- Email: [email protected]
Automatic tailoring of the lowest PEEP to abolish tidal expiratory flow limitation in seated and supine COPD patients
Respiratory Medicine 155 (2019) 13–18
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Automatic tailoring of the lowest PEEP to abolish tidal expiratory flow limitation in seated and supine COPD patients
T
Ilaria Milesia, Roberto Portab, Luca Barbanob, Simona Cacciatorea, Michele Vitaccab, Raffaele L. Dellacàa,* a b
Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy Pneumologia Riabilitativa dell' Istituto di Lumezzane, Istituti Clinici Scientifici Maugeri IRCCS, Brescia, Italy
ARTICLE INFO
ABSTRACT
Keywords: Lung function monitoring Forced oscillation technique Non-invasive ventilation Intrinsic PEEP Supine position
Rationale: In COPD patients, the development of tidal expiratory flow limitation (EFLT) results in intrinsic positive end-expiratory pressure (PEEPi), leading to increased work of breathing and worsening patient-ventilator interaction. An external PEEP can mitigate these consequences, but how to optimize its value it is still unknown. Objective: To measure the minimum PEEP able to abolish EFLT by a new automatic non-invasive ventilation (NIV) mode in stable hypercapnic COPD patients in the seated and supine positions. Methods: Twenty-six hypercapnic COPD patients (mean ± SD: FEV1%pred = 39.2 ± 16.1, FEV1/FVC% pred = 46.3 ± 16.3%) were studied while receiving NIV during two consecutive 15-min periods, with patients studied seated in the first and supine in the second. A ventilator able to identify EFLT breath-by-breath by using the forced oscillation technique optimized in real-time PEEP to the lowest pressure able to abolish EFLT (PEEPO). Results: The ventilator was always able to identify a PEEPO. Its values were highly variable among patients and increased from median(iqr) 4.0 (0.03) (range: 4.0–8.3cmH2O) to 6 (6.1) cmH2O (range: 4.0–15.7 cmH2O) when patients moved from the seated to the supine position, respectively. PEEPO in supine position did not correlate to any spirometric or anthropometric variable. Conclusions: PEEPO in COPD patients is highly variable and increases in supine position. It is not predicted by spirometric nor anthropometric variables, but had a considerable variability among the patients. We suggest that PEEPo may be used as a phenotyping variable in COPD patients.
1. Introduction Patients with chronic obstructive pulmonary disease (COPD) can experience dynamic hyperinflation (DH) [1], leading to an increase of the end-expiratory recoil pressure of the respiratory system compared to the one measured in resting condition at the end of a prolonged and relaxed expiration. This pressure, commonly referred to as the “intrinsic positive end expiratory pressure” (PEEPi), constitutes a relevant additional threshold load to breathing [2]. Two studies on mechanically ventilated COPD found that it accounted up to more than 50% of the overall increase of work of breathing (WOB) due to the disease [3,4]. DH is associated with the presence of tidal expiratory flow-limitation (EFLT), a non-linear flow regimen which limits the maximal expiratory flow regardless of the driving pressure developed by the subject [5]. In this condition the maximal expiratory flow depends only on lung volume, requiring patients to breathe at a higher lung volume in order to restore proper minute ventilation. *
Non-invasive mechanical ventilation (NIV) is commonly used for the treatment of acute respiratory failure in COPD with the aim of improving pulmonary gas exchange and unloading the respiratory muscles [6]. During this treatment, the use of an externally-applied PEEP allows to counteract the detrimental effects of PEEPi, leading to a reduction of the WOB, normalization of the pattern of breathing, improvement of blood gases, and reduction of patient-ventilator asynchronies[3, 7–9]. However, in order to be effective, PEEP needs to be tailored to match patient PEEPi. If the ventilator PEEP is lower than PEEPi, its positive effects are impaired. Conversely, if it exceeds PEEPi, it results in a further increase of the end expiratory lung volume and, therefore, increases patient's WOB. Excessive PEEP also may results in adverse effects on hemodynamics, as uselessly large continuous intrathoracic pressures may decrease venous return and cardiac output, depending upon intravascular volume status, myocardial function and other factors [9–11].
Corresponding author. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Via Giuseppe Colombo, 40, I-20133, Milano, Italy. E-mail address: [email protected] (R.L. Dellacà).
https://doi.org/10.1016/j.rmed.2019.06.022 Received 9 April 2019; Received in revised form 14 June 2019; Accepted 22 June 2019 Available online 23 June 2019 0954-6111/ © 2019 Elsevier Ltd. All rights reserved.
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
The clinical use of PEEP would need, therefore, an accurate tailoring for each individual patient and condition. Such tailoring would also require continuous adjustment, considering that EFLT markedly changes not only with time [12, 13], but also when changing body posture, such as when patient moves from the seated position to the supine or lateral ones. EFLT can be non-invasively detected in real time breath-by-breath using the forced oscillation technique (FOT) by measuring the intrabreath variations of respiratory reactance (ΔXrs) at 5Hz [14]. This method has proven high sensitivity and specificity when compared to the invasive Mead and Wittenberger [15] gold standard during quiet breathing and nasal CPAP [16] and it is in good agreement with the negative expiratory pressure (NEP) technique [17]. As this method is suitable for real time use during NIV, it potentially allows the implementation of real time ventilation strategies that continuously optimises PEEP by adjusting it to the lowest pressure able to abolish EFLT. We hypothesized that a novel NIV ventilation mode which continuously optimizes PEEP in order to abolish EFLT could identify an optimal PEEP for each patient and according to different postures. The aims of the present study were 1) to investigate the prevalence of EFLT and the PEEP values needed to abolish EFLT by the novel NIV approach on a group of stable hypercapnic COPD patients, and 2) to characterise the effects of changing posture from the seated to the supine position on both prevalence of EFLT and optimal PEEP.
connected to the patient's mask, and the ventilator was set to avoid PEEP settings below 4 cmH2O. The pressure support, i.e. the difference between the inspiratory pressure (inspiratory positive airway pressure, IPAP) and PEEP, is kept constant by the ventilator at the initial prescribed (baseline) value by continuously adjusting IPAP by the same changes applied to PEEP. Digital values of pressure (Pao), flow and volume at the airway opening, as well as Rrs and Xrs data, were sent by the ventilator to a laptop at a sampling rate of 100 Hz and recorded for subsequent analysis.
2. Material and methods
2.4. Data analysis
2.1. Subjects
Breathing pattern and respiratory input impedance data analysis: For each posture, the last minute of recorded data was considered, allowing the selection of 5–10 breaths free from artifacts for evaluation. From these breaths, breathing pattern parameters, average PEEP, inspiratory and expiratory resistance (Rrs) and reactance (Xrs) and the ΔXrs were computed, averaged and used for statistical comparison. Estimation of the sample size: Differences in ΔXrs from seated to supine posture in COPD patient is characterized by a standard deviation equal to 3.4 cmH2O*s/L with a matched difference within posture of 2.3 cmH2O*s/L [16]. To reject the null hypothesis (ΔXrs seated is equal to ΔXrs supine) with power equals to 0.90 at a significance level of 0.05, the sample size is 25 patients. Statistical analysis: All data were tested for normality (D'Agostiono & Pearson normality test). T-test was used to compare two groups and one-way ANOVA was used to compare three groups any time the dataset passed the normality test, otherwise Wilcoxon matched-pairs signed rank test was used. A logistic regression model was used to assess the probability of becoming EFLT moving from seated to supine position on the basis of spirometric parameters and body max index (BMI). Only differences/relationships with p < 0.05 were considered statistically significant. Data are reported as mean ± SD when normally distributed, and as median (IQR) otherwise.
