Predicting survival infants ventilated with high-frequency oscillation

Citation
B. Simma et al., Predicting survival infants ventilated with high-frequency oscillation, WIEN KLIN W, 112(18), 2000, pp. 804-810
Citations number
25
Categorie Soggetti
General & Internal Medicine
Journal title
WIENER KLINISCHE WOCHENSCHRIFT
ISSN journal
00435325 → ACNP
Volume
112
Issue
18
Year of publication
2000
Pages
804 - 810
Database
ISI
SICI code
0043-5325(20000929)112:18<804:PSIVWH>2.0.ZU;2-Y
Abstract
This retrospective study identifies factors which may predict outcome in pr eterm infants and infants born at term, ventilated with high-frequency osci llation (HFO). In a 16-bed neonatal and paediatric intensive care unit (lev el ill), 58 consecutive preterm and term infants with a median gestational age of 30 (24-41) weeks and a median birth weight of 1200 (520-3660) g suff ered respiratory failure and were managed with HFO as rescue therapy. Forty -nine patients (84%) received exogenous surfactant before HFO which was ini tiated after a median interval of 20 (1-910) hours following birth. The ove rall survival rate was 70%. No significant differences existed between surv ivors and nonsurvivors in respect to demographic data. A greater proportion of patients with respiratory distress syndrome survived (76%) than with lu ng hypoplasia (20%) or with air-leak syndromes (pulmonary interstitial emph ysema 60%, pneumothorax 28%). In survivors, the mean oxygenation index (OI) before HFO was significantly lower than that in nonsurvivors (16 +/- 1.3 v s. 26 +/- 3, p < 0.01) and showed a significant reduction of 32% within 4 h ours. in contrast, mean OI increased to 68% over the first 4 hours in nonsu rvivors and the difference between survivors and nonsurvivors remained sign ificant during this time (after 2 hours: 15 +/- 1.5 vs. 30 +/- 6, p < 0.01; after 4 hours: 11 +/- 1 vs. 43 +/- 1, p < 0.01). A receiver of operator ana lysis revealed that an initial OI < 25 or alveolar-arterial oxygen differen ce (AaDO2) < 450 mmHg predicted survival with a sensitivity of 93% and a sp ecificity of 41%. The positive predictive value was 79%; the negative predi ctive value, 70%. Conclusion: A low OI and AaDO(2) at the beginning of HFO, improvement in ox ygenation over the first four hours of HFO ventilation and no development o f air-leak syndromes were associated with a high predicted survival. This a llows early identification of infants who may not survive and may benefit f rom established and alternative modes of respiratory support such as extrac orporeal membrane oxygenation, nitric oxide and liquid ventilation.