Secondary Logo

Journal Logo

Original article

Right and left ventricular functions during proportional assist ventilation in patients ready for weaning from mechanical ventilation

Elansary, Mohamed Gamal Lotfy*; Mowafy, Hossam; Zakarya, Ahmed Yehia; Soliman, Randa Aly; Nassar, Yasser Sadek

Author Information
The Egyptian Journal of Critical Care Medicine: August 2018 - Volume 6 - Issue 2 - p 33-46
doi: 10.1016/j.ejccm.2018.09.001
  • Open

1. Introduction

New modes of mechanical ventilation have been designed in an attempt to improve the quality of patient-ventilator synchrony with fewer clinical interventions and thus decrease unnecessary human workload. No randomized controlled trial has clearly shown that one conventional ventilation mode leads to better outcomes than another mode, with the exception of PSV compared with synchronized intermittent mandatory ventilation (SIMV) to wean patients from mechanical ventilation [1].

2. Aim of the study

  • Our primary aim was to evaluate the changes in left & right ventricular functions during Proportional Assist Ventilation and other modes of ventilation (Pressure control – Assisted Control Ventilation, & Pressure Support Ventilation) and during spontaneous breathing.
  • Our secondary aim was to compare the hemodynamic impact of mechanical ventilation in patients with normal & impaired left ventricular systolic function.

3. Patients & methods

The study was a prospective observational study conducted on patients admitted to the Critical Care Medicine department of Cairo University in the time period between January 2015 and July 2016.

The protocol was approved by the ethical committee, and informed consent was obtained from the patients or their next of kin.

We included all adult patients admitted to the Critical Care Medicine department and required mechanical ventilation and fulfilled all the eligibility criteria for a spontaneous breathing trial (SBT) (improvement of the underlying cause of acute respiratory failure, body temperature < 39 °C, hemoglobin level > 7 G/dl, PaO2 > 60 mmHg, FIO2 < 40%, positive end expiratory pressure (PEEP) under or equal to 8 cm H2O, respiratory rate less than 35 breathes/minute, systolic arterial pressure >90 mmHg and <160 mmHg without need for vasoactive drugs, no sedation and a stable neurological status) [2] and passed the SBT successfully were recruited in our study.

We excluded patients who failed weaning within less than 24 h post extubation, patients with atrial fibrillation & patients on vasopressors or inotropic support.

We gathered 129 patients, 29 patients failed SBT were excluded from the study). A total of 100 patients who passed SBT successfully were included in our study.

  1. Patients were classified according to their ejection fraction (EF) into:
    • Impaired systolic function: patients with ejection fraction less than 50%
    • Normal systolic function: patients with ejection fraction 50% or above.

All patients were subjected to detailed history taking, Thorough clinical examination with calculation of body surface area using the Mosteller's formula [3]

  • Index of Contractility (ICoN) was used to measure hemodynamics
  • ICoN measurments:
  • Non invasive assessment of hemodynamic parameters was done using the ICoN device (Osypka Cardiotronic, Albert-Einstein strasse, Berlin, Germany).
    • Cardiac Index (CI): Volume of blood pumped per minute per BSA.
    • Stroke Volume Index (SI): Volume of blood pumped per heart beat per BSA.
    • Systemic Vascular Resistance (SVR/SVRI): afterload indicator.

3.1. Transthoracic Echocardiography

  • Echocardiography was done to all patients using Siemens Acuson X300 PE Ultrasound Machine (Siemens – Germany).
  • Standard 2D, M mode, pulsed and color Doppler using parasternal and apical views to measure dimensions and evaluate global and regional left ventricular function.
  • The following parameters were measured:
    1. Assessment of left ventricular systolic & diastolic function.
    2. Assessment of left ventricular ejection fraction was done using M-mode.method.
      • Assessment of Mitral inflow patterns
        • Measurments
        • Measurements of mitral inflow E/A ratio.[4]
      • Assessment of Tissue Doppler annular early and late diastolic velocities.
        • Meaurments
        • Measurements include the systolic (S), DTI on the lateral mitral annulus and calculating the ratio of the mitral inflow E velocity to tissue Doppler E′ (E/E′) which plays an important role in the estimation of LV filling pressures.[5]
    3. Assessment of Right ventricular systolic & diastolic function
      1. TAPSE or Tricuspid annular Motion.
      2. TAPSE was obtained by placing M-mode cursor along the lateral tricuspid annulus and measuring the amount of longitudinal movement during peak systole [6].
      3. Doppler Tissue Imaging
      4. we assessed the tricuspid annulus & RV basal free wall by tissue Doppler imaging to measure the longitudinal velocity of excursion. This velocity has been termed systolic excursion velocity or RV S'. To obtain this velocity, apical 4 chamber view was used with tissue doppler highlighting the RV free wall with pulsed wave sample volume at tricuspid annulus or middle of the basal segment of RV free wall [7].
      5. Measurement of RV diastolic function
      6. From the apical four chamber view, the Doppler beam was aligned to the tricuspid inflow. The sample volume was placed between the tips of the tricuspid leaflets to obtain the transtricuspid flow velocities E, A, E/A [8].
        Also the Doppler tissue imaging velocities of the lateral tricuspid annulus and the tricuspid E/E' was computed.
      7. Measurment of Pulmonary artery systolic pressure (PASP)

3.2. Mechanical ventilation & weaning protocol

All patients with acute respiratory failure in our study were mechanically ventilated using the Puritan Bennet 840 ventilator (Medtronic C0vidien – USA).

