Carlos Fajardo

Automated versus Manual Oxygen Control with Different Saturation Targets and Modes of Respiratory Support in Preterm Infants.

OBJECTIVE To determine the efficacy and safety of automated adjustment of the fraction of inspired oxygen (FiO2) in maintaining arterial oxygen saturation (SpO2) within a higher (91%-95%) and a lower (89%-93%) target range in preterm infants. STUDY DESIGN Eighty preterm infants (gestational age [median]: 26 weeks, age [median] 18 days) on noninvasive (n = 50) and invasive (n = 30) respiratory support with supplemental oxygen, were first randomized to one of the SpO2 target ranges and then treated with automated FiO2 (A-FiO2) and manual FiO2 (M-FiO2) oxygen control for 24 hours each, in random sequence. RESULTS The percent time within the target range was higher during A-FiO2 compared with M-FiO2 control. This effect was more pronounced in the lower SpO2 target range (62 ± 17% vs 54 ± 16%, P < .001) than in the higher SpO2 target range (62 ± 17% vs 58 ± 15%, P < .001). The percent time spent below the target or in hypoxemia (SpO2 <80%) was consistently reduced during A-FiO2, independent of the target range. The time spent above the target range or at extreme hyperoxemia (SpO2 >98%) was only reduced during A-FiO2 when targeting the lower SpO2 range (89%-93%). These outcomes did not differ between infants on noninvasive and invasive respiratory support. Manual adjustments were significantly reduced during A-FiO2 control. CONCLUSIONS A-FiO2 control improved SpO2 targeting across different SpO2 ranges and reduced hypoxemia in preterm infants on noninvasive and invasive respiratory support. TRIAL REGISTRATION ISRCTN 56626482.

Surfactant versus saline as a vehicle for corticosteroid delivery to the lungs of ventilated rabbits.

Local administration of steroids to the lungs in ventilated newborn infants can minimize the harmful side effects that occur with systemic administration. An efficient system of drug delivery that provides uniform distribution within the lungs is essential for the treatment of bronchopulmonary dysplasia. In this study we compare surfactant with 0.9% saline solution as vehicles for the direct instillation of a steroid (budesonide) into the lungs. Twenty-two anesthetized, ventilated rabbits received [3H]budesonide in either 0.9% saline or surfactant, administered through an endotracheal tube. Before drug administration, bronchial lavage was performed on half of the animals to serve as a model for surfactant deficiency. Lung samples were analyzed by both autoradiography (alveolar versus airway distribution) and liquid scintillation counting (central versus peripheral deposition). As expected, the delivered concentration of [3H]budesonide decreased as airway size decreased and branching increased. Significantly less[3H]budesonide was deposited in the alveolar spaces of all study groups compared with that deposited in the small and large airways (p < 0.05). However, both vehicles were equally efficient in delivering[3H]budesonide to the lungs. Although the alveolar and peripheral areas received less (4-11%) of the drug than the central tissue (14-28%), this was consistent among all the groups and was not affected by altered lung compliance. Therefore, either surfactant or saline could be used to efficiently and reliably deliver budesonide to the lungs at a level greater than that reported using nebulizers. Because normal saline is currently used for tracheal toilette, it is likely to be preferred considering the increased costs associated with surfactant.

Effects of central apnea on cerebral blood flow velocity in healthy term infants.

We evaluated a new method of monitoring cerebral blood flow velocity (CBFV) and described changes in CBFV in relation to central apnea in 17 healthy term infants. The area under the velocity curve during apnea did not change, whereas area under the velocity curve per the waveform showed a significant difference, suggesting that stability is maintained through an increase in CBFV with each heartbeat. The maintenance of cerebral hemodynamics during isolated central apnea supports the assumption that these episodes are benign.

Neurodevelopmental Outcomes of Infants Born at <29 Weeks of Gestation Admitted to Canadian Neonatal Intensive Care Units Based on Location of Birth

