Robert A. Drongowski

An analysis of factors contributing to the development of total parenteral nutrition-induced cholestasis

The risk of developmental of total parenteral nutrition (TPN)-associated cholestatic jaundice in neonates receiving intravenous hyperalimentation is high. Numerous factors have been cited as contributing to TPN cholestasis; however, the exact etiology remains obscure. This retrospective study was undertaken in order to identify any factors which might contribute to this syndrome. The hospital records of 172 neonates requiring TPN for a minimum of 1 week were reviewed. In addition, a subgroup of 32 infants requiring TPN for a minimum of 7 weeks was also examined. Cholestasis was defined as a direct serum bilirubin greater than 2.0 mg/dl during the course of TPN therapy. Significant factors for the development of cholestasis in both groups (n = 172, n = 32) included: number of operations (2.56 vs. 1.08, p = 0.0000), the number of days the patients received antibiotics (40.3 vs 12.9, p = 0.0000), and delayed start of enteral feedings (33.8 vs 14.1, p = 0.0000). Fifteen of the 32 patients who received TPN for at least 7 weeks did not develop cholestasis. In this subgroup (n = 32), there were no differences in birth weight, gestational age, days from birth to the start of TPN, or respiratory distress between those who developed cholestasis and those who remained anicteric. In contrast, there were significant differences between the cholestasis and noncholestasis groups in number of operations (13 vs 6, p = 0.0407), and days until enteral feedings were started (33.1 vs 18.9, p = 0.0289). This study suggests that the factor(s) contributing to the development of TPN-associated cholestasis are likely multifactorial. There appears to be a direct correlation between increasing severity of cholestatic jaundice and duration of TPN. This review does add a new parameter to the various causative factors suggested, namely the number of operative procedures. This new variable could be related to the stress of surgery itself or to the repeated administration of anesthetic agents.

Effect of probiotics on enterocyte bacterial translocation in vitro

Abstract Enteral probiotics such as Lactobacillus casei GG (LGG) have been used in the treatment of a variety of intestinal disorders in infants and children, including diarrhea, malabsorption, and Clostridium difficile colitis. We have previously demonstrated that the probiotic bacterium LGG has an inhibitory effect on bacterial translocation (BT) in a neonatal rabbit model. However, this in-vivo model is limited for investigating the cellular and molecular mechanisms responsible for probiotic inhibition of BT. The purpose of this study was to determine the efficacy of LGG in reducing the rate of Escherichia coliC25 (E. coli C25) translocation using an in-vitro enterocyte cell-culture model. Human colonic carcinoma (Caco-2) enterocytes were seeded in porous filters in the apical chamber of a two-chamber cell- culture system and grown for 14 days to confluence. The monolayers were incubated at 37 °C with LGG for 180 min. Non-adherent LGG was washed away prior to a 120-min incubation period with 105 CFU E. coli C25. E. coli that had translocated across the enterocyte monolayer were quantified by growing basal-chamber media samples on gram-negative bacteria-specific MacConkeys agar. In order to determine monolayer integrity, transepithelial electrical resistance (TEER) was measured across Caco-2 cells treated with LGG and E. coli. Statistical analysis was by ANOVA with P < 0.05 considered significant. LGG inhibited E. coli translocation at all LGG concentrations tested. The TEER ratio was not significantly altered by addition of LGG or E. coli (0.9 ± 0.03 vs 0.8 ± 0.05). These results demonstrate that the probiotic bacterium LGG inhibits BT of E. coli C25 in a dose-dependent manner in an in-vitro cell-culture model. This model should be valuable in investigating the cellular and molecular mechanisms involved in the inhibition of pathological enteral bacteria by probiotic agents.

