In reply:
Responding to my editorial1 that accompanied the research by Delisle et al,2 Sinderby raises 3 main questions. First, Sinderby asks why I summarize the outcome of the study with: “What is the clinical importance of this study; the ‘So what?’ if you will.” A well designed physiological study such as that by Delisle et al2 provides essential evidence of a treatment's functional effects. Table 1 identifies other primary research that has identified statistically significant changes relevant to the implementation of neurally adjusted ventilatory assist (NAVA) and to the care of critically ill patients.
Variables Demonstrating a Statistically Significant Difference in Primary Research Comparing NAVA to an Alternative Ventilation Strategy
As the body of the physiological evidence relating to NAVA has grown, authors and researchers have hypothesized expected clinical outcomes, such as a decreased number of ventilator days16 and decreased ICU stay and hospital stay.23 A smaller number of studies have investigated any association between patient-ventilator synchrony and clinical outcomes. Two projects have identified an association between patient-ventilator trigger asynchrony and longer duration of mechanical ventilation.24,25 De Wit et al demonstrated an association between ineffective triggering and longer duration of mechanical ventilation, longer ICU and hospital stay, and a reduced likelihood of discharge to home.26 De Wit et al have also identified a statistically significant relationship between ineffective triggering index and deeper sedation level: a finding that may result in false conclusions of weaning intolerance and longer duration of mechanical ventilation.27 A recent study including adult trauma subjects identified no association between patient-ventilator synchrony and ventilator days, ICU and hospital stay, proportion of subjects who were discharged to home, or mortality.28
I certainly acknowledge that consistently demonstrated outcomes of NAVA (such as enhanced patient-ventilator synchrony) also happen to be those factors identified with desired clinical outcomes (such as a reduced number of ventilator days), though I have identified no studies to date that have investigated a direct link between the two. Though preliminary research outcomes related to NAVA are promising, well designed studies are needed to directly measure clinical outcomes.3,6,7,10,12,13,17–20,22,25,29–31
Referring to the failure of conventional ventilators to provide ventilation synchronous with patient efforts, Sinderby asks, “is it not in the best interest of respiratory therapists to reduce this problem?” I would respond with a resounding Yes! Respiratory therapists are on the front lines, witnessing trigger, cycle, and flow asynchrony at the bedside every day. We have also, however, been party to the premature application of intermittent positive pressure ventilation: a technology that seemed to make sense but that eventually was rejected for lack of evidence of improved clinical outcomes.
The primary interest of respiratory therapists is patient safety. NAVA has been granted regulatory approval by the United States Food and Drug Administration, and evidence suggests NAVA can be employed safely with neonatal, infant, pediatric, and adult populations. Since primary safety considerations have been addressed, economic considerations are relative to Sinderby's question. NAVA is currently licensed to a single ventilator manufacturer: Maquet Medical Systems. Table 2 indicates expenses related to the purchase of one NAVA-ready Servo-i ventilator and 12 NAVA-ready Servo-i ventilators: the mean number of ventilators owned by acute care hospitals in the United States.32 Table 3 indicates costs relating to the purchase of NAVA software and accessory hardware for facilities that already own Servo-i ventilators. Additional departmental expenses would include time for staff training and equipment maintenance.11 Considering that the United States gross domestic product used by critical care medicine recently increased by 13.7%,33 considering the unique challenges associated with cost containment in critical care,34 and considering that federal funding from the Centers for Medicare and Medicaid Services in the United States is increasingly dependent upon demonstrated effectiveness of therapeutic interventions, using technology without demonstrated improvement of patient outcomes imposes fiscal risk related to uncertain future reimbursement.
Purchase Price of NAVA-Ready Servo-i Ventilator and EAdi Catheter
Purchase Price of NAVA Software and Accessory Hardware Upgrade
The best interests of respiratory therapists also consider the sleep quality of our patients. In addition to addressing the problem of patient-ventilator asynchrony, the research by Delisle et al2 addresses a deficit in our understanding of sleep disturbances in the ICU.35 Pressure support ventilation (PSV) in particular has been examined as an alternative approach to reducing apnea and improving sleep quality of mechanically ventilated patients. Roche-Campo et al studied tracheostomized subjects and concluded that PSV was associated with a higher total sleep time than spontaneous ventilation, though they identified no significant differences in measures of sleep quality.36 Cabello et al examined clinician-adjusted PSV, automatically adjusted PSV, and continuous mandatory ventilation, and identified no statistically significant differences among the interventions with regard to rapid-eye-movement (REM) sleep duration, fragmentation index, number of ineffective breath efforts, or central apneas.37 When PSV was used as a common comparison, improvements related to sleep have been demonstrated with other mechanical ventilation strategies, including: proportional assist ventilation (decreased patient-ventilator asynchrony events,38,39 decreased sleep fragmentation,39 increased REM sleep time,39 increased slow wave sleep time39), NAVA (increased REM sleep time, decreased sleep fragmentation index, decreased ineffective breath efforts, decreased frequency of central apnea),40 titration of PSV to inspiratory muscle effort at night (improved gas exchange, increased sleep efficiency, increased REM sleep, decreased ineffective breath efforts),41 and continuous mandatory ventilation (decreased arousals and awakenings, decreased central apneas).42 All studies were insufficiently powered (sample size range 9–16) to yield statistically significant, generalizable findings. None included consideration of broad clinical outcomes.
Sinderby states, “To suggest that only outcome data from randomized controlled trials are required before new modes can be used simply perpetuates a myth and blinds us to other approaches to rationally selecting the best treatment options.” In point of fact, the 2 primary suggestions of my editorial were: there is insufficient evidence of improved outcomes with NAVA, and until improved outcomes are demonstrated, clinical managers are unlikely to invest in NAVA.
Sinderby's final comments and questions relate to the methodological challenges of randomized controlled trials (RCTs), and echo concerns specific to RCTs in the critical care environment, which have been articulated by Vincent.43 Because well designed RCTs permit cause-and-effect conclusions, they are likely to remain the gold standard of experimental research methods in healthcare research, despite the fact that they are fraught with challenges. Three of the challenges enumerated by Vincent are particularly relevant to the application of NAVA in the critical care environment. First, adequately powered research will likely require multiple research centers and a lengthy enrollment period. Second, statistically significant differences are more likely to be identified if the study subjects are at greater risk of patient-ventilator asynchrony, and the literature suggests this distinction would include pediatric patients3,7 and patients with COPD.11,17,21,24,44 Third, end points should be chosen carefully, and we might give special consideration to the current healthcare reimbursement context. Despite these challenges, I would join with the authors of multiple smaller physiologic studies3,6,7,10,13,17–20,22,23,26–28 to call for appropriately powered, prospective RCTs comparing the effects of conventional ventilation and NAVA on important patient outcomes, especially given the promising research on NAVA in the past decade.
Footnotes
The author has disclosed no conflicts of interest.
- Copyright © 2013 by Daedalus Enterprises
