- Email: [email protected]
The significance of gastric emptying in children with intestinal malrotation
The Significance of Gastric Emptying in Children With Intestinal Malrotation By Stephen G. Jolley, William P. Tunell, Sharlotte Thomas, Jan Young, and E. Ide Smith Oklahoma City, Oklahoma 9 Persistent gastrointestinal symptoms are common postoperatively in children with intestinal malrotation. W e investigated this problem in 14 children with intestinal malrotation who had a Ladd procedure (3 patients), gastroschisis repaired (6 patients), or omphalocele repaired (5 patients) between one month and 15 years prior to study. In 13 patients, gastric emptying was measured at 30 minutes (%GE30) and at 60 minutes (%GE60) following ingestion of 99m-Tc sulfur colloid in apple juice. W e estimated the degree of gastric peristalsis using the %GE corrected for immediate postcibal gastroesophageal reflux (corrected %GE). Patients with vomiting exhibited slow gastric emptying compared to patients without vomiting ( % G E 3 0 : 1 4 . 0 _+ 5,5 v 32.5 _+ 4.2, P < .005). The slow gastric emptying was related to slow gastric peristalsis (corrected % G E 3 0 : 2 0 . 3 ~+ 5.0 v 47.1 _+ 6.0, P < .005). In all 5 patients with persistent bloating and diarrhea, gastric p e r i s t a l s i s w a s rapid at 3 0 m i n u t e s ( c o r r e c t e d %GE30 = 56.7 _+ 4.2) and at 60 minutes (corrected %GE60 = 69.5 +_ 5.3). To assess the role of gastroesophageal reflux (GER) in persisting symptoms, all children had extended (18 to 24 hours) esophageal pH monitoring. Eleven (79%) of the 14 patients demonstrated GER by esophageal pH monitoring, including four of six patients without reflux symptoms. All ten children under t w o years of age demonstrated GER regardless of symptoms or congenital anatomic abnormality. In conclusion, GER is common in patients under t w o years of age with intestinal malrotation, but clinical symptoms seem related more to extreme variations in gastric peristalsis than to GER. 9 1985 by Grune & Stratton, Inc. INDEX WORDS: Gastric emptying; gastric peristalsis; gastroesophageal reflux; sudden infant death syndrome.
NTESTINAL MALROTATION is usually asso-
I ciated with acute problems when discovered by the clinician. The acute problem may be duodenal obstruc-
tion, with or without midgut volvulus, or related to an associated congenital abnormality such as gastroschisis, omphalocele, and diaphragmatic hernia. Chronic problems with intermittent vomiting, abdominal pain, bloating, or constipation have been reported in 18% to 38% of patients operated upon for intestinal malrotation as the only congenital abnormality. l The chronic problems of intestinal malrotation associated with repaired gastroschisis or repaired omphalocele have not been delineated clearly. Barium gastrointestinal studies in these latter children usually show nonrotated bowel without duodenal obstruction. 2 Clinically, some patients with intestinal malrotation have had symptoms related to gastroesophageal reflux (GER) that are significant enough to require antireflux surgery.3'4 Journal of Pediatric Surgery, Vol 20, No 6 (December), 1985: pp 627-631
In the present study, we describe a group of patients with intestinal malrotation who were selected for study because of chronic symptoms. A comprehensive evaluation was performed in these patients and in a small group of asymptomatic patients to assess the relationship of abnormal gastroesophageal function to chronic symptoms. MATERIALS AND METHODS Fourteen children between the ages of one month and 18 years with intestinal malrotation were studied at Oklahoma Children's Memorial Hospital. The study was approved by the Institutional Review Board, University of Oklahoma Health Sciences Center, and informed consent was obtained from the patients' parents or legal guardians. There were nine boys and five girls in the study group. Each patient had a Ladd procedure (three patients), gastroschisis repaired (six patients), or omphalocele repaired (five patients) one month to 15 years prior to study. Five of the nine patients with a gastrostomy tube placed at the initial operation still had a tube in place when studied. Two patient s with omphalocele and one patient with gastroschisis had a staged repair of the abdominal wall defect. Additional abnormalities present in four patients included perinatal asphyxia (one patient), hydrocephalus (one patient), Cantrell's pentalogy (one patient), and repaired small bowel atresia (one patient). The response by the patients' parents to a standardized questionnarie indicated the presence of chronic symptoms in 12 of the 14 children. Two patients were completely asymptomatic. Repeated episodes of nonbilious vomiting were reported in six patients (two Ladd procedures, one repaired gastroschisis, and three repaired omphaloceles). Troublesome postcibal bloating and watery diarrhea were reported in five patients (one Ladd procedure and four repaired gastroschisis). The diarrhea was usually unassociated with infections and did not consist of stools with elevated levels of reducing substances. Small bowel biopsies were not performed. No patient reported both vomiting and bloating with diarrhea. Significant growth retardation was present in five patients (two Ladd procedures, one repaired gastroschisis, and two repaired omphaloceles), and repeated respiratory symptoms (choking or wheezing) were present in three patients (one Ladd procedure and two repaired omphaloceles). The presence of gastric outlet obstruc-
From the Department of Surgery, University of Oklahoma College of Medicine, Oklahoma Oty. Supported in part by Children's Medical Research, Oklahoma Oty. Presented before the 16th Annual Meeting of the American Pediatric Surgical Association, Kohala Coast, Hawaii, May 1-4, 1985. Address reprint requests to William P. Tunell, MD, Section of Pediatric Surgery, University of Oklahoma Health Sciences Center, PO Box 26307, Oklahoma City, OK 73126. 9 1985 by Grune & Stratton, Inc. 0022-3468/85/2006-001 l $03.00/0 627
628
JOLLEY ET AL
tion or duodenal obstruction was excluded in each patient with a barium upper gastrointestinal series. 700 S Malrotation
Esophageal pH Monitoring Continuous monitoring of distal esophageal pH for 18 to 24 hours was performed in each patient to document the presence of GER. The pH score, the pattern of reflux, the mean duration of sleep reflux (ZMD), and the percent time pH < 4 during the immediate two hours after apple juice feedings were extracted from the pH recording as reported previously. 5-8 The ZMD was used to assess the relationship of respiratory symptoms to GER.