3.2.1. The study protocol consisted of 4 phases:

  • Phase 1: initially all patients were ventilated using assisted control ventilation, pressure control mode, using preset airway pressure that is sufficient to maintain a tidal volume of 4–6 ml/kg & preset respiratory rate that is sufficient for CO2 elimination in arterial blood gases, PEEP 5, FiO2: 40–50%.
  • Phase 2: the patients were then shifted to proportional assisted ventilation + Automated tubal compensation with pressure assist 20–30%, PEEP 5 & FiO2 40–50% and were kept on this mode for 2 h.
  • Phase 3: The patients were then shifted to pressure support ventilation, with pressure support 10, PEEP 5 & FiO2 40–50% and were kept on this mode for 2 h.
  • Phase 4: finally the patients were disconnected from mechanical ventilation to spontaneous breathing with oxygen mask 5–10 L/min.

Echocardiographic and ICON measurements were taken after two hours of stabilization on each ventilatory phase.

3.2.2. Statistical methods

Data were coded and entered using the statistical package SPSS (Statistical Package for the Social Sciences) version 23. Data was summarized using mean, standard deviation, median, minimum and maximum in quantitative data and using frequency (count) and relative frequency (percentage) for categorical data. Comparisons between groups were done using unpaired t test [9]. For comparison of serial measurements within each patient repeated measures ANOVA was used [10]. For comparing categorical data, Chi square (χ2) test was performed. Exact test was used instead when the expected frequency is less than 5 [11]. P-values less than 0.05 were considered as statistically significant.

4. Results

The study was conducted between January 2015 to June 2016, it included 100 patients, 61 were females and 39 were males, the patients were classified into 2 groups, Normal left ventricular systolic function group (72 patients) & Heart failure group (28 patients).

4.1. Distribution of the studied cases regarding the general Demographic data & Comorbid conditions (Table 1)

Table 1
Table 1:
Showing distribution of the studied group regarding demographic data & Comorbid conditions.

The majority of the study group were females representing 61% of the study population. The mean age of the studied population was 55.12 and the mean BSA of the studied population was 1.92 m2. Hypertension was the most common comorbidity in the studies population (41%) followed by Diabetes meillitus (28%), Ischemic heart disease (29%), Heart failure (28%). Bronchial asthma, interstitial pulmonary fibrosis were less frequently encountered (Table 1).

4.2. Distribution of the studied cases regarding the cause of respiratory failure (Table 2)

Table 2
Table 2:
Showing the distribution of the studied cases regarding the cause of respiratory failure.

Pneumonia was the most frequent cause of acute respiratory failure (56%), followed by ARDS (12%) & acute pulmonary edema (8%) (Table 2).

4.3. Comparison between the hemodynamic parameters during different ventilation phases

4.3.1. Comparing the changes in SBP & DBP during different ventilation phases

In Total population, mean ± SD for SBP was (128.1 ± 15.4 Vs 129 ± 16.3 Vs 127.8 ± 15.1 Vs 128.7 ± 16.4) for PC-Acv, PAV+, PSV & spontaneous breathing respectively with no statistical significance (P 0.715).

Also, In Total population, mean ± SD for DBP was (76.8 ± 8.8 Vs 76.7 ± 9.4 Vs 76.5 ± 10.2 Vs 77.4 ± 8.2) for PC-Acv, PAV+, PSV & spontaneous breathing respectively with no statistical significance (P 0.43) (see Fig. 1).

Fig. 1.
Fig. 1.:
Changes in SBP & DBP in total population. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, SBP: systolic Blood Pressure, DBP: Diastolic Blood Pressure, P: P-value.

In patients with Normal systolic function, mean ± SD for SBP was (128.1 ± 15.5 Vs 129 ± 16.3 Vs 127.3 ± 14.8 Vs 127.9 ± 16) for PC-ACV, PAV+, PSV & spontaneous breathing respectively with no statistical significance (0.758).

In patients with Normal systolic function, mean ± SD for DBP was (74.6 ± 7.9 Vs 75.4 ± 8.8 Vs 75.1 ± 10.7 Vs 77.2 ± 8.6) for PC-ACV, PAV+, PSV & spontaneous breathing respectively with no statistical significance (P 0.54) (see Fig. 2).

Fig. 2.
Fig. 2.:
Changes in SBP & DBP in patients with Normal systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, SBP: systolic Blood Pressure, DBP: Diastolic Blood Pressure, P: P-value.

In patients with impaired systolic function, mean ± SD for SBP was (127.9 ± 15.5 Vs 128.9 ± 16.6 Vs 128.9 ± 15.9 Vs 130.5 ± 17.5) for PC-ACV, PAV+, PSV & spontaneous breathing respectively with no statistical significance (P 031).

