Causes of near misses in critical care of neonates and children.

AIM: The primary goal of this study was to examine the nature and causes of medical errors known as almost adverse events (AAEs) and potential adverse events (PAEs) in intensive care units. METHODS: Observations were conducted in the Neonatal Intensive Care Unit and in the Pediatric Intensive Care Unit in a large hospital in Israel. The AAEs and PAEs were classified into three main categories: environmental, system and human factors. Data encoding and analysis was based on a Bayesian model previously developed to analyse causes of traffic accidents, and the categories were based on systems and ergonomics approaches. RESULTS: 'Workload' (a system factor) was the main cause of AAEs and 'communication failures' (a human factor) was the second main cause of AAEs. Among the environmental factors, 'failures in medical devices' was the most cited cause of AAEs. Environmental factors accounted for most of PAEs and among them 'form failures' was the most 'AAE'-prone factor. CONCLUSIONS: Environmental factors (mainly 'failures in medical device') and system factors (mainly 'workload') accounted for most of AAEs in the intensive care units studied. The systems and the ergonomics approaches to error analysis can be useful in creating a comprehensive error management programme in order to minimize the gap between work demands and individual capabilities.

Lung-surfactant-meconium interaction: in vitro study in bulk and at the air-solution interface.

Lung surfactants (LSs) form a monolayer at the lung's alveoli air-solution interface and play a crucial role in making normal breathing possible by reducing the surface tension. LS are affected by various agents that hamper their normal functioning. Tobacco smoke [Bringezu, F.; Pinkerton, K. E.; Zasadzinski, J. A. Langmuir 2003, 19, 2900-2907] and meconium, the first excrement of the newborn, are examples for such LS poison. In neonates, intrauterine aspiration of meconium is a known cause for morbidity and mortality. We studied in vitro the interactions between modified porcine LSs (Curosurf), used as LS replacement, and meconium, as well as between their artificial analogues, phospholipids mixture, and taurocholic acid (TA), respectively. The interactions were examined both in the bulk solution and at the air-water interface, representing the pre- and postnatal situations. It was found that the artificial analogues represent the natural system reliably and exhibit similar effects. TA, a principle component of bile, is an amphiphilic sterol compound in which the hydrophilic and hydrophobic moieties are presented at different faces of the sterol plane. Here we found that TA affects the structure of both monolayers at the interface and surfactant aggregates in solution. A likely poisoning mechanism is by stereoselective penetration of TA into the lamellar or monolayer structures, thus disrupting the contiguous structure of the intact monolayer or the bilayer vesicle structure.

Successful control of an Acinetobacter baumannii outbreak in a neonatal intensive care unit.

We describe an outbreak of Acinetobacter baumannii in a neonatal intensive care unit (NICU), and our investigation to determine the source and mode of transmission and identify the population at risk. A case (infected infant) was defined as a patient hospitalized in the NICU during the outbreak period, with clinical signs of sepsis and isolation of A. baumannii. In colonized infants, A. baumannii was isolated from body surfaces without signs of infection. Infected infants were separated and treated by a different medical team. Cultures were taken from working surfaces and along the infant's admission passage from the delivery room to the NICU. The outbreak strain was identified by pulsed-field gel electrophoresis (PFGE). Nine cases and eight colonized infants met the definition criteria. Cases were younger than colonized infants with regard to gestational age and age of diagnosis and had lower birthweights (P<0.01). The outbreak strain was only isolated from hygroscopic bandages used on skin under the ventilation tube and umbilical catheters. Discontinuing the use of the bandages put an end to the outbreak. We conclude that a rapid and thorough investigation of the environment during an outbreak of A. baumannii is essential to finding the source of the infection, and that hygroscopic bandages may be a source of such outbreaks.

Ethics and neonatology in Israel.

Israel is a country of controversies: with high-quality medical care available to all and a high antenatal detection rate of congenital anomalies followed by abortion, the incidence of infants born with malformations has been reduced dramatically in the last decade. On the other hand, religious and strong traditional ethnic attitudes on fertility have led to a world record rate of ARTs and multiple births, which have increased the incidence of VLBW infants and the long-term handicap that follows their survival.

Oligohydramnion, renal failure and no pulmonary hypoplasia in glomerulocystic kidney disease.