Radionuclide Gastric Emptying Measurement of gastric emptying was possible in 13 of the 14 patients. All patients were studied after a four-hour fasting period and within 24 hours of the esophageal pH study. The method utilized 99m-Tc sulfur colloid in 300 m L / m 2 BSA of apple juice (radioactive concentration = 0.5 uCi/mL) taken orally. Serial gamma camera images were taken over the gastric fundus and body areas for 60 minutes following ingestion of the apple juice. Gastric emptying was expressed as the percentage of radionuclide emptied from the monitored stomach area after 30 minutes (%GE30) and after 60 minutes (%GE60) compared to time zero, and with appropriate correction for background and decay. No patient vomited during the gastric emptying study.
Estimation of Gastric Peristalsis Gastric peristalsis was estimated by correcting the measured gastric emptying for postcibal gastroesophageal reflux (Corrected % GE). Simply stated, this measurement represented the predicted gastric emptying if the patient had no postcibal gastroesophageal reflux. The correction was possible because of an inverse linear relationship between measured gastric emptying and immediate postcibal gastroesophageal reflux in asymptomatic children with a normal pH score (unpublished observations). A linear regression with variance estimates for measured gastric emptying and postcibal gastroesophageal reflux was derived from the group of asymptomatic children with a normal pH score to obtain the following formulasg: A = % time esophageal pH < 4 while upright for two hours after apple juice feedings. Corrected % GE30 = ((4.1(% GE30 - 38.1 + 0.35A))/ (3.6(1.1 + ((A - 33.6)2/7336.7))1/2)) + 38.1 Corrected % GE60 = ((10.5(% GE60 - 55.2 + 0.38A))/ (9.3(1.1 + ((A - 33.6)2/7336.7))1/2)) + 55.2
Statistical Analysis The unpaired Student's t-test was used to compare the patient subgroups. All values are expressed as the mean • standard error of the mean (SEM). RESULTS
Eleven of the 14 patients demonstrated GER by extended esophageal pH monitoring, including four of six patients without reflux symptoms. All five children studied with a gastrostomy tube in place had GER. The ten children less than two years of age had GER regardless of the congenital anatomic abnormality
o Gastroschisis
6OO
z~ Omphalocele
500 o rl
2: 04
400
o
3o0
-r
8
2 3 4 5 6 Age (Years)
7 8 9 10 11 14 16 18
Fig 1. The relationship between age and gastroesophageal reflux (GER} as determined by a pH score obtained from 18- to 24-hour esophageal pH recordings in 14 children with intestinal malrotation. The shaded area represents the normal range for the pH score (< 64}, which is calculated from recording intervals beginning after the two hour postcibal period.