2.3. Study protocol Patients were asked to sit on a reclining armchair and connected to the ventilator for delivering their prescribed ventilator settings. After a few minutes for allowing the stabilization of breathing pattern, the automatic tailoring system was enabled and the ventilator started adjusting PEEP iteratively in order to abolish EFLT. After 15 min of spontaneous breathing in automatic mode, patients were moved in the supine position and ventilated for further 15 min. For each posture, a patient was considered as tidal flow limited (EFLT) or not (NEFLT) whether the optimal PEEP identified by the system was ≥ 4 cmH2O, i.e. the minimum allowed PEEP setting. Patients that were eventually EFLT at lower PEEPs but abolished EFLT with PEEP = 4 cmH2O were therefore classified as NEFLT.
Moderate, severe or very severe hypercapnic COPD patients diagnosed by spirometry [18] who were habitual users of nocturnal NIV in stable state condition were eligible for this study. Inclusion criteria were: age below 85 years; ability to provide consent. Exclusion criteria: patients who were acutely ill, medically complicated or who are medically unstable, as determined by the investigator; patients suffering from a COPD exacerbation or who suffered and exacerbation within the past two months. Indication for NIV was based on presence of severe symptoms, clinical history of frequent exacerbations/hospitalizations due to acute hypercapnic respiratory failure and stable hypercapnia (> 54 mmHg) post the acute use of NIV. The study was approved by the Ethical Review Boards of ICS S. Maugeri (No. 897 CEC/2013). All subjects gave written informed consent before the beginning of the study. Patients were recruited from April 2015 to April 2017. 2.2. Measurements and set-up Patients were connected to a non-commercial version of a NIV ventilator (Synchrony, Philips-Respironics) by an unvented facial mask (AMARA, Philips-Respironics). The ventilator incorporates real-time FOT single–frequency within-breath measurement of the total respiratory impedance (Zrs, expressed as resistance, Rrs, and reactance, Xrs) at 5 Hz, and uses these measurements to automatically adjust PEEP in order to abolish EFLT. In details, for each single breath free from artifacts (automatically detected by empirical rules on impedance values), the ventilator computes the difference between the average Xrs measured during inspiration (Xinsp) and the one measured during expiration, namely ΔXrs [14], and it increases/decreases PEEP by 0.1 cmH2O whether ΔXrs is greater or smaller than the threshold of 2.8 cmH2Os/L for detecting EFLT [14,16]. This procedure allows for delivery of the minimum PEEP value able to abolish EFLT and continuously adjusts this value to changes in patient respiratory mechanics. As for all single-limb ventilators, in order to avoid CO2 rebreathing an intentional leak (whisper swivel, Philips Respironics, US) was
3. Results A total of twenty-six patients were studied. Table 1 summarizes their anthropometric and spirometric characteristics at baseline. Fig. 1 reports the time course for one representative patient of the pressure at the mask and ΔXrs during the trial. In the seated position, the ventilator decreased PEEP from the prescribed value of 7 cmH2O to the minimum value of 4 cmH2O, as the patient was not showing EFLT in this posture. When the patient moved to supine, PEEP was gradually increased by the ventilator to approximatively 10 cmH2O in order to abolish the EFLT induced by change in posture. Fig. 2 represents the optimal automatically-adjusted PEEP (PEEPO) required to abolish EFLT in each posture for all patients. Seven patients were EFLT both in the seated and the supine 14
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
Table 1 Main anthropometric and baseline lung function data. BMI: body mass index; FEV1: forced expiratory volume in the first second in liters and as percentage of the predicted value; FEV1/FVC%: ratio of FEV1 on forced vital capacity expressed as the percentage of the predicted value; TLC: total lung capacity, expressed in liters and as percentage of the predicted value. Numbers of patients classified according to GOLD stages are also reported.
Sex (F: M) Age (yrs) BMI (kg/m2) FEV1 (L) FEV1% pred FEV1/FVC % TLC (L) TLC % pred GOLD 1: GOLD 2: GOLD 3: GOLD 4:
ALL (n = 26)
NEFLT (n = 8)
EFLT (n = 7)
EFLSUPINE (n = 11)
F = 7: M = 19 72.4 ± 6.8 30.5 ± 6.3 1.40 ± 0.60 39.2 ± 16.1 46.27 ± 16.33 6.28 ± 1.23 109.33 ± 21.73 0 8 8 10
F = 2: M = 6 73.4 ± 6.3 30.1 ± 5.4 1.58 ± 0.55 43.0 ± 18.5 47.50 ± 14.26 6.30 ± 1.18 109.60 ± 28.19 0 3 3 2
F = 1: M = 6 70.0 ± 8.4 31.1 ± 5.7 1.16 ± 0.46 36.2 ± 12.9 44.00 ± 18.20 6.04 ± 1.08 112.80 ± 27.25 0 1 3 3
F = 4: M = 7 72.0 ± 6.4 31.6 ± 6.7 1.48 ± 0.66 39.9 ± 16.4 48.27 ± 17.80 6.35 ± 1.48 107.2 ± 18.63 0 4 2 4
positions, eight patients never showed EFLT and eleven developed EFLT only in the supine position (EFLT-SUP). No significant differences were found in FEV1, GOLD grades, FEV1/FVC, or TLC between the groups. As the number of NEFLT patients reduced by 43% when moving from seated to supine (from 19 to 8), the median (IQR) PEEPO increased significantly (p < 0.0001) from 4 (0.03) cmH2O in seated to 6.10 (6.13) cmH2O in the supine position. If we consider only patients EFLT when seated, the average PEEPo increased after moving to the supine position from 5.99 ± 1.77 cmH2O to 11.54 ± 5.10 (p = 0.057). Fig. 3 shows the relationship of PEEPo vs the clinically prescribed PEEP for each patient in the supine position. Interestingly, even if the average values for the tailored and for the prescribed PEEP was very similar (7.42 ± 2.10 vs 7.45 ± 3.78 cmH2O, respectively), they were very different considering each individual patient. BMI and PEEPO had no statistically significant relationship, despite a slightly positive trend can be observed in Fig. 4a. Forced expiratory volume in 1s expressed as percentage of predicted value (FEV1%) and FEV1/FVC% did not show significant correlation with PEEPO (Fig. 4). Similarly, spirometric parameters and BMI were non-significant predictors of changing EFLT status from seated to supine position, according to logistic regression output. Similar lack of linear correlation was found after removing patients not exhibiting EFLT in the seated position. Table 2 reports respiratory system parameters and breathing pattern. Inspiratory reactance, Xinsp, showed a non-statistically significant tendency to be lower in NEFLT compared to EFLT in both postures.
Fig. 1. Pressure at the airway opening (Pao) and ΔXrs tracings recorded during the trial for a representative patient.
Fig. 3. Clinically prescribed values of PEEP vs PEEPo. Open circles: patients who are never flow limited (NEFLT); closed circles: patients who are flow limited both in the seated and supine positions (EFLT); closed triangles: patients flow limited only in supine position (EFLT-SUP). Regression line is reported together with r2 and p value. Means and standard deviations are reported for both prescribed PEEP and PEEPO on the horizontal and vertical axis, respectively.