In patients with impaired systolic function, mean ± SD for DBP was (82.1 ± 8.8 Vs 80 ± 10.2 Vs 80 ± 7.7 Vs 77.9 ± 7.4) for PC-ACV, PAV+, PSV & spontaneous breathing respectively with no statistical significance (P 031) (see Fig. 3).

Fig. 3.
Fig. 3.:
Changes in SBP & DBP in patients with impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, SBP: systolic Blood Pressure, DBP: Diastolic Blood Pressure, P: P-value.

4.3.2. Comparing the changes in heart rate during different ventilation phases

In total population, Mean ± SD for HR was (91.9 ± 16.8 Vs 92 ± 16.68 Vs 91.8 ± 15.9 Vs 95.8 ± 14.6) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, HR was significantly higher during spontaneous breathing (P < 0.001) (see Fig. 4).

Fig. 4.
Fig. 4.:
Changes in heart rate during PC-ACV, PAV+, PSV & spontaneous breathing in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, HR: Heart Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for HR was (91.9 ± 18.4 Vs 92.2 ± 18.1 Vs 91.2 ± 17.4 Vs 95.5 ± 15.3) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, HR was significantly higher during spontaneous breathing (P 0.001).

In patients with impaired systolic function, Mean ± SD for HR was (91.9 ± 12.1 Vs 91.4 ± 12.7 Vs 93.3 ± 11.2 Vs 96.4 ± 13.1) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, HR was significantly higher during spontaneous breathing (P 0.015).

4.3.3. Comparing the changes in respiratory rate during different ventilation phases

In Total population, Mean ± SD for RR was (17.9 ± 3 Vs 25.1 ± 5.8 Vs 20.8 ± 4.6 Vs 24.7 ± 3.9) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, RR was higher during PAV+ & spontaneous breathing (P < 0.001) (see Fig. 5).

Fig. 5.
Fig. 5.:
Changes in respiratory rate during PC-ACV, PAV+, PSV & spontaneous breathing in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for RR was (17.69 ± 3 Vs 24.6 ± 6.1 Vs 20.4 ± 4.7 Vs 24.6 ± 3.6) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, RR was higher during PAV+ & spontaneous breathing (P < 0.001).

In patients with impaired systolic function, Mean ± SD for RR was (18.7 ± 2.9 Vs 26.3 ± 4.7 Vs 21.7 ± 4.1 Vs 24.7 ± 4.5) for PC-ACV, PAV+, PSV & spontaneous breathing respectively, RR was higher during PAV+ & spontaneous breathing (P < 0.001).

4.3.4. Comparing the changes in Tidal volume during different ventilation phases

In Total population, Mean ± SD for Tidal volume was (526 ± 85 Vs 337 ± 65 Vs 431 ± 67) for PC-ACV, PAV+ & PSV respectively. Tidal volume was higher during PC-ACV (P < 0.001) (see Fig. 6).

Fig. 6.
Fig. 6.:
Changes in Tidal volume during PC-ACV, PAV+ & PSV in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control- Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, P: P-value.

In patients with Normal systolic function, Mean ± SD for Tidal volume was (530 ± 93 Vs 331 ± 69 Vs 428 ± 72) for PC-ACV, PAV+ & PSV respectively. Tidal volume was higher during PC-ACV (P 0.001).

In patients with impaired systolic function, Mean ± SD for Tidal volume was (515 ± 61 Vs 354 ± 54 Vs 438 ± 54) for PC-ACV, PAV+ & PSV respectively. Tidal volume was higher during PC-ACV (P < 0.001).

4.3.5. Comparing the changes in Minute ventilation during PCdifferent ventilation phases

In Total population, Mean ± SD for Minute ventilation was (9.3 ± 1.7 Vs 8.5 ± 1.5 Vs 8.9 ± 2) for PC-ACV, PAV+ & PSV respectively. Minute ventilation was higher during PC-ACV (P < 0.001) (see Fig. 7).

Fig. 7.
Fig. 7.:
Changes in Minute ventilation during PC-ACV, PAV+ & PSV in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for Minute ventilation was (9.1 ± 1.8 Vs 8.2 ± 1.5 Vs 8.8 ± 2.2) for PC-ACV, PAV+ & PSV respectively. Minute ventilation was higher during PC-ACV (P 0.001).

In patients with impaired systolic function, Mean ± SD for Minute ventilation was (9.7 ± 1.6 Vs 9.2 ± 1.6 Vs 9.5 ± 1.2) for PC-ACV, PAV+ & PSV respectively. Minute ventilation was higher during PC-ACV, yet there was no statistical significance (P 0.258).

4.3.6. ICON in different ventilation phases

In Total population, Mean ± SD for Stroke index was (43 ± 12 Vs 46 ± 15 Vs 46 ± 15 Vs 45 ± 14 ml/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. SI showed no significant changes with P value 0.222 (see Fig. 8).

Fig. 8.
Fig. 8.:
Changes in stroke volume index during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviation: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value, NS: Non significant.