(Fig 1). The pattern of GER was type I in five infants, type II in four infants, and type III in one infant. None of the three patients with repeated respiratory symptoms had a prolonged ZMD, whereas four patients without respiratory symptoms had a prolonged ZMD. Medical antireflux therapy was recommended in the latter group of patients, but in one asymptomatic child with repaired gastroschisis and no other anatomic abnormality, the recommendation was not followed by the family. This single child died suddenly three weeks after study with postmortem findings characteristic of the sudden infant death syndrome (SIDS). A Nissen fundoplication was performed in one patient with a repaired omphalocele who had an abnormal pH score, growth retardation, vomiting, and subsequent documentation of advanced esophagitis by endoscopy. The patients with persistent nonbilious vomiting had slow gastric emptying compared to patients without vomiting (% GE30:14.0 _+ 5.5 v 32.5 _+ 4.2, P < .005 and % GE60:30.2 _+ 8.8 v 44.7 _+ 4.3, P < .05). Slow gastric peristalsis appeared to be the reason for slow gastric emptying (Fig 2) rather than the extent of GER (pH score: 183.0 _+ 55.1 v 272.0 ___90.2, NS). The presence of a tube gastrostomy did not decrease gastric peristalsis (Corrected % GE30 = 42.4 +_ 7.6 and Corrected % GE60 = 58.6 _ 6.9). In the patients with troublesome postcibal bloating and diarrhea, gastric emptying was not significantly faster than in patients without bloating or diarrhea (% GE30:
GASTRIC EMPTYING WITH INTESTINAL MALROTATION
629 9 Malrotation 100 --
o Gastroschisis
I-" O < .005 " 1
A Omphalocele
r~p
< .02--.- I
90o
80O 70uJ (.9 6 0 -
o
Q ~ W
o
~
-o 50
8 t-
O o
40 Fig 2. Estimation of gastric peristalsis in intestinal rnalrotation patients w i t h nonbilious vomiting (N ~ 5) and w i t h no vomiting (N = 8| using gastric emptying at 30 minutes (% GE30) and at 60 minutes (% GE60) corrected for postcibal gastroesophageal reflux. The horizontal bars represent the mean and the vertical bars the standard error of the mean,
"+
20 10
I
I
I
I
No Vomiting
Vomiting
No Vomiting
associated with vomiting as was slow gastric peristalsis. Therefore, the logical treatment of patients with chronic vomiting should be directed at the improvement of gastric peristalsis in addition to medical antireflux therapy. Another chronic problem associated with gastric peristalsis was postcibal bloating and diarrhea. In these patients, gastric peristalsis appeared to be rapid. At least two situations may explain the association of bloating and diarrhea with rapid gastric peristalsis in intestinal malrotation. The presence of rapid gastric peristalsis may be an indicator of generalized gut hypermotility. Another possibility may include an impairment of fluid absorption in the small and large bowel, which is triggered by rapid and discoordinated gastric emptying. The latter explanation is particularly appealing in the patients with repaired gastroschisis who have previously thickened and foreshortened bow-
Chronic gastrointestinal symptoms associated with intestinal malrotation pose a challenge in the longterm management of three children. The present study delineates two upper alimentary tract abnormalities related to repeated nonbilious vomiting in the absence of gastric outlet obstruction. Incompetence of the gastroesophageal junction was a constant finding in children under two years of age and, although present in most patients with vomiting, was not as clearly
100
r--P < .005--1
o
Vomiting
DISCUSSION
oGastroschisis
o
0
34.8 + 4.5 v 18.7 _+ 5.6, NS and % GE60:45.8 _+ 2.8 v 35.0 _+ 7.0, NS). However, gastric peristalsis was rapid in the patients with bloating and diarrhea (Fig 3). Respiratory symptoms were not clearly related to abnormalities in gastric emptying and peristalsis, but the child with SIDS findings showed both slow gastric emptying and slow gastric peristalsis.
9 Malrotation
o
z~ e
30
= Omphalocele i'~
P < .05
9
90 8o
o LU 70
O
o
(5
+
LLI
~
o
o
60 "O
o o ~
50
~
a0 2C
10
o
I Bloating & Diarrhea
I No Bloating or Diarrhea
I Bloating & Diarrhea
I No Bloating or Diarrhea
Fig 3. Estimation of gastric peristalsis in intestinal malrotation patients with postcibal bloating and w a t e r y diarrhea (N = 5) and with no bloating or diarrhea (N ~ 8) using gastric emptying at 30 minutes (% GE30} and at 60 minutes (% GE60) corrected for postcibal gastroesophagaal reflux, The horizontal bars represent the mean and the vertical bars the standard error of the mean,
630
JOLLEY E-I" AL
el. Regardless of the mechanism, some relief of the bloating and diarrhea can be achieved by reducing the free carbohydrate load of feedings and by providing smaller, more frequent feedings. A comment should be made regarding the single child with uncomplicated repair of gastroschisis who became a SIDS victim. Despite the absence of vomiting and respiratory symptoms, this patient demonstrated slow gastric emptying with slow gastric peri-
stalsis, and a type I pattern of GER with a prolonged ZMD. Such findings should alert the clinician about the possibility of sudden death in an otherwise asymptomatic patient. In conclusion, GER is a common finding in children under two years of age with intestinal malrotation, but chronic vomiting and bloating with diarrhea seem related more to extreme variations in gastric peristalsis than to GER.
REFERENCES 1. Bill AH: Malrotation of the intestine, in Ravitch MM, Welch K J, Benson CD, et al (eds): Pediatric Surgery. Chicago, Year Book Medical, 1982, pp 912-923. 2. Touloukian R J, Spackman T J: Gastrointestinal function and radiographic appearance following gastroschisis repair. J Pediatr Surg 6:427-434, 1971
3. Johnson DG, Jolley SG, Herbst J J, et al: Surgical selection of infants with gastroesophageal reflux. J Pediatr Surg 16:587--594, 198l 4. Pennell RC, Lewis JE, Cradock TV, et al: Management of severe gastroesophageal reflux in children. Arch Surg 119:553-557, 1984 5. Jolley SG, Johnson DG, Herbst J J, et al: An assessment of
gastroesophageal reflux in children by extended pH monitoring of the distal esophagus. Surgery 84:16-24, 1978 6. Jolley SG, Herbst J J, Johnson DG, et al: Patterns of postcibal gastroesophageal reflux in symptomatic infants. Am J Surg 138:946-950, 1979 7. Jolley SG, Herbst J J, Johnson DG, et al: Esophageal pH monitoring during sleep identifies children with respiratory symptoms from gastroesophageal reflux. Gastroenterol 80:1501-1506, 1981 8. Jolley SG, Tunell WP, Carson JA, et al: The accuracy of abbreviated esophageal pH monitoring in children. J Pediatr Surg 19:848-854, 1984 9. Huntsberger DV, Leaverton PE: Regression and correlation, in Huntsberger DV, Leaverton PE (eds): Statistical Inference in the Biomedical Sciences. Boston, Allyn & Bacon, 1970, pp 167-189.
