doi:10.1111/j.1440-1754.2006.00813.x ORIGINAL ARTICLE Does measuring respiratory function improve neonatal ventilation? Jan Klimek, 1 Colin John Morley, 2 Rosalind Lau 2 and Peter Graham Davis 2 1 NICU, Nepean Hospital, New South Wales, 2 Neonatal Services, Royal Women’s Hospital, Victoria, Australia. Aims: To determine whether using a respiratory function monitor alters clinicians’ choice of ventilator settings, tidal volumes or blood gases in the first 48 h of ventilation. Methods: Clinicians were trained to use a respiratory function monitor to optimize neonatal ventilation. Thirty-five infants, weighing < 2 kg, treated with the Infant Star ventilator were randomized to have a respiratory function monitor display visible or concealed. All reasons for altering ventilator settings were noted. Data on ventilator parameters and clinical care were collected hourly. The primary outcome was the mean peak pressure used during the first 48 h. Results: There were no statistically significant differences in peak pressures, tidal volumes or arterial carbon dioxide levels between the two groups. Conclusions: Using the Florian respiratory function monitor in the first 48 h of ventilation with the Infant Star ventilator did not alter the choice of ventilator settings, tidal or minute volumes or arterial blood gases. Possible explanations for this result include lack of power due to the small numbers recruited and bias due to the unblinded nature of the trial. Key words: infant newborn; infant premature; pulmonary ventilation; tidal volume; ventilators mechanical. Until recently clinicians have used observation of chest wall move- ment and blood gases to guide their choice of rates and pressures to ventilate newborn infants. 1,2 In an effort to reduce volutrauma and bronchopulmonary dysplasia, newer modes of ventilation targeting tidal volumes have been introduced and widely accepted. Respiratory function monitors (RFMs) measure and display airway pressure and gas flow in and out of the endotracheal tube using a pressure transducer and a small, light, low dead space, pneumota- chograph. In addition the monitor’s software calculates and displays respiratory rate, inspired and expired tidal volumes, leak around the endotracheal tube, minute volume, peak pressure, mean pressure and positive end expiratory pressure. RFMs also provide information on inspiratory and expiratory times as well as indicating the degree of synchrony between the infant’s spontaneous breaths and ventilator inflations, potentially allowing clinicians to adjust ventilator settings to optimize assisted ventilation (3–6). Although most new neonatal ventilators now measure many of these parameters as standard, the Infant Star ventilator (Puritan Ben- nett, Pleasanton, CA) does not (7,8). The Infant Star ventilator is widely used in Australasian neonatal units. The aim of this study was to de- termine whether the use of a RFM, as an adjunct to the Infant Star ventilator, during the first 48 h of neonatal ventilation would alter ventilator settings, tidal volumes and arterial blood gases. Correspondence: Dr Jan Klimek; NICU, Nepean Hospital, PO Box 63, Penrith, NSW 2751, Australia. Fax: +61 2 4734 2698. email: KlimekJ@wahs. nsw.gov.au Accepted for publication 8 September 2005. Materials and Methods The RFM used was the Florian Infant Graphic Monitor, (Acutronic Medical Systems AG, Zug, Switzerland). This is a portable, stand- alone monitor that weighs 3.1 kg and measures 25 × 15 × 15 cm. It measures and displays flow volume and pressure curves, flow- pressure loops, pressure-volume loops, airway pressures, expired tidal volume, endotracheal tube leak, peak inflating pressure, pos- itive end expiratory pressure, respiratory rate and minute volume on intubated patients. A reusable hot-wire flow sensor is positioned between the ventilator circuit Wye piece and the endotracheal tube connector. It has a 1 mL dead-space and is accurate during neonatal ventilation to ± 8%. Prior to the study, staff were trained in the use of the RFM and interpretation and use of the information it provided. They were also given a purpose-made booklet with this information. Guidelines targeting tidal volumes to ∼5 mL/kg and minute volumes to ∼200– 300 mL/kg were provided. Babies of birth weight <2000 g, ventilated with the Infant Star and less than 24 h old were eligible for the study. We excluded infants with air leak or congenital anomalies potentially affecting ventilation. All infants had the Florian sensor in place within 3 h of starting ventilation. Infants were randomized to a ‘Florian visible’ or ‘Florian concealed’ group. Randomization was based on a random number table and the group of allocation placed in a sequentially numbered, sealed, opaque envelope. All the infants were treated using unit protocols including at least 6-hourly blood gases. The ventilation mode (Inter- mittent Mandatory Ventilation, Synchronous Intermittent Mandatory Ventilation, Assist Control or High Frequency Oscillatory Ventilation) was chosen by the attending neonatologist. Data were collected for the first 48 h of ventilation. 140 Journal of Paediatrics and Child Health 42 (2006) 140–142 C  Paediatrics and Child Health Division (Royal Australasian College of Physicians)