Fig. 2. Individual values of PEEP required to abolish EFLT (PEEPO) in the seated and the supine positions. 15
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
Fig. 4. Linear correlation of a) BMI, b) FEV1% and c) FEV1/FVC% vs. the automatically tailored PEEP (PEEPO). Regression lines are reported in each panel with correspondent r2 and p values.
4. Discussion
were asked to move from the seated to the supine position. Eleven patients had a sudden increase in ΔXrs after they changed from sitting to supine position. This response to a posture change is likely the consequence of the lung volume reduction promoted by the supine position, where the different direction of the gravitational force acting on the abdominal contents promotes a cranial shift of the diaphragm [21]. In the eight patients that did not develop EFLT in supine position, ΔXrs increased from 0.53 (2.15) to 1.84 (1.27) cmH2O*s/L, suggesting that, even if EFLT did not develop in those patients, some choke points started to appear in the bronchial tree during expiration [19]. In all patients who became EFLT moving to supine, the ventilator reacted by increasing the PEEP breath by breath until a value able to abolish EFLT was reached. It is noteworthy that EFLT was abolished in all patients simply by increasing PEEP, i.e. without changing any other ventilator parameter, within 10.1 (5.3) min from the posture changes. PEEPo showed greater range of values compared to commonly clinical prescribed ones. Interestingly, no significant correlations were found between spirometric variables and PEEPo in the seated or supine positions, suggesting that the degree of obstruction per se it is not a major determinant of EFLT. This is also in line with previous studies that showed that EFLT is present in COPD over a wide range of FEV1%, even if it is more common in patients with more severe airway obstructions [22]. Similarly, multiple linear regression analysis did not find any combination of parameters able to predict PEEPo, suggesting that the pressure needed to abolish EFLT is the result of a very complex interaction between anthropometric variables and the functional alteration due to the disease. Our results are in line with those of a previous study where it was found that, despite the expertise of the healthcare team, the clinical setting based only on patients comfort and arterial blood gasses response resulted in differences when compared to a physiological setting
In this study we introduce a new NIV setting mode that automatically and continuously detects EFLT breath-by-breath using FOT and adjusts PEEP accordingly in order to deliver the lowest PEEP able to abolish EFLT. This device has been applied on a population of moderate to very severe COPD patients in order to characterise the prevalence of EFLT for each posture and to measure PEEPo and how this value changes when patients move from the seated to the supine position. The main findings are: 1) the ventilator was always able to identify a PEEP value which abolished EFLT for all patients who showed EFLT in any posture (85% of patients); 2) the PEEPo was highly variable between patients, with a maximum PEEPo identified in this study of 15.71 cmH2O, a value greater than the commonly prescribed PEEPs for this group of patients; 3) when moving from the seated to the supine position, the prevalence of EFLT markedly increased (from 33% to 85%). Also PEEPo increased when moving to supine, with the amplitude of such increase showing high inter-individual variability; 4) PEEPo cannot be predicted by either BMI, spirometric parameters or a combination of those. We suggest that PEEPo represents an independent feature of patients’ conditions and, therefore, potentially identifies different phenotypes. EFLT is a relatively common condition in COPD, and several previous studies reported a prevalence ranging from 29.4% to 52.3% in a wide variety of patient conditions [17, 19, 20]. In our study population, we found the incidence of EFLT in the seated posture of 26,9%, slightly lower, but still comparable, with previous findings. This lower incidence may be the consequence of the need of applying, for technical reasons (see methods), a minimum PEEP of 4 cmH2O, that may be enough to abolish EFLT in some of the patients. The prevalence of EFLT increased significantly to 69% when patients
Table 2 Respiratory mechanics and breathing pattern. Data are reported by grouping patients according to their EFLT condition. ΔXrs: within breath variation of respiratory reactance; Xinsp: inspiratory reactance; Rinsp: inspiratory resistance; Vt: tidal volume; RR: respiratory rate, Ti/Ttot: respiratory duty cycle; VE: minute ventilation.
ALL Seated Supine NEFLT Seated Supine EFLT Seated Supine EFLT-SUP Seated Supine
ΔXrs (cmH2O*s/L)
Xinsp (cmH2O*s/L)
Rinsp (cmH2O*s/L)
Vt (L)
RR (Breaths/min)
Ti/Ttot
VE (L/min)
1.99(2.25) 2.73(1.11)
−3.49(1.93) −5.17(3.53)
7.54(4.37) 8.46(4.46)
0.85(0.51) 0.74(0.52)
18.07(8.14) 17.03(8.16)
0.33(0.07) 0.31(0.06)
12.91(5.35) 10.83(4.15)
0.53(2.15) 1.84(1.27)
−3.41(2.26) −3.51(1.68)
5.18(0.75) 5.11(2.39)
0.83(0.34) 0.59(0.43)
22.25(4.98) 17.51(7.54)
0.35(0.05) 0.34(0.09)
14.59(8.55) 11.19(5.50)
2.75(0.63) 2.95(0.74)
−6.71(2.85) −6.58(2.02)
7.97(2.41) 8.81(1.66)
0.72(0.47) 0.70(0.28)
16.04(7.50) 15.54(3.50)
0.32(0.06) 0.29(0.03)
10.71(3.82) 10.50(4.71)
1.81(1.36) 2.97(0.87)
−3.15(1.59) −4.38(1.19)
8.63(2.30) 9.20(2.69)
1.09(0.63) 1.00(0.52)
14.49(5.88) 12.79(7.20)
0.33(0.04) 0.30(0.05)
12.04(4.00) 10.72(2.89)
16
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
oriented by assessing PEEPi using esophageal pressure in almost 50% of the patients [23]. These discrepancies were associated to an increased presence of ineffective efforts, suggesting that an empirical setting may be not appropriate for a generalized use. The same study also reported that, in line with our results, despite the PEEP determined by the clinical optimization approach was often different to the one determined by esophageal manometry, their average values were very similar [23]. Another study reported that, when ventilator PEEP is tailored to patient's respiratory mechanics by esophageal manometry, the inspiratory muscle workload due to PEEPi is reduced by NIV at a greater extent compared to clinical settings [24]. Even if the use of esophageal manometry has a strong rationale, the invasiveness and the complexity of the method allowed its use only for small physiological studies. Moreover, this approach allows the identification of the optimal PEEP at a specific time point, requiring delivery of the same PEEP until a further assessment is performed, impeding a prompt response to changes in patients’ respiratory mechanics. On the contrary, our method allows for the first time a continuous PEEP titration aimed to provide the lowest PEEP able to abolish EFLT during ventilation. This generates a new form of interaction between the ventilator and the patient: when the ventilator optimizes PEEP, the neural respiratory control of the patient reacts to any change of the mechanical operating conditions of the respiratory system by adjusting the spontaneous breathing pattern. Changes in breathing pattern, in turn, impacts also on the presence of EFLT, leading to a continuous patient-ventilator interaction lead by the neurological control system which cannot be obtained when single occasional adjustments of ventilator parameters are performed.