In patients with Normal systolic function, Mean ± SD for Stroke index was (46 ± 14 Vs 44 ± 12 Vs 46 ± 15 Vs 45 ± 12 ml/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. SI showed no significant changes with P value 0.463.

In patients with impaired systolic function, Mean ± SD for Stroke index was (47 ± 19 Vs 43 ± 11 Vs 45 ± 15 Vs 46 ± 18 ml/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. SI showed no significant changes with P value 0.384 (see Fig. 9).

Fig. 9.
Fig. 9.:
Changes in cardiac output cardiac index different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value, NS: Non significant.

In Total population, Mean ± SD for Cardiac Index was (4.2 ± 1.2 Vs 4.2 ± 1.2 Vs 4.2 ± 1.3 Vs 4.2 ± 1.2 L/min/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. CI showed no significant changes with P value 0.969.

In patients with Normal systolic function, Mean ± SD for Cardiac Index was (4.2 ± 1.2 Vs 4.3 ± 1.3 Vs 4.2 ± 1.4 Vs 4.1 ± 1.1 L/min/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. CI showed no significant changes with P value 0.788.

In patients with impaired systolic function, Mean ± SD for Cardiac Index was (4.3 ± 1.2 Vs 4.1 ± 0.9 Vs 4.1 ± 1.1 Vs 4.4 ± 1.4 L/min/m2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. CI showed no significant changes with P value 0.457 (see Fig. 10).

Fig. 10.
Fig. 10.:
Changes in systemic vascular resistance during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value, NS: Non significant.

In Total population, Mean ± SD for SVR was (929 ± 342 Vs 909 ± 238 Vs 926 ± 390 Vs 943 ± 388) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. No statistically significant differences (P 0.725).

In patients with Normal systolic function, Mean Mean ± SD for SVR was (960 ± 389 Vs 940 ± 263 Vs 951 ± 442 Vs 1008 ± 427) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. No statistically significant differences (P 0.421).

In patients with impaired systolic function, Mean ± SD for SVR was (850 ± 158 Vs 828 ± 131 Vs 861 ± 192 Vs 776 ± 183) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. No statistically significant differences (P 0.088).

4.3.7. Echocardiography during different modes

4.3.7.1 Changes in left ventricular systolic function

In Total population, Mean ± SD for EF was (56.9% ± 12.7% Vs 57.5% ± 12.4% Vs 56.9% ± 12.8% Vs 56.2% ± 12.2%) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. EF was higher during PAV + mode (P 0.018).

In patients with Normal systolic function, Mean ± SD for EF was (63% ± 8.6% Vs 63.8% ± 7.8% Vs 63.1 ± 8.5% Vs 62.1 ± 7.9%) for PC-ACV, PAV+ & PSV respectively. EF was higher during PAV + mode (P 0.049).

In patients with impaired systolic function, Mean ± SD for EF was (41.3% ± 6.8% Vs 41.9% ± 6.6% Vs 41.2% ± 7.7% Vs 40.9% ± 7%) for PC-ACV, PAV+ & PSV respectively. EF was found higher during PAV + mode but no statistical significance (P 0.251) (see Fig. 11).

Fig. 11.
Fig. 11.:
Changes in LV ejection fraction during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In Total population, Mean ± SD for “S” wave was (0.12 ± 0.03 Vs 0.12 ± 0.02 Vs 0.12 ± 0.03 Vs 0.12 ± 0.02) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. “S” wave showed no statistical significance (P 0.595).

In patients with Normal systolic function, Mean ± SD for “S” wave was (0.13 ± 0.03 Vs 0.13 ± 0.03 Vs 0.12 ± 0.03 Vs 0.12 ± 0.02) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. “S” wave was higher during PC-ACV & PAV + mode (P 0.024).

In patients with Impaired systolic function, Mean ± SD for “S” wave was (0.1 ± 0.02 Vs 0.11 ± 0.02 Vs 0.11 ± 0.02 Vs 0.1 ± 0.02) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. “S” wave was higher during PAV & PSV modes (P 0.002) (see Fig. 12).

Fig. 12.
Fig. 12.:
Changes in Tissue doppler derived mitral S wave during different ventilation phases in total population, Normal systolic function group & impaired systolic function group. Abbreviations: PC-ACV: Pressure Control- Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.
4.3.7.2 Changes in left ventricular diastolic function

In Total population, Mean ± SD for E/A ratio was (1.18 ± 0.62 Vs 1.22 ± 0.53 Vs 1.26 ± 0.65 Vs 1.35 ± 0.74) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. E/A ratio was higher during spontaneous breathing (P 0.012) (see Fig. 13).

Fig. 13.
Fig. 13.:
Changes in mitral E/A ratio during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for E/A ratio was (1.22 ± 0.64 Vs 1.23 ± 0.45 Vs 1.26 ± 0.59 Vs 1.40 ± 0.72) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. E/A ratio was higher during spontaneous breathing (P 0.018).

In patients with Impaired systolic function, Mean ± SD for E/A ratio was (1.1 ± 0.54 Vs 1.21 ± 0.69 Vs 1.26 ± 0.79 Vs 1.23 ± 0.81) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. E/A ratio was higher during spontaneous breathing (P 0.014).