Discussion Eric Fonkalsrud (Los Angeles): During the past few years, improved isotope techniques for calculating gastric emptying with incorporation of technetium colloid into solid as well as liquid feedings have shown that as many as 50% of infants and children with symptomatic reflux may have this problem. This is not limited to children with malrotation, and is particularly common in patients with neurologic disorders, who experience symptoms of reflux. This has occasionally led to the "gas-bloat" problem in children following fundoplication, who may develop a type of "closed loop" obstruction with the fundoplication at one end and slow gastric emptying at the other. We have had to . reoperate to perform a pyloroplasty on eight children who had previous fundoplication, and have performed simultaneous pyloroplasty and fundoplication on 11 children. Another 14 have had pyloroplasty alone, for gastric dysmotility. Three hundred and eight patients have had fundoplication alone. Thus, in over 11% of our overall refluxing children, we have performed a pyloroplasty. We now think of reflux as a syndrome, and have found that an algorithm using esophageal pH monitoring, gastric emptying studies, (including those with solid feeds), and esophageal manometry, can be quite helpful in sorting out which patient's needs which procedure(s). Although not a definitive way to approach this problem, it's the best we have found. I
would ask the authors if, in their large experience with reflux, they have seen dysmotility in children other than those with malrotation.? Robert Touloukian (New Haven): Malrotation appears to me to be the weakest link in establishing a common identity between omphalocele, gastroschises, and congenital duodenal bands and drawing any differences or similarities on that basis is a temptation I would try to avoid. Most of the babies with bloating and diarrhea followed gastroschisis repair, raising the possibility of mild to severe malabsorption of carbohydrate, but also of other nutrients. Transit time and sodium and potassium absorption studies would be important to further characterize this group. Also, has a lactose-free diet been attempted to try to reverse some of the clinical manifestations? The long-term outlook in the gastroschisis is encouraging as demonstrated by our series of patients. GI function recovered and patients were growing at the 25th to 97th percentile within one year. To test for the possible effects of gastric compression, is there any correlation between attempting a primary closure or staged repair in your symptomic patients with omphalocele or gastroschisis ? Have contrast studies shown either the liver overlying the stomach or some positional abnormality of the stomach that may help to account for the delayed gastric emptying in the omphalocele group? Our expe-
GASTRIC EMPTYING WITH INTESTINAL MALROTATION
rience with vomiting after a Ladd procedure always makes us concerned about gastroesophageal reflux related to the previous, but now relieved, obstruction. Hopefully, some recovery of lower esophageal sphincter tone may be anticipated after the primary operation. Since so many patients were studied during the first year, I would be very interested in learning of a late follow-up in the medically treated group. Michael Matlak (Salt Lake City): What's the etiology of the gastrointestinal distress that occasionally follows corrective surgery for malrotation? Dr Jolley and his colleagues have started to objectively address this question by studying esophageal and gastric physiology; but I have at least three major concerns regarding the authors' study design and conclusions. First, only two end organs were evaluated--the stomach and the esophagus. What about the small intestine? What was its functional and anatomic status? Second, the authors have made intestinal malrotation the common denominator in this study. From a clinical standpoint, however, we are dealing with three different diseases. In mid-gut volvulus, the primary event is malrotation; but in omphalocele and gastroschisis, malrotation is simply an associated anomaly, rather than the primary event. Each of these study groups should be studied separately, rather than lumping them together based upon rotational states or symptoms. Third, if we look at the subset with gastroschisis and malrotation, we're dealing with a dynamic, perhaps transient physiologic process. Were any of these children studied longitudinally to see if the gastric emptying changed or improved with time? I suggest that we look a little bit lower than the gastroesophageal junction or gastric antrum for the answer to this perplexing problem. Stephen Jolley (closing): Dr Fonkalsrud, we have found the same things in our children with gastroesophageal reflux in Oklahoma that you talked about. We're seen groups of children who have very slow
631
gastric peristalsis and those who have very rapid gastric peristalsis. We feel that this is something that should be examined and watched with regard to some of the bloating problems that these children have postoperatively--not only gas bloating which may require pyloroplasty to avoid, but also in children who have dumping symptoms, which is very difficult to deal with postoperatively without changing diets. Dr Touloukian, I think you hit the nail on the head in that these children are a rather heterogenous group. The children with repaired gastroschisis have very thickened and foreshortened small bowel early on in life and have abnormal small bowel absorption. One would presume that they would have difficulty absorbing nutrients. One of the things that surprised us about these patients is that they emptied their stomachs rather quickly. We think that is probably a factor in triggering the malabsorption problems, although not the total problem in producing the bloating and diarrhea. The numbers of patients that we studied with staged closures v primary closure was only three, so the numbers were really too small to evaluate that as a factor in some of the chronic symptoms. Dr Matlak, your comments were very much similar to Dr Touloukian's, in that the answer to a majority of the gastrointestinal problems in these children is farther downstream, in the small intestine and perhaps even the large bowel. That's something that, hopefully, we'll look at in the future as we are better able to define small bowel peristalsis. The three study groups are heterogeneous, and part of the reason is that its's very difficult to get enough children back who are otherwise doing okay for these kinds of studies. That's one of the reasons why our study group was selected toward those with chronic gastrointestinal symptoms. We have not done longitudinal studies on these children, with the exception of one patient in whom slow gastric peristalsis improved after a 4-month period.
