic content of the intestina1 fluid or in the nerve
supply of the bowel.
Clinical separation of neurogenic small bowei obstructions from
megacolon may be difEcult. Usually the gas shadows and distribu-
ucn of meconium shadows on the roentgenogram will yield a pre-
sumptive diagnosis. A barium or iodized oil enema can be used in the
newborn to establish the diagnosis of aganglionic megacolon (Plate
13, A and B). However, although megacolon may be positively
diagnosed, absence of the typical lesion in the rectum and descend-
ing colon does not rule out the involvement of a larger segment of
the large bowel or of the colon and small bowel.
The appearance of the intestine at the junction of normal bowel
with the aganglionic segment differs according to the anatomic site
of the transition. This in turn is probably due to the consistency of
the intestinal contents at a given level. For example, in high jejunal
obstruction due to agenesis of the myenteric plexus there may be no
demonstrable change in the contour of the bowel. In the usual
lesion of Hirschsprung's disease, on the other hand, the aganglionic
segment of colon is preceded by a markedly hypertrophied and di-
lated segment of bowel in which longitudinal muscle bundles partic-
ularly are seen to be hyperfrophied.
The aganglionic segment of bowel is commonly termed spastic or
in spasm. This is incorrect Actually, the affected sepent is deprived
both of parasympathetic innervation and its local reflexes, resulting in
its inability to undergo peristaltic action. This explanation is con-
sistent with the generally accepted view that preganglionic parasym-


90                                   C. EVERETT KOOP
path&c fibers\ end around the neurons of the myenteric plexus and
that intestinal reflex patterns are begun and controlled by the neurons
in this plexus. The attempt by the proximal bowel to propel its con-
tents through the aganglionic segment obviously accounts for the
hypertrophy (Plate 13, C). The degree of hypertrophy seems to be
related not only to the consistency of the bowel content at that par-
ticular level but also to the length of time the infant has survived.

                 MEGACOLON
In their fist report on aganglionic megacolon in 26 patients, Swen-
son et al. (43) stated that in all but one case the onset was in the
first 2 weeks of life. Hirschsprung had reported 2 such cases and
Barrington, in a study of 19 autopsies, considered that 16 cases had
been symptomatic from birth. In our series of more than 60 cases,
symptoms were present from birth in a large number. The great
majority of these have limped along with rectal irrigations and a
variety of nostrums by mouth. Some in the age group of a few weeks
to a few months have a history `of an exploratory laparotomy shortly
after birth, at which time the aganglionic megacolon was seen but
not recognized. These infants received no surgical relief of ob-
struction or there was an attempt to milk meconium! out of the anus
through the aganglionic segment.
A small number of patients with neonatal obstruction due to
megacolon have come to us a few hours to a few days old without
previous treatment. In such cases, a colostomy proximal to the point
of obstruction is performed immediately after establishing the di-
agnosis, and definitive surgery is deferred until a later date.
The diagnosis of megacolon in the neonatal period is suggested
when the flat film of the abdomen shows fecal shadows in the colon
of an infant who is passing no stools and is vomiting, or becoming
distended, or both. Digital examination may show the rectum to be
empty. An enema with an opaque medium reveals on x-ray a distal
collapsed segment of dyskinetic bowel above which there is a hyper-
trophied and dilated segment.
Our clinical and roentgen findings in megacolon parallel those
of Swenson and co-workers, and our histologic observations on the
absence of ganglion cells in the myenteric plexus of the collapsed
segment of bowel are in accord with those of Swenson et al., and
Bodian et al. (2, 3). The histologic basis of neurogenic obstruction 3


