0
Original Research: Critical Care |

Patient-Controlled Positive End-Expiratory Pressure With Neuromuscular DiseaseSpeech and Mechanical Ventilation: Effect on Speech in Patients With Tracheostomy and Mechanical Ventilation Support FREE TO VIEW

Marine Garguilo, SP; Karl Leroux; Michèle Lejaille, MS; Sophie Pascal, SP; David Orlikowski, MD, PhD; Frédéric Lofaso, MD, PhD; Hélène Prigent, MD, PhD
Author and Funding Information

From the EA4497 of the University of Versailles Saint-Quentin-en-Yvelines (Mss Garguilo and Lejaille and Drs Orlikowski, Lofaso, and Prigent), Versailles; Association d’Entraide des Polios et Handicapés (ADEP Assistance) (Mr Leroux), Suresnes; Service de Médecine Physique et Réadaptation (Ms Pascal), Centre d’Investigation Clinique Innovations Technologiques (Ms Lejaille and Dr Orlikowski), Home Ventilation Unit (Intensive Care Department) (Dr Orlikowski), and Physiology Department (Drs Lofaso and Prigent), Hôpital Raymond Poincaré, Assistance Publique-Hôpitaux de Paris, Garches; and INSERM U492 (Dr Lofaso), Créteil, France.

Correspondence to: Hélène Prigent, MD, PhD, Service de Physiologie-Explorations Fonctionnelles, Hôpital Raymond Poincaré, 92380 Garches, France; e-mail: helene.prigent@rpc.aphp.fr


Funding/Support: Ms Garguilo received a PhD research grant from the Fondation Garches and the AXA Research Fund. The study was supported by the Association d’Entraide des Polios et Handicapés (ADEP Assistance).

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2013;143(5):1243-1251. doi:10.1378/chest.12-0574
Text Size: A A A
Published online

Objective:  Communication is a major issue for patients with tracheostomy who are supported by mechanical ventilation. The use of positive end-expiratory pressure (PEEP) may restore speech during expiration; however, the optimal PEEP level for speech may vary individually. We aimed to improve speech quality with an individually adjusted PEEP level delivered under the patient’s control to ensure optimal respiratory comfort.

Methods:  Optimal PEEP level (PEEPeff), defined as the PEEP level that allows complete expiration through the upper airways, was determined for 12 patients with neuromuscular disease who are supported by mechanical ventilation. Speech and respiratory parameters were studied without PEEP, with PEEPeff, and for an intermediate PEEP level. Flow and airway pressure were measured. Microphone speech recordings were subjected to both quantitative and qualitative assessments of speech, including an intelligibility score, a perceptual score, and an evaluation of prosody determined by two speech therapists blinded to PEEP condition.

Results:  Text reading time, phonation flow, use of the respiratory cycle for phonation, and speech comfort significantly improved with increasing PEEP, whereas qualitative parameters remained unchanged. This resulted mostly from the increase of the expiratory volume through the upper airways available for speech for all patients combined, with a rise in respiratory rate for nine patients. Respiratory comfort remained stable despite high levels of PEEPeff (median, 10.0 cm H2O; interquartile range, 9.5-12.0 cm H2O).

Conclusions:  Patient-controlled PEEP allowed for the use of high levels of PEEP with good respiratory tolerance and significant improvement in speech (enabling phonation during the entire respiratory cycle in most patients). The device studied could be implemented in home ventilators to improve speech and, therefore, autonomy of patients with tracheostomy.

Trial registry:  ClinicalTrials.gov; No.: NCT01479959; URL: clinicaltrials.gov

Figures in this Article

Communication is a major issue for patients with tracheostomy supported by mechanical ventilation (MV).13 Tracheostomy ventilation induces important modifications of speech physiology. When leak ventilation can be achieved with cuffless or deflated tracheostomy tubes,4 patients are able to speak. However, speech becomes essentially inspiratory when leaks around the tracheostomy tube during ventilator pressurization lead to a rise in tracheal pressure, allowing phonation. The ventilator rhythms phonation, which quickly stops after the beginning of expiration because of rapid tracheal pressure drop.57 Several interventions have been proposed to improve speech during MV. Adjustments of ventilation parameters, such as an increase in inspiratory time (Ti) and use of positive end-expiratory pressure (PEEP), may lengthen phonation time on inspiration and expiration.5,810

Most home ventilators maintain PEEP by using a balloon or a diaphragm that behaves as a threshold pressure valve during expiration. The pressure maintained inside the expiratory valve occludes the expiratory circuit until the pressure inside reaches this targeted pressure.11,12 If airway pressure remains beneath the set PEEP level throughout expiration, patients expire completely through the upper airways, mimicking the effect of a one-way speaking valve. Therefore, expiratory volume can be dedicated to speech. Ideally, PEEP adjustment should be just above the pressure required to completely expire through the upper airways during quiet respiration, which could be considered as the optimal PEEP level (PEEPeff) to improve speech. However, it does not take into consideration the potential deleterious effects of permanent high PEEP level, such as increased end-expiratory volume and hyperinflation, barotrauma, interaction with cardiac output, and humidification impairment of the airways due to permanent expiration through the upper airways. Therefore, PEEP adjustment is a compromise between speech efficiency and tolerance and usually set to < 5 cm H2O.9

To address this issue, we developed a prototype ventilator with a separate control of the expiratory valve to apply the pressure-defining PEEP level only when activated by a key-pinch switch under the patient’s control so that PEEP is only used when needed (ie, for speech). We evaluated speech and respiratory comfort at PEEPeff and at a PEEP level representing 50% of PEEPeff (PEEP50).