In Total population, Mean ± SD for E/E′ ratio was (6.2 ± 2.3 Vs 6.4 ± 2.3 Vs 6.6 ± 2.5 Vs 6.5 ± 2.1) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. E/E' ratio showed no significant difference (P 0.146).

In patients with Normal systolic function, Mean ± SD for E/A ratio was (5.9 ± 2.1 Vs 6.4 ± 2.5 Vs 6.4 ± 2.4 Vs 6.4 ± 2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. E/E' was lower during PC-ACV (P 0.037).

In patients with Impaired systolic function, Mean ± SD for E/A ratio was (7 ± 2.6 Vs 6.5 ± 1.7 Vs 7.1 ± 2.6 Vs 6.8 ± 2.2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively E/E' ratio showed no significant difference (P 0.166) (see Fig. 14).

Fig. 14.
Fig. 14.:
Changes in mitral E/E′ ratio during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.
4.3.7.3 Changes in right ventricular systolic function

In Total population, Mean ± SD for TAPSE was (22.1 ± 5.7 Vs 22.1 ± 4.4 Vs 21.4 ± 4.3 Vs 21.4 ± 4.4) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. TAPSE was higher during PC-ACV & PAV+ (P 0.004).

In patients with Normal systolic function, Mean ± SD for TAPSE was (22.7 ± 5.1 Vs 22.6 ± 4 Vs 21.9 ± 3.9 Vs 21.9 ± 3.9) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. TAPSE was higher during PC-ACV & PAV+ (P 0.008).

In patients with Impaired systolic function, Mean ± SD for TAPSE was (20.5 ± 6.9 Vs 20.9 ± 5.1 Vs 20.1 ± 5.3 Vs 20 ± 5.3) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. TAPSE showed no statistical significance (P 0.2) (see Fig. 15).

Fig. 15.
Fig. 15.:
Changes in TAPSE during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value, TAPSE: Tricuspid Annular Peak Systolic Excursion.

In Total population, Mean + SD for Rt “S” wave was (0.18 ± 0.04 Vs 0.19 ± 0.05 Vs 0.18 ± 0.04 Vs 0.18 ± 0.04) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Rt “S” wave was highest during PAV + mode but there was no statistical significance (P 0.369).

In patients with Normal systolic function, Mean ± SD for Rt “S” wave was (0.19 ± 0.04 Vs 0.2 ± 0.05 Vs 0.19 ± 0.05 Vs 0.19 ± 0.04) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Rt “S” wave was highest during PAV mode but there was no statistical significance (P 0.71).

In patients with Impaired systolic function, Mean ± SD for Rt “S” wave was (0.16 ± 0.04 Vs 0.17 ± 0.04 Vs 0.16 ± 0.04 Vs 0.16 ± 0.03) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Rt “S” wave was higher during PAV + mode (P 0.024) (see Fig. 16).

Fig. 16.
Fig. 16.:
Changes in Tissue doppler derived tricuspid S wave during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.
4.3.7.4 Changes in right ventricular diastolic function

In Total population, Mean ± SD for Tricuspid E/E′ ratio was (4.1 ± 1.4 Vs 3.9 ± 1.2 Vs 4 ± 1.6 Vs 4.3 ± 1.9) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Tricuspid E/E′ ratio was highest during spontaneous breathing (P 0.002) (see Fig. 17).

Fig. 17.
Fig. 17.:
Changes in tricuspid E/E' ratio during different ventilation phases in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for Tricuspid E/E′ ratio was (4.1 ± 1.3 Vs 4 ± 1.2 Vs 4.1 ± 1.4 Vs 4.3 ± 1.8) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Tricuspid E/E′ ratio was highest during spontaneous breathing (P 0.017).

In patients with Impaired systolic function, Mean ± SD for Tricuspid E/E′ ratio was (3.9 ± 1.5 Vs 3.7 ± 1.2 Vs 4 ± 2 Vs 4.2 ± 2.1) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. Tricuspid E/E′ ratio was highest during spontaneous breathing (P 0.041).

4.3.7.5 Changes in Pulmonary Artery systolic Pressure (PASP)

In Total population, Mean ± SD for PASP was (28.5 ± 13.5 Vs 28.6 ± 13.2 Vs 28.54 ± 13.6 Vs 28.5 ± 13.5) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. PASP showed no significant differences (P 0.668) (see Fig. 18).

Fig. 18.
Fig. 18.:
Changes in pulmonary artery systolic pressure during different ventilation in total population, Normal systolic function & impaired systolic function. Abbreviations: PC-ACV: Pressure Control-Assist control ventilation, PAV+: Proportional Assist Ventilation, PSV: Pressure Support Ventilation, RR: Respiratory Rate, P: P-value.

In patients with Normal systolic function, Mean ± SD for PASP was (25 ± 8.5 Vs 24.6 ± 7.2 Vs 24.5 ± 7.9 Vs 24.6 ± 7.6) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. PASP showed no significant differences (P 0.695).

