Elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants.

backgroundRespiratory failure due to lung immaturity is a major cause of mortality in preterm infants. Although intermittent positive pressure ventilation (IPPV) saves lives, lung distortion during its use is associated with lung injury and chronic lung disease (CLD). Conventional IPPV is provided at 30-80 breaths per minute while a newer form of ventilation called high frequency oscillatory ventilation (HFOV) provides 'breaths' at 10-15 seconds. This has been shown to result in less lung injury in experimental studies.

objectivesThe objective of this review is to determine whether the elective use of high frequency oscillatory ventilation (HFOV) as compared to conventional ventilation in preterm infants who are mechanically ventilated for the respiratory distress syndrome decreases the incidence of chronic lung disease (CLD) without adverse effects. SEARCH STRATEGY: Searches were made of the Oxford Database of Perinatal Trials, MEDLINE, EMBASE, previous reviews including cross references, abstracts, conferences and symposia proceedings, expert informants, journal handsearching by the Cochrane Collaboration, mainly in the English language. Expert informant's search in the Japanese language was made by Prof. Y. Ogawa. SELECTION CRITERIA: Randomized controlled trials comparing HFOV and CV in preterm or low birth weight infants with pulmonary dysfunction, mainly due to RDS, who are to be given IPPV. Randomization and commencement of treatment should have been as soon as possible after the start of IPPV and usually in the first 12 hours of life. DATA COLLECTION AND ANALYSIS: The methodological quality of each trial was independently reviewed by the various authors. Each author extracted data separately; they were compared and differences were resolved. The standard method of the Cochrane Neonatal Review Group was used to synthesize the data using relative risk (RR) and risk difference (RD). From 1/RD the number needed to treat (NNT) for benefits, and number needed to harm (NNH) for adverse effects, were calculated. MAIN

resultsMeta-analysis of the six eligible studies comparing HFOV with CV revealed that there is no difference in mortality. There are trends toward decreases in CLD in survivors at 28-30 days, 'death or CLD at 28-30 days' and CLD in survivors at 36-37 weeks postmenstrual age or discharge in the HFOV group. However, there are trends towards increases in severe (grades 3 & 4) intraventricular hemorrhage (IVH) and in periventricular leukomalacia (PVL) in the HFOV group. HFOV results in a small increase in any air leak syndrome (ALS), [summary RR 1.20 (1.03, 1.39)]. Only 2 trials have included neurodevelopmental follow up and more survivors in the HFOV group are abnormal [summary RR 1.26 (1.01, 1.58)]. In the subgroup of four trials where a high volume strategy (HVS) was used, HFOV results in more favourable pulmonary outcomes. There are significantly lower rates of CLD in survivors at 28-30 days [summary RR 0.53 (0.36, 0.76)] and of 'death or CLD at 28-30 days' [summary RR 0.56 (0.40, 0.77) with a non-significant trend towards a reduction in oxygen use at 36-37 weeks postmenstrual age or discharge [summary RR 0.74 (0.55, 1.01)]. There were no differences in the rates of IVH or PVL. Of the four trials in the subgroup using surfactant routinely, three also used the HVS. The trends in results were similar with surfactant to those for the HVS subgroup analysis. One trial suggests that HFOV may reduce the cost of in-hospital care. In the subgroup of two trials (HIFI 1989, Rettwitz-Volk 1998) not using a HVS there is no effect of HFOV on the rate of CLD; however, there is an increase in the rate of PVL [summary RR 1.64 (1.02, 2.64). REVIEWER'S

conclusionsThe overall meta-analyses is dominated by the large HIFI study which did not use the HVS recommended on the basis of animal studies, and in which surfactant was not available. Studies which used HVS have shown some benefits in short term measures of CLD without an in