Objective To compare mortality and neurodevelopmental outcomes of outborn and inborn preterm infants born at <29 weeks of gestation admitted to Canadian neonatal intensive care units (NICUs). Study design Data were obtained from the Canadian Neonatal Network and Canadian Neonatal Follow‐up Network databases for infants born at <29 weeks of gestation admitted to NICUs from April 2009 to September 2011. Rates of death, severe neurodevelopmental impairment (NDI), and overall NDI were compared between outborn and inborn infants at 18‐21 months of age, corrected for prematurity. Results Of 2951 eligible infants, 473 (16%) were outborn. Mean birth weight (940 ± 278 g vs 897 + 237 g), rates of treatment with antenatal steroids (53.9% vs 92.9%), birth weight small for gestational age (5.3% vs 9.4%), and maternal college education (43.7% vs 53.9%) differed between outborn and inborn infants, respectively (all P values <.01). The median Score for Neonatal Acute Physiology‐II (P = .01) and Apgar score at 5 minutes (P < .01) were higher in inborn infants. Severe brain injury was more common among outborn infants (25.3% vs 14.7%, P < .01). Outborn infants had higher odds of death or severe NDI (aOR 1.7, 95% CI 1.3‐2.2), death or overall NDI (aOR 1.6, 95% CI 1.2‐2.2), death (aOR 2.1, 95% CI 1.5‐3.0), and cerebral palsy (aOR 1.9, 95% CI 1.1‐3.3). Conclusions The composite outcomes of death or neurodevelopmental impairment were significantly higher in outborn compared with inborn infants admitted to Canadian NICUs. Adverse outcomes were mainly attributed to increased mortality and cerebral palsy in outborn neonates.

CEREBRAL BLOOD FLOW VELOCITY (CBFV) CHANGES AFTER INDOMETHACIN (INDO) ARE DEPENDENT ON PDA CLOSURE. † 1204

INDO prophylaxis may reduce incidence of IVH in preterm infants. INDO reduces CBFV. Hypothesis: The fall in CBFV after IV INDO is associated with PDA closure. Methods: Continuous recording of CBFV by transcranial pulsed doppler ultrasound, for 30 min. before, and 120 min. after, IV INDO given for hemodynamically significant PDA (n = 12); simultaneous recording of arterial O2 saturation, BP, heart rate and TcPCO2. Echocardiography at 120 min (n = 11). Results: INDO did not change BP. CBFV fell for up to 120 min (maximum 45%, P<0.001); pulsatility index (PI) rose (maximum 77%, P<0.05). Closure of PDA at 120 min. was associated with fall in CBFV (P<0.05) and increase in BP/CBFV ratio (P<0.0001). High LA/AO ratio at 120 min. was associated with rise in PI (P<0.01) and fall in CBFV (P<0.05). Large residual PDA was associated with fall in CBFV (P<0.02), rise in PI (P<0.01), and rise in BP/CBFV ratio (P<0.0001). Conclusions: Effects of INDO on CBFV depend on PDA status. Effects of early INDO prophylaxis on CBFV deserve further investigation.

PS-278 Automated Versus Manual Fio2 Control At Different Saturation Targets In Preterm Infants

Background Preterm infants spend only 50% of time within the target oxygen saturation (SpO2) during manual FiO2 control (M-FiO2). Automated FiO2 control (A-FiO2) improves SpO2 targeting but it is uncertain if this applies to different SpO2 target ranges and during non-invasive support (NIVS) and mechanical ventilation (MV). Objective To compare the efficacy of A-FiO2 vs M-FiO2 in keeping two different SpO2 targets during NIVS or MV. Design/methods Preterm infants on FiO2 >0.21 receiving NIVS or MV were randomised to SpO2 targets 89–93% or 91–95% and underwent M-FiO2 and A-FiO2 for 24 h each, in random sequence. Results 80 infants (GA:26 w, age:18 d) were included (NIVS = 48, MV = 32). Time within target increased and below target decreased during A-FiO2 compared with M-FiO2, especially in the lower target range. There was a reduction in time and hypoxemia episodes with SpO2 < 80% during A-FiO2. Outcomes did not differ between NIVS or MV. Conclusions Automated FiO2 control improved SpO2 targeting across different SpO2 ranges and reduced hypoxemia with less workload during both NIVS and MV. Abstract PS-278 Table 1 Target 89–93% Target 91–95% A-FiO2 M-FiO2 A-FiO2 M-FiO2 %-time in target 62 (17) 54 (16)* 62 (17) 58 (15)* %-time >target 21 (13) 25 (10)* 22 (13) 19 (8) %-time < target 17 (11) 21 (8)* 17 (10) 23 (9)* %-time SpO2 >98% 0.2 (0.0–0.8) 0.7 (0.1–1.6)* 0.7 (0.2–2.1) 1.7 (0.7–4.3) %-time SpO2 < 80% 1.2 (0.2–2.2) 2.6 (1.0–4.3)* 0.8 (0.3–2.1) 2.0 (0.9–5.0)* Episodes < 80%, >1 min/24 h 4 (1–12) 15 (5–24)* 4 (1–11) 13 (3–24)* Manual FiO2 adjustments/24h 1 (0–3) 102 (72–173)* 1 (0–3) 109 (79–156)* * p < 0.05