Hirschsprung's disease in the newborn

Hirschsprungs disease has become a more common cause of newborn intestinal obstruction in the past 30 years. In a group of 137 newborns with intestinal obstruction the most common diagnosis was necrotizing enterocolitis. The second most common cause, however, was Hirschsprungs disease. We have reviewed 26 infants with aganglionosis who presented at 32 days of age or less between 1972 and 1978. The average age at presentation was 8.3 days. There were 21 males and five females. Five infants had long-segment or total colonic disease. There was only one premature infant (34 weeks, 1840 g). The mean birthweight in the series was 3.6 kg. Six children had a family history of a congenital anomaly (23%). Three of these had a family history of Hirschsprungs disease (12%). Nine infants (35%) had associated congenital anomalies. Four of these newborns had Downs syndrome, and all four had a cardiac anomaly as well. Fifteen newborns presented with emesis (58%) which was bilious in nine (35%) cases. Seventeen babies (65%) had abdominal distension at the time of presentation. Eleven infants passed a meconium stool by 24 hours of age (42%), and 15 had passed meconium by 48 hours (58%). Twenty-two of 24 (92%) barium enema examinations available prior to diagnosis were diagnostic of Hirschsprungs disease. All of the 23 suction rectal biopsies were positive. All 26 patients underwent a colostomy or ileostomy following diagnosis. There was no enterocolitis and no mortality. All 26 patients have had an endorectal pullthrough performed at a mean age of 11.8 months without major complications. This review highlights several features of Hirschsprungs disease, some of which refute, and some of which substantiate currently held concepts. The low incidence of prematurity and the increased incidence of Downs syndrome with aganglionosis is supported by this study. The high incidence (35%) of associated congenital anomalies has not been emphasized previously. Importantly, the passage of a meconium stool before 48 hours of life does not rule out Hirschsprungs disease.

The development of necrotizing enterocolitis following repair of gastroschisis: A suprisingly high incidence

We recently observed the development of necrotizing enterocolitis (NEC) in two consecutive newborn infants after gastroschisis repair. Because this association was unexpected, a retrospective review of our 11-year experience was performed using a multivariant computer analysis. The cohort consisted of 54 newborns with gastroschisis. All infants with omphalocele were excluded. Ten of 54 infants (18.5%) developed NEC as defined by classical clinical findings and pneumatosis intestinalis. Twenty-one distinct episodes of NEC occurred with up to three episodes (mean, 2.1) per patient. Twenty of the 21 episodes were successfully treated nonoperatively. Two infants developed pneumoperitoneum, one of whom underwent laparotomy upon which no perforation or intestinal infarction was found. Eight of the ten patients survived--a survival rate no different than for the remainder of the study group. Neither of the two deaths was attributable to NEC. The NEC was atypical in that no significant relationship was established with known predisposing conditions such as prematurity or low Apgar scores. Suspected risk factors such as time of feeding, type of closure, type of formula, total parenteral nutrition (TPN), and composition of TPN were not statistically related. Significant associations included concurrent TPN associated cholestatic liver disease in nine of the ten infants, antecedent intestinal surgery other than abdominal wall closure in five of the ten infants, and delay in initiation of enteral feedings (greater than 30 days) in eight of ten infants. In addition, the NEC occurred significantly later (range, 32 to 79 days; mean, 52 days) in the clinical course than usual; in fact, three of ten infants were rehospitalized with NEC following discharge. We conclude that a relationship exists between NEC and gastroschisis.(ABSTRACT TRUNCATED AT 250 WORDS)

The Kasai portoenterostomy for biliary atresia: a review of a 27-year experience with 81 patients

