Avoidance of leakage and strictures when creating an invisible conduit for antegrade colonic enemas.

BACKGROUND: Antegrade enemas have been proven to minimize problems with stool control for many incontinent patients; however, foreign bodies are unsightly, and native tissues have been shown to leak or stricture. METHODS: Using a limited laparotomy incision, the appendix or a rolled cecal tube was sutured to the apex of the everted umbilical skin using a V-shaped incision. The skin then was inverted to hide the mucosa. The cecum was plicated around the base of the appendix or cecal tube and then tested by filling the cecum with saline to a pressure of 30 cm H2O. RESULTS: Twenty patients (aged 4.8 to 41 years) with fecal incontinence caused by imperforate anus (17 cases) and other causes underwent this procedure. Two patients had minor strictures that resolved with home dilatations, and one cecal tube necrosed, and the patient has refused reoperation. One patient had minor prolapse and underwent revision. None of the conduits leak. Two patients achieved continence and stopped cannulating their stomas. With adjustment and customization of each enema regimen, stool accidents are infrequent. Follow-up is 22 +/- 14 (mean +/- SD) months. CONCLUSION: A catheterizeable colocutaneous conduit has been developed that allows for an invisible, leak-proof, and relatively stricture-free means through which antegrade enemas can be given.

Short-term complications of central line placement in children with the human immunodeficiency virus.

PURPOSE: The aim of this study was to characterize the perioperative complications of central venous catheter placement in children infected with human immunodeficiency virus (HIV) METHODS: A retrospective chart review was conducted of all central venous catheters placed by the surgical service into HIV-infected children from 1988 to 1998 at a large urban children's hospital. Complications occurring within 1 month of catheter placement were analyzed for several host and environmental factors. RESULTS: Forty HIV-positive patients underwent 60 central venous access procedures. Thirty-two of the patients were severely immunosuppressed. Eight catheter placements (13%) resulted in perioperative complications, including hemorrhage (n = 2), site infection (n = 2), catheter sepsis (n = 2), thrombotic occlusion (n = 1), and a pleural effusion secondary to catheter malposition (n = 1). Only 3 patients required catheter removal. There was no significant relationship between either hemophilia or thrombocytopenia and perioperative hemorrhage. No significant relationship was found between infectious complications and preoperative white blood cell count, absolute neutrophil count, CD4% and CD4#, suggesting that a patient's compromised immune status should not be considered a contraindication to central venous catheter placement. CONCLUSION: The complication rate of central venous catheter placement into HIV-infected children is low (<15%), but is still higher than that of the general pediatric population. With careful preoperative preparation this procedure can be performed safely, even in patients with advanced HIV disease. J Pediatr Surg 36:1777-1780.

Technical advances in pediatric laparoscopy have had a beneficial impact on splenectomy.

PURPOSE: The aim of this study was to demonstrate the effects of recent technical advances on the safety and benefits of pediatric laparoscopic splenectomy. METHODS: A retrospective review was conducted of patients undergoing laparoscopic splenectomy from January 1998 to January 2000. Technical advances utilized during this period included the harmonic scalpel, a specialized flexible hilar retractor, a larger, wire-rimmed specimen bag, and lateral patient positioning. RESULTS: Laparoscopic splenectomy was performed successfully on 18 patients aged 3 to 17 years (median, 9). The indications were hereditary spherocytosis (n = 10), idiopathic thrombocytopenic purpura (n = 5), and other (n = 3). Eight patients had concomitant procedures including cholecystectomy (n = 3), resection of an accessory spleen (n = 3), and other (n = 2). The median operating time, including the concomitant procedures, was 125 minutes (range, 70 to 235). Patients tolerated a regular diet on median postoperative day 1 (range, 1 to 3), and 16 were discharged home on or before postoperative day 2. None of the patients required blood product transfusion or conversion to an open technique. There were no complications, and all patients had returned to usual activity by 2 weeks. CONCLUSION: With recent technological advances, the laparoscopic approach has become easy to perform, safe, and should be considered the procedure of choice for pediatric splenectomy.

