Introduction

Partial ventilatory support is increasingly used in the intensive care unit [1]. The most widely used mode of partial ventilation is pressure support (PSV) [1], in which a constant preset level of pressure assists each inspiration regardless of the patient’s inspiratory effort. Mismatch between patient respiratory drive and level of assistance can therefore occur and can be potentially harmful by causing lung overdistension and subsequent lung injury [2], respiratory discomfort [3, 4] and patient–ventilator asynchrony [5]. Neurally adjusted ventilatory assist (NAVA) delivers a pressure that is proportional to the electrical activity of the diaphragm (EAdi), a surrogate of the neural drive to breathe (see Table E1 in the Online Data Supplement for a brief description of main differences between NAVA and PSV). Compared to PSV, NAVA can prevent lung overdistension [6], usually reduces patient–ventilator asynchrony [7] improves gas exchange [8] and restores breathing variability [9]. However, although they pledge various substantial benefits of NAVA, all available NAVA studies were single-centre short-term studies conducted on small populations and limited to physiological outcomes. Large multicentre controlled studies with relevant clinical outcome evaluating the potential translation of these results into daily clinical practice are currently lacking. In addition, although NAVA has been successfully applied in critically ill patients for a short period of time [6–9], it is not known whether this mode could be used efficiently and safely as the main mode of partial ventilatory support.

The flexibility of NAVA and the link that it establishes between neural breathing control and ventilatory assistance could provide an advantage in terms of the patient’s ability to tolerate mechanical ventilation during the early phase of partial ventilator support. The present study was therefore designed to compare NAVA and PSV during the first 48 h following the first transition from controlled mechanical ventilation to partial ventilator support. We chose this outcome considering that weaning should be a straightforward process from controlled mechanical ventilation to partial ventilator support and eventually extubation and that any turn back would be a failure. The primary outcome was the likelihood for a patient to consistently remain in a partial mode during the first 48 h without having to return to a controlled mode. Secondary outcomes were the prevalence of patient–ventilator asynchrony and major outcomes such as ventilator-free days, length of stay and 28-day mortality.

Patients and methods

Patients

Patients receiving endotracheal mechanical ventilation for more than 24 h for acute respiratory failure of respiratory cause (de novo hypoxaemic respiratory failure, acute cardiogenic pulmonary oedema or acute-on-chronic respiratory failure) were eligible when they met the following criteria: ability to sustain PSV for at least 30 min with a total level of inspiratory pressure below 30 cmH2O, estimated remaining duration of mechanical ventilation greater than 48 h, level of sedation at most 4 on the Ramsay scale in the absence of medical decision to increase the level of sedation, fraction of inspired oxygen (FiO2) at most 50 % with positive end-expiratory pressure (PEEP) no greater than 8 cmH2O and absence of administration of high-dose vasopressor therapy defined as norepinephrine above 0.3 µg kg−1 min−1 or dopamine above 10 µg kg−1 min−1. Exclusion criteria were age younger than 18 years, known pregnancy, participation in another trial within the 30 days preceding completion of the eligibility criteria, contraindication to placement of the NAVA oesophageal tube (i.e. any contraindication to placement of a gastric tube or repositioning of a tube already in place, recent gastrointestinal suture, rupture of oesophageal varices with gastrointestinal bleeding during the 4 days prior to inclusion) and decision to withhold life-sustaining treatment.

Design

Patients were enrolled from April 2012 to June 2013 at 11 intensive care units in France, comprising six medical ICU, four medical surgical ICUs and one surgical ICU. The study protocol was approved for all centres by the Comité de Protection des Personnes Ile de France 6 (independent ethics committee), according to French law. Written informed consent was obtained from the patients or their surrogates before study inclusion. The trial was monitored by an independent data and safety monitoring board. Designated investigators at each centre enrolled patients. Blinding of treating clinicians to group assignment was not feasible, but all analyses were conducted blindly.

As soon as patients were included, the standard nasogastric feeding tube was removed and replaced by an EAdi catheter. Patients were then randomly assigned to blocks stratified by centre to receive either PSV or NAVA. Allocation concealment was achieved using sequentially numbered, opaque, sealed envelopes.