PURPOSE The aim of this study was to utilize clinical outcome methodology through multivariable analysis of perioperative factors to predict a successful Kasai-portoenterostomy (PE). METHODS Records of 81 patients treated for biliary atresia (BA) were reviewed. Outcome was defined as successful if the patient was alive and had no liver transplant (LT). To predict future successful or failed PE, patients were categorized at 6 months post-PE into 2 groups: Success: direct bilirubin (DB) less than 2.0 mg/dL; Failure: DB greater than 2 mg/dL, or the patient was listed/had undergone LT, or had died. Groups were analyzed for positive or negative predictive values (PPV, NPV) at 2 and 5 years after PE. Cox regression was used to determine risk factors for PE. RESULTS PE was successful in 38% and failed in 62%. PPV of future success was 96% at 2 years post-PE and 95% at 5 years post-PE, NPV of failure was 76% and 74%, respectively. Bridging liver fibrosis at the time of PE and postoperative cholangitic episodes were interdependent risk factors for a failed PE (P <.05). Other covariates showed no significant relationship for PE outcome. CONCLUSION Classifying of patients 6 months postoperatively allowed us to determine a successful PE outcome. Bridging liver fibrosis at the time of the Kasai, and the increased number of postoperative cholangitic episodes were predictive of a poor PE outcome.

Acute kidney injury in congenital diaphragmatic hernia requiring extracorporeal life support: an insidious problem

PURPOSE Patients with congenital diaphragmatic hernia (CDH) requiring extracorporeal life support (ECLS) are at increased risk for acute kidney injury (AKI). We hypothesized that AKI would be associated with increased mortality. We further hypothesized that vasopressor requirement, nephrotoxic medications, and infections would be associated with AKI. METHODS We performed a retrospective chart review in all patients with CDH requiring ECLS from 1999 to 2009 (n = 68). Patient variables that could potentiate renal failure were collected. We used a rise in creatinine from baseline by the RIFLE (risk, 1.5×; injury, 2×; failure, 3×; loss; and end-stage renal disease) criteria to define AKI. Statistical analysis was performed via SPSS (SPSS, Chicago, IL) using Student t test and χ(2) analysis, with P < .05 being considered significant. RESULTS Survival to hospital discharge was 37 (54.4%) of 68. Acute kidney injury was identified in 48 (71%) of 68 patients, with 15 (22% of all patients) qualifying as injury and 33 (49% of all patients) qualifying as failure by the RIFLE criteria. Patients who qualified as failure by the RIFLE criteria had a significant decrease in survival (27.3% with failure vs 80% without failure; P = .001). Patients who qualified as failure also had increased length of ECLS (314 ± 145 vs 197 ± 115 hours; P = .001) and decreased ventilator-free days in the first 60 days (1.39 ± 5.3 vs 20.17 ± 17.4 days; P = .001). There was no significant difference in survival when patients qualified as risk or injury. CONCLUSIONS This is the first report using a systematic definition of AKI in patients with CDH on ECLS. There is a high incidence of AKI in these patients, and when it progresses to failure, it is associated with higher mortality, increased ECLS duration, and increased ventilator days. This highlights the importance of recognizing AKI in patients with CDH requiring ECLS and the potential benefit of preventing progression of AKI or early intervention.

Intragastric pressure measurement: A guide for reduction and closure of the silastic chimney in omphalocele and gastroschisis

In newborn infants with omphalocele or gastroschisis, traditional criteria for reduction of the herniated viscera either primarily or after application of a Silastic chimney have been the babys color, respiratory rate, and lower extremity turgor. These are not always accurate or immediately apparent. In order to define more objective guidelines for reduction, measurements of intragastric pressure through a gastrostomy tube using a water manometer were carried out. The validity of this pressure measurement was demonstrated in five puppies where intra-abdominal pressure correlated well with inferior vena cava pressure and intragastric pressure measured through a gastrostomy tube (R = .98 and .99, respectively). Over a 3.5-yr period, 25 newborn infants with omphalocele (9) or gastroschisis (16) were treated. Ten underwent primary closure, and 15 were treated by placement of a Silastic chimney with serial reduction and closure. Manual reductions were performed once or twice daily to a maximum intragastric pressure of 20 cm water. Greater pressures demonstrated cardiovascular and respiratory comprise both experimentally and clinically. The mean time required for removal of the Silastic chimney was 4.7 days. There were no infections related to the chimney. There were 2 early and 5 late deaths, a 28% mortality rate. The remaining patients are alive and well. Intragastric pressure measurement in patients with omphalocele or gastroschisis provides objective criteria for safe primary closure and Silastic chimney reduction, shortens the time of reduction, and reduces the number of associated circulatory, respiratory, and septic complications.