From June 2010 to July 2011, adult patients with neuromuscular disease (NMD) admitted for their usual follow-up of home ventilation were included if they were receiving long-term MV with a cuffless or deflated cuff tracheostomy tube. Exclusion criteria were inability to read and unstable respiratory state. The study protocol was approved by our institutional review board (CPP Île de France XII no. 2009-A0132364, Agence Francaise de Securite Sanitaire des Produits de Sante No. B91637-80), and written informed consent was obtained from all patients.

Prototype Ventilator

A volume-controlled ventilator (Eole3; ResMed) was associated with a constant-pressure generator (Puritan Bennett GoodKnight 420 Evolution; Covidien) connected to the balloon of the expiratory valve using a Y circuit and a solenoid valve under the control of a key-pinch switch (Micro Leger 7C02; AbleNet, Inc) (Fig 1). The pressure generator only applied pressure to the balloon of the expiratory valve when the patient willingly activated and maintained pressure on the switch.

Figure Jump LinkFigure 1. Experimental device. The balloon of the external expiratory valve of the ventilator circuit is connected to a pressure generator that applies a constant pressure when activated and, therefore, generates a resistive positive end-expiratory pressure (PEEP) during expiration. The pressure generator can only apply pressure (and, therefore, PEEP) in the expiratory valve balloon when the patient presses a switch, which activates a solenoid valve that opens the circuit. When the patient activates the switch, the pressure generator immediately applies the prescribed pressure to the balloon of the expiratory valve. Expiration can occur through the expiratory valve only if the pressure in the circuit rises above the pressure inside the valve balloon (PEEP level); therefore, expiratory flow is directed toward the upper airways, allowing the patient to speak. To determine optimal PEEP level, the pressure delivered by the pressure generator is increased progressively by 1-cm H2O steps and tested at each level by pressing the switch until no expiratory volume is observed through the ventilator or until the pressure rise is limited by the patient’s respiratory comfort. Once obtained, the optimal PEEP level is set on the pressure generator, and the patient can obtain it instantaneously by pressing the switch.Grahic Jump Location

In this system, the pressure inside the balloon depends on the ventilator regulation when the solenoid valve is closed. Specifically, during inspiration, this pressure is equal to the inspiratory pressure (Pi) delivered just at the exit of the ventilator (before the inspiratory circuit) to avoid expiratory valve leakage during insufflation. When the solenoid valve is open, the pressure inside the balloon is equal to the highest pressure between the pressure applied by the pressure generator and the pressure at the exit of the ventilator. Accordingly, during inspiration, the highest pressure is the pressure generated by the ventilator, whereas during expiration, the highest pressure could be the pressure delivered by the pressure generator to voluntarily increase the PEEP level above the baseline PEEP level determined by the ventilator.

With this pressure applied to the expiratory valve balloon, the expiratory valve works like a pressure threshold valve, which allows an expiratory flow through the tracheostomy tube only when the positive pressure inside the end of the ventilator circuit is at least equal to the pressure inside the balloon. The volumetric ventilator used did not have a flow-by system to maintain PEEP. In this system, PEEP is defined as the pressure threshold of the expiratory valve, which corresponds to the pressure applied to the expiratory valve balloon. When the patient is ventilated without leaks, the expiratory pressure inside the tube is equal to the prescribed PEEP level, whereas when the patient is ventilated with an uncuffed tracheostomy tube, this system is unable to maintain the prescribed PEEP level because the pressure decreases accordingly to the degree of air leaks. A picture of the prototype is provided in e-Figure 1.

Experimental Setup

Ventilator-delivered volume (VI) and expiratory volume through the tracheostomy tube (VTe) were obtained from the integration of the inspiratory and expiratory flow signal measured using a pneumotachograph (Fleisch #2; Electromedsis); tracheal pressure was measured at the tracheostomy tube proximal end using a differential pressure transducer (Validyne MP 45 ± 100 cm H2O; Validyne Engineering). Upper airway resistance (Raw) was measured based on the expiratory flow obtained with a second pneumotachograph placed on a face mask while occluding the expiratory circuit and measuring pressure at the proximal end of the tracheostomy tube.

Acoustic speech signals recorded from a microphone (SBC MD150; Philips) positioned 20 cm from the patient’s mouth were routed to a microcomputer with an analog-to-digital converter (MP150 and AcqKnowledge; BIOPAC Systems, Inc) that synchronized respiratory and acoustic data. Respiratory signals were digitized at 128 Hz and speech signals at 20,000 Hz. Speech was also recorded on a digital recorder (Olympus DS-55; Olympus Corporation) with a signal bandwidth between 50 and 19,000 Hz for subsequent analysis by speech therapists. Oxygen saturation was monitored using pulse oximetry (Ohmeda Biox; BOC Healthcare).

Experimental Protocol

Patients were ventilated in their usual MV mode. VI, Ti, and backup rate remained unchanged. PEEP level was increased by 1-cm H2O steps (starting at 4 cm H2O, with a maximum of 15 cm H2O) until PEEPeff was reached (defined as the absence of VTe during quiet ventilation with no speech) or until patients expressed respiratory discomfort (in which case the last level of PEEP reached was considered PEEPeff, even if there remained a VTe). Once PEEP level (PEEPeff and PEEP50) was adjusted, it remained constant throughout the speech protocol. If autotriggering occurred during PEEP rise, trigger sensitivity was adjusted.

Patients were initially studied without PEEP (PEEP0); PEEP conditions (PEEPeff and PEEP50) were then randomly tested. Patients uttered the A sound as long as possible, read a list of words and a standard text passage, and pronounced the same sentence with different utterance forms (ie, question, command).

