Adverse neurodevelopmental outcome after extracorporeal membrane oxygenation among neonates with bronchopulmonary dysplasia.

OBJECTIVE: The relationship between bronchopulmonary dysplasia (BPD) and neurodevelopmental outcome after extracorporeal membrane oxygenation (ECMO) has not been extensively reported. We compared the outcomes in a large series of infants with and without BPD after ECMO. STUDY DESIGN: Hospital charts and follow-up records of 145 infants treated with ECMO (1985 through 1990) were reviewed. Complete long-term respiratory and follow-up outcome data were available in 64 infants. BPD occurred in 17 survivors; the remaining 47 did not have BPD. RESULTS: Babies with BPD were more likely to have had respiratory distress syndrome. Mean (+/- SD) age at ECMO initiation was later for the BPD group (127+/-66 vs 53+/-39 hours, p < 0.001), and the duration of ECMO treatment was longer (192+/-68 vs 119+/-53 hours, p < 0.001). Bayley Scales of Infant Development scores at <30 months were lower in infants with BPD (p < 0.001), as were three of four Mullen Scales of Early Learning scores (> or = 30 months, p < 0.001 or p = 0.01). At 57+/-16 months 11 (64%) patients with BPD had mild neurologic disabilities, and 3 (18%) had severe disabilities. At a similar age (53+/-16 months, p = NS) 16 (34%) patients without BPD had mild disabilities, whereas 2 (4%) had severe disabilities (p < 0.01). CONCLUSIONS: The occurrence of BPD after ECMO is associated with adverse neurodevelopmental outcome. Patients with BPD after ECMO merit close long-term follow-up.

Effect of circumcision status on periurethral bacterial flora during the first year of life.

To determine whether bacterial colonization is a factor in the pathogenesis of urinary tract infection, we compared the periurethral bacterial flora of uncircumcised versus circumcised boys during the first year of life. Intraurethral and circumferential glans cultures were obtained from 25 circumcised and 25 uncircumcised infants at 2 days, 2 weeks, 2 months, 4 months, 6 months, and 12 months of age. Different children were used at each age (300 total). The types of bacteria and the total and specific colony counts were compared. The results of the glans cultures were similar to those from the urethra. Uncircumcised boys had significantly higher total colony counts (p less than 0.003) at all ages except 12 months. Escherichia coli was present significantly more often (p less than 0.01) in the urethras of uncircumcised boys at 2 weeks, 2 months, 4 months, and 6 months. Gram-negative uropathogenic organisms (Klebsiella-enterobacter, Proteus mirabilis, and Pseudomonas aeruginosa) were cultured more frequently (p less than 0.0005) from the urethras of uncircumcised boys at 2 months, 4 months, and 6 months. The specific colony counts for E. coli and the other uropathogenic organisms were significantly higher (p less than 0.05) at all ages except 12 months. We conclude that during the first 6 months of life, the presence of a foreskin is associated with a greater quantity of periurethral bacteria and a greater likelihood for the presence of, as well as higher concentrations of, potentially uropathogenic organisms.

Tracheal and bronchial injury in high-frequency oscillatory ventilation and high-frequency flow interruption compared with conventional positive-pressure ventilation.

We compared the histopathologic changes in the airways of premature baboons treated with conventional positive-pressure ventilation (PPV) with those seen after high-frequency oscillatory ventilation (HFOV) and high-frequency flow interruption (HFFI). Twenty-six animals were treated with ventilation for 24 hours (five PPV, 10 HFOV, 11 HFFI), and 18 were treated with ventilation for 96 hours (six PPV, six HFOV, six HFFI). A semiquantitative scoring system was used to grade tissue changes in the trachea, carina, and both main-stem bronchi. Alterations were produced by all forms of mechanical ventilation. The degree of injury was similar and relatively mild for the PPV- and HFOV-treated animals at both 24 and 96 hours. Eleven of 17 baboons treated with HFFI ventilation (8/11 at 24 hours; 3/6 at 96 hours) had severe airway damage characterized by diffuse submucosal necrosis, extensive hemorrhage, dense polymorphonuclear leukocyte infiltration, sloughed epithelium, focal basophilia, and intraluminal debris. HFOV resulted in no greater degree of airway damage than did PPV. The use of HFFI, with the particular strategy we employed, resulted in a far greater degree of damage than either PPV (P less than 0.01) or HFOV.

Tracheal and bronchial injury in high-frequency oscillatory ventilation compared with conventional positive pressure ventilation.

We compared airway histopathologic findings in premature baboons given standard positive pressure ventilation with those seen after high-frequency oscillatory ventilation. Six animals received standard frequency conventional ventilation for a mean of 9.2 days; seven received high-frequency oscillatory ventilation at 10 Hz using a piston oscillator for a mean of 10.2 days; five baboons served as controls, and were killed immediately after birth. A semiquantitative histopathologic scoring system was used to grade tissue changes in the trachea, carina, and both mainstem bronchi. Compared with the nonventilated control animals, injury was produced with both forms of mechanical ventilation (P less than 0.01 for both instruments); however, the degree of damage was mild, with no significant difference in the extent of injury between the two treatment groups. High-frequency oscillatory ventilation appears to result in no greater degree of airway damage than conventional positive pressure ventilation.

Estimated blood volumes in polycythemic neonates as a function of birth weight.

This prospective study was designed to test the correlation of blood volume per kilogram of body weight with birthweight in polycythemic neonates. One hundred and sixty-four neonates with venous hematocrits of 65% or greater were treated with partial exchange transfusion. Follow-up venous hematocrits were obtained, and BV/kg was estimated for each infant based on the exchange volume used and the observed change in hematocrit. Birth weight ranged from 1,210 to 5,080 gm. Individual blood volume estimates ranged from 44 to 176 ml/kg, with a mean of 88.4 +/- 23.5 ml/kg. There was a highly significant negative linear correlation of BV/kg with birth weight (P less than 0.001), the regression line intersecting 100 ml/kg at 2,000 gm and 70 ml/kg at 4,500 gm. This correlation has clinical application in the management of polycythemic neonates with partial exchange transfusion. A nomogram is provided for estimating BV/kg in polycythemic neonates based on birth weight.