Ventilation strategies

The pressure support level in the PSV group and the NAVA level in the NAVA group were set to obtain a tidal volume of 6–8 ml/kg of ideal body weight. In both groups, the physician in charge set partial pressure of arterial oxygen (FiO2) and positive end-expiratory pressure (PEEP) according to local guidelines. NAVA or PSV were continued for 14 days unless the patients met criteria for switching to controlled mechanical ventilation or for weaning and subsequent extubation. After 14 days, patients of the NAVA group were switched to PSV.

Criteria for switching to assist-control mechanical ventilation in the two groups were (1) respiratory distress, hypoxaemia or hypercapnic acidosis despite optimization of ventilator settings; (2) severe hypotension, shock or arrhythmias; (3) increased need for sedation for agitation or for patient–ventilatory asynchrony; (4) an investigation requiring increased sedation; (5) procedures that necessitated a level of sedation requiring controlled mechanical ventilation (gastrointestinal endoscopy, transoesophageal echocardiography, surgery). As soon as the criteria for switching to assist-control mechanical ventilation and the inclusion criteria were restored, partial ventilatory support with either NAVA or PSV was re-instituted according to the randomization group.

Weaning protocol

Weaning from the ventilator was performed according to the same protocol in both groups. Although weaning could be started before the second day following randomization if deemed appropriate by the attending physician, the protocol required daily screening of prerequisite criteria for weaning at least from the second day following randomization: improvement of the acute respiratory failure causative factor, SpO2 greater than 92 % with FiO2 less than 50 % and PEEP less than 5 cmH2O, no infusion of vasopressor agents or sedatives and adequate responses to simple commands. If prerequisite criteria for weaning were present, a spontaneous breathing trial was performed. This test consisted of breathing spontaneously for 30 min to 1 h disconnected from the ventilator on a T piece, or in pressure-support ventilation with an inspiratory pressure of 7 cmH2O and zero end-expiratory pressure. The test was interrupted if any of the following signs of poor tolerance were observed: respiratory rate greater than 35/min, SpO2 less than 90 %, heart rate below 140/min, arterial systolic blood pressure above 180 or below 90 mmHg. If no signs of poor tolerance were observed, the patient was extubated. Post-extubation NIV could be instituted according to the decision of the physician in charge.

Data collection and outcome measures

The primary outcome was the proportion of patients who remained continuously in partial ventilatory support (either NAVA or PSV) during the first 48 h following inclusion or until extubation without any return to assist-control ventilation, which was defined as “success” (of partial ventilator support). In contrast, “failure” was defined as at least one episode of return to assist-control mechanical ventilation. To accurately assess the primary outcome, we precisely determined the time spent in each ventilator mode each day: data were collected each day from the ventilator with a ventilation record card, and were then further processed and stored.

Secondary outcomes were the prevalence of patient–ventilator asynchrony, the incidence of ventilator-acquired pneumonia according to predefined criteria [10], ventilator-free days, duration of mechanical ventilation from either intubation or inclusion to successful extubation, defined as extubation not followed by another intubation within 48 h, ICU and hospital length of stay, ICU and 28-day mortality. Application and duration of post-extubation noninvasive mechanical ventilation was also collected. There were no predefined criteria for the institution of post-extubation noninvasive mechanical ventilation, which was based on the decision of the physician in charge of the patient.

Data were collected at the time of randomization to characterize the severity of the underlying medical conditions, the severity of the acute illness, ventilatory settings, arterial blood gases and the cause of acute respiratory failure. Patients were monitored daily for dyspnoea as quantified by a visual analogue scale, comfort as evaluated by the ATICE score [11], ventilatory conditions and arterial blood gases.

In order to quantify patient–ventilator asynchrony, airway pressure, flow and EAdi were acquired at 100 Hz for 20 min from the ventilator via an RS232 interface connected to a computer using commercially available software (Servo-i® RCR, version 3.6.2, Maquet Critical Care) 12, 24, 36 and 48 h after inclusion. The five main types of asynchronies were quantified off-line by the two same investigators (A.D. and C.R.D.) according to previously published definitions [7] and are described in Table E2 (see the Online Data Supplement): (1) ineffective efforts, (2) auto-triggering, (3) premature cycling (4) late cycling and (5) double triggering. The number of episodes of each type of asynchrony was reported as the total number of each event per minute. A global asynchrony index (AI) was computed as previously published [5], defined as the number of asynchrony events divided by the total respiratory rate computed as the sum of the number of ventilator cycles (triggered or not) and of wasted efforts: asynchrony index (expressed in percentage) = number of asynchrony events/total respiratory rate (ventilator cycles + wasted efforts) × 100.