Data Analysis

Respiratory rate (RR), Ti, VI, and VTe were measured on the computerized flow signal. Raw was calculated with the mouth expiratory flow and tracheal pressure signals at 0.2 L/s.

Speech evaluation included mean duration of speech during the respiratory cycle, time required to read the text, phonation flow (number of syllables/s), and qualitative analyses by two speech therapists with no prior experience with ventilated patients and blinded to speech condition. The speech therapists assessed intelligibility with a French adaptation of the Frenchay Dysarthria Assessment,13 assessed prosody with a score based on the correctly identified expressions, and determined a perceptual score.14 In each condition, respiratory and speech comfort were evaluated with a visual analog scale.6,15,16

Sample Size Calculation

A previous study10 allowed us to estimate the SD of 60 s for the text reading duration and a correlation of 0.8 between two speech improvement techniques. To have 80% power to detect a difference in means across three levels of repeated measures by an effect size of 0.4 and two-sided risk of 0.05, we calculated that a sample size of at least 10 patients was required. Because the three conditions impose six order possibilities, the statistician chose the lowest multiple of 6 above 10. Accordingly, to avoid an order effect, 12 patients were included and randomly subdivided into two subgroups of six patients, and for each subgroup, the six possibilities were designed.

Bench Study Evaluation

To evaluate whether our system was adaptable to other home ventilators, we conducted a bench study of five life-support home ventilators. Details are available in e-Tables 1 and 2.

Statistical Analysis

Unless otherwise specified, results are expressed as median and interquartile range (25%-75%). Because sample sizes were too small for assessment of normality, differences among the three conditions were assessed using the Friedman repeated-measures nonparametric test. If a significant difference appeared, conditions were compared two by two using Wilcoxon tests. Analyses were performed with StatView (SAS Institute, Inc). P < .05 was considered significant.

Population

Twelve patients with NMD and tracheostomy supported by MV were included in the study period. The patients’ characteristics as well as their usual MV parameters are presented in Table 1. All patients had severe respiratory failure (vital capacity, 620 [587-657] mL) and were invasively ventilated for 13.0 (8.7-16.0) years.

Table Graphic Jump Location
Table 1 —Demographic and Respiratory Characteristics of the Patient Population

BR = ventilator backup rate; CM = congenital myopathy; DMD = Duchenne muscular dystrophy; F = female; FKRP = limb girdle muscular dystrophy type 2I with a fukutin related protein deficiency; M = male; NA = not available; PEEPeff = optimal positive end-expiratory pressure level; Pemax = maximal expiratory pressure; Pimax = maximal inspiratory pressure; Pt = patient; Raw = airway resistance; SMA = spinal muscular atrophy; VC = vital capacity; VI = ventilator-delivered volume.

a 

Shiley unfenestrated/uncuffed tracheostomy tube (Covidien).

b 

Puritan Bennett Legendair ventilator (Covidien).

c 

Trachéoflex unfenestrated/uncuffed tracheostomy tube (Teleflex Medical).

d 

Tracoé unfenestrated/uncuffed tracheostomy tube (Pouret Médical).

e 

Rüsch unfenestrated/uncuffed tracheostomy tube (Teleflex Medical).

PEEPeff was 10.0 (9.5-12.0) cm H2O. Only three patients (1, 10, and 11) were limited in PEEP rise by respiratory tolerance; all remaining patients reached PEEPeff (ie, no expiratory flow through the tracheostomy tube) and tolerated its use continuously for several minutes. Expiratory upper Raw was evaluated in nine patients and varied widely, but no relation was found between upper Raw and PEEPeff.

Respiratory Parameter

At rest, VI was similar in the three conditions (P = .37). During PEEP level adjustments, seven patients required a rise in trigger level to avoid autotriggering. Despite trigger adjustments, the RR at rest showed a slight, but significant rise (P = .0045) in PEEP0 (17.5 [15.0-20.0] cycles/min) compared with PEEP50 (17.5 [16.8-20.3] cycles/min) and PEEPeff (19.0 [17.8-21.0] cycles/min).

Respiratory parameters during speech are presented in Table 2. RR increased significantly with PEEP use (P = .04). Consequently, a significant increase in minute ventilation insufflation (MVI), resulting from the product of VTI and RR, was also observed with PEEP50 (P = .04) and PEEPeff (P = .03) compared with PEEP0. Although only two patients (8, 12) raised RR during speech with PEEP0, eight increased RR values by 3 cycles/min above the backup rate with PEEP50, and nine did so with PEEPeff.

Table Graphic Jump Location
Table 2 —Respiratory Characteristics During Speech (n = 12)

Data are presented as median (interquartile range). Pe = expiratory pressure; Pi = inspiratory pressure; MVI = minute ventilation insufflation; PEEP0 = without positive end-expiratory pressure; PEEP50 = positive end-expiratory pressure level representing 50% of PEEPeff; RR = respiratory rate; Spo2 = oxygen saturation as measured by pulse oximetry; Te = expiratory time; Ti = inspiratory time; TTOT = total respiratory time; VTe = expiratory volume through the tracheostomy tube. See Table 1 legend for expansion of other abbreviation.

a 

Friedman test.

b 

P < .04 vs PEEP0 by Wilcoxon test.

c 

P < .003 vs PEEP0 by Wilcoxon test.

d 

P < .007 vs PEEP0 by Wilcoxon test.

e 

P < .03 vs PEEP50 by Wilcoxon test.

f 

P < .002 vs PEEP50 by Wilcoxon test.

