Lung Recruitment and Breathing Pattern During Variable Versus Continuous Flow Nasal Continuous Positive Airway Pressure in Premature Infants: An Evaluation of Three Devices

DISCUSSION

Few comparative studies of nasal CPAP devices have been published. In particular, we are unaware of any studies that compare lung recruitment with different nasal CPAP devices as they are currently being used in infants. The main objective of this study was to compare the efficacy of 3 distinct devices—2 continuous flow devices and 1 variable flow device—in recruiting lung volume at various nasal CPAP levels.

Factors determining the effectiveness of any nasal CPAP device include its associated work of breathing, flow characteristics, ease of application, and the comfort level of the infant once the device is in place. Continuous flow nasal CPAP is increased or decreased by varying the resistance to exhalation at the exhalation valve on an infant ventilator. Nasal prongs are commonly used to provide continuous flow nasal CPAP. Concerns exist, however, about increased work of breathing with nasal prongs, compared with face mask CPAP.¹¹Additionally, nasal prongs often become dislodged making care of these infants difficult. Locke et al¹² demonstrated that inadvertent CPAP can be provided with flow through a nasal cannula. Because nasal cannulas are easy to apply and keep in place, their use had become popular in our neonatal intensive care unit to provide nasal CPAP. This cannula system had not been investigated, and the amount of CPAP actually generated was not clear.

The variable flow nasal CPAP device, in contrast to the continuous flow devices, generates CPAP at the airway. It uses jet flows at high velocity, which can entrain gas to assist inspiration on demand and keep the CPAP level constant. On exhalation, the design of the nasal prongs results in gas flow being shunted through an expiratory outlet rather than continue toward the nares, which can increase expiratory work.^13–15 Two physical model studies of the variable flow device have been published. Moa et al¹³ compared it with a continuous flow system using a lung model and found that the variable flow device showed less variation in mean airway pressure and external workload. Using a simulated breathing apparatus, Klausner et al¹⁵ compared work of breathing via nasal prongs with the variable flow system and a commonly used continuous flow device. They concluded that the imposed work of breathing with the variable flow system was approximately one fourth that of the continuous flow system. We are aware of only one clinical study comparing a variable flow device with a continuous flow device. Ahluwalia et al¹⁶ compared the variable flow system with nasal CPAP delivered by an endotracheal tube inserted into one nostril. In a crossover design in 20 infants over a total study time of 8 hours, no differences were found in Fio₂, respiratory rate, heart rate, blood pressure, or comfort score between devices.

To evaluate lung recruitment with the 3 nasal CPAP devices, we applied DC-coupled RIP to monitor static lung volume changes. Inline pneumotachography cannot be used to measure lung volume changes once a nasal CPAP device is applied to the infant, and changes in breathing synchrony cannot be assessed with pneumotachograpy. Respiratory inductance plethysmography, however, is a noninvasive technique for obtaining real-time data on breathing patterns. Accurate calibration of RIP using a face mask with pneumotachograph has been demonstrated in both lambs and premature infants.^6,7 DC-coupled RIP allows for a constant baseline so that ΔV_L can be measured. These changes can then be converted to volume (in mL) based on previous calibration by facemask pneumotachography. The sum of the rib cage and abdominal motion equals the tidal volume, and the phase angle between the rib cage and abdominal motion indicates the degree of paradoxical breathing. An increase in paradoxical breathing may indicate an increased work of breathing.¹⁷ Additionally, Locke et al⁴ have shown that in infants with respiratory insufficiency, increasing CPAP results in a decreasing phase angle that is directly and significantly correlated with changes in esophageal pressures.

Our results indicate that the variable flow nasal CPAP system recruits lung volume better than both continuous flow devices. The θ (hence breathing synchrony) obtained with the continuous flow nasal prong system was, surprisingly, similar to that obtained with the variable flow system despite significantly higher lung recruitment with the latter. There are 2 possible, albeit related,explanations for this intriguing finding. First, the decrease in θ with lung recruitment (or nasal CPAP) may be nonlinear. Second, the decrease in θ with recruitment may change or possibly reverse when the lungs are either overdistended (decreased compliance) or above an optimal lung volume. The presence or absence of these conditions is not available from our data, which only show ΔV_Lfrom baseline (on nasal CPAP = 0).

The decreased variability in the mean airway pressure during nasal CPAP provided with the variable flow device^13–15 compared with conventional nasal CPAP systems is perhaps the critical factor leading to the increased lung recruitment with this system. Another factor that alters the efficacy of any nasal CPAP delivery system is the airway leak around the prongs. Because the prongs are mechanically in parallel with the lungs, a larger leak around them results in lower effective nasal CPAP or less recruitment. The use of relatively larger prongs is possible with variable flow systems, both because of the design of the delivery system and the prongs themselves. When the nasal CPAP level is reached at the proximal airway (nares), the inflow from the device is shunted away from the infant.^13–15 The prongs are made of a thin, soft material, which may flare out during gas inflow, thus increasing the effective internal diameter and decreasing the leak around the prongs. In contrast to conventional nasal CPAP, the wider ID and thinner walls of the prongs coupled with no gas inflow during exhalation all may effectively reduce the imposed work of breathing because of the mechanical device.

Differences in flow rate inherent between the devices may also affect lung recruitment. Although variable flow devices by definition vary CPAP attained by varying the flow rate, actual flow delivered to the nares of the infant has not been assessed. Much of the flow with these systems is shunted away from the infant and out the expiratory limb of the CPAP circuit.

Continuous flow nasal CPAP via the modified nasal cannula recruited lung volume equally to the nasal CPAP prongs, but at a very high cost: RR, θ, and Fio₂ all increased significantly with the modified nasal cannula. In the study by Locke et al,¹² no CPAP was generated with a nasal cannula similar in size to the one we used; however, they did not increase the flow rate beyond 2 L/minute. CPAP provision with the cannula as used in our nursery may possibly occur not only because of the high flow rates used but also because of the design of the cannula base, which often covers much of the nares and may thus obstruct expiration. These findings with the modified nasal cannula device not only indicate that use of such a system for provision of nasal CPAP should be discouraged but raise some questions as well about the respiratory effects of providing oxygen to the preterm neonate by nasal cannula.