A standardized feeding regimen for hypertrophic pyloric stenosis decreases length of hospitalization and hospital costs.

BACKGROUND/PURPOSE: Before the institution of a standardized feeding regimen (SFR) for hypertrophic pyloric stenosis (HPS) at the authors' institution, the postoperative feeding regimen and, thus, the length of hospitalization for HPS patients was variable. The aim of this study was to evaluate whether a SFR would affect the length of hospitalization or hospital costs for HPS patients. METHODS: A 5-year retrospective analysis was performed on 242 patients who underwent pyloromyotomy via a standard right upper quadrant incision. The length of hospitalization and hospital costs were compared in these patients before and after the institution of a standardized postoperative feeding regimen. RESULTS: The SFR decreased total length of hospitalization by 19.4% (3.1 days v2.5 days, P = .002), postoperative length of stay by 21% (1.9 days v 1.5 days, P< .001), total costs by 11.9% (P= .05), direct costs by 7.7% (P= .22), and indirect costs by 18.6% (P= .003). This occurred despite a small increase in costs per day. The SFR did not change the complication rate (5.3% before SFR v6.1% after SFR, P = 1.0). CONCLUSION: A postoperative standardized feeding regimen for patients with HPS decreased length of hospitalization and hospital costs without adverse effects.

Venous ischemia as a cause of ureteral necrosis in transplanted ureters.

BACKGROUND: Urologic complications after pediatric renal transplantation can adversely effect the outcome and may result in decreased graft survival. Efforts to prevent these complications are worthwhile. This study investigates the incidence of these complications in a clinical transplant program and reports on an animal model used to investigate one possible cause. METHODS: In the clinical study, the results of a pediatric renal transplant program at a large children's hospital for a 5(1/2)-year period were reviewed with special attention paid to patients suffering ureteral necrosis. In the experimental study, 9 swine underwent laparotomy, bilateral complete infrahilar ureteric dissection, and extravesical ureteroneocystostomy. On the left side only, the renal and adrenal veins were ligated. The arterial supply remained intact. The right side did not undergo vessel ligation and served as the control. Three pigs each were killed at 3, 8, and 15 days. Kidneys, ureters and a cuff of bladder were examined histologically. RESULTS: In the clinical study 75 renal transplants were performed with a total of 5 cases of early ureteral necrosis. Two of these 5 displayed venous congestion and ischemia, and 2 were associated with kidneys displaying primary nonfunction of the graft. Seventy-one of 75 grafts are continuing to function. One of the 4 early graft losses also had an ischemic ureter. In the experimental study all right kidneys and ureters were normal. All left kidneys had complete hemorrhagic necrosis. Necrosis also was found in 5 of 9 proximal left ureters and in 7 of 9 distal left ureters. Viable left ureters displayed moderate to severe submucosal and periureteric hemorrhage. Four of 9 ureters displayed more damage distally than proximally. The extent of necrosis was similar at 3, 8, and 15 days. CONCLUSION: In both clinical and experimental studies, venous congestion and subsequent ischemia have been shown to be important causes of ureteral necrosis after renal transplantation.

The umbilical fold approach to pyloromyotomy: is it a safe alternative to the right upper-quadrant approach?