Although MVI increased, VTe reduced significantly as PEEP level rose, with a significant difference between PEEP50 and PEEPeff compared with PEEP0. During rest, all patients except for the three limited by respiratory comfort during PEEPeff (among whom two had a decreased VTe of < 60 mL) expired entirely through the upper airways (VTe < 20 mL). During speech, seven of nine patients still expired entirely through the upper airways with PEEPeff. Accordingly, mean expiratory pressure increased significantly with PEEP rise (Table 2). Although comparable in the three conditions at rest, mean IP rose significantly with PEEP during speech, whereas peak IP and VI remained stable.

Speech

Speech parameters improved significantly with PEEP rise (Table 3). The sustained A sound duration throughout the respiratory cycle increased with PEEP level. Although none of the patients was able to sustain the vowel for > 55% of the respiratory cycle with PEEP0, three could sustain the sound for > 90% of the respiratory cycle with PEEP50. With PEEPeff, nine patients used 100% of the respiratory cycle, and one patient could sustain the sound for 78% of the cycle and two for > 90% of the cycle. Two of the patients limited in PEEPeff setting by respiratory tolerance were nonetheless able to use 100% of their respiratory cycle with their limited PEEPeff.

Table Graphic Jump Location
Table 3 —Phonation Characteristics (n = 12)

Data are presented as median (interquartile range). Perceptual score, 0-128 scale; best score, 0.14 Intelligibility score, 0-8 scale; best score, 8.13 Prosodia score, 0-4; best score, 4. See Table 1 and 2 legends for expansion of abbreviations.

a 

Friedman test.

b 

P < .006 vs PEEP0 by Wilcoxon test.

c 

P < .003 vs PEEP0 by Wilcoxon test.

d 

P < .006 vs PEEP50 by Wilcoxon test.

PEEP use improved the text reading task for the reading time, with a significant difference between PEEP0 and PEEP50 (P = .006), between PEEP50 and PEEPeff (P = .006), and for the phonation flow between PEEP0 and PEEP50 (P = .005) and between PEEP50 and PEEPeff (P = .005). Intelligibility and prosodia did not change significantly among the different conditions, but there was a trend of improvement of the perceptual score with PEEP use. Audio samples of patient 3 reading the two first sentences of the text with PEEP0 and PEEPeff (set at 10 cm H2O) are provided as Audio 1 and 2, respectively.

Tolerance

All patients tolerated PEEP throughout the trial. Respiratory comfort evaluated by patients with the visual analog scale remained stable regardless of PEEP level, whereas speech comfort increased significantly with each PEEP level (Fig 2).

Figure Jump LinkFigure 2. Evolution of speech and respiratory comfort in 12 patients according to PEP level as evaluated using a VAS (0 = extremely uncomfortable, 10 = extremely comfortable). Data are presented as median and interquartile range. *P = .01, **P = .04, and ***P = .003 by Wilcoxon test. NS = nonsignificant; PEP0 = without positive end-expiratory pressure; PEP50 = positive end-expiratory pressure level representing 50% of PEPeff; PEPeff = optimal positive end-expiratory pressure level; VAS = visual analog scale.Grahic Jump Location

Nine patients used the switch fewer than three times throughout the entire text reading, holding it almost continuously, whereas the three remaining patients used the switch > 10 times during text reading (patient 5, 28 times; patient 7, 11 times; patient 8, 38 times). These three patients had no respiratory limitation during the PEEPeff setting at rest and showed stable respiratory comfort throughout the different conditions, with speech comfort improvement during PEEP use.

The use of a patient-controlled PEEP allowed for the use of higher PEEP levels with a significant improvement of speech efficiency and preserved respiratory comfort in patients with NMD and tracheostomy supported by MV. Speech quality was significantly improved by a PEEP level individually set for optimal speech.

Optimal PEEP

We sought to determine individually optimal PEEP levels that allow complete expiration through the upper airways and, therefore, speech during expiration. We were limited by respiratory tolerance for only three patients, whose speech nevertheless benefited from the PEEP rise that could be achieved. For the other patients, we reproduced successfully the effect of a one-way speaking valve. Placement of a speaking valve on the ventilator circuit allows the patient to speak as it closes during expiration, but it imposes a complete expiration through the upper airways.10,17,18 In a previous comparison of PEEP use and speaking valves, there was no significant difference in terms of efficiency of speech, but the improvement observed resulted from different adaptations according to the technique used.10 Although PEEP is unsuccessful with some patients, it is usually used at low levels (≤ 5 cm H2O) for safety’s sake because patients frequently experience cardiac failure related to neuromuscular disease. Hoit et al9 evaluated higher levels of PEEP (15 cm H2O) in three patients used to speaking valves and observed comparable improvement of speech. Given the proper level, PEEP achieves the effect of a speaking valve but is potentially more secure because the ventilator circuit remains open, allowing expiration as soon as airway pressure rises above PEEP level.

PEEPeff is likely to vary individually according to patient anatomy and the relationship between tracheostomy and trachea diameters. However, in the present study, upper Raw was not related to PEEPeff; therefore, PEEP adjustment could be carried out clinically, seeking the level at which expiration occurs exclusively through the upper airways while considering respiratory tolerance.

PEEP level was determined during rest, and it is questionable whether it remains adequate during speech. Indeed, phonation increases leaks in the patient-ventilator system and modifies resistance and impedance of the system. Accordingly, among the nine patients able to expire exclusively through the upper airways during rest, only seven could still do so with the same PEEP level during speech. Indeed, during speech, glottis closure occurs intermittently, and, therefore, the subglottic pressure can increase above the maximal subglottic pressure obtained during quiet breathing without speech, thus attaining the adjusted PEEPeff and inducing a partial expiration through the expiratory valve. Because the experimental design did not allow us to modify PEEPeff once it was obtained, we did not test an increase of PEEP level, but it is clear to us that it is possible to try to further increase the PEEP level to obtain the absence of VTe during speech or until the patient expresses respiratory discomfort. Furthermore, special attention should be given to the type and size of tracheostomy tube the patient uses because sometimes small modifications may result in improvements of speech quality (ie, switching from a deflated tracheostomy tube to a cuffless tube).