Statistical analysis

Sample size calculation was based on a previous report [12] as follows: assuming 78 % of patients remaining in partial ventilator support in the PSV group during the first 48 h, 58 patients per group would provide 80 % power at a two-sided α level of 0.05 to detect a 17 % absolute increase in the probability of continuously remaining on partial ventilator support in the NAVA group without any return to assist-control ventilation. With an estimated 10 % failure of ventilator data collection, the final calculated sample size was 128 patients.

All analyses were conducted on an intention-to-treat basis at a 5 % level of statistical significance. Continuous data are reported as median (interquartile range [IQR]) and categorical data as number of events (percentages). Differences between groups were assessed with the t test or the Mann–Whitney test for continuous variables and the χ 2 test for categorical variables. Adjustments were performed with the Cochran–Mantel–Haenszel test. Probability of remaining continuously in an assisted mode, time to successful extubation defined as extubation not followed by reintubation within 48 h, and time to unassisted breathing (including noninvasive ventilation) curves were constructed (Kaplan–Meier method) and differences between groups were compared using the log rank test. Patients’ characteristics and baseline variables between failure and success patients were compared. All reported p values are two-sided. Statistical analysis was conducted using SAS version 9.2 statistical software (SAS Institute Inc., Cary NC, USA).

Results

Study population

Among 3896 patients receiving mechanical ventilation, 128 were enrolled (median [IQR] age, 66 [57–77] years) (Fig. 1, see also Table E3 for volume of patients and enrolment rate per centre). Baseline characteristics of the patients are indicated in Table 1 (see also Table E4 in the Online Data Supplement for an extensive description of cause of acute respiratory failure). The two study groups were balanced at baseline with regards to age, severity of illness, cause of acute respiratory failure, ventilation and oxygenation parameters. The proportion of men was higher in the NAVA group and patients of the PSV group had a higher degree of comorbidity according to the Charlson score.

Fig. 1
figure 1

Flow diagram of the trial. EAdi electromyographic activity of the diaphragm, NIV noninvasive ventilation

Full size image
Table 1 Baseline characteristics of the patients
Full size table

Only 6/62 patients (10 %) in the NAVA group and 5/66 patients (8 %) in the PSV group had successfully passed the weaning test before the second day following randomization (P = 0.71).

Primary outcome

For technical reasons, ventilator data collection failed in four patients of the NAVA group and six patients of the PSV group (Fig. 1). Therefore, data for assessment of the primary outcome were available for 118 patients, 58 in the NAVA group and 60 in the PSV group. The proportion of patients with successful partial ventilator support who were therefore not switched at least once to assist-control ventilation during the first 48 h following inclusion was 67.2 % (n = 39/58) in the NAVA group vs. 63.3 % (n = 38/60) in the PSV group (relative risk [RR] = 1.19, 95 % confidence interval [CI] = 0.56–2.54; P = 0.66, Table 2). After adjustments based on the gender and on the Charlson score that were unbalanced between the two groups, the difference between the two groups remained non-significant (P = 0.25 and P = 0.32, respectively). The probability of successful partial ventilator mode is shown in Fig. 2 and was similar in the two groups (Log-rank test, P = 0.56). The main reasons for switching to assist-control mechanical ventilation were similar in the two groups (see Table E5 in the Online Data Supplement). Table E6 (see Online Data Supplement) displays the patients’ characteristics and baseline variables that differed significantly between failure and success in univariate analysis. None of these factors independently predicted the failure of partial ventilator support (either NAVA or PSV).

Table 2 Primary and secondary endpoints
Full size table
Fig. 2
figure 2

Kaplan–Meier estimates of probability of continuously remaining in partial ventilatory support (either NAVA or PSV) without any return to assist-control ventilation during the first 48 h following randomization. NAVA neurally adjusted ventilatory assist, PSV pressure support ventilation

Full size image

The time spent in each mode of mechanical ventilation during the first 48 h following inclusion is shown in Table E7 (see Online Data Supplement). The time spent in a partial ventilator mode (NAVA or PSV) was 46.6 [43.0–48.0] h in the NAVA group, including 44.1 (33.0–47.8) h in the NAVA mode, vs. 47.1 [39.8–48.0] h in the PSV group (P = 0.55).