-AL OBSTRUCTION IN THE NEONATAL PERIOD       91
apparent$ still ignored by a number of workers, who continue to use
suq$cal procedures which leave an aganglionic segment of bowel in
the patient. If there is a true megacolon, such procedures can only
lead to a recurrence of the clinical picture of obstruction. At least
half the infants with megacolon have an aganglionic segment ex-
tending distally below the pelvic floor; it therefore cannot be ade-
quately removed by an anterior resection of the colon.
Fortunately, the emergency aspects of relieving obstruction in
the newborn are paramount; few, if any, surgeons experienced in the
treatment of aganglionic megacolon attempt definitive procedures in
he first few days of life. In the less severely affected infants, who
can be eased along out of the. neonatal period without surgery, a
choice of treatment may be possible at a later date.
Some believe, notably Potts ( 32), that because infants tolerate
a colostomy so poorly a definitive procedure should be carried out as
early as possible. All but 1 of our patients treated with colostomy in
the neonatal period or in early infancy have thrived. The single
fatality was not due to fluid loss. In our experience, even a trans-
verse colostomy is not difficult to handle, so far as fluid balance is
concerned. Of greater importance would seem to be the poor tol-
erance of all infants with megacolon to any gastrointestinal i&&ion.
A child with megacolon who has not had a colostomy or any other
operative procedure is laid low by a bout of diarrhea out of propor-
tion to the severity of the disease. Should an infant with megacolon
and a colostomy have a bowel infection, the colostomy may perhaps
exaggerate his problems, but the blame for fluid balance diEculties
cannot fairly be placed on the colostomy itself. The infant with an
obstruction due to megacolon demanding surgical relief would do far
better with a transverse colostomy than with an immediate attempt
at a definitive procedure, such as excising the aganglionic segment
of colon.

                SMALL Bown,
When the aganglionic segment is not confined to the terminal
colon, the symptoms closely resemble those of megacolon, but the
problems of management are far greater. Indeed, when the agang-
lionic segment is proximal to the cecum, no treatment so far offers
the hope of survival.
Forshall et al. (11) presented clinical material and an interpreta-


92                                   C. EVFiRElT KOOP
tion of it which is in line with the earlier report of Zuelxer and
Wilson (47). Bi kh
              c  am's (37) experience with small bowel agang-
lionic obstruction closely resembles ours: these cases are difficult to
separate from megacolon preoperatively; they do not respond to
spinal anesthesia; there is no external clue to the site of aganglionosis;
and lesions above the terminal ileum are uniformly fatal.
In treating an.infant with neurogenic bowel obstruction, laparotomy
is advisable, in order to ascertain the level of agenesis. If a typical case
of Hirschsprung's disease is found, a proximal colostomy is indicated,
with a definitive treatment postponed to a later date. We prefer a trans-
verse colostomy; this leaves a normally innervated segment of bowel,
distal to the colostomy, which through disuse will contract to a size
suitable for later anastomosis with the rectum. But if the dyskinetic
area is proximal to the transverse colon, the choice is between doing
nothing, and facing the infant's certain death, and performing an
ileostomy, which most infants tolerate poorly. Swenson has performed
a successful terminal ileostomy; and at a time when fluid balance
was much less well understood than it is now, Dr. W. E. Lee per-
formed terminal ileostomies with surprising success. Nevertheless,
infants tolerate poorly any type of small bowel fistula, and the in-
genuity of the attending staff is greatly taxed in keeping up with
fluid and electrolyte losses. If the infant can be carried over the
Srst 2 weeks, there is apparently a compensatory adjustment in fluid
loss which eases the problems of management.
Paralytic ileus of the small intestine must be considered in the
differential diagnosis of neurogenic obstruction of the bowel. It may
occur with an infarct of the kidney or with retroperitoneal hemorrhage
resulting from a traumatic delivery or with spontaneous hemorrhage
soon after birth. It is di0kult to assess paralytic ileus in the neonatal
period. In the presence of apparent obstruction, it is probably safer
to perform an exploratory laparotomy, which in retrospect will have
been unnecessary, than to procrastinate when laparotomy could have
lead to definitive treatment.
Zuelzer and Wilson (47) have called attention to the familial
tendency in congenital, neurogenic obstructions of the bowel. Among
our cases, in each of 3 families, there have been 3 children with
neurogenic obstruction. In 2 of the families, the colon only was af-
fected; in the third, 2 children had small bowel myenteric plexus
agenesis and 1 had typical Hirschsprung's disease.


