Brief Reports Indian Pediatrics 2000;37: 522-526 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Use
of High Frequency Ventilation as a Rescue Measure inPremature Babies
with Severe Respiratory Failure
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rajiv
Aggarwal
Lynette Downe*
Respiratory failure is still a major cause of morbidity and mortality in newborn infants. Various modes of ventilation have been tried with an aim to reduce this mortality. High frequency ventilation (HFV) has been trialed in preterm babies with respiratory distress syndrome (RDS)(1-4). Although, HFV was associated with improved oxygenation, these studies did not show any difference in the outcome of their patients. The mortality rate, incidence of air leak, bronchopulmonary dysplasia (BPD), intraventricular hemorrhage (IVH) and treatment failure in patients treated with HFV did not differ significantly as compared to patients treated with conventional ventilation. However HFV may have a role in those babies who do not respond initially to conventional ventilation. We report a case series where high frequency oscillatory ventila-tion (HFOV) was utilized as a rescue measure in 9 preterm infants with respiratory failure who did not respond to conventional mechanical ventilation (CMV).
Nine premature babies who developed respiratory failure while on conventional mechanical ventilation between January 1996 and June 1996 are reported here. Respiratory failure was considered as arterial carbon dioxide tension (PaCO2) above 55 mmHg with pH below 7.25 or arterial oxygen tension (PaO2) below 50 mmHg. These infants were initially treated on CMV with respiratory rates of 60-100 cycle/m, FiO2 up to 1.0 and peak inspiratory pressures (PIP) of 25-35 cm H2O. If satisfactory blood gases were unable to be maintained on optimal CMV, they were switched to HFOV. The decision to change over to HFOV was made by the clinician in charge. HFOV was provided by a Drager Babylog 8000 oscillator fitted with controls for Amplitude, Frequency and Mean Airway Pressure (MAP). Initial MAP used on the HFOV was 1-2 cm H2O higher than that of CMV. Frequency was set at 10 Hz and amplitude was gradually increased until adequate chest movements could be seen. An effort was made to maintain on oscillatory volume of at least 2 ml/kg. Adequate inflation of the chest was assessed by a CXR done within 30-60 minutes of commencing HFOV. Lung inflation was considered optimum if chest X-ray showed the eighth posterior rib level expansion and decreased lung opacification. MAP was adjusted to obtain adequate oxy-genation and FiO2 gradually weaned. When the FiO2 was less than 0.6, MAP was given equal priority in the weaning process. The settings were adjusted to maintain the blood gases within the desired range and at the same time avoiding lung overdistention. Infants were transferred back to CMV on minimal settings when MAP became stable below 8 cm H2O and FiO2 below 0.3. Patients were evaluated for PDA and IVH using standard protocols and techniques. Blood pressure was monitored continuously using an arterial line. All infants were sedated with fentanyl and paralyzed with vecuronium only if required. After stabilization of blood gases on HFOV, the babies were allowed to recover from the paralysis and breathe spontaneously. Outcome variables documented included survival, duration of ventilation and oxygen requirement, incidence of chronic lung disease (CLD), PDA, necrotising enterocolitis (NEC), retinopathy of prematurity (ROP) and results of cerebral ultrasound examinations.