Two newborns with glomerulocystic kidney disease manifesting as late onset oligohydramnion and neonatal anuria, yet without severe respiratory distress, are presented. They had a similar perinatal course and associated clinical manifestations. No associated congenital or inherited malformation syndrome could be defined. Both infants' parents were first degree cousins and belonged to the same small Bedouin tribe, and neither they nor the infants' siblings had polycystic kidneys or renal insufficiency, pointing to either a possible genetic etiology or a common external toxic exposure.

Heart rate variability in the neonate and infant: analytical methods, physiological and clinical observations.

Naturally occurring oscillations in heart rate have long been considered to reflect the modulating influences of the autonomic nervous system. Individual reports of heart rate variability in healthy and sick neonates and infants have provided valuable information as to pathophysiological changes in autonomic cardiovascular control. It is evident that prematurity and poor health are reflected in attenuated heart rate variability. However, at present, there are few standard criteria for the analysis of heart rate variability, preventing precise comparisons among the various studies. Until recommendations for standardized analytical methods are made, clinical application of heart rate variability in neonatal and infant prognosis and therapy remains premature.

Failure of early postnatal dexamethasone to prevent chronic lung disease in infants with respiratory distress syndrome.

OBJECTIVE: To study the effect of early postnatal dexamethasone (days 1-3) on the incidence and severity of chronic lung disease in preterm infants with respiratory distress syndrome. METHODS: A multicentre, randomised, placebo controlled, blinded study was carried out in 18 neonatal intensive care units in Israel. The primary outcome measure was survival to discharge without requirement for supplemental oxygen therapy beyond 28 days of life. The secondary outcome measures were requirement for mechanical ventilation at 3 and 7 days, duration of ventilation or oxygen therapy, need for subsequent steroids for established chronic lung disease and incidence of major morbidities. RESULTS: The study consisted of 248 infants (dexamethasone n = 132; placebo n = 116). No differences were found in the outcome variables except for a reduction in requirement for mechanical ventilation at age 3 days in treated infants (dexamethasone 44%, placebo 67%; P = 0.001). Gastrointestinal haemorrhage, hypertension, and hyperglycaemia were more common in treated infants, but no life threatening complications, such as gastrointestinal perforation, were encountered. CONCLUSIONS: These data do no support the routine use of early postnatal steroids, but may justify further study in a selected, high risk group of infants.

Pulmonary hypertension in respiratory distress syndrome.

Pulmonary hypertension was associated with nonresponse to surfactant in six premature infants with respiratory distress syndrome. The diagnosis was suspected on the basis of a discrepancy between the X-ray findings and the severity of the clinical status as reflected by hypoxia despite maximal ventilatory support. The diagnosis of pulmonary hypertension was made by pre- and postductal oxygen saturation differences or by echodoppler cardiography, showing suprasystemic right ventricular pressures or right to left shunts through a patent foramen ovale or the ductus arteriosus. The response to surfactant was quantified by the arterial/alveolar (a/A) ratio difference before and 1 hr after therapy ("delta a/A ratio"); the delta a/A ratio was 0 +/- 0.01, which indicates a nonresponse. A single dose of 1 mg/kg tolazoline was administrated and the response assessed by a/A difference. A delta a/A ratio of 0.11 +/- 0.11 (range 0.02-0.32) represented a dramatic response and enabled oxygenation in these severely ill infants. No significant side effects were observed. We conclude that pulmonary hypertension may be an important and reversible condition in certain cases of respiratory distress syndrome and has to be considered in infants who do not respond to surfactant.

Beta-adrenergic blocking agents in the treatment of pregnancy-induced hypertension.

Fifty-one women with pregnancy-induced hypertension (PIH) were randomly allocated to one of three treatment groups: A: hydralazine (13); B: hydralazine and propranolol (17); and C: hydralazine and pindolol (19). All women fulfilled the pretreatment criteria and were of similar age, numbers of previous pregnancies and had systolic blood pressure (SBP) of between 140 and 160 mmHg and diastolic blood pressure (DBP) of between 95 and 110 mmHg. Hypertension was treated equally well by all three regimens (mean SBP was 133.6, 130 and 134 mmHg, respectively). Heart rate was significantly higher than baseline in group A and lower in groups B and C, as is to be expected with beta-blocker treatment. Side-effects were more frequent in group A than in groups B and C, 62% of the patients on hydralazine monotherapy complained of palpitations compared to 35% on combination treatment. Fetal outcome differed in the various groups. Birth weight was significantly lower in group B, where regimen included propranolol, compared to that of group C, for whom the regimen included pindolol (3,044.7 +/- 443.8 and 2,709.6 +/- 485.5 gm, p < 0.05). Mean blood glucose of the newborns were similar in groups A and C (76.5 +/- 16.5 and 78.6 +/- 15 gm%) and significantly lower in group B (62.6 +/- 14 gm%, p < 0.02). In conclusion, blood pressure was equally well treated in all three treatment groups. However, more maternal side-effects occurred in group A, the group treated with hydralazine monotherapy, while propranolol in combination with hydrazaline (group B) had some negative effects on fetal development which did not occur in pindolol/hydrazaline combination.