~~TI.I'JJLL OBSTRUCTION IN TIiJ3 NEONATAL PERIOD       93

IMPERFORATE ANUS

The diagnosis of imperforate anus is usually made by observation
alone, and usually only a few minutes after birth. Paradoxically, this
prompt diagnosis permits the delay of a day, sometimes of 2, before
surgery.

There are several varieties of imperforate anus. (1) A transparent
membrane behind which meconium can be seen; its perforation is
&rple, completely effective, and requires no further discussion. (2)
A normal anus and rectal canal ending in a blind pouch a few centi-
meters proximal to the sphincter; this is really an atresia of the colon,
and its management is similar to that of other colonic obstructions.
(3) Imperforate anus, in which the distance between anal dimple
and blind end of colon is 1.5 cm. or less. (4) Imperforate anus in
which the distance is greater than 1.5 cm. About the differential di-
agnosis of the last two and their management are centered the prob-
lems of imperforate anus.

To digerentiate the high from the low, gas-containing bowel is
essential for roentgenographic contrast, and hence the need for some
delay after diagnosis before surgical treatment is carried out. If the
infant cries and swallows, suihcient air will be taken into the gastro-
intestinal &act in the i&t 8 to 12 hours to permit accurate x-ray di-
agnosis. Roentgenograms at intervals of 8 hours usually show the
level of the lesion within 24 hours or occasionally in 36. To prevent
excessive distention, suction drainage via a catheter placed in the
stomach should be instituted after 12 hours.

Traditionally, roentgenograms for the diagnosis of imperforate anus
are taken with the infant held by his heels and with an opaque ob-
ject over the anal dimple, as described by Wangensteen and Rice
(46). However, flexing the thighs on the belly, as advocated by
Rhodes (35), may serve better to push gas into the distal bowel
(Plate 14). A tight abdominal binder is also helpful. As soon as it
is established that the distance between colon and anal dimple is less
than 1.5 cm., or when the caliber of the colon increases on successive
iYms without further descent toward the anus, the decision to operate
can be made. The procedure is dictated by the distance between
anal dimple and colon.

For lesions in which the separation is 1.5 cm. or less, the perineal
approach is not only satisfactory, but entails little risk and permits


94                                   C. EVEBETr KOOP
excellent results. When the separation is more than 1.5 cm., the peri-
neal approach should not be used; a colostomy or definitive surgery
by a combined abdominoperineal approach are then the procedures
of choice.
A colostomy is lifesaving, simple, and postpones definitive surgery
indefinitely. The combined abdominoperineal procedure should not
be done unless reliance can be placed on all phases of surgical as-
sistance, including anesthesia and postoperative care. In our practice,
the combined operation is used for any infant with a gap of more than
1.5 cm. and a weight of over 4% lb., in the absence of medical con-
traindications (22, 34).
Too frequently, the perineal approach is used, with undesirable
sequelae, when the combined operation or colostomy would have
been preferable. It is possible to bridge a gap of more than 1.5 cm., at
the expense of nerve and blood supply to adjacent structures, and
pull the mucosa of the colon to the skin edge for suturing. Unfortu-
nately, under tension, these sutures do not hold, the bowel retracts,
and a raw canal remains which forms a stricture. Such infants survive
their lesion and the surgery, but lifelong prob1ems.m defecation and,
particularly, in fecal incontinence are initiated,
The foregoing, by and large, refers to males only. In the female,
a rectovaginal fistula is commonly associated with an imperforate anus.
This prevents intestinal obstruction and may be managed as a fistula
by a variety of surgical maneuvers. Males frequently have fist&s
also, but these are between the blind pouch of the colon and some
portion of the lower urinary tract. Such associated anomalies can be
corrected at the time of surgery for imperforate anus by either ap-
proach.
                    REFERENCES
1. Bilderback, J. B., and Rosenblatt, M. S.: Acute intestinal obstruction caused
  by clamping of intestine in umbilical cord clamp, Arm. Surg. H&4:146, 1946.
2. Bodian, M.: Hirschsprung's disease, Gastrointestology 76:339, 1950-51.
3. Bodian, M.; Carter, C. O., and Ward, B. C. H.: Hirschsprung's disease with
  radiological observations, Lancet 1:302, 1951.
4. Brunkow, C. W, et al.: Fetal meconium peritonitis; abscess; intestinal ob-
  struction; disappearance of radiopaque meconium bodies, West. J. Surg
  57 ~424, 1949.
5. Buchanan, D. J., and Rapoport, S.: Chemical comparison of normal meconium
  and meconium from a patient with meconium ileus, J. Pediat. 9:304, 1952.
6. Bush, J. A. K.; Lenox, C. C., and Myers, H. C.: Volvulus neonatorum, South-
  Surgeon 13:204, 1947.
7. Dannenberg, M., et ~2.: High gastrointestinal obstruction in newborn infant;
  radiological interpretation, J. Pediat. 37:380, 1950.