NTESTINAL MALROTATION is usually asso-
I ciated with acute problems when discovered by the clinician. The acute problem may be duodenal obstruc-
tion, with or without midgut volvulus, or related to an associated congenital abnormality such as gastroschisis, omphalocele, and diaphragmatic hernia. Chronic problems with intermittent vomiting, abdominal pain, bloating, or constipation have been reported in 18% to 38% of patients operated upon for intestinal malrotation as the only congenital abnormality. l The chronic problems of intestinal malrotation associated with repaired gastroschisis or repaired omphalocele have not been delineated clearly. Barium gastrointestinal studies in these latter children usually show nonrotated bowel without duodenal obstruction. 2 Clinically, some patients with intestinal malrotation have had symptoms related to gastroesophageal reflux (GER) that are significant enough to require antireflux surgery.3'4 Journal of Pediatric Surgery, Vol 20, No 6 (December), 1985: pp 627-631
In the present study, we describe a group of patients with intestinal malrotation who were selected for study because of chronic symptoms. A comprehensive evaluation was performed in these patients and in a small group of asymptomatic patients to assess the relationship of abnormal gastroesophageal function to chronic symptoms. MATERIALS AND METHODS Fourteen children between the ages of one month and 18 years with intestinal malrotation were studied at Oklahoma Children's Memorial Hospital. The study was approved by the Institutional Review Board, University of Oklahoma Health Sciences Center, and informed consent was obtained from the patients' parents or legal guardians. There were nine boys and five girls in the study group. Each patient had a Ladd procedure (three patients), gastroschisis repaired (six patients), or omphalocele repaired (five patients) one month to 15 years prior to study. Five of the nine patients with a gastrostomy tube placed at the initial operation still had a tube in place when studied. Two patient s with omphalocele and one patient with gastroschisis had a staged repair of the abdominal wall defect. Additional abnormalities present in four patients included perinatal asphyxia (one patient), hydrocephalus (one patient), Cantrell's pentalogy (one patient), and repaired small bowel atresia (one patient). The response by the patients' parents to a standardized questionnarie indicated the presence of chronic symptoms in 12 of the 14 children. Two patients were completely asymptomatic. Repeated episodes of nonbilious vomiting were reported in six patients (two Ladd procedures, one repaired gastroschisis, and three repaired omphaloceles). Troublesome postcibal bloating and watery diarrhea were reported in five patients (one Ladd procedure and four repaired gastroschisis). The diarrhea was usually unassociated with infections and did not consist of stools with elevated levels of reducing substances. Small bowel biopsies were not performed. No patient reported both vomiting and bloating with diarrhea. Significant growth retardation was present in five patients (two Ladd procedures, one repaired gastroschisis, and two repaired omphaloceles), and repeated respiratory symptoms (choking or wheezing) were present in three patients (one Ladd procedure and two repaired omphaloceles). The presence of gastric outlet obstruc-
From the Department of Surgery, University of Oklahoma College of Medicine, Oklahoma Oty. Supported in part by Children's Medical Research, Oklahoma Oty. Presented before the 16th Annual Meeting of the American Pediatric Surgical Association, Kohala Coast, Hawaii, May 1-4, 1985. Address reprint requests to William P. Tunell, MD, Section of Pediatric Surgery, University of Oklahoma Health Sciences Center, PO Box 26307, Oklahoma City, OK 73126. 9 1985 by Grune & Stratton, Inc. 0022-3468/85/2006-001 l $03.00/0 627
628
JOLLEY ET AL
tion or duodenal obstruction was excluded in each patient with a barium upper gastrointestinal series. 700 S Malrotation
Esophageal pH Monitoring Continuous monitoring of distal esophageal pH for 18 to 24 hours was performed in each patient to document the presence of GER. The pH score, the pattern of reflux, the mean duration of sleep reflux (ZMD), and the percent time pH < 4 during the immediate two hours after apple juice feedings were extracted from the pH recording as reported previously. 5-8 The ZMD was used to assess the relationship of respiratory symptoms to GER.