BACKGROUND: Hypertrophic pyloric stenosis has been approached using two standard incisions for pyloromyotomy: the circumumbilical (UMB) and the right upper quadrant (RUQ). The UMB approach produces an almost undetectable scar but has been associated with more complications. STUDY DESIGN: A 5-year retrospective analysis was performed on 344 patients (90 UMB and 254 RUQ) to compare technical and wound complications. The effect of prophylactic antibiotics on wound infection was also evaluated. RESULTS: The intraoperative complication rate was 5.5% (13.3% UMB versus 2.8% RUQ; p = 0.001). The mucosal perforation rate was 3.5% (8.9% UMB versus 1.6% RUQ; p = 0.003). Mucosal perforations increased the mean +/- SD length of hospitalization in UMB patients (3.9 +/- 0.8 versus 2.4 +/- 1.1 days; p < 0.001). The serosal tear rate was 2.0% (4.4% UMB versus 1.2% RUQ; p=0.08). The postoperative complication rate was 5.8% (14.4% UMB versus 2.8% RUQ; p < 0.001), and the wound infection rate was 2.6% (6.7% UMB versus 1.2% RUQ; p = 0.01). Antibiotic prophylaxis decreased the rate of wound infection to 1.8% and eliminated the statistical difference between the groups (4.5% UMB versus 0% RUQ; p=0.16). The rate of other postoperative complications was 3.2% (7.8% UMB versus 1.6% RUQ; p = 0.009). Duration of hospital stay did not differ between the groups overall (2.6+/-1.12 days for UMB versus 2.7+/-1.5 days for RUQ; p = 0.35). CONCLUSIONS: The UMB approach to pyloromyotomy was cosmetically superior but increased complication rates. Technical complications were easily corrected and length of stay was not affected. Wound infections were decreased in both groups by the use of prophylactic antibiotics.

Neurogenic bladder in infants born with anorectal malformations: comparison with spinal and urologic status.

BACKGROUND/PURPOSE: Spinal dysraphism and neurovesical dysfunction (NVD) frequently are associated in children with anorectal malformations (ARM). This study compares the urodynamic data from a selected group of patients with the results of their spinal and urologic imaging studies. METHODS: Twenty-six children (20 with isolated imperforate anus and six with persistent cloacal malformations) were investigated. All patients were evaluated with leak point pressures (LPP), renal ultrasound scan, and voiding cystourethrography. Eight children had urodynamics performed before and after posterior sagittal anorectoplasty (PSARP). The spinal cord was assessed using ultrasonography or magnetic resonance imaging. Current urologic status was obtained to provide long-term follow-up. RESULTS: Twenty-one of 26 children demonstrated elevated LPPs above the established normal value of 40 cm H2O, and 15 of these children had normal spinal imaging study findings. Uroradiographic studies findings showed that 12 of 21 children with elevated LPPs had hydronephrosis or vesicoureteral reflux with seven of these patients having normal spinal cords. LPPs in the eight patients with pre- and postoperative studies were 74 +/- 14.7 cm H2O and 68 +/- 31.8 cm H2O (mean +/- SD), respectively. CONCLUSIONS: These urodynamic and radiographic data confirm that NVD (elevated LPP) is common in patients with anorectal malformations despite normal spinal cords. Bladder dysfunction does not appear to be a sequelae of a properly performed PSARP. Patients with ARM and any uroradiographic or clinical urologic abnormality should undergo urodynamic testing even though the spinal studies are normal.

Intestinal stoma complications in immunocompromised children.

BACKGROUND/PURPOSE: When performing an urgent gastrointestinal operation on an immunocompromised child, the pediatric surgeon may have to decide between performing an intestinal anastomosis (and risk leakage or sepsis) or creating an intestinal stoma. This study evaluates the postoperative course of those patients treated with intestinal stomas and the long-term survival rate of such patients. METHODS: A 13-year retrospective review of immunocompromised children with intestinal stomas was performed. Patients were assessed as to their diagnosis, indication for surgery, stoma type, postoperative complications (within 30 days of surgery), ostomy-related complications, and survival. RESULTS: 19 stomas (8 ileostomies and 11 colostomies) were created in 18 patients. Six children had immunodeficiency disorders; 12 were immunosuppressed from chemotherapy treatment for cancer. Indications for surgery included infectious complications (n = 8); neoplasm-induced bowel obstruction, perforation, or invasion (n = 10); and Hirschsprung's disease (n = 1). Postoperative complications occurred in 13 cases (68%); two warranted reoperation. Four of six patients with neutropenia had serious postoperative infectious complications. Stoma complications occurred in 6 cases (32%); 1 required revision. All 3 patients in whom bleeding developed from their stoma site were thrombocytopenic. Nine of 18 patients (50%) died, yet no patient died of complications attributable to their stomas. Of the surviving 9 children, 6 underwent stoma takedown at a mean of 19 months after creation; 1 has a permanent colostomy, and 2 currently are undergoing chemotherapy. CONCLUSION: Although immunocompromised children who require intestinal stomas frequently die of their underlying illnesses and their stomas often produce considerable morbidity, stoma creation does not jeopardize their chance of survival.