S.

9.

10.

11.


I 9.

13.

14.

15.

16.

Ii.

IS.

19.

20.
21.

22.

23.
24.

25.


26.


27.


28.

Farber, s.: GngeGtal atresia of the alimentary tract; diagnosis by micro-
smpic examination of the meconium, J.A.M.A. 100:1753, 1933.
F&CS, s.: PmmatiC insufficiencY and cehac syndrome, New England J.
hied. 229:682, 1943.

Farber, S.: Relation of pancreatic achylia to meconium ileus, J. &&at. 24:

387, 1944.

Forshall, I.; Rkhkm, P. P., and Mossman, D. B.: Functional intestinal ob
struction in the newborn, Arch. Dis. Childhood 26:123, 1951.
Gardner, C. E., Jr., and Hart, D.: Anomalies of intestinal rotation as a cause
of intestinal obstruction, Arch. Surg. 29942, 1934.
Gross, Il. E., and Chisholm, T. C.: Annular pancreas producing duodenal ob-
struction, Ann. Surg. 119:759, 1944.

Gross, R. E.; Holcomb, G. W., and Farber, S.: Duplications of the alimen-
tary tract, Pediatrics 9:449, 1952.

Raight, C., and Towsley, H. A.: Congenital atresia of esophagus with tracheo-
esophageal ilstula; extrapleural ligation of fist& and end-to-end anastomosis
of esophageal segments, Surg., Gynec. & Obst. 76:672, 1943.
Hiatt, R. B., and Wilson, P. E.: Celiac syndrome; therapy of meconium ileus;
report of 8 cases with review of literature, Surg., Gynec. & Obst. 87:317, 1948.
Hinden, E.: Meconium ileus with no pancreatic abnormality, Arch. Dis. Child-
hood 25:99, 1950.
Hoppu, K.: On the technique of intestinal anastomosis with regard to atresia
intestini treatment of the newborn, Ann. chir. et gynaec. Fenmae,
Supp. 1, 1951. (Also separately published, Helsinki, 1951.)     Vol. 40,
Hurwitt, E. S., and Amheim, E. E.: Meconiurn ileus associated with stenosis
of pancreatic ducts: clinical, pathologic and embryologic study, Am. J. Dis.
Child. 64:443, 1942.