Gestational diabetes among Bedouins in southern Israel: comparison of prevalence and neonatal outcomes with the Jewish population.

Differences in the prevalence of gestational diabetes mellitus (GDM) have recently been reported between various ethnic populations. In the Negev region of Israel, a universal free screening programme for GDM was implemented in 1985. Between 1 March 1987 and 31 July 1988 11,003 deliveries occurred at the Soroka Medical Center, which provides free delivery and postnatal care to the whole Jewish and Bedouin population of the region. GDM was found in 5.7% of Jewish and in 2.4% of Bedouin women (odds ratio, 2.3, 95% confidence interval (CI) 1.8-2.9; P < 0.0001). Ethnicity was unrelated to maternal outcome, perinatal mortality or to any of the examined morbidity conditions of the newborn. The incidence of major congenital malformations was significantly higher in Jewish than in Bedouin infants of GDM women (Fisher's Exact Test, P < 0.03). Conversely, Jewish infants had fewer minor congenital anomalies (odds ratio 0.26, 95% CI 0.09-0.73). In a multivariate logistic regression model, gestational age, mode of delivery and insulin requirement during pregnancy were the only factors independently associated with neonatal morbidity. The results of this study suggest that in our health care system, among women with GDM ethnicity is not associated with an excess of unfavourable maternal or infant outcomes.

Respiratory tract colonization with Ureaplasma urealyticum and bronchopulmonary dysplasia in neonates in southern Israel.

Ureaplasma urealyticum has been recognized as an important potential pathogen in premature neonates. Reported rates of colonization of the respiratory tract vary. Data on neonatal ureaplasma colonization outside the United States and Western Europe are rare. Therefore we prospectively studied nasopharyngeal and endotracheal colonization in a cohort of 114 preterm and 100 full term infants within 48 hours of birth. The colonization rate was 24% in the premature infants and zero in the full term infants. Bronchopulmonary dysplasia developed in 19% of infants with nasopharyngeal Ureaplasma colonization and in 4.6% of noncolonized infants (P < 0.03). Bronchopulmonary dysplasia developed in 40% of intubated infants with positive endotracheal Ureaplasma cultures and only in 9.8% of infants with negative endotracheal cultures (P < 0.04). Thus Ureaplasma colonization of either the nasopharynx or the trachea was associated with an increased risk for the development of bronchopulmonary dysplasia (relative risk, 4.0 and 4.1, respectively).

[Surfactant replacement therapy for respiratory distress syndrome: a pilot study].

A pilot study of the effect of exogenous surfactant (ES) on premature infants with respiratory distress syndrome (RDS) is reported. Each of the first 15 infants in this study received 200 mg/kg of natural surfactant (Curosurf) during the first day of life. Controls were 56 infants with RDS seen in the 15 months prior to the study. Within 5 minutes of starting ES, in all infants there was rapid and dramatic improvement in oxygenation and improvement in the average arterial/alveolar ratio of 169%. They had lower oxygen and ventilatory requirements than the control group throughout the first 5 days of life. No treated infant suffered from pulmonary air leak, while in the control group 21% developed pneumothorax and 11% had pulmonary interstitial emphysema. Mortality was 13% in the treated group as compared to 27% in the control group (p less than 0.01). There were no differences between the groups in the incidence of sepsis, patent ductus arteriosus, necrotizing enterocolitis, intraventricular hemorrhage, or bronchopulmonary dysplasia, nor were there side-effects of therapy. Dosage, timing and composition of the ideal surfactant are important questions for future studies.

Effects of insulin and glucose on pulmonary insulin receptors in late gestation fetal rats.