Definitive localization of isolated tracheoesophageal fistula using bronchoscopy and esophagoscopy for guide wire placement.

PURPOSE: To aid in identification of isolated tracheoesophageal fistulas (TEF), many surgeons have recommended the bronchoscopic placement of a ureteric or Fogarty catheter. This method can fail because of intraoperative dislodgment of the catheter. The authors present a new technique that enables us to definitively isolate and treat all H-type fistulas. METHODS: Six cases of isolated TEF are presented consisting of 4 H-type fistulas, a proximal pouch fistula, and a recurrent TEF. Three of the patients had undergone a total of four prior failed operations at outside institutions using attempted bronchoscopic catheter placement. On all six patients, bronchoscopy was first performed where the fistula tract was noted in the trachea and a guide wire was passed through the fistula. After orotracheal intubation, the authors performed rigid esophagoscopy; the guide wire was identified and brought out through the mouth. This created a wire loop through the fistula. With the use of x-ray we were then able to visualize the level of the fistula and determine whether a cervical or thoracic approach should be used. Identification of the fistula intraoperatively was then facilitated by traction on the loop by the anesthesiologist. RESULTS: Five of the six TEFs were repaired with neck exploration; one required right thoracotomy. In all patients, the fistula was identified and divided. There were no recurrences or other complications. CONCLUSION: This new technique is a simple and definitive method in identification and treatment of isolated TEF.

Risk factors for early infection of central venous catheters in pediatric patients.

BACKGROUND: In an effort to avoid infections that can lead to the premature removal of indwelling central venous catheters (CVCs), the surgical technique and host factors present in pediatric recipients of permanent CVCs were reviewed. STUDY DESIGN: All patients receiving CVCs over a 17-month period were identified. Those patients with fever and positive blood cultures drawn through the CVC within 45 days of line placement were labeled as having early infection. A case-control design was used to select two control patients for each infected patient. Charts from both the infection and control groups were reviewed for several factors present at the time of CVC placement, including fever, neutropenia (absolute neutrophil count [ANC] < 500 and ANC < 1,000), use of perioperative antibiotics, diagnosis, CVC site, and type of CVC. Chi-square test with Yates correction was used to compare the groups. Odds ratios (ORs) and 95% confidence intervals were derived. RESULTS: Among the 473 CVCs placed, early infections developed in 53 patients (12%). The control group consisted of 106 patients. Neutropenia was present in 16 of 53 infected patients versus 8 of 106 controls (p = 0.004, OR = 5.30). Perioperative antibiotics were given to 25 of 53 infected patients versus 72 of 106 controls (p = 0.02, OR = 0.42). Fever was present in 12 of 53 infected patients versus 14 of 106 controls (p = 0.19, OR = 1.92). Factors that were equally prevalent between the groups and that did not appear to influence the CVC infection rate included a diagnosis of malignancy, CVC type, and site of placement. Of the 53 infected catheters, 16 (30%) could not be cleared of infection and were removed. CONCLUSIONS: This study documents that neutropenia and failure to administer prophylactic antibiotics are risk factors for the development of early CVC infection in pediatric patients. To avoid early infection and possible premature CVC removal, we recommend that placement of permanent CVCs be postponed until the ANC is > 1,000. Perioperative antibiotics should be given. A trend toward higher infection rates was seen in patients with preoperative fever.