Jeffrey, F. W.: Intussusception in newborn, Canad. M. A. J. 54:271, 1946.
F;;P;F~~E.: Certain aspects of surgery of the newborn, Delaware M. J. 24:
      .
$pif50E.: Tha,n&agement of imperforate anus, Pennsylvania M. J. 53:
Kodp, C. E.: Unpublished data.
KOOP, C. E., and Johnson, S.: Transthoracic repair of diaphragmatic hernia
in infants, Ann. Surg. 136:1007, 1952.
KOOP, C. E.; Perlingiero, J. G., and Weiss, W.: Cicatrizing enterocohtis in a
newborn infant, Am. J. hf. SC. 214:27, 1947.
Ladd, W. E., and Gross, R. E.: Abdominal Surgery
hood (Philadelphia: W. B. Saunders Company, 1941). of Infaw and Chi&
McIntosh, R., and Donovan, E. J.: Disturbances of rotation of intestinal tract,
Am. J. Dfs. Child. 57:116, 1939.
Moretz, W. H., and Morton, J. J.: Acute volvuIus of small intestine, Ann.
Surg. 132:899, 1950.

28a. Neale, A. V.: Obstetric-paediatrtc relationships, Proc. Roy. Sot. Med. 42:910,
  1949.

29. Neuhauser, E. B. D.: The roentgen diagnosis of fetal meconiurn peritonitis,
Am. J. Roentgenol. 51:421, 1944.
30. Overall, J. C., and Kirtley, J. A., Jr.: Intestinal obstruction in infant and
newborn, J. Tennessee M. A. 43:397, 1950.
31. Packard, G. B., and Reynolds, L. E.: Meconium peritonitis, Ann. Surg. 133:
548, 1951.


96                              C. Em KOOp
32. Potts, W. J.; Boggs, J. D., and White, H.: Intestinal obstruction in the new-
born infant due to agenesis of the myenteric plexes, Pediatrics 10253, 1952,
33. Rapoport, S., and Buchanan, D. J.: Composition of meconium: Isolation of
  blood-group-specific polysaccharides: Abnormal composition of meconium b
  meconium ileus, Science 112:150, 1950.
34. Rhoads, J. E.; Pipes, R. L., and Randall, J. P.: Simultaneous abdominal and
  perineal approach in operations for imperforate anus with atresia of rectum
  and rectosigmoid, Ann. Surg. 127:572, 1948.
35. Rhodes, R. L.: Further observations upon imperforate anus, Ann. Surg. 123:
  877, 1946.
36. Richmond, J. B, and Moore, H. R.: Abdominal abscess as cause of intestinal
  obstruction in newborn infant, J. Pediat. 31:343, 1947.
37. Rickham, P. P.: Personal communication.
38. Roberts, M. H.: Diagnosis of obstructive lesions of gastrointestinal tract of
  newborn infant, J. M. A. Georgia 39:320, 1950.
39. Shwachman, H.; Patterson, P. R., and Laguna, J.: Studies in pancreatic fibro-
  sis; a simple diagnostic gelatin film test for stool trypsin, Pediatrics 4:222,
   1949.
40. Starch, C.; Redner, B., and Turin, R. D.: Small bowel obstruction in infancy
  and childhood; radiological interpretation, J. Pediat. 35:386, 1949.
41. Swenson, 0.: End to end anastomosis of the esophagus for airesia of the
  esophagus, Surgery 22:324, 1947.
42. Swenson, 0.: Congenital megacolon, Pediatrics 8:542, 1951.
43. Swenson, 0.; Neuhauser, E. B. D., and Pickett, L. K.: New concepts of eti-
ology, diagnosis and treatment of congenital megacolon (Hirschsprung's dis-
  ease), Pediatrics 4:201, 1949.
44. Swenson, 0.; Rheinlander, H. F., and Diamond, I.: Hirschsprung's disease:
  New concept of etiology, New England J. Med. 241:551, 1949.
45. Wallgren, A.: Preclinical stage of infantile hypertrophic pyloric stenosis, Am.
  J. Dis. Child. 72:371, 1948.
46. Wangensteen, 0. H., and Rice, C. 0.: Imperforate anus-a method of deter-
  mining the surgical approach, Ann. Surg. 92:77, 1930.
47. Zuelzer, W. W., and Wilson, J. L.: Functional intestinal obstruction on @p
  genital neurogenic basis in infancy, Am. J. Dis. Child. 75:40, 1948.