In patients with Impaired systolic function, Mean ± SD for PASP was (37.7 ± 18.9 Vs 38.6 ± 18.8 Vs 38.3 ± 19.2 Vs 38.3 ± 19.2) for PC-ACV, PAV+, PSV & spontaneous breathing respectively. PASP was significantly lower during PC-ACV (P 0.005).

5. Discussion

Mechanical ventilation induces changes in lung volumes and intrathoracic pressures which are transmitted to the heart, great arteries and veins and therefore independently affect the key determinants of cardiovascular performance, atrial filling or preload, the impedance to ventricular emptying or afterload and consequently heart rate and myocardial contractility [12].

We hypothesized that new ventilatory modes with more patient synchrony may have better hemodynamic effects, PAV + differs from PSV in delivering ventilatory assistance based on the patient's respiratory effort [13].

5.1. Regarding hemodynamic data (SBP, DBP & HR)

We showed no statistically significant changes in systolic & diastolic blood pressures during different modes of ventilation in Normal systolic function & Impaired systolic function groups.

In agreement with our findings, no significant changes were found in SBP, DBP & MAP between PSV, PAV & PAV+ (PAV + Automated tubal compensation), in 12 patients without chronic heart or lung problems during SBT, MAP was (83.9 ± 15.3 Vs 81.5 ± 15.9 Vs 82.2 ± 14.8) in PSV, PAV & PAV + respectively. PSV, PAV & PAV + ATC were received in random order while maintaining mean inspiratory airway pressure constant as studied by Dirk Varelmann et al. (2005) [14].

We showed that HR was highest during spontaneous breathing (95.8 + 14.6) in comparison to the PC-ACV, PAV+ & PSV (91.9 + 16.8 Vs 92 + 16.68 Vs 91.8 + 15.9) in Normal EF group & Heart failure group (P < 0.001), But no significant changes in HR in-between PC-ACV, PAV+ & PSV.

Similar findings were reported in a study which showed a significant increase in HR & RR between high PSV 15–20 cmH2O & low PSV 5 cmH2O during the weaning of 42 surgical patients, in both successful and failure groups. It is to be noted that HR & RR increased significantly in the unsuccessful group in comparison to the successful group (106.2 ± 13.5 Vs 88.3 ± 5.4 and 31.3 ± 7.5 Vs 23.6 ± 6.6 respectively) P < 0.001 as reported by Vasilios E. Papaioannou et al. (2011) [15].

In studies comparing PSV & PAV+, no significant changes in HR were found between PSV, PAV & PAV+ in the Dirk Varelmann study (96.8 ± 19 VS 92.9 ± 19 VS 93.1 ± 20.5 respectively). When ventilatory demand was increased by adding dead space to ventilatory circuit, Work of breathing & HR increased significantly during PAV in comparison to PSV (from 93.1 ± 20.5 to 96.6 ± 15.3 in PAV + VS from 96.8 ± 19 to 96 ± 20.5 in PSV) [14]. It is claimed that dynamic inspiratory pressure assistance with PAV will better adapt the degree of ventilatory support to patients' actual demands than constant PSV [16]. In the Dirk Varelmann study, this was not observed and work of breathing & HR increased significantly with increasing ventilatory demands during PAV. The explanation of this findings maybe is their use of the Evita 4, Dräger ventilator which lacks the automated measurement of dynamic respiratory system resistance (Rrs) and elastance (Ers), and Rrs & Ers were determined during CMV, then were used as a surrogate for setting PAV. These measured Ers and Rrs do not necessarily reflect the values expected during SB. This limited approach explains the unexpected lower adaptation of dynamic inspiratory pressure assistance after artificial increase in ventilatory demand during PAV. In our study we used the Puritan Bennet 840 ventilator which has a software that automatically measures Rrs & Ers and adapts the inspiratory pressure assistance according to ventilatory demands. Similarly, No significant changes in HR were found between PSV & PAV+ (90.2 ± 11.6 Vs 88.7 ± 9.3 respectively) during the weaning of 60 COPD patients as reported by Anwar Elganady et al. (2014) [17].

5.2. Regarding breathing & ventilatory patterns (RR, VT & MV)

In our study, RR was found higher in PAV+ mode (25 ± 5.7) in comparison to ACV, PSV & spontaneous breathing (17.9 + 3 Vs 20.8 + 4.6 Vs 24.7 + 3.9 respectively) in Normal EF group & Impaired Systolic function group (P < 0.001).

Giannouli E et al. (1999) studied 14 patients to determine whether alterations in level of support (during PSV & PAV) results in different ventilatory, breathing pattern & gas exchange. The highest PSV support (PSVmax) was that associated with a tidal volume (VT) of 10 ml/kg and the highest level of PAV assist (PAVmax) was the one which maintained the same VT. Level of assist was decreased in steps to the lowest tolerable level (PSVmin, PAVmin). There was no difference between PSVmin and PAVmin in any of the variables. [18]. This study didn't consider standard ventilatory parameters for the whole study group, instead each patient received individual level of support during PSVmin & PAVmin which maintained their tidal volume. In our study, We used a standardized protocol for all patients during different phases, SBT with PSV (PS10 & PEEP 5) and during PAV+ (pressure assist 20–30% & PEEP 5).