Respiratory variables and dyspnoea

Respiratory variables on days 1, 2 and 5 following randomization are reported in Table E8 (see Online Data Supplement). The initial level of assistance was a PS level of 12 (10–14) cmH2O in the PSV group and a gain of 2.0 (1.0–2.5) cmH2O µV−1 in the NAVA group. Tidal volume, respiratory rate, minute ventilation, PaO2/FiO2 and PaCO2 did not differ between the two groups.

In the NAVA group, fewer patients reported dyspnoea on day 1 (P = 0.03). ATICE comfort score did not differ between groups.

Patient–ventilator asynchrony

For technical reasons, flow, pressure and EAdi recording failed in nine patients of the NAVA group and 16 patients of the PSV group. Data concerning the prevalence of the main asynchronies were therefore available for 103 patients, 53 in the NAVA group and 50 in the PSV group. Figure 3 shows the prevalence of the main patient–ventilator asynchronies observed in the two groups. The asynchrony index was significantly lower in the NAVA group than in the PSV group (26.7 [15.8–45.1] % vs. 14.7 [12.3–21.7] %, P < 0.001). Ineffective efforts and late cycling were more frequent in the PSV group, whereas double triggering was more frequent in the NAVA group.

Fig. 3
figure 3

Rates and prevalence of main patient–ventilator asynchronies. Box plots showing the median and interquartile range for the prevalence of the main patient–ventilator asynchronies (left) and the asynchrony index (right ). *p < 0.05

Full size image

Other secondary outcomes

The spontaneous breathing trial was performed in 115 patients on pressure-support ventilation and disconnected from the ventilator on a T piece in 13 patients. Differences between groups were not significant for ICU mortality (RR for NAVA, 0.61 [95 % CI, 0.27–1.35]), hospital mortality (RR, 0.68 [95 % CI, 0.32–1.47]), 28-day mortality (RR 0.64 [95 % CI, 0.30–1.35]), ICU length of stay and hospital length of stay, and the relative risk of acquiring ventilator-acquired pneumonia was 2.66 [95 % CI, 0.54–13.22]) (Table 2). The number of ventilator-free days was lower in the NAVA group on day 7, but not on day 14 or day 28. The probability of successful extubation and the probability of breathing without assistance are shown in Fig. 4 and were similar in the two groups (Log-rank test, p = 0.49). In the NAVA group compared to the PSV group, fewer patients received post-extubation noninvasive mechanical ventilation (RR, 0.65 [95 % CI, 0.47–0.91]) and for a shorter duration (Table 2).

Fig. 4
figure 4

Kaplan–Meier estimates of probability of successful extubation (left) and probability of breathing without assistance (right) from the day of randomization to day 28. Successful extubation was defined as extubation not followed by reintubation within 48 h. NAVA neurally adjusted ventilatory assist, PSV pressure support ventilation

Full size image

Adverse events

Four adverse events possibly related to the NAVA catheter were recorded: one in the NAVA group (one pneumothorax) and three in the PSV group (one pneumothorax, one gastrointestinal haemorrhage and one laryngeal oedema). The pneumothorax occurring in the PSV group was directly related to a misplacement of the NAVA tube. No adverse events related to the NAVA mode were recorded.

Discussion

The main findings of our study can be summarized as follows: (1) NAVA did not increase the probability of remaining in a partial ventilator mode compared to PSV; (2) NAVA improved patient–ventilator synchrony and was associated with less dyspnoea and less frequent application for post-extubation NIV; (3) NAVA did not reduce the duration of mechanical ventilation or ICU mortality; (4) NAVA can be delivered safely and efficiently for several days.

