Delivery room repair of gastroschisis.

BACKGROUND: Outcome for most abdominal wall defects is related to the presence or absence of additional anomalies or prematurity. In gastroschisis, outcome is almost as closely related to the severity of the inflammatory "peel" on bowel that is thought to result from direct contact with amniotic fluid. Improving eviscerated bowel quality would be expected to reduce morbidity in these patients. METHODS: From 1986 to 1991, 32 patients with the antenatal diagnosis of gastroschisis were treated. All were delivered by cesarean section; 13 surgical repairs were made immediately in the delivery room. Surgical repairs in 19 patients were made at less than 6 hours of age after transfer from the delivering hospital to the pediatric surgery center. RESULTS: Thirty percent of infants who underwent surgical repair in delivery room and 32% of infants who underwent urgent surgical repair were either premature or had significant associated anomalies. Seventy-three percent of delivery room repair group had fascial repairs compared with 37% in the transferred group. When infants more than 34-weeks' gestation without associated anomalies are compared with transferred infants, delivery room repair group underwent more frequent fascial repair (8 of 9 vs 5 of 13, p < 0.03), were extubated sooner (2.9 vs 7.4 days, p < 0.04), tolerated enteral feedings earlier (8.1 vs 22.2 days, p < 0.009), and required fewer hospital days (13.6 vs 31.3 days, p < 0.01). Eviscerated bowel of infants who underwent immediate surgical repair lacked the characteristic matted, edematous, and fibrinous coated appearance seen in transferred patients. CONCLUSIONS: Immediate delivery room repair of gastroschisis results in increased fascial repairs and earlier extubation, feeding, and hospital discharge. These benefits appear to be due to the minimal reactive peel on eviscerated bowel at birth.

Delayed repair of congenital diaphragmatic hernia.

Early reports of improved survival in newborns with congenital diaphragmatic hernias (CDH) utilizing extracorporeal membrane oxygenation (ECMO) and/or a delayed repair (DR) approach have been tempered by recent failures to document such an improvement. We have used ECMO to salvage emergently repaired patients with CDH since January 1984. From January 1990 to January 1991, we treated 16 CDH patients with mechanical ventilation and other supportive techniques until persistent pulmonary hypertension of the newborn resolved. We compared this DR group to 19 patients emergently repaired from February 1987 to December 1989. Of the 19 emergently repaired patients, 16 had a best post ductal (BPD) PO2 > 50 mm Hg. Eight patients survived (42 per cent of all and 50 per cent of those with a BPD PO2 > 50). Thirteen required ECMO and six of these survived. Five of six ECMO survivors had significant ECMO complications, and ECMO was "inappropriately" performed on three of six nonsurvivors. Of the 16 DR patients, nine achieved a BPD PO2 > 50 mm Hg and seven survived (44 per cent of all and 78 per cent of those with a BPD PO2 > 50). Seven required ECMO and four of these survived. All survivors in both groups had a BPD PO2 > 90 mm Hg. Delayed repair did not improve survival statistics for CDH with early respiratory distress at our hospital but it has allowed stratification of potential survivors, fewer inappropriate ECMO cannulations, and many fewer ECMO complications.

ECMO without heparin: laboratory and clinical experience.

To evaluate the feasibility of long-term extracorporeal membrane oxygenation (ECMO) without heparin, we placed six lambs on standard venoarterial ECMO for 71 to 96 hours. Group 1 (3 animals) was given doses of heparin to maintain activated clotting times (ACT) greater than 400 seconds. No form of anticoagulant was used for the three animals in group 2. Blood flow was maintained at 60 mL/kg/min. No histological evidence of thrombosis was noted at necropsy. ACT, prothrombin time, and partial thromboplastin time were higher in group 1, and much lower, although still above normal in group 2. Fibrinogen was significantly lower in group 2 (75 +/- 35 v 219 +/- 64 mg/dL group 1), and, although the platelet count was lower in group 2 (142 +/- 76 x 10(3)/mm3 v 225 +/- 167 x 10(3)/mm3), it was clinically acceptable. These results encouraged us to discontinue heparin when faced with severe hemorrhage in four patients on ECMO, rather than withdraw support at a time when there was little chance of survival. Heparin was discontinued for 10.5 +/- 6 hours. The mean ACT was reduced from 220 +/- 23 seconds to 144 +/- 22 seconds. One patient, who required repair of gastric necrosis while on ECMO following repair of a congenital diaphragmatic hernia, survived and had a decrease in blood loss from 2 to 0 mL/kg/h after the heparin was discontinued. One of the three patients who died had an autopsy with no evidence of thrombosis. We conclude that it may be reasonable to discontinue heparin in the face of life-threatening hemorrhage while on ECMO.