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments We acknowledge Cristina Zanoni 1 e Luca Gitti1 for their support in performing the measurements, we also acknowledge Bob Romano2, Jim McKenzie2 and Peter Hill2 for their technical support and Mar Janna Dahl3 for language editing and proofreading. 1 Pneumologia Riabilitativa, ICS S. Maugeri IRCCS, Lumezzane (Brescia), Italy; 2 Philips Respironics. Pittsburgh (PA), USA; 3 University of Utah (UT), USA and University of Western Australia (WA), Australia. List of abbreviations BMI COPD CPAP DH EFLT EFLT-SUP FEV1 FVC IPAP IQR PEEP PEEPi PEEPo
5. Conclusion In this study we found that a new ventilation modality that uses FOT for continuously tailoring PEEP to the lowest pressure able to abolish EFLT was able to identify, in all patients, an optimal PEEP. This optimal PEEP was very variable among patients and independent from the prescribed values. Moreover, it could not be predicted by spirometric nor anthropometric variables. For these reasons, we suggest that PEEPo may be used as a phenotyping variable for COPD. We believe that this automatic, continuous optimization of PEEP may empower the performance of the neural respiratory control of the patient by minimizing the mechanical load and providing, therefore, a continuously individualized ventilator support opening new scenarios in the management of COPD patients. Future studies are warranted to evaluate whether this personalized NIV ventilation mode has impacts on arterial blood gases, patient comfort, adherence, exacerbations rate and sleep quality.
FOT NEP NEFLT NIV Rrs WOB Xrs Xinsp Zrs ΔXrs
Body Mass Index Chronic Obstructive Pulmonary Disease Continuous Positive Airway Pressure Dynamic Hyperinflation tidal Expiratory Flow Limitation EFLT only in the supine position Forced Expiratory Volume in the 1st second Forced Vital Capacity inspiratory positive airway pressure Inter-Quartile Range Positive End Expiratory Pressure intrinsic Positive End Expiratory Pressure automatically-adjusted optimal Positive End Expiratory Pressure Forced Oscillation Technique Negative Expiratory Pressure Non Tidal Expiratory Flow Limited Non-Invasive Ventilation Total respiratory system input resistance Work Of Breathing Total respiratory system input reactance Xrs measured during inspiration Total respiratory system input impedance difference between the average Xrs measured during inspiration and the one measured during expiration
References [1] B.R. Celli, W. MacNee, A. ATS/ERS Task Force, et al., Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper, Eur. Respir. J. 23 (2004) 932–946. [2] B.P. Krieger, Hyperinflation and intrinsic positive end-expiratory pressure: less room to breathe, Respiration 77 (2009) 344–350. [3] L. Appendini, A. Purro, A. Patessio, et al., Partitioning of inspiratory muscle workload and pressure assistance in ventilator-dependent COPD patients, Am. J. Respir. Crit. Care Med. 154 (1996) 1301–1309. [4] M.L. Coussa, C. Guérin, N.T. Eissa, et al., Partitioning of work of breathing in mechanically ventilated COPD patients, J. Appl. Physiol. 75 (1993) 1711–1719. [5] S.V. Dawson, E.A. Elliott, Wave-speed limitation on expiratory flow-a unifying concept, J. Appl. Physiol. 43 (1977) 498–515. [6] N. Ambrosino, L. Xie, The use of non-invasive ventilation during exercise training in COPD patients, COPD J. Chronic Obstr. Pulm. Dis. 14 (2017) 396–400. [7] M.W. Elliott, D.A. Mulvey, J. Moxham, et al., Inspiratory muscle effort during nasal intermittent positive pressure ventilation in patients with chronic obstructive airways disease, Anaesthesia 48 (1993) 8–13. [8] S. Nava, C. Bruschi, C. Fracchia, et al., Patient-ventilator interaction and inspiratory effort during pressure support ventilation in patients with different pathologies, Eur. Respir. J. 10 (1997) 177–183. [9] F. Baigorri, A. de Monte, L. Blanch, et al., Hemodynamic responses to external counterbalancing of auto-positive end-expiratory pressure in mechanically ventilated patients with chronic obstructive pulmonary disease, Crit. Care Med. 22
Sources of support and conflicts of interest Politecnico di Milano University, Institution of IM, SC and RLD, owns a patent on the use of FOT for the detection of expiratory flow limitation licensed to Philips Respironics. Istituti Clinici Scientifici Maugeri received a free loan of part of the equipment used for this study and a research grant from Philips. Authors’ contribution I.M. participated in the conception and design of the study, data collection, data analysis, interpretation of the data and preparation of the manuscript; R.P. and L.B. participated in the data collection and analysis and preparation of the manuscript; S.C. participated in the data collection and data analysis; M.V. and R.L.D. participated in the conception and design of the study, data analysis, interpretation of the data and preparation of the manuscript. 17
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al. (1994) 1782–1791. [10] N. Ambrosino, S. Nava, A. Torbicki, et al., Haemodynamic effects of pressure support and PEEP ventilation by nasal route in patients with stable chronic obstructive pulmonary disease, Thorax 48 (1993) 523–528. [11] V.M. Ranieri, M. Dambrosio, N. Brienza, Intrinsic PEEP and cardiopulmonary interaction in patients with COPD and acute ventilatory failure, Eur. Respir. J. 9 (1996) 1283–1292. [12] E. Boni, L. Corda, D. Franchini, et al., Volume effect and exertional dyspnoea after bronchodilator in patients with COPD with and without expiratory flow limitation at rest, Thorax 57 (2002) 528–532. [13] H. Patel, K.L. Yang, Variability of intrinsic positive end-expiratory pressure in patients receiving mechanical ventilation, Crit. Care Med. 23 (1995) 1074–1079. [14] R.L. Dellacà, P. Santus, A. Aliverti, et al., Detection of expiratory flow limitation in COPD using the forced oscillation technique, Eur. Respir. J. 23 (2004) 232–240. [15] J. Mead, J.L. Whittenberger, Physical properties of human lungs measured during spontaneous respiration, J. Appl. Physiol. 5 (1953) 779–796. [16] R.L. Dellacà, M. Rotger, A. Aliverti, et al., Noninvasive detection of expiratory flow limitation in COPD patients during nasal CPAP, Eur. Respir. J. 5 (2006) 983–991. [17] R.L. Dellacà, N. Duffy, P.P. Pompilio, et al., Expiratory flow limitation detected by forced oscillation and negative expiratory pressure, Eur. Respir. J. 29 (2007) 363–7418.
[18] GOLD, Global Strategy for the Diagnosis, Management and Prevention of COPD, Global Initiative for Chronic Obstructive Lung Disease (GOLD), 2017. [19] R.L. Dellacà, P.P. Pompilio, P.P. Walker, et al., Effect of bronchodilation on expiratory flow limitation and resting lung mechanics in COPD, Eur. Respir. J. 33 (2009) 1329–1337. [20] T. Akita, T. Shirai, K. Mori, et al., Association of the forced oscillation technique with negative expiratory pressure in COPD, Respir. Physiol. Neurobiol. 220 (2016) 62–68. [21] A. Aliverti, R. Dellacà, P. Pelosi, et al., Compartmental analysis of breathing in the supine and prone positions by optoelectronic plethysmography, Ann. Biomed. Eng. 29 (2001) 60–70. [22] B.B. Aarli, P.M. Calverley, R. Jensen, et al., The association of tidal EFL with exercise performance, exacerbations, and death in COPD, Int. J. Chronic Obstr. Pulm. Dis. 12 (2017) 2179–2188. [23] M. Vitacca, S. Nava, M. Confalonieri, et al., The appropriate setting of noninvasive pressure support ventilation in stable COPD patients, Chest 118 (2000) 1286–1293. [24] M.1 Vitacca, B. Lanini, S. Nava, L. Barbano, R. Portal, E. Clini, N. Ambrosino, Inspiratory muscle workload due to dynamic intrinsic PEEP in stable COPD patients: effects of two different settings of non-invasive pressure-support ventilation, Monaldi Arch. Chest Dis. 61 (2) (2004 Apr-Jun) 81–85.
