Abstract 1807 Poster Session I, Saturday, 5/1 (poster 32)

Rapid changes in lung compliance in ventilated preterm infants may result in excessive tidal volume (TV) leading to volutrauma and hypocarbia. Volume-targeted ventilatory modes servo-regulate peak pressure in response to changing compliance to deliver a target TV. We hypothesized that the volume-targeted mode would maintain the TV and PaCO2 within a target range of 4-6 ml/kg and 35-45 torr, respectively, more consistently than conventional modes.

Methods: We recorded exhaled TV at 1 min. intervals during the first 24h of life or until extubation, whichever came first. After informed consent was obtained, 10 infants were randomly assigned to receive assist/control (A/C) ventilation or A/C with volume guarantee (A/C+VG) using the Draeger Babylog infant ventilator (Lubeck, Germany). TV was measured by the hot wire anemometer flow sensor of the ventilator and downloaded directly from the RS232 communication interface. Arterial blood gases (ABG) were obtained q2-4h clinically indicated. In the A/C mode peak pressure was set to obtain an initial TV of 4-6 ml/kg and adjusted according to ABG. In the A/C+VG mode target TV was set at 5ml/kg and lowered if hypocarbia developed. No sedation or muscle relaxants were used. The trigger sensitivity was kept at the maximum setting. The number of breaths and PaCO2 values outside the target range were compared by Chi square. Birth weight (BW), gestational age (GA) and blood gas values were analyzed by unpaired t-test. A p value <0.05 was deemed significant.

Results: Five infants were assigned to each group. The mean BW and GA (1463±391 g and 31.4±1.9 wk for A/C vs. 1634±508 g and 31.2±2.4 wk for A/C+VG) were similar. No autotriggering was observed. The table below shows the number of breaths and PaCO2 values outside the target range. *p<0.0001, compared to A/C.

Table 1 No caption available

We conclude that A/C+VG, as used in this study, keeps TV within the target range more consistently than A/C alone, but does not reliably prevent hypocarbia. Unsedated preterm infants often spontaneously hyperventilate in the first 24h of life for reasons that are poorly understood. It is unlikely that the 5 ml/kg target TV was too high, because the infants were often generating TV above the target range with good spontaneous effort while peak pressure was far below the set limit. Most infants were also tachypneic (mean respiratory rate of 65.2 and 72.6 br./min, respectively, with set rate of 40). Frequent handling and procedures in the first hours of life (including blood gas drawing) aggitate the infants. This appears to contribute to the spontaneous hyperventilation observed in this study.

Supported in part by Draeger Inc.