Speech Improvement

Several mechanisms may account for speech improvement. First, RR rose with PEEP use. Some patients use triggering to generate more-frequent Tis, which comprise the main period for speech when no intervention is carried out.5,8 Because patients often are ventilated with volume-controlled ventilation and cannot vary the inspiratory period architecture (as Ti and VI are preset), to increase their phonation time, they only can trigger the ventilator if respiratory muscle strength is sufficiently preserved. Thus, RR rise during speech may be a physiologically adaptive behavior in this population.5,7,19 Only two patients increased RR during speech with PEEP0, suggesting that the other patients were unable to trigger their ventilator. With PEEP rise, nine patients had RR values of 3 cycles/min above the backup rate. PEEP level may influence the triggering system in the presence of leaks, reducing trigger delay and favoring autotriggering,20,21 which explains why several patients required trigger adjustments during PEEP setting. Although these modifications may have been adequate during rest, phonation involves increased air leaks and changes the impedance of the system, which can modify ventilator performance.22 This may generate new trigger impairment; however, in these circumstances, it might be favorable to help patients to increase RR and, therefore, to improve speech. Indeed, in the present study, increased RR and MVI did not impair respiratory comfort during speech. This finding supports that this mechanism may also participate in speech quality improvement. Additionally, rise in PEEP level resulted in a significant decrease in VTe. Therefore, airflow dedicated to speech increased with PEEP level, allowing nine patients to use the entire respiratory cycle for phonation at PEEPeff. This situation reproduced the effect of a speaking valve, and during speech, seven patients expired entirely through the upper airways with PEEPeff.

Moreover, median Pe significantly decreased with PEEP rise during speech, although it was comparable in the three conditions at rest. This suggests that because high PEEP level allows phonation throughout the respiratory cycle, patients do not have to wait for the beginning of insufflation to speak. Therefore, insufflated gas was fully used for phonation without any latency at the beginning of inspiration, increasing even more leaks during inspiration and, consequently, reducing median Pe. This resulted in an improvement in phonation flow and increased patients’ speech autonomy.

Because patients used the switch only when PEEP was needed, they tolerated the entire PEEP trial. Interestingly, both patients who triggered for speech reduced their RR as PEEP level rose, as if this new method of speech improvement allowed them to rely less on their usual respiratory adaptative mechanisms.

Although optimal PEEP level achieves an effect comparable to a speaking valve, it offers a safer option. Indeed, in situations of accidental tube cuff inflation or upper airway occlusion, a one-way speaking valve exposes patients to hyperinflation. Accordingly, the different trials evaluating one-way speaking valves in ventilator-dependent patients reported the inability for some patients to tolerate the use of this type of device.10,17,18 PEEP use decreases these risks because the ventilator expiratory circuit remains open and allows expiration as soon as airway pressure exceeds PEEP level. Using a patient-controlled PEEP further minimizes this risk because the patient may discontinue PEEP as soon as he or she feels discomfort.

Technological Issues

Issues regarding the PEEP production techniques, the choice of interfaces adapted to a population with major motor impairment,23,24 and the home ventilators bench study are addressed in e-Appendix 1 and e-Tables 1 and 2.

Limitations

One of the main limitations of this study was the short period of time in a hospital setting; therefore, the technique needs to be validated for long-term use for both its tolerance and its adaptability in an outpatient environment because it is intended for patients receiving long-term home ventilation. The safety of the device’s long-term use and capacity to fit in patients’ daily lives and activities (adaptation to a wheelchair, efficiency and stability of the switch while moving, etc) need to be evaluated in the patients’ environment.

The device in its current layout requires the simultaneous use of two devices, which were not initially designed to function together. Although in its current disposition the combination of the two devices did not alter the delivery of the ventilation by a bellow ventilator, we observed that this was not the case with some turbine home ventilators as exhibited in the bench study (e-Appendix 1). Therefore, the use of the pressure generator to regulate the delivered PEEP cannot be generalized to any home ventilator without a prior evaluation. This takes into consideration the medical and legal implications should one of the devices malfunctions because with this modified use, the warranties of the devices may be challenged. It is obvious that for both safety and warranty implications, it would be preferable to integrate the technique within a single device that provides ventilation and allows the option for trigger use. With interest from ventilator manufacturers, this seems feasible because some home ventilators already offer patients direct access to different preset programs of ventilation. Such devices could be adapted with a program that allows patients to customize the PEEP level by switching rapidly from one setting to another. However, as far as we know, the ventilation mode can only be accessed through the ventilator’s central monitor and, thus, is not directly accessible to patients with motor impairments. Implementing the option of rapid access (on/off) to secondary settings through a switch controlled by the patient would possibly broaden the use of this technique and improve patient speech while preserving a warranted use of the ventilator.

In conclusion, the use of a patient-controlled PEEP individually adjusted to achieve optimal speech showed a significant improvement of speech parameters and a good respiratory tolerance in a short-term evaluation. High PEEP level allowed most patients to speak throughout the respiratory cycle, giving them autonomy of speech with regard to ventilation and putting them in a situation where they exploit respiration even more than normal subjects. Optimization of the PEEP control interface is necessary to ensure access to the technique in different environments, especially in the outpatient care setting. Integration of this device into home ventilators could contribute to improved patient communication and, therefore, autonomy and quality of life.