The effects of high glucose and insulin concentrations on fetal lung insulin receptors and tyrosine kinase activity were studied in an in vitro system utilizing 19- or 20-d fetal rat lung explants. Exposure of the explants to 100 mM glucose and insulin (0.1 unit/mL) for 72 h resulted in a significant decrease in specific binding of insulin to partially purified receptors [5.78% +/- 0.66 (SEM) vs. 9.64% +/- 1.68; P less than .01] when compared with lung explants exposed to 10 mM glucose alone. When individual effects of high insulin and glucose were studied, down-regulation of specific insulin binding was also observed, but to a lesser extent than that observed using both high glucose and insulin. Differences in insulin receptor affinity were not noted. Insulin receptor tyrosine kinase activity was also significantly decreased (52% of control values) under high-glucose/high-insulin conditions. Total phosphatidylcholine and disaturated phosphatidylcholine concentrations were significantly decreased in explants grown under high-glucose/high-insulin conditions, consistent with delayed pulmonary maturation. High glucose and insulin levels thus result in down-regulation of fetal lung insulin receptors and insulin receptor tyrosine kinase activity late in gestation. These results may have implications for substrate availability in the developing fetal lung.

Effect of vitamin K on neonatal erythrocytes.

This study investigated the possible oxidative effect of vitamin K3 (menadione) and Vitamin K1 (Konakion) on neonatal erythrocytes by controlled in vitro exposure. Menadione caused only mild morphological changes and did not decrease ATP levels. However, it oxidized intracellular hemoglobin to methemoglobin in neonatal cells more than in adult cells. Reduced glutathione contents were higher in neonatal cells, but less available for antioxidant protection. Konakion did not increase methemoglobin levels in newborn infants after a prophylactic injection. In vitro exposure to Konakion did not affect reduced glutathione and ATP levels, nor did it oxidize hemoglobin. However, extensive morphological changes were observed, attributed to the effect of its solvent. Therefore, it seems that menadione, which is no longer administered to newborns, causes oxidative stress in neonatal cells whereas Konakion, the current vitamin K1, does not, either in in vitro studies or by therapeutic administration.

Pseudohypoaldosteronism in a preterm infant: intrauterine presentation as hydramnios.

After a pregnancy complicated by severe hydramnios, a preterm infant had clinical and biochemical evidence of pseudohypoaldosteronism. Fetal polyuria probably caused the hydramnios.

Pseudooutbreak of Candida guilliermondii fungemia in a neonatal intensive care unit.

During a 3-week period multiple blood cultures obtained from 14 Neonatal Intensive Care Unit infants and 3 Newborn Unit babies grew Candida guilliermondii, a yeast rarely associated with infections in humans. At the time of detection of positive cultures, most infants had been hospitalized for days or weeks for serious perinatal conditions and treated with antibiotics and intravenous hyperalimentation. Two critically ill premature infants from whom the yeast was isolated were given amphotericin B. In 7 other infants, however, yeasts were recovered on the day of birth, raising the question of pseudofungemia. Exhaustive interrogation on the blood culture practices revealed that when drawing blood for a culture from small infants, "butterfly" needles were often flushed with a diluted heparin solution to prevent blood clotting. Culture of a single lot of diluted heparin vials, prepared at the hospital pharmacy and distributed to the Neonatal Intensive Care Unit and Newborn Unit shortly before the onset of the epidemic, grew between 10,000 and 15,000 colony-forming units of Candida guilliermondii/ml. Removal of contaminated heparin vials and discontinuation of heparinization of needles used for blood cultures resulted in cessation of the epidemic. The present outbreak illustrates the difficulties in recognizing pseudoinfections in sick premature infants and the importance of intensive investigation and intervention during such an outbreak.

Oxidative stress in newborn erythrocytes.

Phenylhydrazine (PHZ) exposure is used to study in vitro red cell aging mechanisms dependent on Hb oxidation. The effect of PHZ on normal neonatal red blood cells was studied in unseparated blood and after separation into light and heavy cells. PHZ caused more extensive morphologic changes in neonatal than in adult red blood cells. PHZ exposure of neonatal cells caused less reduced glutathione depletion than in adult cells. Although we found the same total amount of oxidized Hb in both cells, a well-defined oxidation product of Hb was demonstrated by Mössbauer spectra only in neonatal cells. This oxidation product was not methemoglobin but a trivalent, high-spin iron compound. All neonatal cell populations were more sensitive to PHZ than were adult ones, as demonstrated by the presence of Heinz bodies at low PHZ concentration, which did not affect adult cells. These studies demonstrate greater sensitivity of neonatal cells to PHZ in all density-separated populations.