In contrast to our findings, RR was lower during PAV 80% in comparison to PSV & PAV 50% during weaning of 13 COPD patients using PSV, PAV + 50% & PAV + 80% as reported by Wrigge et al. (1999) [19].

The disagreement can be attributed to the different methodology, as in our study we used PAV+ with much lower pressure assist 20–30% which requires a higher work of breathing, lower tidal volume and higher RR. On the other hand he used PAV 80% which showed lower work of breathing, Tidal volume & RR. Furthermore, all their patients were COPD who are different from our cohort of patients without COPD. Also in the Dirk Varelmann study, no significant changes in RR were found between PSV, PAV & PAV+ (17.6 ± 4.8 Vs 17 ± 5.6 Vs 17 ± 6.5 respectively) [14]. Different study designs should be noted beginning with the different ventilator types and the mode of interpretation of Rrs & Ers. In addition, they used PAV & PAV+ with pressure assist that compensates for 50% of Rrs & Ers. In our study we used PAV+ with pressure assist 20–30%.

In agreement with our findings, Eumorfia Kondili et al. (2006), compared the short term cardiorespiratory effects of PSV & PAV in 12 patients admitted with ARDS secondary to sepsis. With PAV, RR was slightly but significantly higher than the corresponding values with PSV (24.5 ± 6.9 vs. 21.4 ± 6.9 breaths/min respectively) [20].

In our study, The mean VT was lowest during PAV+ mode (337.7 ± 65.4) in comparison to PC-ACV & PSV (526 + 85 Vs 431 + 67 respectively).

Eumorfia Kondili also reported lower VT during PAV+ in comparison to PSV (7.7 ± 1.9 Vs 8.0 ± 1.6 respectively) while comparing PSV & PAV in 12 ARDS patients [20].

In contrast to our findings, There were no significant changes in RR or VT during weaning of 60 COPD patients using PAV & PSV (16.6 ± 1.63 & 530 ± 52 Vs 16 ± 1.61 & 539 ± 45 respectively) as reported by Anwar Elganady et al. (2014) [17]. The discrepancy can be explained by the different methodology, as they used PAV with much higher pressure assist 70% and all the patients in the study were COPD. In the present study we used PAV 20–30% & none of the patients was COPD.

Although it is believed that high RR signifies excessive work of breathing and inadequate assist level, studies have shown that respiratory rate is not a good predictor of work of breathing or the pressure–time product during assisted modes of support [21]. Particularly during PAV, high RR may not indicate distress, but it may represent the spontaneously selected pattern of breathing [22]. We clearly observed that despite the lower VT during PAV+ than that with PS, none of the patients exhibited clinical signs of excessive work of breathing.

5.3. Regarding hemodynamic variables of ICON device (SV, SI, CO, CI & SVR)

In the present study using the non invasive ICON device, We didn't find any statistically significant difference in SV, SI, CO & CI between the different modes of mechanical ventilation (PC-ACV, PAV+, PSV) and spontaneous breathing in both Normal EF group & Impaired systolic function group.

In agreement with our study, No significant change in CO & CI (calculated through pulmonary artery catheter) was found between CPAP 5cmH2O, CPAP 10cmH2O, Bilevel noninvasive positive-pressure ventilation & spontaneous breathing in 6 patients with acute pulmonary edema as reported by Karim Chadda et al. (2002) [23].

In contrast to our findings, Eumorfia Kondili et al. (2006), showed better hemodynamic profile during PAV in comparison to PSV in 12 ARDS patients. CI was significantly higher during PAV in comparison to PSV (4.4 ± 1.6 vs. 4.1 ± 1.3 L/min/m2, respectively) [20]. The different findings can be explained by the accuracy of the different methods used for evaluation of CI where Kondili used the pulmonary artery catheter thermodilution technique (Vigilance Monitor; Edwards Lifesciences, Irvine, CA) and in our study we used the noninvasive impedance cardiography ICON device.

Xu L et al. (2006), studied the effect of mechanical ventilation using BIPAP, PSV & proportionate pressure support PPS (similar to PAV mode in Drager ventilator) in 32 patients with normal and impaired cardiac function. Non Invasive Cardiac Output Monitor (NICOM) [a thoracic electrical impedance bioreactance velocimetry (Cheetah medical) similar to ICON (cardiotronics)] was used for evaluation of hemodynamics. CO & CI were significantly higher during PPS & PSV in normal cardiac function patients. And in impaired cardiac function patients CO & CI were highest during PPS [24]. Different study methodology can explain the contradictory results, where they used different ventilator machine which is the Drager Evita 4 with the absent technology of automatically and continuously monitoring of Ers & Rrs, Also they used the CI for classification of cardiac function and considered CI less than 2 L/min/m2 as impaired cardiac function. We used the Puritan Bennet 840 machine and we used EF% for classification of cardiac function which is the method used in previous studies.

