2006 OPEN FORUM Abstracts
High-frequency Oscillatory Ventilation in Adults: The Initial Experience in Singapore General Hospital
Peggy Ler, RRT, Nancy Lew, RRT, Tan Herng Lee, RRT, Constance Lo, MBBS, MRCP, FAMS, FCCP, FACP, MRCP, FAMS, FCCP,
Kenneth Chan, MBBS, Loo Chian Min, MBBS M Med (Int Med), MRCP, FAMS, Philip Eng, MBBS, M Med (Int Med)
From Department of Respiratory and Critical Care Medicine, Singapore General Hospital, Singapore
Background: High-Frequency Oscillatory Ventilation (HFOV) is a ventilatory option in patients with Acute Respiratory Distress Syndrome (ARDS) who have failed conventional mechanical ventilation (CMV). HFOV was first introduced into the Singapore General Hospital medical intensive care unit in 2004. The objective of the study is to evaluate the safety and improvements in oxygenation in our preliminary group of patients who were commenced on HFOV.
Methods: We conducted a retrospective review of medical records on all patients who received HFOV from February 2004 to October 2005. 8 patients, all Chinese males, were enrolled in this study. The data collected included: patient demographics, co-morbidities, etiology of respiratory failure, duration of CMV, ventilation settings, duration of ARDS prior to HFOV initiation, blood gases, hemodynamic parameters, complications, successful weaning to CMV, as well as, from mechanical ventilation and outcome.
Results: The mean age (+/- SD) of the patients was 63 +/- 19 years and their mean APACHE II score was 30 +/- 6. The primary etiology for respiratory failure was severe pneumonia for all patients, 3 of the patients were in septic shock. These patients received a mean of 4.1 days of CMV and were diagnosed with ARDS for a mean of 3.0 days, before HFOV was initiated. The patients were placed on HFOV for a mean of 2.2 days. The average value of mean airway pressure (MAP) during HFOV was markedly higher than during CMV (p<0.03). MAP increased from 26 +/- 3 to 33 +/- 6 at 24 hours post HFOV initiation (p=0.014). The mean PaO2/FiO2 increased within an hour HFOV was started. By 24 hours of HFOV initiation, there were significant improvements of PaO2/FiO2 from 109 +/- 45 to 138 +/- 47 (p=0.04). 8 hours after HFOV was initiated, the mean oxygenation index decreased from 31 +/- 13 to 27 +/-13, but this did not reach statistical significance. By 8 hours post institution of HFOV, a statistically significant drop in PCO2 was observed (p=0.022). No significant changes were seen in the hemodynamic parameters although we observed an increase in vasopressor support during HFOV in 3 patients. 2 cases of barotrauma were observed: first case of pneumothorax and pneumomediastinum, and second case of subcutaneous emphysema. The latter case was withdrawn from HFOV due to increase in subcutaneous emphysema. The mortality rate at 30 days was 50%. Of the 5 patients who were eventually weaned to CMV; 3 were extubated, 1 died from overwhelming sepsis whilst the other died from an acute bowel perforation. The other 2 patients died from refractory hypoxemia and 1 from cardiac arrest.
Conclusions: Using high constant mean airway pressure with very small tidal volumes, HFOV has the theoretical advantage of minimizing volutrauma and atelectrauma. Thus, HFOV is especially appealing as an ideal lung protective ventilatory strategy. Improved lung recruitment from the high mean airway pressures allows better oxygenation. HFOV is a promising tool in our armamentarium of ventilatory strategies for ARDS. However, due to the different respiratory physiology involved in applying this mode of ventilation, HFOV should be ideally used only in institutions with experienced personnel.