Classification of anorectal malformations--initial approach, diagnostic tests, and colostomy.

The optimal surgical care of patients with imperforate anus begins with appropriate decision making in the critical newborn period. In most cases the decision to create a colostomy should be delayed until the infant is 18 to 24 hours old. Except in cases of a rectoperineal fistula, most neonates are best treated with a completely divided left-lower-quadrant colostomy between the descending and sigmoid colons. Female patients with cloacal anomalies must be recognized at birth so that all urgent urologic evaluations can be performed. Hydrocolpos and obstructive uropathy are common in these neonates and warrant urgent decompression of the urinary tract with a vaginostomy and/or vesicostomy as well as a colostomy. Renal ultrasonography and voiding cystourethrography are mandatory for all patients regardless of the height of the defect. It is critical to discover the important precursors to renal insufficiency including renal agenesis, renal dysplasia, and vesicoureteral reflux in the neonate. The presence of these anomalies mandates early consultation with a pediatric urologist because the morbidity and mortality of these lesions often exceed those of the imperforate anus. Spinal cord anomalies are common and can be found even in patients who have normal plain films and low defects. Spinal ultrasonography or magnetic resonance imaging should be performed in all neonates to rule out occult spinal pathology such as tethered cord or lipoma of the cord. Efficacious and cost-effective care of patients with imperforate anus begins with a carefully thought out plan in the neonate. Optimal execution of the evaluation and surgical treatment at this phase sets the stage for the best possible outcome later in life.

Surgical treatment of testicular trauma: effects on fertility and testicular histology.

Unilateral testicular trauma in the postpubertal male can lead to alterations in semen analysis, but it is not clear what effect this has on fertility. To better understand how surgical treatment of testicular trauma affects both fertility and testicular histology the following study was performed. Eighty postpubertal Lewis rats were divided into eight equal groups with one group serving as a control. In the 70 remaining rats the left testicle was subjected to blunt or penetrating injury. The testicles were either left untreated, were removed, or were repaired with sutures or mesh before being returned to the scrotum. Following recovery, each male was allowed to mate to determine fertility. Fertility rates were significantly lower in all postinjury groups except the postinjury orchiectomy group. Histological analyses showed nonspecific inflammation, smaller tubules, and impaired spermatogenesis in all postinjury testicles regardless of the type of treatment. Contralateral testicles had no evidence of autoimmune injury and were essentially identical to the control group. In the postpubertal Lewis rat, unilateral testicular trauma leads to impaired fertility unless the injured testicle is removed soon after the injury. Various methods of repairing the injury did not improve fertility. In spite of the impaired fertility, the contralateral testicle appears histologically normal.

Choledocholithiasis in a premature neonate.

Choledocholithiasis in neonates and infants has been reported only rarely. Infants with complications of prematurity are more predisposed to development of biliary calculi. With the current widespread use of diagnostic ultrasonography, more neonates may be found to have gallstones and common bile duct stones. We describe a case of choledocholithiasis and cholelithiasis in a premature neonate successfully treated by surgical placement of a cholecystotomy tube and irrigation of the biliary system.

Use of tubularized peritoneal free grafts as urethral substitutes in the rabbit.