Radionuclide Gastric Emptying Measurement of gastric emptying was possible in 13 of the 14 patients. All patients were studied after a four-hour fasting period and within 24 hours of the esophageal pH study. The method utilized 99m-Tc sulfur colloid in 300 m L / m 2 BSA of apple juice (radioactive concentration = 0.5 uCi/mL) taken orally. Serial gamma camera images were taken over the gastric fundus and body areas for 60 minutes following ingestion of the apple juice. Gastric emptying was expressed as the percentage of radionuclide emptied from the monitored stomach area after 30 minutes (%GE30) and after 60 minutes (%GE60) compared to time zero, and with appropriate correction for background and decay. No patient vomited during the gastric emptying study.
Estimation of Gastric Peristalsis Gastric peristalsis was estimated by correcting the measured gastric emptying for postcibal gastroesophageal reflux (Corrected % GE). Simply stated, this measurement represented the predicted gastric emptying if the patient had no postcibal gastroesophageal reflux. The correction was possible because of an inverse linear relationship between measured gastric emptying and immediate postcibal gastroesophageal reflux in asymptomatic children with a normal pH score (unpublished observations). A linear regression with variance estimates for measured gastric emptying and postcibal gastroesophageal reflux was derived from the group of asymptomatic children with a normal pH score to obtain the following formulasg: A = % time esophageal pH < 4 while upright for two hours after apple juice feedings. Corrected % GE30 = ((4.1(% GE30 - 38.1 + 0.35A))/ (3.6(1.1 + ((A - 33.6)2/7336.7))1/2)) + 38.1 Corrected % GE60 = ((10.5(% GE60 - 55.2 + 0.38A))/ (9.3(1.1 + ((A - 33.6)2/7336.7))1/2)) + 55.2
Statistical Analysis The unpaired Student's t-test was used to compare the patient subgroups. All values are expressed as the mean • standard error of the mean (SEM). RESULTS
Eleven of the 14 patients demonstrated GER by extended esophageal pH monitoring, including four of six patients without reflux symptoms. All five children studied with a gastrostomy tube in place had GER. The ten children less than two years of age had GER regardless of the congenital anatomic abnormality
o Gastroschisis
6OO
z~ Omphalocele
500 o rl
2: 04
400
o
3o0
-r
8
2 3 4 5 6 Age (Years)
7 8 9 10 11 14 16 18
Fig 1. The relationship between age and gastroesophageal reflux (GER} as determined by a pH score obtained from 18- to 24-hour esophageal pH recordings in 14 children with intestinal malrotation. The shaded area represents the normal range for the pH score (< 64}, which is calculated from recording intervals beginning after the two hour postcibal period.
(Fig 1). The pattern of GER was type I in five infants, type II in four infants, and type III in one infant. None of the three patients with repeated respiratory symptoms had a prolonged ZMD, whereas four patients without respiratory symptoms had a prolonged ZMD. Medical antireflux therapy was recommended in the latter group of patients, but in one asymptomatic child with repaired gastroschisis and no other anatomic abnormality, the recommendation was not followed by the family. This single child died suddenly three weeks after study with postmortem findings characteristic of the sudden infant death syndrome (SIDS). A Nissen fundoplication was performed in one patient with a repaired omphalocele who had an abnormal pH score, growth retardation, vomiting, and subsequent documentation of advanced esophagitis by endoscopy. The patients with persistent nonbilious vomiting had slow gastric emptying compared to patients without vomiting (% GE30:14.0 _+ 5.5 v 32.5 _+ 4.2, P < .005 and % GE60:30.2 _+ 8.8 v 44.7 _+ 4.3, P < .05). Slow gastric peristalsis appeared to be the reason for slow gastric emptying (Fig 2) rather than the extent of GER (pH score: 183.0 _+ 55.1 v 272.0 ___90.2, NS). The presence of a tube gastrostomy did not decrease gastric peristalsis (Corrected % GE30 = 42.4 +_ 7.6 and Corrected % GE60 = 58.6 _ 6.9). In the patients with troublesome postcibal bloating and diarrhea, gastric emptying was not significantly faster than in patients without bloating or diarrhea (% GE30:
GASTRIC EMPTYING WITH INTESTINAL MALROTATION
629 9 Malrotation 100 --
o Gastroschisis
I-" O < .005 " 1
A Omphalocele
r~p
< .02--.- I
90o
80O 70uJ (.9 6 0 -
o
Q ~ W
o
~
-o 50
8 t-
O o
40 Fig 2. Estimation of gastric peristalsis in intestinal rnalrotation patients w i t h nonbilious vomiting (N ~ 5) and w i t h no vomiting (N = 8| using gastric emptying at 30 minutes (% GE30) and at 60 minutes (% GE60) corrected for postcibal gastroesophageal reflux. The horizontal bars represent the mean and the vertical bars the standard error of the mean,