Author contributions: Dr Prigent had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Ms Garguilo: contributed to the study design, data collection and analysis, speech analysis, writing of the manuscript, and critical revision of the manuscript.

Mr Leroux: contributed to the conception and construction of the ventilator prototype and critical revision of the manuscript.

Ms Lejaille: contributed to the data collection and critical revision of the manuscript.

Ms Pascal: contributed to the speech analysis and critical revision of the manuscript.

Dr Orlikowski: contributed to the data collection and critical revision of the manuscript.

Dr Lofaso: contributed to the study design, conception and construction of the ventilator prototype, data analysis, writing of the manuscript, and critical revision of the manuscript.

Dr Prigent: contributed to the study design, conception and construction of the ventilator prototype, data collection and analysis, writing of the manuscript, and critical revision of the manuscript.

Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Mr Leroux is employed by ADEP Assistance, which was initially part of a patients’ association (ADEP) but became a part of Air Liquide Santé by the end of the study. Drs Orlikowski, Lofaso, and Prigent are part of the EA 4497 Research Laboratory of the University of Versailles Saint-Quentin-en-Yvelines, which has received some equipment from ResMed for research projects independent from the present study. Mss Garguilo, Lejaille, and Pascal have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or in the preparation of the manuscript.

Other contributions: We thank Natacha Bracco, SP, for her help in the speech analysis. This work was performed at the Raymond Poincaré Hospital, Assistance Publique-Hôpitaux de Paris, Garches, France.

Additional information: The e-Appendix, e-Figure, e-Tables, and Audio can be found in the “Supplemental Materials” area of the online article.

MV

mechanical ventilation

MVI

minute ventilation insufflation

NMD

neuromuscular disease

PEEP

positive end-expiratory pressure

PEEP0

without positive end-expiratory pressure

PEEP50

positive end-expiratory pressure level representing 50% of optimal positive end-expiratory pressure

PEEPeff

optimal positive end-expiratory pressure level

Pi

inspiratory pressure

Raw

airway resistance

RR

respiratory rate

Ti

inspiratory time

VI

ventilator-delivered volume

VTe

expiratory volume through the tracheostomy tube

Laakso K, Markström A, Hartelius L. Communication and quality of life in individuals receiving home mechanical ventilation. Int J Ther Rehabil. 2009;16(12):648-655.
 
Raphaël JC, Dazord A, Jaillard P, et al. Assessment of quality of life for home ventilated patients with Duchenne muscular dystrophy [in French]. Rev Neurol (Paris). 2002;158(4):453-460. [PubMed]
 
Leder SB. Importance of verbal communication for the ventilator-dependent patient. Chest. 1990;98(4):792-793. [CrossRef] [PubMed]
 
Bach JR, Alba AS. Tracheostomy ventilation. A study of efficacy with deflated cuffs and cuffless tubes. Chest. 1990;97(3):679-683. [CrossRef] [PubMed]
 
Hoit JD, Shea SA, Banzett RB. Speech production during mechanical ventilation in tracheostomized individuals. J Speech Hear Res. 1994;37(1):53-63. [PubMed]
 
Prigent H, Samuel C, Louis B, et al. Comparative effects of two ventilatory modes on speech in tracheostomized patients with neuromuscular disease. Am J Respir Crit Care Med. 2003;167(2):114-119. [CrossRef] [PubMed]
 
Shea SA, Hoit JD, Banzett RB. Competition between gas exchange and speech production in ventilated subjects. Biol Psychol. 1998;49(1-2):9-27. [CrossRef] [PubMed]
 
Hoit JD, Banzett RB. Simple adjustments can improve ventilator-supported speech. Am J Speech Lang Pathol. 1997;6:87-96.
 
Hoit JD, Banzett RB, Lohmeier HL, Hixon TJ, Brown R. Clinical ventilator adjustments that improve speech. Chest. 2003;124(4):1512-1521. [CrossRef] [PubMed]
 
Prigent H, Garguilo M, Pascal S, et al. Speech effects of a speaking valve versus external PEEP in tracheostomized ventilator-dependent neuromuscular patients. Intensive Care Med. 2010;36(10):1681-1687. [CrossRef] [PubMed]
 
Lofaso F, Aslanian P, Richard JC, et al. Expiratory valves used for home devices: experimental and clinical comparison. Eur Respir J. 1998;11(6):1382-1388. [CrossRef] [PubMed]
 
Marini JJ, Culver BH, Kirk W. Flow resistance of exhalation valves and positive end-expiratory pressure devices used in mechanical ventilation. Am Rev Respir Dis. 1985;131(6):850-854. [PubMed]
 
Auzou P, Ozsancak C, Jan M, et al. Clinical assessment of dysarthria: presentation and validation of a method [in French]. Rev Neurol (Paris). 1998;154(6-7):523-530. [PubMed]
 
Ozsancak C, Parais AM, Auzou P. Perceptual analysis of dysarthria: presentation and validation of a clinical scale. Preliminary study [in French]. Rev Neurol (Paris). 2002;158(4):431-438. [PubMed]
 
Fauroux B, Pigeot J, Polkey MI, Isabey D, Clément A, Lofaso F. In vivo physiologic comparison of two ventilators used for domiciliary ventilation in children with cystic fibrosis. Crit Care Med. 2001;29(11):2097-2105. [CrossRef] [PubMed]
 
McCormack HM, Horne DJ, Sheather S. Clinical applications of visual analogue scales: a critical review. Psychol Med. 1988;18(4):1007-1019. [CrossRef] [PubMed]
 