18
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Automatic tailoring of the lowest PEEP to abolish tidal expiratory flow limitation in seated and supine COPD patients
T
Ilaria Milesia, Roberto Portab, Luca Barbanob, Simona Cacciatorea, Michele Vitaccab, Raffaele L. Dellacàa,* a b
Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy Pneumologia Riabilitativa dell' Istituto di Lumezzane, Istituti Clinici Scientifici Maugeri IRCCS, Brescia, Italy
ARTICLE INFO
ABSTRACT
Keywords: Lung function monitoring Forced oscillation technique Non-invasive ventilation Intrinsic PEEP Supine position
Rationale: In COPD patients, the development of tidal expiratory flow limitation (EFLT) results in intrinsic positive end-expiratory pressure (PEEPi), leading to increased work of breathing and worsening patient-ventilator interaction. An external PEEP can mitigate these consequences, but how to optimize its value it is still unknown. Objective: To measure the minimum PEEP able to abolish EFLT by a new automatic non-invasive ventilation (NIV) mode in stable hypercapnic COPD patients in the seated and supine positions. Methods: Twenty-six hypercapnic COPD patients (mean ± SD: FEV1%pred = 39.2 ± 16.1, FEV1/FVC% pred = 46.3 ± 16.3%) were studied while receiving NIV during two consecutive 15-min periods, with patients studied seated in the first and supine in the second. A ventilator able to identify EFLT breath-by-breath by using the forced oscillation technique optimized in real-time PEEP to the lowest pressure able to abolish EFLT (PEEPO). Results: The ventilator was always able to identify a PEEPO. Its values were highly variable among patients and increased from median(iqr) 4.0 (0.03) (range: 4.0–8.3cmH2O) to 6 (6.1) cmH2O (range: 4.0–15.7 cmH2O) when patients moved from the seated to the supine position, respectively. PEEPO in supine position did not correlate to any spirometric or anthropometric variable. Conclusions: PEEPO in COPD patients is highly variable and increases in supine position. It is not predicted by spirometric nor anthropometric variables, but had a considerable variability among the patients. We suggest that PEEPo may be used as a phenotyping variable in COPD patients.
1. Introduction Patients with chronic obstructive pulmonary disease (COPD) can experience dynamic hyperinflation (DH) [1], leading to an increase of the end-expiratory recoil pressure of the respiratory system compared to the one measured in resting condition at the end of a prolonged and relaxed expiration. This pressure, commonly referred to as the “intrinsic positive end expiratory pressure” (PEEPi), constitutes a relevant additional threshold load to breathing [2]. Two studies on mechanically ventilated COPD found that it accounted up to more than 50% of the overall increase of work of breathing (WOB) due to the disease [3,4]. DH is associated with the presence of tidal expiratory flow-limitation (EFLT), a non-linear flow regimen which limits the maximal expiratory flow regardless of the driving pressure developed by the subject [5]. In this condition the maximal expiratory flow depends only on lung volume, requiring patients to breathe at a higher lung volume in order to restore proper minute ventilation. *
Non-invasive mechanical ventilation (NIV) is commonly used for the treatment of acute respiratory failure in COPD with the aim of improving pulmonary gas exchange and unloading the respiratory muscles [6]. During this treatment, the use of an externally-applied PEEP allows to counteract the detrimental effects of PEEPi, leading to a reduction of the WOB, normalization of the pattern of breathing, improvement of blood gases, and reduction of patient-ventilator asynchronies[3, 7–9]. However, in order to be effective, PEEP needs to be tailored to match patient PEEPi. If the ventilator PEEP is lower than PEEPi, its positive effects are impaired. Conversely, if it exceeds PEEPi, it results in a further increase of the end expiratory lung volume and, therefore, increases patient's WOB. Excessive PEEP also may results in adverse effects on hemodynamics, as uselessly large continuous intrathoracic pressures may decrease venous return and cardiac output, depending upon intravascular volume status, myocardial function and other factors [9–11].
Corresponding author. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Via Giuseppe Colombo, 40, I-20133, Milano, Italy. E-mail address: [email protected] (R.L. Dellacà).
https://doi.org/10.1016/j.rmed.2019.06.022 Received 9 April 2019; Received in revised form 14 June 2019; Accepted 22 June 2019 Available online 23 June 2019 0954-6111/ © 2019 Elsevier Ltd. All rights reserved.
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
The clinical use of PEEP would need, therefore, an accurate tailoring for each individual patient and condition. Such tailoring would also require continuous adjustment, considering that EFLT markedly changes not only with time [12, 13], but also when changing body posture, such as when patient moves from the seated position to the supine or lateral ones. EFLT can be non-invasively detected in real time breath-by-breath using the forced oscillation technique (FOT) by measuring the intrabreath variations of respiratory reactance (ΔXrs) at 5Hz [14]. This method has proven high sensitivity and specificity when compared to the invasive Mead and Wittenberger [15] gold standard during quiet breathing and nasal CPAP [16] and it is in good agreement with the negative expiratory pressure (NEP) technique [17]. As this method is suitable for real time use during NIV, it potentially allows the implementation of real time ventilation strategies that continuously optimises PEEP by adjusting it to the lowest pressure able to abolish EFLT. We hypothesized that a novel NIV ventilation mode which continuously optimizes PEEP in order to abolish EFLT could identify an optimal PEEP for each patient and according to different postures. The aims of the present study were 1) to investigate the prevalence of EFLT and the PEEP values needed to abolish EFLT by the novel NIV approach on a group of stable hypercapnic COPD patients, and 2) to characterise the effects of changing posture from the seated to the supine position on both prevalence of EFLT and optimal PEEP.
connected to the patient's mask, and the ventilator was set to avoid PEEP settings below 4 cmH2O. The pressure support, i.e. the difference between the inspiratory pressure (inspiratory positive airway pressure, IPAP) and PEEP, is kept constant by the ventilator at the initial prescribed (baseline) value by continuously adjusting IPAP by the same changes applied to PEEP. Digital values of pressure (Pao), flow and volume at the airway opening, as well as Rrs and Xrs data, were sent by the ventilator to a laptop at a sampling rate of 100 Hz and recorded for subsequent analysis.
2. Material and methods
2.4. Data analysis
2.1. Subjects
Breathing pattern and respiratory input impedance data analysis: For each posture, the last minute of recorded data was considered, allowing the selection of 5–10 breaths free from artifacts for evaluation. From these breaths, breathing pattern parameters, average PEEP, inspiratory and expiratory resistance (Rrs) and reactance (Xrs) and the ΔXrs were computed, averaged and used for statistical comparison. Estimation of the sample size: Differences in ΔXrs from seated to supine posture in COPD patient is characterized by a standard deviation equal to 3.4 cmH2O*s/L with a matched difference within posture of 2.3 cmH2O*s/L [16]. To reject the null hypothesis (ΔXrs seated is equal to ΔXrs supine) with power equals to 0.90 at a significance level of 0.05, the sample size is 25 patients. Statistical analysis: All data were tested for normality (D'Agostiono & Pearson normality test). T-test was used to compare two groups and one-way ANOVA was used to compare three groups any time the dataset passed the normality test, otherwise Wilcoxon matched-pairs signed rank test was used. A logistic regression model was used to assess the probability of becoming EFLT moving from seated to supine position on the basis of spirometric parameters and body max index (BMI). Only differences/relationships with p < 0.05 were considered statistically significant. Data are reported as mean ± SD when normally distributed, and as median (IQR) otherwise.
