Air leaks from the respiratory tract in mechanically ventilated children with severe respiratory disease.

Our objectives were to evaluate the frequency of air leaks (AL) from the respiratory tract (pneumothorax, pneumomediastinum, pneumoperitoneum, subcutaneous emphysema) in critically ill children on mechanical ventilation (MV) for severe respiratory diseases, and to examine whether AL could be correlated with specific clinical events or ventilator settings. The study constitutes a retrospective cohort of 80 consecutive critically ill children receiving MV for severe respiratory diseases between 1986 and 1993. Patients (mean age 2.9 +/- 0.6 years, 49 males and 31 females), were admitted to the Pediatric Intensive Care Unit (PICU) with acute respiratory syndrome (ARDS) (27%), asthma (15%), bronchiolitis (10%), pneumonia (21%), pulmonary congenital diseases (9%), or foreign body aspiration (18%). Patients were divided into two groups; those with AL (n=22) and those without air-leaks (non-AL) (n = 58). Air leaks developed in 22 of 80 patients or in 27.5%. Survival was significantly lower in the AL group, compared to the non-AL group (41% vs. 76%, P < 0.01). The odds ratio that a patient with multiple organ system failure (MOSF) or infection would develop AL was 2.96 and 2.19, respectively. Candida and Pseudomonas species were recovered with significantly higher frequency in the AL group compared with the non-AL group (P < 0.025). There was a strong positive correlation between the incidence of AL and high ventilatory pressures (PIP 36 vs. 29.7 cm H(2)O, P < 0.001), or large tidal volumes (V(T) 12 vs. 9 mL/kg, P < 0.05), suggesting that large volumes might elicit injury to the pulmonary epithelium. Multiple logistic regression analysis showed that only V(T) was independently associated with the development of AL in children with primary severe respiratory disease (r(2) = -0.38, P = 0.01). In conclusion, MV will produce AL, particularly when high peak airway pressures (barotrauma) or large tidal volumes (volotrauma) are delivered by the ventilator. Sepsis, MOSF, and lung superinfection with Pseudomonas or Candida species may be also important factors in the development of AL in critically ill children.

Effectiveness and safety of a protocol for promotion of early intragastric feeding in critically ill children.

OBJECTIVES: To investigate the feasibility, adequacy, and efficacy of early poststress intragastric feeding (EPIGF) in critically ill children. DESIGN: A prospective clinical study. SETTING: Pediatric intensive care unit in a tertiary care children's hospital. PATIENTS: Seventy-one consecutively enrolled critically ill children requiring prolonged mechanical ventilation. INTERVENTIONS: Full-strength intragastric tube feedings (Nutrison Pediatric, Standard) were initiated within 12 hrs of the study-entry event. Enteral feedings were advanced to a target volume of energy intake = 1/2, 1, 5/4, 6/4, and 6/4 of the predicted basal metabolic rate (PBMR) on days 1-5, respectively. MEASUREMENTS AND MAIN RESULTS: Nutritional status by the caloric intake, recommended dietary allowances, PBMR, predicted energy expenditure (PEE), anthropometry, and clinical indices were evaluated on days 1 and 5. Safety was assessed by the clinical course of disease, laboratory findings, and occurrence of complications. Success was determined by accomplishment of the PEE target. The early success rate was 94.4% and predicted late enteral feeding success accurately (p =.0001). Caloric intake approached PBMR the second day (43 +/- 1.7 kcal/kg/day vs. 43.2 +/- 1.1 kcal/kg/day) and PEE the fifth day (66.2 +/- 2.7 kcal/kg/day vs. 67.7 +/- 6.4 kcal/kg/day). Multivariate stepwise regression analysis showed that poor outcome and a high Therapeutic Intervention Scoring System score correlated with failure of EPIGF (p <.0001). Patients who succeeded EPIGF had significantly higher myocardial ejection (65% vs. 43%; p <.0001) or shortening fractions (34% vs. 20%; p =.0001) on day 1 than those who failed. Patients tolerated EPIGF well; 9.9% developed nosocomial pneumonia, 5.6% developed diarrhea, and 8.5% needed treatment with cisapride because of a delay of gastric emptying. The mortality rate (5.6%) was different between initial and final success and failure groups (p <.0001) and was lower than predicted by the admission severity scores (12% +/- 2%). CONCLUSIONS: This study showed that increases of caloric intake during the acute phase of a critical illness are well tolerated and may approach PBMR by the second day and PEE by the fourth day in critically ill children. Caloric intake lower than PBMR is associated with higher mortality and morbidity rates.