The clinical data of the nine infants transferred to HFOV is presented in Table I. The 8 babies with a diagnosis of RDS received 2 doses of surfactant; the first within 4 hours of birth and the second 12 hours after the first dose. When the infant was on HFOV at the time of surfactant administration, this was instilled during hand bagging after which the HFOV was recommenced. All infants responded well to HFOV (Table II). Within 24 hours of commencing HFOV, there was an improvement in the oxygenation index (OI), ventilation index (VI) and a progressive decrease in the MAP and FiO2 required for normal blood gases. Five of the nine babies had changes of PIE on the chest X-ray prior to commencing HFOV. These babies showed improvement in their PIE on HFOV. One of the babies with PIE developed a pneumothorax within 24 hours of commenc-ing HFOV and required intercostal drainage and subsequently improved. One baby died of candida sepsis after switching back to CMV on day 14. This baby had been weaned to CMV on day 8 of life. Two of the surviving eight babies (25%) continued to require oxygen and developed chronic lung disease (CLD). One of these babies continued to require oxygen at 40 weeks corrected age and was discharged on home oxygen therapy. Postnatal steroids were used in 6 of the surviving 8 babies. All babies were extubated by 35 weeks corrected post menstrual age. Three babies developed PDA and three babies developed NEC. Retinopathy of prematurity was diagnosed in 2 of the eight babies which regressed spontaneously. One baby developed a grade IV IVH and one baby was diagnosed to have periventricular leukomalacia (PVL) at two weeks of age. Table I - Clinical Data
Abbreviations: PIE = Pulmonary interstitial emphysema; PTX = Pneumothorax; Pneum = Pneumonia; RDS = Respiratory distress syndrome. * died at 14 days of Candida septicemia. Table II - Changes in Ventilation with Time on HFV
FiO2
= fraction inspired oxygen; HFVL = last parameters on HFOV prior to
CMV; MAP = mean airway pressure; OI = (MAPXFiO2 ´ 100) ¸ PaO2; VI
= MAP ´ CO2
High frequency oscillatory ventilation (HFOV) is a viable option as rescue therapy for premature infants "failing" optimal CMV. In our series of nine babies, all infants showed improvement in their respiratory status after commencing HFOV. Previous studies done in the pre surfactant era have also shown HFV to be effective in the management of respiratory failure in premature infants(5,6). Chan et al.(7) have used HFOV and surfactant in the treatment of respiratory failure in preterm newborns and shown progressive improvement in their respiratory status. However they reported very high mortality with 47% of their babies dying of respiratory failure. McDougall et al.(8) in their series of 39 babies with respiratory failure managed on HFOV reported 28% mortality. However, their mortality in the group weighing below 1500 g was 47%. In our series, the single mortality was due to candida septicemia after having been weaned to CMV. HFV has been found to be beneficial in the treatment of PIE. Keszler et al(9). had used HFV and shown that newborn infants with PIE were more likely to respond to HFV than to CMV (61% vs 37%; p <0.01). Other trials done without the use of surfactant have also indicated that HFV may be beneficial in the treatment of air leak syndromes in small babies(5,6). In our series with the additional use of surfactant, all babies showed improvement in their interstitial emphysema while on HFOV. The newborn with PIE developed a pneumothorax soon after initiation of HFV. However this baby also showed consistent improvement in his respiratory condition with HFV. Although babies showed improvement in their respiratory failure and PIE, HFOV did not have any beneficial effect on CLD and day of extubation. Babies were weaned to CMV but continued to require minimal ventilation for bronchopulmonary dysplasia (BPD) and the eight surviving babies were extubated only by 35 weeks corrected age. Hence in our series, HFV did not have any beneficial effect on the long term morbidity with 75% babies needing postnatal steroids for BPD and 25% babies developing CLD. During HFOV the oxygenation is depen-dant upon mean airway pressure and a higher MAP than that of CMV may initially be required for adequate oxygen values. In our series we increased the MAP from CMV by at least 1-2 cm H2O. Increase in the MAP was not associated with hypotension. This step of slightly increasing the MAP when changing over from CMV to HFOV and also post suctioning seems to be of vital importance because unless the MAP overcomes the critical opening pressure of the diseased lung, adequate gas exchange cannot be ensured. MAP was weaned according to improvement in oxygenation. There has been concern that HFV has been associated with increased intraventricular hemorrhage(1,4). Two babies in our series were diagnosed to have serious intracerebral complications. We share the concern of previous authors regarding the intracerebral complications. However, these may also be due to the severity of the respiratory distress rather than the mode of ventilation used. Associated hypotension requiring the use of inotropes would increase the risk of IVH and PVL in these neonates with respiratory failure. Our preliminary experience suggests that HFOV may be safely and effectively used as rescue treatment in preterm babies who have failed to respond to conventional ventilation, especially in those with pulmonary interstitial emphysema. Contributors: RA coordinated the study (particularly its design, data collection and interpretation) and drafted the paper; he will act as the guarantor for the paper. LD helped in designing the study and drafting the paper. Funding:
None.
|