Passy V, Baydur A, Prentice W, Darnell-Neal R. Passy-Muir tracheostomy speaking valve on ventilator-dependent patients. Laryngoscope. 1993;103(6):653-658. [CrossRef] [PubMed]
 
Manzano JL, Lubillo S, Henríquez D, Martín JC, Pérez MC, Wilson DJ. Verbal communication of ventilator-dependent patients. Crit Care Med. 1993;21(4):512-517. [CrossRef] [PubMed]
 
Hoit JD, Lohmeier HL. Influence of continuous speaking on ventilation. J Speech Lang Hear Res. 2000;43(5):1240-1251. [PubMed]
 
Konyukov YA, Kuwayama N, Fukuoka T, et al. Effects of different triggering systems and external PEEP on trigger capability of the ventilator. Intensive Care Med. 1996;22(4):363-368. [CrossRef] [PubMed]
 
Sassoon CSH. Triggering of the ventilator in patient-ventilator interactions. Respir Care. 2011;56(1):39-51. [CrossRef] [PubMed]
 
Lofaso F, Fodil R, Lorino H, et al. Inaccuracy of tidal volume delivered by home mechanical ventilators. Eur Respir J. 2000;15(2):338-341. [CrossRef] [PubMed]
 
Pellegrini N, Guillon B, Prigent H, et al. Optimization of power wheelchair control for patients with severe Duchenne muscular dystrophy. Neuromuscul Disord. 2004;14(5):297-300. [CrossRef] [PubMed]
 
Pellegrini N, Pelletier A, Orlikowski D, et al. Hand versus mouth for call-bell activation by DMD and Becker patients. Neuromuscul Disord. 2007;17(7):532-536. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1. Experimental device. The balloon of the external expiratory valve of the ventilator circuit is connected to a pressure generator that applies a constant pressure when activated and, therefore, generates a resistive positive end-expiratory pressure (PEEP) during expiration. The pressure generator can only apply pressure (and, therefore, PEEP) in the expiratory valve balloon when the patient presses a switch, which activates a solenoid valve that opens the circuit. When the patient activates the switch, the pressure generator immediately applies the prescribed pressure to the balloon of the expiratory valve. Expiration can occur through the expiratory valve only if the pressure in the circuit rises above the pressure inside the valve balloon (PEEP level); therefore, expiratory flow is directed toward the upper airways, allowing the patient to speak. To determine optimal PEEP level, the pressure delivered by the pressure generator is increased progressively by 1-cm H2O steps and tested at each level by pressing the switch until no expiratory volume is observed through the ventilator or until the pressure rise is limited by the patient’s respiratory comfort. Once obtained, the optimal PEEP level is set on the pressure generator, and the patient can obtain it instantaneously by pressing the switch.Grahic Jump Location
Figure Jump LinkFigure 2. Evolution of speech and respiratory comfort in 12 patients according to PEP level as evaluated using a VAS (0 = extremely uncomfortable, 10 = extremely comfortable). Data are presented as median and interquartile range. *P = .01, **P = .04, and ***P = .003 by Wilcoxon test. NS = nonsignificant; PEP0 = without positive end-expiratory pressure; PEP50 = positive end-expiratory pressure level representing 50% of PEPeff; PEPeff = optimal positive end-expiratory pressure level; VAS = visual analog scale.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1 —Demographic and Respiratory Characteristics of the Patient Population

BR = ventilator backup rate; CM = congenital myopathy; DMD = Duchenne muscular dystrophy; F = female; FKRP = limb girdle muscular dystrophy type 2I with a fukutin related protein deficiency; M = male; NA = not available; PEEPeff = optimal positive end-expiratory pressure level; Pemax = maximal expiratory pressure; Pimax = maximal inspiratory pressure; Pt = patient; Raw = airway resistance; SMA = spinal muscular atrophy; VC = vital capacity; VI = ventilator-delivered volume.

a 

Shiley unfenestrated/uncuffed tracheostomy tube (Covidien).

b 

Puritan Bennett Legendair ventilator (Covidien).

c 

Trachéoflex unfenestrated/uncuffed tracheostomy tube (Teleflex Medical).

d 

Tracoé unfenestrated/uncuffed tracheostomy tube (Pouret Médical).

e 

Rüsch unfenestrated/uncuffed tracheostomy tube (Teleflex Medical).

Table Graphic Jump Location
Table 2 —Respiratory Characteristics During Speech (n = 12)

Data are presented as median (interquartile range). Pe = expiratory pressure; Pi = inspiratory pressure; MVI = minute ventilation insufflation; PEEP0 = without positive end-expiratory pressure; PEEP50 = positive end-expiratory pressure level representing 50% of PEEPeff; RR = respiratory rate; Spo2 = oxygen saturation as measured by pulse oximetry; Te = expiratory time; Ti = inspiratory time; TTOT = total respiratory time; VTe = expiratory volume through the tracheostomy tube. See Table 1 legend for expansion of other abbreviation.

a 

Friedman test.

b 

P < .04 vs PEEP0 by Wilcoxon test.

c 

P < .003 vs PEEP0 by Wilcoxon test.

d 

P < .007 vs PEEP0 by Wilcoxon test.

e 

P < .03 vs PEEP50 by Wilcoxon test.

f 

P < .002 vs PEEP50 by Wilcoxon test.