"+
20 10
I
I
I
I
No Vomiting
Vomiting
No Vomiting
associated with vomiting as was slow gastric peristalsis. Therefore, the logical treatment of patients with chronic vomiting should be directed at the improvement of gastric peristalsis in addition to medical antireflux therapy. Another chronic problem associated with gastric peristalsis was postcibal bloating and diarrhea. In these patients, gastric peristalsis appeared to be rapid. At least two situations may explain the association of bloating and diarrhea with rapid gastric peristalsis in intestinal malrotation. The presence of rapid gastric peristalsis may be an indicator of generalized gut hypermotility. Another possibility may include an impairment of fluid absorption in the small and large bowel, which is triggered by rapid and discoordinated gastric emptying. The latter explanation is particularly appealing in the patients with repaired gastroschisis who have previously thickened and foreshortened bow-
Chronic gastrointestinal symptoms associated with intestinal malrotation pose a challenge in the longterm management of three children. The present study delineates two upper alimentary tract abnormalities related to repeated nonbilious vomiting in the absence of gastric outlet obstruction. Incompetence of the gastroesophageal junction was a constant finding in children under two years of age and, although present in most patients with vomiting, was not as clearly
100
r--P < .005--1
o
Vomiting
DISCUSSION
oGastroschisis
o
0
34.8 + 4.5 v 18.7 _+ 5.6, NS and % GE60:45.8 _+ 2.8 v 35.0 _+ 7.0, NS). However, gastric peristalsis was rapid in the patients with bloating and diarrhea (Fig 3). Respiratory symptoms were not clearly related to abnormalities in gastric emptying and peristalsis, but the child with SIDS findings showed both slow gastric emptying and slow gastric peristalsis.
9 Malrotation
o
z~ e
30
= Omphalocele i'~
P < .05
9
90 8o
o LU 70
O
o
(5
+
LLI
~
o
o
60 "O
o o ~
50
~
a0 2C
10
o
I Bloating & Diarrhea
I No Bloating or Diarrhea
I Bloating & Diarrhea
I No Bloating or Diarrhea
Fig 3. Estimation of gastric peristalsis in intestinal malrotation patients with postcibal bloating and w a t e r y diarrhea (N = 5) and with no bloating or diarrhea (N ~ 8) using gastric emptying at 30 minutes (% GE30} and at 60 minutes (% GE60) corrected for postcibal gastroesophagaal reflux, The horizontal bars represent the mean and the vertical bars the standard error of the mean,
630
JOLLEY E-I" AL
el. Regardless of the mechanism, some relief of the bloating and diarrhea can be achieved by reducing the free carbohydrate load of feedings and by providing smaller, more frequent feedings. A comment should be made regarding the single child with uncomplicated repair of gastroschisis who became a SIDS victim. Despite the absence of vomiting and respiratory symptoms, this patient demonstrated slow gastric emptying with slow gastric peri-
stalsis, and a type I pattern of GER with a prolonged ZMD. Such findings should alert the clinician about the possibility of sudden death in an otherwise asymptomatic patient. In conclusion, GER is a common finding in children under two years of age with intestinal malrotation, but chronic vomiting and bloating with diarrhea seem related more to extreme variations in gastric peristalsis than to GER.
REFERENCES 1. Bill AH: Malrotation of the intestine, in Ravitch MM, Welch K J, Benson CD, et al (eds): Pediatric Surgery. Chicago, Year Book Medical, 1982, pp 912-923. 2. Touloukian R J, Spackman T J: Gastrointestinal function and radiographic appearance following gastroschisis repair. J Pediatr Surg 6:427-434, 1971
3. Johnson DG, Jolley SG, Herbst J J, et al: Surgical selection of infants with gastroesophageal reflux. J Pediatr Surg 16:587--594, 198l 4. Pennell RC, Lewis JE, Cradock TV, et al: Management of severe gastroesophageal reflux in children. Arch Surg 119:553-557, 1984 5. Jolley SG, Johnson DG, Herbst J J, et al: An assessment of
gastroesophageal reflux in children by extended pH monitoring of the distal esophagus. Surgery 84:16-24, 1978 6. Jolley SG, Herbst J J, Johnson DG, et al: Patterns of postcibal gastroesophageal reflux in symptomatic infants. Am J Surg 138:946-950, 1979 7. Jolley SG, Herbst J J, Johnson DG, et al: Esophageal pH monitoring during sleep identifies children with respiratory symptoms from gastroesophageal reflux. Gastroenterol 80:1501-1506, 1981 8. Jolley SG, Tunell WP, Carson JA, et al: The accuracy of abbreviated esophageal pH monitoring in children. J Pediatr Surg 19:848-854, 1984 9. Huntsberger DV, Leaverton PE: Regression and correlation, in Huntsberger DV, Leaverton PE (eds): Statistical Inference in the Biomedical Sciences. Boston, Allyn & Bacon, 1970, pp 167-189.