As an alternative to bladder mucosa, free grafts of tubularized peritoneum were used as urethral substitutes in a rabbit model of hypospadias. In group 1, six mature rabbits underwent partial penile urethrectomy followed by interposition of a 2-cm-long peritoneal-lined tube graft. These animals had urethrograms performed at 3 months and were killed at 6 (n = 4) or 12 (n = 2) months. In group 2, six rabbits underwent total penile urethrectomy with placement of 3-cm-long grafts. These animals were killed 1 to 4 weeks after surgery. Clinical assessment and gross examination of the 12 rabbits showed no urinary retention, two small fistulas at the proximal anastomosis, and no strictures or diverticular. At 6 and 12 months (group 1) the urethra had healed completely and the graft edges were not visible. In group 2, 1 to 4 weeks after surgery the graft was intact and the interface between the graft and native urethra was visible. Histological studies of the grafts were compared with control peritoneum. At 1 week, a high-density single-cell layer was present. Beginning at 2 weeks, a multilayered epithelium was present, which became more organized in the older grafts. Neovascularity became visible in the subepithelial layer at 2 weeks. Acute inflammatory cells were present early and were replaced by a palisading layer of lymphocytes and plasma cells in the older grafts. Minimal fibrosis was observed. Tubularized peritoneal free grafts are a promising new urethral substitute. Graft placement is technically simple. Replacement of the peritoneum by a multilayered epithelium resembling transitional epithelium occurs early after graft placement. As much as 1 year later, minimal scarring is present.

Spinal osteomyelitis after TPN catheter-induced septicemia.

Osteomyelitis of the spine is a well-recognized delayed manifestation of septicemia but has not been recognized as a complication of total parenteral nutrition. We report five cases of spinal osteomyelitis that were clinically recognized 1 to 13 months after total parenteral nutrition catheter-induced septicemia. Radiographic evidence of osteomyelitis was seen in all five patients. In three patients, culture of bony aspirates was positive for the same organism as from the blood. In one case, the diagnosis was established by histology, and in one the diagnosis was based on radiographic and radionuclide evidence of osteomyelitis. The organism responsible was Staphylococcus aureus in two cases, Candida albicans in another two cases and C tropicalis in one case. The septic episode that preceded osteomyelitis was treated with systemic antibiotics and catheter removal in four patients, and antibiotics without catheter removal in one patient. Nevertheless, osteomyelitis occurred, requiring bracing or operative debridement as well as prolonged antibiotic therapy. Spinal osteomyelitis may occur as a delayed manifestation of total parenteral nutrition catheter-induced septicemia. Prompt and effective treatment of septicemia is indicated but may not always be sufficient. Clinical suspicion is the key to the correct and early diagnosis of osteomyelitis and therefore to adequate treatment.

Infant pyeloplasty is a low-risk procedure.

Early pyeloplasty for the treatment of congenital ureteropelvic junction obstruction to maximize nephron salvage is only justified if the potential hazards of operating on small infants are avoided. The records of all infants who underwent pyeloplasty by the authors over a 7-year period were analyzed. The outcome of surgery in the younger infant group (YIG; patients less than 2 months of age) was compared with that of the older infant group (OIG; patients more than 2 months of age). Preoperative evaluation in cases of mild or moderate hydronephrosis was directed toward ruling out a nonobstructed collecting system and included voiding cysto-urethrography and serial ultrasonography and/or dual isotope diuretic renography. Open pyeloplasty was performed if collecting systems had deteriorated or were demonstrated to be obstructed; it was also performed for severe cases of hydronephrosis. Postoperative assessment consisted of serial ultrasonography and/or nuclear imaging to confirm decompression and relief of obstruction. Thirty three pyeloplasties were performed in 31 patients in the YIG (two bilateral), and 33 were performed in 32 infants in the OIG (one bilateral). The only significant differences between the groups were as follows. Patients in the YIG were more likely to present in utero (87% v 53%; P < .01), whereas those in the OIG were more likely to present with a urinary tract infection (44% v 0%; P < .01). The YIG was more likely to have nephrostomy drainage postoperatively (67% v 45%; P < .01). In the overall series, there were five complications, all of which were postoperative infections requiring intravenous antibiotic therapy. Two occurred in the YIG and three in the OIG.(ABSTRACT TRUNCATED AT 250 WORDS)

Treatment of bilateral Wilms' tumor: comparison of initial biopsy and chemotherapy to initial surgical resection in the preservation of renal mass and function.