Protamine selectively inhibits collagen synthesis by human intestinal smooth muscle cells and other mesenchymal cells.

Collagen synthesis is a major function of human intestinal smooth muscle (HISM) cells and contributes to intestinal fibrosis in chronic inflammatory bowel disease. As an extension of previous in vitro studies of the role of heparin in regulating HISM cell proliferation and collagen synthesis, the effect of protamine sulfate was studied. Protamine decreased collagen production by 50% in confluent and proliferating cultures. This effect was concentration-dependent and was selective for collagen in that neither noncollagen production nor DNA accumulation in the culture plates was affected. Other human mesenchymal cells which produce collagen, such as dermal fibroblasts and aortic smooth muscle cells, responded to protamine in a similar fashion. Protamine has a strong cationic charge and is rich in lysine and arginine. To determine which of these properties was important in decreasing collagen production, the effect of protamine was compared to that of other polyionic compounds. Poly-L-lysine decreased collagen production to a lesser degree than protamine. Poly-L-arginine was toxic to the cells. Poly-L-glutamic acid, which has an opposite charge to protamine, had no effect. These findings suggest that both the number and the arrangement of lysyl residues, in addition to positive charge, are important. Binding assays demonstrated that protamine did not inhibit collagen production by binding to ascorbate in the culture medium. Electrophoretic separation and chromatography of collagen types expressed following protamine treatment showed that the ratio of type I to type III collagen remained 2:1. This observation suggests that suppression of collagen production is not specific to a particular collagen type. The selective inhibition of collagen production by protamine provides an important tool to study the regulation of collagen production in human cells and may also provide potential therapy of fibrotic disorders.

Transthoracic Nissen fundoplication for gastroesophageal reflux in patients with severe kypho-roto-scoliosis.

Transthoracic Nissen fundoplication was used for the correction of gastroesophageal reflux in five mentally retarded patients with severe kypho-roto-scoliosis. This deformity may result in a gastroesophageal junction so high above the left costal margin that transabdominal fundoplication is extremely difficult. Operative morbidity was minimal. One wrap disruption occurred that required reoperation. Results have been satisfactory 6 to 24 months postoperatively.

Use of the Greenfield filter in adolescents for deep vein thrombosis and pulmonary embolism.

Serious venous thromboembolic disease is now recognized more frequently in the pediatric age group. Caval interruption is indicated most commonly for prophylaxis against life-threatening or recurrent pulmonary embolism (PE) when anticoagulation is ineffective or contraindicated. Greenfield vena caval filters have been utilized locally in 415 adult patients with 97% long-term patency and 5% recurrent embolization. Its application in adolescents is reported herein. Standard adult (30-mm) vena caval filters were placed in ten patients, ages 13 to 18. Four filters were required following PE, six were used for deep venous thrombosis (DVT) when anticoagulation was contraindicated, and one was inserted prophylactically. In eight patients, filter insertion was accomplished with local anesthesia, while two underwent filter placement under general anesthesia administered for other procedures. One filter was misplaced into the right renal vein, requiring a second filter insertion. All patients have been followed from 1 to 11 years with yearly vascular duplex imaging and radionuclide venograms documenting caval patency without clinical embolic sequelae. This duplicates the adult experience in safety and efficacy. As recognition of venous thromboembolism becomes more frequent in the pediatric age group, safe caval interruption may be necessary for those excluded from or not responsive to anticoagulation. This technique may be extended to smaller patients with miniaturization of both filter and carrier.

Mortality with extracorporeal membrane oxygenation following repair of congenital diaphragmatic hernia in 93 infants.

The results of extracorporeal membrane oxygenation (ECMO) following repair of congenital diaphragmatic hernia (CDH) are analyzed for 93 neonates reported to the Neonatal ECMO Registry. Each infant was symptomatic at birth and developed life-threatening hypoxemia after herniorrhaphy. ECMO was used after neonates met criteria predictive of death in individual centers. These included (1) failure of medical therapy to reverse hypoxemia (35 neonates); (2) acute clinical deterioration making death appear likely (37 neonates); (3) AaDO2 greater than 600 for 8 hours (13 neonates); (4) oxygen index score of 51 for 4.8 hours (5 neonates); and (5) cardiac arrest (1 neonate). With ECMO, 52 infants (58%) survived and were discharged. Pre-ECMO arterial blood gas analysis was not predictive of outcome. Mortality was higher in small premature infants. ECMO corrected the hypoxemia, which usually causes death following congenital diaphragmatic herniorrhaphy. Unfortunately, 47 major bleeding complications occurred with 29 deaths. Therefore, improved survival may depend on new methods designed to prevent bleeding.