2.3. Study protocol Patients were asked to sit on a reclining armchair and connected to the ventilator for delivering their prescribed ventilator settings. After a few minutes for allowing the stabilization of breathing pattern, the automatic tailoring system was enabled and the ventilator started adjusting PEEP iteratively in order to abolish EFLT. After 15 min of spontaneous breathing in automatic mode, patients were moved in the supine position and ventilated for further 15 min. For each posture, a patient was considered as tidal flow limited (EFLT) or not (NEFLT) whether the optimal PEEP identified by the system was ≥ 4 cmH2O, i.e. the minimum allowed PEEP setting. Patients that were eventually EFLT at lower PEEPs but abolished EFLT with PEEP = 4 cmH2O were therefore classified as NEFLT.
Moderate, severe or very severe hypercapnic COPD patients diagnosed by spirometry [18] who were habitual users of nocturnal NIV in stable state condition were eligible for this study. Inclusion criteria were: age below 85 years; ability to provide consent. Exclusion criteria: patients who were acutely ill, medically complicated or who are medically unstable, as determined by the investigator; patients suffering from a COPD exacerbation or who suffered and exacerbation within the past two months. Indication for NIV was based on presence of severe symptoms, clinical history of frequent exacerbations/hospitalizations due to acute hypercapnic respiratory failure and stable hypercapnia (> 54 mmHg) post the acute use of NIV. The study was approved by the Ethical Review Boards of ICS S. Maugeri (No. 897 CEC/2013). All subjects gave written informed consent before the beginning of the study. Patients were recruited from April 2015 to April 2017. 2.2. Measurements and set-up Patients were connected to a non-commercial version of a NIV ventilator (Synchrony, Philips-Respironics) by an unvented facial mask (AMARA, Philips-Respironics). The ventilator incorporates real-time FOT single–frequency within-breath measurement of the total respiratory impedance (Zrs, expressed as resistance, Rrs, and reactance, Xrs) at 5 Hz, and uses these measurements to automatically adjust PEEP in order to abolish EFLT. In details, for each single breath free from artifacts (automatically detected by empirical rules on impedance values), the ventilator computes the difference between the average Xrs measured during inspiration (Xinsp) and the one measured during expiration, namely ΔXrs [14], and it increases/decreases PEEP by 0.1 cmH2O whether ΔXrs is greater or smaller than the threshold of 2.8 cmH2Os/L for detecting EFLT [14,16]. This procedure allows for delivery of the minimum PEEP value able to abolish EFLT and continuously adjusts this value to changes in patient respiratory mechanics. As for all single-limb ventilators, in order to avoid CO2 rebreathing an intentional leak (whisper swivel, Philips Respironics, US) was
3. Results A total of twenty-six patients were studied. Table 1 summarizes their anthropometric and spirometric characteristics at baseline. Fig. 1 reports the time course for one representative patient of the pressure at the mask and ΔXrs during the trial. In the seated position, the ventilator decreased PEEP from the prescribed value of 7 cmH2O to the minimum value of 4 cmH2O, as the patient was not showing EFLT in this posture. When the patient moved to supine, PEEP was gradually increased by the ventilator to approximatively 10 cmH2O in order to abolish the EFLT induced by change in posture. Fig. 2 represents the optimal automatically-adjusted PEEP (PEEPO) required to abolish EFLT in each posture for all patients. Seven patients were EFLT both in the seated and the supine 14
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
Table 1 Main anthropometric and baseline lung function data. BMI: body mass index; FEV1: forced expiratory volume in the first second in liters and as percentage of the predicted value; FEV1/FVC%: ratio of FEV1 on forced vital capacity expressed as the percentage of the predicted value; TLC: total lung capacity, expressed in liters and as percentage of the predicted value. Numbers of patients classified according to GOLD stages are also reported.
Sex (F: M) Age (yrs) BMI (kg/m2) FEV1 (L) FEV1% pred FEV1/FVC % TLC (L) TLC % pred GOLD 1: GOLD 2: GOLD 3: GOLD 4:
ALL (n = 26)
NEFLT (n = 8)
EFLT (n = 7)
EFLSUPINE (n = 11)
F = 7: M = 19 72.4 ± 6.8 30.5 ± 6.3 1.40 ± 0.60 39.2 ± 16.1 46.27 ± 16.33 6.28 ± 1.23 109.33 ± 21.73 0 8 8 10
F = 2: M = 6 73.4 ± 6.3 30.1 ± 5.4 1.58 ± 0.55 43.0 ± 18.5 47.50 ± 14.26 6.30 ± 1.18 109.60 ± 28.19 0 3 3 2
F = 1: M = 6 70.0 ± 8.4 31.1 ± 5.7 1.16 ± 0.46 36.2 ± 12.9 44.00 ± 18.20 6.04 ± 1.08 112.80 ± 27.25 0 1 3 3
F = 4: M = 7 72.0 ± 6.4 31.6 ± 6.7 1.48 ± 0.66 39.9 ± 16.4 48.27 ± 17.80 6.35 ± 1.48 107.2 ± 18.63 0 4 2 4
positions, eight patients never showed EFLT and eleven developed EFLT only in the supine position (EFLT-SUP). No significant differences were found in FEV1, GOLD grades, FEV1/FVC, or TLC between the groups. As the number of NEFLT patients reduced by 43% when moving from seated to supine (from 19 to 8), the median (IQR) PEEPO increased significantly (p < 0.0001) from 4 (0.03) cmH2O in seated to 6.10 (6.13) cmH2O in the supine position. If we consider only patients EFLT when seated, the average PEEPo increased after moving to the supine position from 5.99 ± 1.77 cmH2O to 11.54 ± 5.10 (p = 0.057). Fig. 3 shows the relationship of PEEPo vs the clinically prescribed PEEP for each patient in the supine position. Interestingly, even if the average values for the tailored and for the prescribed PEEP was very similar (7.42 ± 2.10 vs 7.45 ± 3.78 cmH2O, respectively), they were very different considering each individual patient. BMI and PEEPO had no statistically significant relationship, despite a slightly positive trend can be observed in Fig. 4a. Forced expiratory volume in 1s expressed as percentage of predicted value (FEV1%) and FEV1/FVC% did not show significant correlation with PEEPO (Fig. 4). Similarly, spirometric parameters and BMI were non-significant predictors of changing EFLT status from seated to supine position, according to logistic regression output. Similar lack of linear correlation was found after removing patients not exhibiting EFLT in the seated position. Table 2 reports respiratory system parameters and breathing pattern. Inspiratory reactance, Xinsp, showed a non-statistically significant tendency to be lower in NEFLT compared to EFLT in both postures.
Fig. 1. Pressure at the airway opening (Pao) and ΔXrs tracings recorded during the trial for a representative patient.
Fig. 3. Clinically prescribed values of PEEP vs PEEPo. Open circles: patients who are never flow limited (NEFLT); closed circles: patients who are flow limited both in the seated and supine positions (EFLT); closed triangles: patients flow limited only in supine position (EFLT-SUP). Regression line is reported together with r2 and p value. Means and standard deviations are reported for both prescribed PEEP and PEEPO on the horizontal and vertical axis, respectively.
Fig. 2. Individual values of PEEP required to abolish EFLT (PEEPO) in the seated and the supine positions. 15
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
Fig. 4. Linear correlation of a) BMI, b) FEV1% and c) FEV1/FVC% vs. the automatically tailored PEEP (PEEPO). Regression lines are reported in each panel with correspondent r2 and p values.