Contribution of Demographic and Environmental Factors to the Etiology of Gastroschisis: A Hypothesis

We examined the clinical literature on congenital abdominal wall defects to confirm our impression that gastroschisis had become more common than omphalocele. We then examined the teratology literature and noted that congenital abdominal wall defects were frequently induced by teratogens. This lead us to review the antenatal history of 19 infants with gastroschisis and 54 control infants born with a congenital anomaly unrelated to gastroschisis. When compared to controls, mothers of infants with gastroschisis were more likely to have used aspirin during pregnancy and to have been taking oral contraceptives at the time of conception. Additionally, an increased incidence of illegal drug use (particularly cocaine) was noted among the study mothers. We conclude that gastroschisis is becoming the more common congenital abdominal wall defect, and that it could be related to exposure to an environmental teratogen.

Fundoplication in 160 children under 2 years of age

The natural history of fundoplication in young children with gastroesophageal reflux (GER) had not been analyzed previously. The authors reviewed the charts of 160 children who underwent gastric fundoplication (GF) before the age of 2 years (mean age [+/- SD], 9 +/- 7 months; range, 1 week to 2 years), from 1974 to 1992. Reflux was documented by upper gastrointestinal series in 124 patients, by 24-hour pH probe monitor in 98 patients, and by both in 68 patients. Clinical indications for GF included failure to thrive (FTT) in 68%, emesis (Ems) in 58%, and aspiration pneumonia (Asp) in 53%. Neurological impairment (NI) was present in 47% of all patients, and 13% had esophageal atresia (EA). The type of GF used was a Nissen fundoplication in 79% and an anterior fundoplication (AF) in 21%. Of the 160 patients, 24 (15%) died of unrelated causes. Of the remaining 136, follow-up of at least 2 years was obtained for 96 (mean follow-up period, 5.3 +/- 3.0 years; range, 2 to 15 years). Clinical resolution of symptoms/findings after GF occurred in 87% of children with FTT, 92% with Ems, 70% with Asp, and 71% overall. A second fundoplication was required for 15 children (16%) because of documented recurrent reflux. The type of GF, the age of the patient, and the presence of EA or NI did not significantly affect the success of GF.(ABSTRACT TRUNCATED AT 250 WORDS)

Use of Cholecystokinin to Prevent the Development of Parenteral Nutrition-Associated Cholestasis

BACKGROUND Neonates are at high risk for the development of parenteral nutrition-associated cholestasis when receiving a prolonged course of total parenteral nutrition (TPN). Although this cholestasis is of unknown etiology, it may result from a lack of gastrointestinal hormone formation, including cholecystokinin, which normally occurs after enteral feedings. METHODS Two groups of neonates were studied. The treatment group consisted of 21 consecutive, prospectively enlisted neonates receiving TPN for > 14 days. The nontreatment group consisted of 21 infants from the 2 years preceding the study who were matched to the treatment group by gestational age, diagnosis, and duration of TPN. The major outcome determinant was direct bilirubin. Cholestasis was defined as a direct bilirubin > 2.0 mg/dL and was considered severe if the direct bilirubin was > 5.0 mg/dL after other causes were ruled out. RESULTS The mean direct bilirubin levels in the nontreated group progressively rose over time, whereas the mean direct bilirubin the treated group remained level. The incidence of infants with a direct bilirubin > 2.0 mg/dL was 24% and 43% in the CCK+ and CCK- groups, respectively, and was not significant (p = .14). The percentage of infants with a direct bilirubin > 5.0 mg/dL was 9.5% and 38% in the treatment and nontreatment groups, respectively, and was significant, p = .015. CONCLUSIONS Levels of direct bilirubin were lower in the treated compared with the nontreated group. These findings suggest that cholecystokinin prophylaxis in high-risk neonates may help prevent the development of parenteral nutrition-associated cholestasis.