Heparin modulates human intestinal smooth muscle cell proliferation, protein synthesis, and lattice contraction.

The effect of heparin on human intestinal smooth muscle cell proliferation, collagen and noncollagen protein synthesis, and collagen lattice contraction was studied in vitro. Proliferation of serum-stimulated cells was inhibited in a concentration-related fashion by continuous exposure to heparin. The inhibition of proliferation was reversible when heparin was removed from the culture medium. Collagen synthesis was inhibited by 24-h exposure to heparin, but only during that phase of culture (8-12 days) when collagen synthesis was maximal. Noncollagen protein synthesis was down-regulated by 24-h exposure to heparin at all phases of culture tested (5-21 days). Heparin also abolished the contraction of collagen lattices populated by human intestinal smooth muscle cells. These studies demonstrate that heparin plays a significant role in the modulation of human intestinal smooth muscle cell behavior in vitro and suggest that a similar role is played by heparinlike components of the extracellular matrix in vivo.

Extracorporeal membrane oxygenation following repair of congenital diaphragmatic hernias.

From 1981 through 1986, 8 newborns with congenital diaphragmatic hernia required herniorrhaphy in the first 8 hours of life. Extracorporeal membrane oxygenation (ECMO) was employed in 7 after they met local criteria predictive of 95% mortality. These criteria were an alveolar-postductal arterial oxygen gradient greater than 600 mm Hg for 12 hours or hemodynamic instability. Four of these 7 patients had unremitting hypoxemia after herniorrhaphy (no "honeymoon" period), 3 of whom survived. One additional patient died, producing a mortality of 29%. ECMO used for 68 to 241 hours (mean, 163 hours) provided reliable oxygenation in all. Deaths resulted from disseminated intravascular coagulation and bleeding, and bleeding and pulmonary failure after ligation of a patent ductus arteriosus. Complications occurred in 6 patients and included bleeding (3), hernia recurrence (3), and air embolism (1). Follow-up ranging from 1 year to 6 years after discharge of the 5 survivors shows normal growth and development in 4. The reported mortality without ECMO following congenital diaphragmatic herniorrhaphy in the first 8 hours of life ranges between 60 and 80%. While bleeding may present problems, survival of newborns with refractory hypoxemia after diaphragmatic repair has improved with ECMO.

Collagen synthesis by human intestinal smooth muscle cells in culture.

Growth kinetics and collagen production were determined in smooth muscle cells isolated from human jejunum and maintained in cell culture. Collagen synthesis increased during the first 15 days in culture at a time when the rate of cell proliferation was maximal. When confluent, these cells produced significantly more collagen than human dermal fibroblasts cultured under identical conditions. The smooth muscle cells required daily replenishment of ascorbate for maximal collagen synthesis. The types of collagen produced by human intestinal smooth muscle cells in culture were the same as those collagens extracted from strictured human bowel (types I, III, and V). These findings suggest that collagen production by human intestinal smooth muscle cells has a role in the repair as well as the fibrosis of the gastrointestinal tract.

Aorto-iliac aneurysms following umbilical artery catheterization.

Aneurysms of the aorta and its major branches are rare in the pediatric population and are usually associated with cardiovascular malformations. Recently, however, a number of children have emerged with aneurysms following umbilical artery catheterization (UAC). In the last 8 years, two such patients have presented at the Medical College of Virginia. These are reported in detail and compared with 18 others found in the literature (12 thoracic, 4 abdominal, 2 iliac). Several conclusions follow from the study of this group of 20 children: The history of umbilical artery catheterization in the presence of bacteremia, especially staphylococcal, is a prime risk factor in the development of aneurysms (90%, P less than .01). In addition, there was a concordance between the location of the catheter tip and the site of future aneurysm formation (P less than .05). UAC associated aneurysms are difficult to diagnose because of their rarity and have often been mistaken for tumors. The diagnosis should be suspected in the workup of a tumor in a child with an appropriate history. Though these aneurysms are probably mycotic, pus has not been encountered at surgery and there have been no reported graft infections. Patients with S aureus bacteremia complicating UAC should be followed closely to detect aneurysm formation.