4. Discussion
were asked to move from the seated to the supine position. Eleven patients had a sudden increase in ΔXrs after they changed from sitting to supine position. This response to a posture change is likely the consequence of the lung volume reduction promoted by the supine position, where the different direction of the gravitational force acting on the abdominal contents promotes a cranial shift of the diaphragm [21]. In the eight patients that did not develop EFLT in supine position, ΔXrs increased from 0.53 (2.15) to 1.84 (1.27) cmH2O*s/L, suggesting that, even if EFLT did not develop in those patients, some choke points started to appear in the bronchial tree during expiration [19]. In all patients who became EFLT moving to supine, the ventilator reacted by increasing the PEEP breath by breath until a value able to abolish EFLT was reached. It is noteworthy that EFLT was abolished in all patients simply by increasing PEEP, i.e. without changing any other ventilator parameter, within 10.1 (5.3) min from the posture changes. PEEPo showed greater range of values compared to commonly clinical prescribed ones. Interestingly, no significant correlations were found between spirometric variables and PEEPo in the seated or supine positions, suggesting that the degree of obstruction per se it is not a major determinant of EFLT. This is also in line with previous studies that showed that EFLT is present in COPD over a wide range of FEV1%, even if it is more common in patients with more severe airway obstructions [22]. Similarly, multiple linear regression analysis did not find any combination of parameters able to predict PEEPo, suggesting that the pressure needed to abolish EFLT is the result of a very complex interaction between anthropometric variables and the functional alteration due to the disease. Our results are in line with those of a previous study where it was found that, despite the expertise of the healthcare team, the clinical setting based only on patients comfort and arterial blood gasses response resulted in differences when compared to a physiological setting
In this study we introduce a new NIV setting mode that automatically and continuously detects EFLT breath-by-breath using FOT and adjusts PEEP accordingly in order to deliver the lowest PEEP able to abolish EFLT. This device has been applied on a population of moderate to very severe COPD patients in order to characterise the prevalence of EFLT for each posture and to measure PEEPo and how this value changes when patients move from the seated to the supine position. The main findings are: 1) the ventilator was always able to identify a PEEP value which abolished EFLT for all patients who showed EFLT in any posture (85% of patients); 2) the PEEPo was highly variable between patients, with a maximum PEEPo identified in this study of 15.71 cmH2O, a value greater than the commonly prescribed PEEPs for this group of patients; 3) when moving from the seated to the supine position, the prevalence of EFLT markedly increased (from 33% to 85%). Also PEEPo increased when moving to supine, with the amplitude of such increase showing high inter-individual variability; 4) PEEPo cannot be predicted by either BMI, spirometric parameters or a combination of those. We suggest that PEEPo represents an independent feature of patients’ conditions and, therefore, potentially identifies different phenotypes. EFLT is a relatively common condition in COPD, and several previous studies reported a prevalence ranging from 29.4% to 52.3% in a wide variety of patient conditions [17, 19, 20]. In our study population, we found the incidence of EFLT in the seated posture of 26,9%, slightly lower, but still comparable, with previous findings. This lower incidence may be the consequence of the need of applying, for technical reasons (see methods), a minimum PEEP of 4 cmH2O, that may be enough to abolish EFLT in some of the patients. The prevalence of EFLT increased significantly to 69% when patients
Table 2 Respiratory mechanics and breathing pattern. Data are reported by grouping patients according to their EFLT condition. ΔXrs: within breath variation of respiratory reactance; Xinsp: inspiratory reactance; Rinsp: inspiratory resistance; Vt: tidal volume; RR: respiratory rate, Ti/Ttot: respiratory duty cycle; VE: minute ventilation.
ALL Seated Supine NEFLT Seated Supine EFLT Seated Supine EFLT-SUP Seated Supine
ΔXrs (cmH2O*s/L)
Xinsp (cmH2O*s/L)
Rinsp (cmH2O*s/L)
Vt (L)
RR (Breaths/min)
Ti/Ttot
VE (L/min)
1.99(2.25) 2.73(1.11)
−3.49(1.93) −5.17(3.53)
7.54(4.37) 8.46(4.46)
0.85(0.51) 0.74(0.52)
18.07(8.14) 17.03(8.16)
0.33(0.07) 0.31(0.06)
12.91(5.35) 10.83(4.15)
0.53(2.15) 1.84(1.27)
−3.41(2.26) −3.51(1.68)
5.18(0.75) 5.11(2.39)
0.83(0.34) 0.59(0.43)
22.25(4.98) 17.51(7.54)
0.35(0.05) 0.34(0.09)
14.59(8.55) 11.19(5.50)
2.75(0.63) 2.95(0.74)
−6.71(2.85) −6.58(2.02)
7.97(2.41) 8.81(1.66)
0.72(0.47) 0.70(0.28)
16.04(7.50) 15.54(3.50)
0.32(0.06) 0.29(0.03)
10.71(3.82) 10.50(4.71)
1.81(1.36) 2.97(0.87)
−3.15(1.59) −4.38(1.19)
8.63(2.30) 9.20(2.69)
1.09(0.63) 1.00(0.52)
14.49(5.88) 12.79(7.20)
0.33(0.04) 0.30(0.05)
12.04(4.00) 10.72(2.89)
16
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al.
oriented by assessing PEEPi using esophageal pressure in almost 50% of the patients [23]. These discrepancies were associated to an increased presence of ineffective efforts, suggesting that an empirical setting may be not appropriate for a generalized use. The same study also reported that, in line with our results, despite the PEEP determined by the clinical optimization approach was often different to the one determined by esophageal manometry, their average values were very similar [23]. Another study reported that, when ventilator PEEP is tailored to patient's respiratory mechanics by esophageal manometry, the inspiratory muscle workload due to PEEPi is reduced by NIV at a greater extent compared to clinical settings [24]. Even if the use of esophageal manometry has a strong rationale, the invasiveness and the complexity of the method allowed its use only for small physiological studies. Moreover, this approach allows the identification of the optimal PEEP at a specific time point, requiring delivery of the same PEEP until a further assessment is performed, impeding a prompt response to changes in patients’ respiratory mechanics. On the contrary, our method allows for the first time a continuous PEEP titration aimed to provide the lowest PEEP able to abolish EFLT during ventilation. This generates a new form of interaction between the ventilator and the patient: when the ventilator optimizes PEEP, the neural respiratory control of the patient reacts to any change of the mechanical operating conditions of the respiratory system by adjusting the spontaneous breathing pattern. Changes in breathing pattern, in turn, impacts also on the presence of EFLT, leading to a continuous patient-ventilator interaction lead by the neurological control system which cannot be obtained when single occasional adjustments of ventilator parameters are performed.
Declaration of interests The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments We acknowledge Cristina Zanoni 1 e Luca Gitti1 for their support in performing the measurements, we also acknowledge Bob Romano2, Jim McKenzie2 and Peter Hill2 for their technical support and Mar Janna Dahl3 for language editing and proofreading. 1 Pneumologia Riabilitativa, ICS S. Maugeri IRCCS, Lumezzane (Brescia), Italy; 2 Philips Respironics. Pittsburgh (PA), USA; 3 University of Utah (UT), USA and University of Western Australia (WA), Australia. List of abbreviations BMI COPD CPAP DH EFLT EFLT-SUP FEV1 FVC IPAP IQR PEEP PEEPi PEEPo
5. Conclusion In this study we found that a new ventilation modality that uses FOT for continuously tailoring PEEP to the lowest pressure able to abolish EFLT was able to identify, in all patients, an optimal PEEP. This optimal PEEP was very variable among patients and independent from the prescribed values. Moreover, it could not be predicted by spirometric nor anthropometric variables. For these reasons, we suggest that PEEPo may be used as a phenotyping variable for COPD. We believe that this automatic, continuous optimization of PEEP may empower the performance of the neural respiratory control of the patient by minimizing the mechanical load and providing, therefore, a continuously individualized ventilator support opening new scenarios in the management of COPD patients. Future studies are warranted to evaluate whether this personalized NIV ventilation mode has impacts on arterial blood gases, patient comfort, adherence, exacerbations rate and sleep quality.