The NAVA mode, first described more than 15 years ago [13], has been available for clinical use for almost 7 years. To our knowledge, the present study is the first to describe the use of NAVA in a large cohort of patients recovering from acute respiratory failure. Although many studies have previously investigated NAVA in ICU patients, they were conducted in small samples of patients, who were rarely studied for more than a few hours [14]. In the present study, NAVA was extensively described during the first 48 h, and was continued for 14 days unless the patients met criteria for switching to controlled mechanical ventilation or for weaning and subsequent extubation. One serious adverse event was observed, a pneumothorax caused by the misplacement of the NAVA catheter in the PSV group. Our data show that NAVA can be applied safely and efficiently in routine care instead of PSV. It is noteworthy that patients in the NAVA group mostly remained on NAVA with few switches to either PSV or assist-control ventilation. However, NAVA did not improve the probability of remaining in an assisted mode of breathing compared to PSV and the reasons for switching back to a controlled mode were similar in the two groups. In addition, the total time spent in a controlled mode was similar in the two groups. Why NAVA failed to improve the probability and the time spent in an assisted mode remains unclear. A number of hypotheses can be proposed. First, various precipitating factors of acute respiratory failure were observed. It is possible that NAVA is mostly beneficial in specific patient groups, such as COPD, patients with severe patient–ventilator asynchrony or patients who are difficult to wean. Second, all participating centres were experienced with mechanical ventilation. Therefore, the proportion of patients in the PSV arm who did not remain constantly in PSV could have been sufficiently low to prevent any additional improvement. After removal of failures of partial ventilator mode (either NAVA or PSV) due to sedation for the purposes of an investigation, comparison with previous reports from experienced teams showed a similar partial ventilator mode failure rate [12, 15]. Finally, NAVA failed to reduce the duration of mechanical ventilation and ICU mortality, but our study was not powered to detect such differences.

NAVA improved patient–ventilator synchrony over the first 48 h. Patient–ventilator asynchrony is potentially an important outcome, as a high prevalence of asynchrony is associated with longer duration of mechanical ventilation and a higher mortality rate [5, 16, 17]. The benefit of NAVA on patient–ventilator interactions has been previously described in small single-centre physiology studies [7, 18]. Our results tend to extend these findings by the multicentre design of this study. We also found a lower prevalence of dyspnoea in the NAVA group on the first day following randomization, but not thereafter, which could be related to improved matching between respiratory neural drive and assistance during a transition period of maximal lability. Finally, we found that fewer patients in the NAVA group received post-extubation noninvasive mechanical ventilation and for a shorter duration. Consequently, the number of ventilator-free days on day 7 was lower in the NAVA group. The relevance of this noticeable finding needs to be confirmed by further studies since there were no predefined criteria for the institution of post-extubation noninvasive mechanical ventilation, which was based on the decision of the physician in charge of the patient.

The strengths of this trial include a rigorous method to avoid bias such as the recording of the ventilator trends with a card that allows precise calculation of the time spent on each mode and predefined criteria for initiation of a spontaneous breathing trial and for extubation. The participation of 11 ICUs enhances the generalisability of our findings. Finally, this is the first study to demonstrate the feasibility and safety of NAVA for several days rather than for just a few hours.

Limitations of the trial include the recommended NAVA level setting that targeted a range of tidal volume. The relevance of this setting has been questioned and alternative rules have been proposed [14, 19]. However, these rules are still quite complex to use and time-consuming to apply. Targeting a tidal volume remains an easy way to set NAVA and is easily compatible with routine clinical practice. In addition, one could argue that our study was underpowered. However, the difference between the two groups was small and a very high sample size would be required to detect such a difference. We feel that our primary outcome may not be the best one to describe a difference between the two groups. The fact that ventilator asynchrony was calculated on the basis of four recordings performed on the 2 days following inclusion is another limitation. Indeed, a recent study shows that asynchrony may occur at anytime and that should be quantified over the whole duration of mechanical ventilation [16]. For technical reasons (failure of recordings), patient–ventilator asynchrony could not be detected in 25 patients. These missing data could alter the veracity of our results. To solve this issue, we performed a simulation. Markov chain Monte Carlo was first preformed to produce monotone missingness followed by multiple imputations with the predictive mean matching method (for asynchrony index, P = 0.062). Finally, although postoperative patients were included, they all had a complication of surgery involving the respiratory system or prolonged mechanical ventilation.

In summary, in patients recovering from acute respiratory failure, we found that NAVA was safe and feasible over a prolonged period of time, but we failed to find a difference in the efficacy of NAVA and PSV to maintain patients in a partial ventilatory mode. We confirmed that NAVA reduces the prevalence of patient–ventilator asynchrony, and observed that it does so in a sustained manner. Our findings of improved dyspnoea on day 1 under NAVA and decreased use of post-extubation noninvasive mechanical ventilation after NAVA provide guidance for the design of future trials.