The National Wilms' Tumor Study 3 (NWTS-3) recommended treatment of bilateral Wilms' tumor with initial biopsy followed by chemotherapy with subsequent operation to resect the remaining tumor. This study was performed to determine if this approach preserves renal mass and function when compared with initial surgical resection followed by chemotherapy. Over a 20-year period (1970 to 1990), 15 patients with synchronous bilateral Wilms' tumor were treated at the Childrens Hospital of Los Angeles. Eight patients in the surgical group underwent initial unilateral nephrectomy with contralateral biopsy, wedge resection, or partial nephrectomy and subsequent chemotherapy. The seven patients in the chemotherapy group underwent bilateral tumor biopsy, followed by chemotherapy and subsequent tumor resection. Patients were assigned to each group in a nonrandomized manner according to the preference of the attending oncologist and surgeon. Comparison of the two groups showed no significant differences in sex distribution, initial renal function, tumor histology, dose and field of radiotherapy, presence or absence of positive surgical margins, and local recurrence rates. Patients in the surgery group were slightly older than those in the chemotherapy group: 3.6 +/- 2.2 versus 2.3 +/- 2.2 years. The percent of renal mass involved by tumor for the surgery group was 52 +/- 12 versus 73 +/- 16 for the chemotherapy group (P = .03). The percent of renal mass preserved following all operations, local recurrence rates, incidence of metastases, and survival was nearly identical between the two groups. There were three cases of renal failure in the surgery group and one case of renal failure in the chemotherapy group.(ABSTRACT TRUNCATED AT 250 WORDS)

Primary repair without routine gastrostomy is the treatment of choice for neonates with esophageal atresia and tracheoesophageal fistula.

Gastrostomy and staged repair are techniques frequently recommended for the management of esophageal atresia with distal tracheoesophageal fistula (EA-TEF), especially for those infants at high risk. We describe 42 consecutive patients with EA-TEF treated during the past 8 years. Staged repair and preliminary gastrostomy were not routinely employed. Fifteen infants were considered to be at high risk (Waterston class C). Surgical treatment via an extrapleural approach consisted of fistula division and primary single-layer end-to-end esophageal anastomosis. Four patients required proximal esophageal circular myotomy. Four patients early in the series received a gastrostomy at or before definitive repair for various life-threatening indications. One patient had fistula division only and died before esophageal anastomosis was possible. Two neonates died before repair and another died after repair. The deaths in this series of patients were unrelated to EA-TEF. One patient developed a clinically significant anastomotic leak. Four patients required multiple dilatations for anastomotic stricture. Fundoplication was necessary in 3 patients with symptomatic gastroesophageal reflux. Our data demonstrate that excellent overall survival (90%) with low morbidity (15%) can be achieved using primary repair without preliminary gastrostomy in most neonates with EA-TEF. We believe that mortality in high-risk patients with EA-TEF is due to associated life-threatening anomalies.

Effect of total parenteral nutrition on hepatic histology.

We evaluated the effect of total parenteral nutrition (TPN) on abnormalities of hepatic histology. Liver biopsies of 93 patients who were concurrently receiving TPN were compared with a control group of 35 patients. The control patients were matched for extent of preexisting liver disease and degree of illness. The liver biopsy specimens were blindly graded on 19 histopathologic findings, including fatty change, portal inflammation, and cholestasis. Twenty-seven clinical variables, such as preexisting liver disease, the presence or absence of sepsis or shock, and number of days receiving TPN before biopsy, were recorded. Individual and partial correlations were established between the clinical variables and histopathologic findings. The control and TPN groups proved to have been closely matched regarding the extent of risk factors for hepatic histopathologic features. Positive correlations were repeatedly found between abnormal hepatic histologic features and preexisting liver disease, abdominal sepsis, renal failure, and blood transfusion but not with the administration of TPN. We conclude that clinical phenomena, such as existing liver disease, renal failure, and abdominal sepsis, rather than administration of TPN, had a predominant effect on histopathologic features in this group of patients.