Table Graphic Jump Location
Table 3 —Phonation Characteristics (n = 12)

Data are presented as median (interquartile range). Perceptual score, 0-128 scale; best score, 0.14 Intelligibility score, 0-8 scale; best score, 8.13 Prosodia score, 0-4; best score, 4. See Table 1 and 2 legends for expansion of abbreviations.

a 

Friedman test.

b 

P < .006 vs PEEP0 by Wilcoxon test.

c 

P < .003 vs PEEP0 by Wilcoxon test.

d 

P < .006 vs PEEP50 by Wilcoxon test.

References

Laakso K, Markström A, Hartelius L. Communication and quality of life in individuals receiving home mechanical ventilation. Int J Ther Rehabil. 2009;16(12):648-655.
 
Raphaël JC, Dazord A, Jaillard P, et al. Assessment of quality of life for home ventilated patients with Duchenne muscular dystrophy [in French]. Rev Neurol (Paris). 2002;158(4):453-460. [PubMed]
 
Leder SB. Importance of verbal communication for the ventilator-dependent patient. Chest. 1990;98(4):792-793. [CrossRef] [PubMed]
 
Bach JR, Alba AS. Tracheostomy ventilation. A study of efficacy with deflated cuffs and cuffless tubes. Chest. 1990;97(3):679-683. [CrossRef] [PubMed]
 
Hoit JD, Shea SA, Banzett RB. Speech production during mechanical ventilation in tracheostomized individuals. J Speech Hear Res. 1994;37(1):53-63. [PubMed]
 
Prigent H, Samuel C, Louis B, et al. Comparative effects of two ventilatory modes on speech in tracheostomized patients with neuromuscular disease. Am J Respir Crit Care Med. 2003;167(2):114-119. [CrossRef] [PubMed]
 
Shea SA, Hoit JD, Banzett RB. Competition between gas exchange and speech production in ventilated subjects. Biol Psychol. 1998;49(1-2):9-27. [CrossRef] [PubMed]
 
Hoit JD, Banzett RB. Simple adjustments can improve ventilator-supported speech. Am J Speech Lang Pathol. 1997;6:87-96.
 
Hoit JD, Banzett RB, Lohmeier HL, Hixon TJ, Brown R. Clinical ventilator adjustments that improve speech. Chest. 2003;124(4):1512-1521. [CrossRef] [PubMed]
 
Prigent H, Garguilo M, Pascal S, et al. Speech effects of a speaking valve versus external PEEP in tracheostomized ventilator-dependent neuromuscular patients. Intensive Care Med. 2010;36(10):1681-1687. [CrossRef] [PubMed]
 
Lofaso F, Aslanian P, Richard JC, et al. Expiratory valves used for home devices: experimental and clinical comparison. Eur Respir J. 1998;11(6):1382-1388. [CrossRef] [PubMed]
 
Marini JJ, Culver BH, Kirk W. Flow resistance of exhalation valves and positive end-expiratory pressure devices used in mechanical ventilation. Am Rev Respir Dis. 1985;131(6):850-854. [PubMed]
 
Auzou P, Ozsancak C, Jan M, et al. Clinical assessment of dysarthria: presentation and validation of a method [in French]. Rev Neurol (Paris). 1998;154(6-7):523-530. [PubMed]
 
Ozsancak C, Parais AM, Auzou P. Perceptual analysis of dysarthria: presentation and validation of a clinical scale. Preliminary study [in French]. Rev Neurol (Paris). 2002;158(4):431-438. [PubMed]
 
Fauroux B, Pigeot J, Polkey MI, Isabey D, Clément A, Lofaso F. In vivo physiologic comparison of two ventilators used for domiciliary ventilation in children with cystic fibrosis. Crit Care Med. 2001;29(11):2097-2105. [CrossRef] [PubMed]
 
McCormack HM, Horne DJ, Sheather S. Clinical applications of visual analogue scales: a critical review. Psychol Med. 1988;18(4):1007-1019. [CrossRef] [PubMed]
 
Passy V, Baydur A, Prentice W, Darnell-Neal R. Passy-Muir tracheostomy speaking valve on ventilator-dependent patients. Laryngoscope. 1993;103(6):653-658. [CrossRef] [PubMed]
 
Manzano JL, Lubillo S, Henríquez D, Martín JC, Pérez MC, Wilson DJ. Verbal communication of ventilator-dependent patients. Crit Care Med. 1993;21(4):512-517. [CrossRef] [PubMed]
 
Hoit JD, Lohmeier HL. Influence of continuous speaking on ventilation. J Speech Lang Hear Res. 2000;43(5):1240-1251. [PubMed]
 
Konyukov YA, Kuwayama N, Fukuoka T, et al. Effects of different triggering systems and external PEEP on trigger capability of the ventilator. Intensive Care Med. 1996;22(4):363-368. [CrossRef] [PubMed]
 
Sassoon CSH. Triggering of the ventilator in patient-ventilator interactions. Respir Care. 2011;56(1):39-51. [CrossRef] [PubMed]
 
Lofaso F, Fodil R, Lorino H, et al. Inaccuracy of tidal volume delivered by home mechanical ventilators. Eur Respir J. 2000;15(2):338-341. [CrossRef] [PubMed]
 
Pellegrini N, Guillon B, Prigent H, et al. Optimization of power wheelchair control for patients with severe Duchenne muscular dystrophy. Neuromuscul Disord. 2004;14(5):297-300. [CrossRef] [PubMed]
 
Pellegrini N, Pelletier A, Orlikowski D, et al. Hand versus mouth for call-bell activation by DMD and Becker patients. Neuromuscul Disord. 2007;17(7):532-536. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement 1

Supporting Data

Online Supplement 2

Supporting Data

Online Supplement 3

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
Speech and mechanical ventilation. Chest 2013;144(5):1739-40.
Response. Chest 2013;144(5):1740-1.
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543