FOT NEP NEFLT NIV Rrs WOB Xrs Xinsp Zrs ΔXrs
Body Mass Index Chronic Obstructive Pulmonary Disease Continuous Positive Airway Pressure Dynamic Hyperinflation tidal Expiratory Flow Limitation EFLT only in the supine position Forced Expiratory Volume in the 1st second Forced Vital Capacity inspiratory positive airway pressure Inter-Quartile Range Positive End Expiratory Pressure intrinsic Positive End Expiratory Pressure automatically-adjusted optimal Positive End Expiratory Pressure Forced Oscillation Technique Negative Expiratory Pressure Non Tidal Expiratory Flow Limited Non-Invasive Ventilation Total respiratory system input resistance Work Of Breathing Total respiratory system input reactance Xrs measured during inspiration Total respiratory system input impedance difference between the average Xrs measured during inspiration and the one measured during expiration
References [1] B.R. Celli, W. MacNee, A. ATS/ERS Task Force, et al., Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper, Eur. Respir. J. 23 (2004) 932–946. [2] B.P. Krieger, Hyperinflation and intrinsic positive end-expiratory pressure: less room to breathe, Respiration 77 (2009) 344–350. [3] L. Appendini, A. Purro, A. Patessio, et al., Partitioning of inspiratory muscle workload and pressure assistance in ventilator-dependent COPD patients, Am. J. Respir. Crit. Care Med. 154 (1996) 1301–1309. [4] M.L. Coussa, C. Guérin, N.T. Eissa, et al., Partitioning of work of breathing in mechanically ventilated COPD patients, J. Appl. Physiol. 75 (1993) 1711–1719. [5] S.V. Dawson, E.A. Elliott, Wave-speed limitation on expiratory flow-a unifying concept, J. Appl. Physiol. 43 (1977) 498–515. [6] N. Ambrosino, L. Xie, The use of non-invasive ventilation during exercise training in COPD patients, COPD J. Chronic Obstr. Pulm. Dis. 14 (2017) 396–400. [7] M.W. Elliott, D.A. Mulvey, J. Moxham, et al., Inspiratory muscle effort during nasal intermittent positive pressure ventilation in patients with chronic obstructive airways disease, Anaesthesia 48 (1993) 8–13. [8] S. Nava, C. Bruschi, C. Fracchia, et al., Patient-ventilator interaction and inspiratory effort during pressure support ventilation in patients with different pathologies, Eur. Respir. J. 10 (1997) 177–183. [9] F. Baigorri, A. de Monte, L. Blanch, et al., Hemodynamic responses to external counterbalancing of auto-positive end-expiratory pressure in mechanically ventilated patients with chronic obstructive pulmonary disease, Crit. Care Med. 22
Sources of support and conflicts of interest Politecnico di Milano University, Institution of IM, SC and RLD, owns a patent on the use of FOT for the detection of expiratory flow limitation licensed to Philips Respironics. Istituti Clinici Scientifici Maugeri received a free loan of part of the equipment used for this study and a research grant from Philips. Authors’ contribution I.M. participated in the conception and design of the study, data collection, data analysis, interpretation of the data and preparation of the manuscript; R.P. and L.B. participated in the data collection and analysis and preparation of the manuscript; S.C. participated in the data collection and data analysis; M.V. and R.L.D. participated in the conception and design of the study, data analysis, interpretation of the data and preparation of the manuscript. 17
Respiratory Medicine 155 (2019) 13–18
I. Milesi, et al. (1994) 1782–1791. [10] N. Ambrosino, S. Nava, A. Torbicki, et al., Haemodynamic effects of pressure support and PEEP ventilation by nasal route in patients with stable chronic obstructive pulmonary disease, Thorax 48 (1993) 523–528. [11] V.M. Ranieri, M. Dambrosio, N. Brienza, Intrinsic PEEP and cardiopulmonary interaction in patients with COPD and acute ventilatory failure, Eur. Respir. J. 9 (1996) 1283–1292. [12] E. Boni, L. Corda, D. Franchini, et al., Volume effect and exertional dyspnoea after bronchodilator in patients with COPD with and without expiratory flow limitation at rest, Thorax 57 (2002) 528–532. [13] H. Patel, K.L. Yang, Variability of intrinsic positive end-expiratory pressure in patients receiving mechanical ventilation, Crit. Care Med. 23 (1995) 1074–1079. [14] R.L. Dellacà, P. Santus, A. Aliverti, et al., Detection of expiratory flow limitation in COPD using the forced oscillation technique, Eur. Respir. J. 23 (2004) 232–240. [15] J. Mead, J.L. Whittenberger, Physical properties of human lungs measured during spontaneous respiration, J. Appl. Physiol. 5 (1953) 779–796. [16] R.L. Dellacà, M. Rotger, A. Aliverti, et al., Noninvasive detection of expiratory flow limitation in COPD patients during nasal CPAP, Eur. Respir. J. 5 (2006) 983–991. [17] R.L. Dellacà, N. Duffy, P.P. Pompilio, et al., Expiratory flow limitation detected by forced oscillation and negative expiratory pressure, Eur. Respir. J. 29 (2007) 363–7418.
[18] GOLD, Global Strategy for the Diagnosis, Management and Prevention of COPD, Global Initiative for Chronic Obstructive Lung Disease (GOLD), 2017. [19] R.L. Dellacà, P.P. Pompilio, P.P. Walker, et al., Effect of bronchodilation on expiratory flow limitation and resting lung mechanics in COPD, Eur. Respir. J. 33 (2009) 1329–1337. [20] T. Akita, T. Shirai, K. Mori, et al., Association of the forced oscillation technique with negative expiratory pressure in COPD, Respir. Physiol. Neurobiol. 220 (2016) 62–68. [21] A. Aliverti, R. Dellacà, P. Pelosi, et al., Compartmental analysis of breathing in the supine and prone positions by optoelectronic plethysmography, Ann. Biomed. Eng. 29 (2001) 60–70. [22] B.B. Aarli, P.M. Calverley, R. Jensen, et al., The association of tidal EFL with exercise performance, exacerbations, and death in COPD, Int. J. Chronic Obstr. Pulm. Dis. 12 (2017) 2179–2188. [23] M. Vitacca, S. Nava, M. Confalonieri, et al., The appropriate setting of noninvasive pressure support ventilation in stable COPD patients, Chest 118 (2000) 1286–1293. [24] M.1 Vitacca, B. Lanini, S. Nava, L. Barbano, R. Portal, E. Clini, N. Ambrosino, Inspiratory muscle workload due to dynamic intrinsic PEEP in stable COPD patients: effects of two different settings of non-invasive pressure-support ventilation, Monaldi Arch. Chest Dis. 61 (2) (2004 Apr-Jun) 81–85.
18