Discussion Eric Fonkalsrud (Los Angeles): During the past few years, improved isotope techniques for calculating gastric emptying with incorporation of technetium colloid into solid as well as liquid feedings have shown that as many as 50% of infants and children with symptomatic reflux may have this problem. This is not limited to children with malrotation, and is particularly common in patients with neurologic disorders, who experience symptoms of reflux. This has occasionally led to the "gas-bloat" problem in children following fundoplication, who may develop a type of "closed loop" obstruction with the fundoplication at one end and slow gastric emptying at the other. We have had to . reoperate to perform a pyloroplasty on eight children who had previous fundoplication, and have performed simultaneous pyloroplasty and fundoplication on 11 children. Another 14 have had pyloroplasty alone, for gastric dysmotility. Three hundred and eight patients have had fundoplication alone. Thus, in over 11% of our overall refluxing children, we have performed a pyloroplasty. We now think of reflux as a syndrome, and have found that an algorithm using esophageal pH monitoring, gastric emptying studies, (including those with solid feeds), and esophageal manometry, can be quite helpful in sorting out which patient's needs which procedure(s). Although not a definitive way to approach this problem, it's the best we have found. I
would ask the authors if, in their large experience with reflux, they have seen dysmotility in children other than those with malrotation.? Robert Touloukian (New Haven): Malrotation appears to me to be the weakest link in establishing a common identity between omphalocele, gastroschises, and congenital duodenal bands and drawing any differences or similarities on that basis is a temptation I would try to avoid. Most of the babies with bloating and diarrhea followed gastroschisis repair, raising the possibility of mild to severe malabsorption of carbohydrate, but also of other nutrients. Transit time and sodium and potassium absorption studies would be important to further characterize this group. Also, has a lactose-free diet been attempted to try to reverse some of the clinical manifestations? The long-term outlook in the gastroschisis is encouraging as demonstrated by our series of patients. GI function recovered and patients were growing at the 25th to 97th percentile within one year. To test for the possible effects of gastric compression, is there any correlation between attempting a primary closure or staged repair in your symptomic patients with omphalocele or gastroschisis ? Have contrast studies shown either the liver overlying the stomach or some positional abnormality of the stomach that may help to account for the delayed gastric emptying in the omphalocele group? Our expe-
GASTRIC EMPTYING WITH INTESTINAL MALROTATION
rience with vomiting after a Ladd procedure always makes us concerned about gastroesophageal reflux related to the previous, but now relieved, obstruction. Hopefully, some recovery of lower esophageal sphincter tone may be anticipated after the primary operation. Since so many patients were studied during the first year, I would be very interested in learning of a late follow-up in the medically treated group. Michael Matlak (Salt Lake City): What's the etiology of the gastrointestinal distress that occasionally follows corrective surgery for malrotation? Dr Jolley and his colleagues have started to objectively address this question by studying esophageal and gastric physiology; but I have at least three major concerns regarding the authors' study design and conclusions. First, only two end organs were evaluated--the stomach and the esophagus. What about the small intestine? What was its functional and anatomic status? Second, the authors have made intestinal malrotation the common denominator in this study. From a clinical standpoint, however, we are dealing with three different diseases. In mid-gut volvulus, the primary event is malrotation; but in omphalocele and gastroschisis, malrotation is simply an associated anomaly, rather than the primary event. Each of these study groups should be studied separately, rather than lumping them together based upon rotational states or symptoms. Third, if we look at the subset with gastroschisis and malrotation, we're dealing with a dynamic, perhaps transient physiologic process. Were any of these children studied longitudinally to see if the gastric emptying changed or improved with time? I suggest that we look a little bit lower than the gastroesophageal junction or gastric antrum for the answer to this perplexing problem. Stephen Jolley (closing): Dr Fonkalsrud, we have found the same things in our children with gastroesophageal reflux in Oklahoma that you talked about. We're seen groups of children who have very slow
631
gastric peristalsis and those who have very rapid gastric peristalsis. We feel that this is something that should be examined and watched with regard to some of the bloating problems that these children have postoperatively--not only gas bloating which may require pyloroplasty to avoid, but also in children who have dumping symptoms, which is very difficult to deal with postoperatively without changing diets. Dr Touloukian, I think you hit the nail on the head in that these children are a rather heterogenous group. The children with repaired gastroschisis have very thickened and foreshortened small bowel early on in life and have abnormal small bowel absorption. One would presume that they would have difficulty absorbing nutrients. One of the things that surprised us about these patients is that they emptied their stomachs rather quickly. We think that is probably a factor in triggering the malabsorption problems, although not the total problem in producing the bloating and diarrhea. The numbers of patients that we studied with staged closures v primary closure was only three, so the numbers were really too small to evaluate that as a factor in some of the chronic symptoms. Dr Matlak, your comments were very much similar to Dr Touloukian's, in that the answer to a majority of the gastrointestinal problems in these children is farther downstream, in the small intestine and perhaps even the large bowel. That's something that, hopefully, we'll look at in the future as we are better able to define small bowel peristalsis. The three study groups are heterogeneous, and part of the reason is that its's very difficult to get enough children back who are otherwise doing okay for these kinds of studies. That's one of the reasons why our study group was selected toward those with chronic gastrointestinal symptoms. We have not done longitudinal studies on these children, with the exception of one patient in whom slow gastric peristalsis improved after a 4-month period.