Effects of Inhaled Nitric Oxide (INO) on Ventilation in Healthy Pretern Infants † 1598

Inhaled Nitric Oxide (INO) improves oxygenation and decreases mortality in newborn infants with persistent pulmonary hypertension. However, the effect of INO on ventilation in preterm infants with healthy lungs is unknown. We hypothesize that INO by preferentially vasodilating capillaries in well ventilated areas would decrease dead space and improve gas exchange in healthy preterm infants. To test this hypothesis, we studied 6 healthy preterm infants[BW 1100±237 g (Mean±SEM); SW 1780±140 g; GA 28±1 wk; PNA 54±14 d] during quiet and REM sleep. A flow-through system and Beckman analyzers were used to measure ventilation and alveolar gases. After a control period in 21% O2, infants inhaled 5 ppm of NO for one hour. In quiet sleep, minute ventilation decreased from 0.305±0.011 (control), to 0.260±0.008 l/min/kg (INO; p<0.001). In REM sleep, it decreased from 0.299±0.007 (control), to 254±0.011 l/min/kg (INO; p<0.001). These changes in ventilation were primarily related to a decrease in frequency during quiet sleep [45±3 breaths/minute (control) vs 36±1 (INO); p<0.01] and to a decrease in tidal volume during REM sleep [7.1±0.3 ml/kg (control) vs 6.6±0.3 (INO); p<0.05]. No significant differences were observed in the number and the density of apneas per hour in any sleep state. Oxygenation, as measured by O2 saturation and TcPO2, did not change significantly during INO in any sleep state. Alveolar PO2 was 100±2 Torr (quiet) and 97±2 (REM) during control and 97±3 Torr (quiet) and 100±4 (REM) during INO (both p=NS). Alveolar PCO2 decreased from 33±1 Torr (control) to 32±1 (INO; p=NS) during quiet sleep and from 31±1 Torr(control), to 29±1 (INO; p<0.05) during REM sleep. These findings suggest that in healthy preterm infants INO makes ventilation more effective; this is indicated by a decrease in minute ventilation during quiet and REM sleep without significant changes in oxygenation and with a slight decrease in alveolar PCO2. We speculate that the more effective ventilation observed during INO is related to capillary recruitment in poorly perfused ventilated alveolar units with a consequent decrease in dead space.

SYSTEMIC AND INTRATRACHEAL STEROIDS DECREASE THE PRODUCTION OF HYALURONAN(HA) AND REDUCE THE EXPRESSION OF THE HA RECEPTOR RHAMM DURING BLEOMYCIN-INDUCED PULMONARY INFLAMMATION. ▴ 1975

We have previously shown that hyaluronan (HA) and its receptor RHAMM(Receptor for HA-Mediated Motility) regulate macrophage motility during bleomycin-induced pulmonary inflammation. Since steroid therapy has been used to limit inflammation in this model, and may limit the severity of bronchopulmonary dysplasia in premature infants, we hypothesized that, in addition to reducing the inflammatory response, steroids would reduce the expression of RHAMM and HA after bleomycin-induced lung injury in rats. Sprague-Dawley rats (n=5/group) were treated with either 30 mg/kg solumedrol intramuscularly (IM), or 1 mg budesonide intratracheally (IT) 1 day prior to and daily for 4 days after intratracheal bleomycin. Control animals received an equal volume of normal saline by the same routes. Four days after injury, bronchoalveolar lavage (BAL) was performed. The cells from the BAL were separated from the lavage fluid by centrifugation. The HA content of the lavage fluid was measured using an ELISA-like assay, while the cells were analyzed for RHAMM content by western blot, and macrophage content by glucosaminidase activity. Steroid treatment decreased HA content of the lavage fluid by 40%, and decreased RHAMM expression by 40% (IM) and 90% (IT) as compared to controls. Both treatments resulted in a 40% reduction in macrophage accumulation in the lungs of injured animals. We conclude that, in addition to decreasing pulmonary inflammation after intratracheal bleomycin, steroids decrease the expression of RHAMM and HA in vivo. We speculate that the anti-inflammatory action of steroids in lung injury may be explained in part by the inhibition of RHAMM and HA.