Urinary leukotriene E(4) excretion during the first month of life and subsequent bronchopulmonary dysplasia in premature infants.

BACKGROUND: Inflammation plays an important role in the pathogenesis of bronchopulmonary dysplasia (BPD), but the exact nature of this inflammatory process is incompletely understood. Older infants with established BPD have higher levels of urinary leukotriene E(4) (LTE(4)) compared to healthy infants of the same age. This suggests that cysteinyl leukotrienes may play a role in the abnormalities seen in BPD. OBJECTIVES: To measure urinary LTE(4) levels during the first month of life in premature infants, and to determine whether there are significant differences in premature infants who develop BPD, as compared to those who do not develop BPD. DESIGN: Prospective, blinded, controlled study. SETTING: Neonatal ICUs of a tertiary-care university hospital. METHODS: Thirty-seven premature infants (< 33 weeks of gestational age) were enrolled prospectively at birth. Urinary LTE(4) levels were measured blinded, using a standard radioimmunoassay technique at 2 days, 7 days, and 28 days of life. At 1 month of age, infants were classified as with or without BPD, based on need for supplemental oxygen, and characteristic chest radiographs. Clinical features and urinary LTE(4) were compared between the two groups. RESULTS: Mean +/- SD gestational age was 29 +/- 2.6 weeks. None of the infants had a family history of asthma. Thirteen of 37 infants were classified as having BPD at 28 days after birth. Mean gestational age in infants who developed BPD was 27 +/- 2.4 weeks, compared to 30 +/- 2 weeks in infants who did not develop BPD (p < 0.05). In infants with BPD, mean urinary LTE(4) levels of urinary creatinine were 1,762 +/- 2,003 pg/mg, 1,236 +/- 992 pg/mg, and 5,541 +/- 5,146 pg/mg at days 2, 7, and 28, respectively, compared to 1,304 +/- 1,195 pg/mg, 1,158 +/- 1,133 pg/mg, and 2,800 +/- 2,080 pg/mg in infants without BPD. LTE(4) levels at 2 days, 7 days, and 28 days did not correlate with the subsequent development of BPD. LTE(4) levels at day 28 were significantly higher than LTE(4) levels at day 2 and day 7 in both groups, even after correcting for gestational age or birth weight (p < 0.05). There was significant inverse correlation between LTE(4) levels at day 2 with gestational age and birth weight (p < 0.05). All 13 infants with BPD received steroid pulses, compared to 3 of 26 infants without BPD. Gestational age and use of postnatal steroid pulses, diuretics, and theophylline (for apnea of prematurity) were significantly associated with each other and with the subsequent development of BPD. CONCLUSION: Urinary LTE(4) levels measured on the second day of life in very-low-birth-weight infants inversely correlate with gestational age and birth weight. Urinary LTE(4) levels may reflect lung injury and/or inflammation in premature infants, not necessarily related to BPD as it is presently defined.

Differential effects of pentobarbital and cocaine on punished and nonpunished responding.

Similar rates of punished and nonpunished responding, maintained with equated rates of reinforcement, were established in pairs of rats. One subject of each pair was exposed to a random-ratio schedule of food presentation. The interreinforcement intervals for this subject comprised the intervals of a random-interval schedule of reinforcement for the other (yoked) rat. The random-ratio schedule maintained rates of responding higher than those maintained by the same rate of reinforcement schedule according to the yoked random-interval contingency. A random-ratio schedule of electric foot shock added to the random-ratio schedule of food presentation suppressed rates of responding such that similar rates of responding were observed in rats of both groups. Pentobarbital (3.0 to 17.0 mg/kg) increased punished responding at doses that had little effect on or decreased nonpunished responding, whereas cocaine (5.6 to 30 mg/kg) increased nonpunished responding at doses that decreased or did not alter punished responding. Qualitatively different effects of pharmacological agents on punished and nonpunished responding can be obtained using procedures that generate similar rates and temporal patterns of punished and nonpunished responding. The effects of pentobarbital and cocaine on responding can be determined by factors other than simply the baseline rate of responding.