Similarly, no differences were noted in CO, CI, SV, SI & SVR during weaning of intubated patients (including heart failure patients) at 3 reading points (baseline, during SBT with pressure support 5 and after extubation) as reported by Saad, Antonio et al. (2015) [25].

In the present study, there was no significant differences between systemic vascular resistance during different modes of mechanical ventilation and during spontaneous breathing.

In agreement with our findings Qin Gu et al. (2007), performed a study during the weaning of 12 critically ill patients. The patients were ventilated using BIPAP mode with RR 15-20 per min, then the patients were shifted to PSV with the same pressures for 20 min. There was no significant difference in HR, MAP, & SVR [26]. Similarly, the results shown by Saad, Antonio et al. (2015) revealed no significant changes in SVR during baseline, SBT & after extubation [25].

5.4. Regarding left ventricular function (EF, E/A & E/E′)

We found the mean LVEF to be higher during PAV+ (63.8% ± 7.8%) in comparison to PC-ACV, PSV & spontaneous breathing (63% ± 8.6% Vs 63.1 ± 8.5% Vs 62.1 ± 7.9% respectively) in the Normal systolic function group. In the impaired systolic function group, the mean LVEF was higher during PAV mode but without statistical significance.

L Vetrugno et al. (2012) studied 13 patients (7 with normal EF & 5 heart failure) with ARDS using transthoracic echocardiography during NIV with PSV & during spontaneous breathing. In heart failure patients, EF was higher during PSV (P < 0.05), while no change in EF was found in the normal EF group [27].

In the present study, we showed that there was a significant increase in mean E/A ratio during spontaneous breathing (1.35 ± 0.74) in comparison to PC-ACV, PAV+ & PSV (1.18 ± 0.62 Vs 1.22 ± 0.53 Vs 1.26 ± 0.65) P < 0.012 in Normal systolic function group & impaired systolic function group. Also mean E/E' ratio was increased during the transition to spontaneous breathing, but with no statistical significance. This signifies a better LV diastolic function during mechanical ventilation and that transition to spontaneous breathing was associated with elevated LV filling pressures.

Similar findings were shown by Hafid Ait-Oufella (2007), who studied 31 patients (including patients with normal EF & heart failure patients) with successful weaning from mechanical ventilation, echocardiography was performed before SBT and 30 min following SBT with T-tube, The E/A ratio increased from 0.91 (0.66–3.56) to 1.17 (0.5–4.76), (p = 0.01), after disconnection of mechanical ventilation [28]. In addition, Vincent Caille et al. (2010) studied 117 patients (58 with normal EF & 59 with heart failure) during weaning from mechanical ventilation, TTE was performed immediately before SBT with PSV and after 30 min of SBT. There was a statistically significant increase of E/A during the SBT. There was increase in E/E' during the SBT without reaching statistical significance [29].

Gerbaud E et al. (2012), studied 44 patients with heart failure during their weaning from mechanical ventilation. Plasma levels of NTproBNP and transthoracic echocardiography indices including cardiac index, E/A ratio and E/Ea ratio were recorded immediately before commencing and just before the end of SBT done with Pressure support 7 & PEEP 0. In patients who passed the weaning there was a significant increase in cardiac index with no significant change in E/A & E/E′. While the patients who failed weaning could not increase their cardiac index, their E/A & E/E′ significantly increased. Also NTproBNP increased significantly in patients who failed weaning [30]. Their study population presented with acute cardiac failure and depressed LVEF, E/A and E/E′ ratio [40.5 (34–50, 1.24 (0.83–2.21) & 9.5 (7–15.7) respectively]. These values are somewhat higher than the values obtained in our study population (41.9 ± 6.5, 1.1 ± 0.5, 6.5 ± 0.7 respectively). In addition, during SBT they used PSV with PS 7 and PEEP 0 which is different from our study with PS 10 & PEEP 5.

Khalaf Eldehily et al. (2016), conducted a study on 50 patients to test whether TAPSE, right ventricular (RV) systolic (S tricuspid) and diastolic (é tricuspid) tissue Doppler imaging (TDI) velocities have correlations with success and duration of weaning in mechanically ventilated patients with acute respiratory failure. Doppler echocardiographic study of right and left ventricular systolic and diastolic function was performed on the day of admission to the ICU on volume controlled ventilation and within 30 min of the 1st spontaneous breathing trial (SBT) on PS/CPAP 10/5 cm H2O.

In successfully weaned patients there was a rise in E/A & E/E′ ratio after liberation from mechanical ventilation (1.22 ± 0.5 & 5.31 ± 2.49 Vs 1.38 ± 0.47 & 6.18 ± 2.29 respectively) [31].

5.5. Regarding right ventricular function (TAPSE, “S” wave & E/E′)

In the present study, Mean TAPSE was higher during PC-ACV & PAV+ modes (22.7 + 5.1 Vs 22.6 + 4) in comparison to PSV & Spontaneous breathing (21.9 + 3.9 Vs 21.9 + 3.9) (P 0.008) in Normal EF group. TAPSE was found slightly higher during PAV mode in impaired systolic function group but didn't reach statistical significance. In addition, Tissue Doppler Tricuspid “S” wave was found higher during PAV (0.19 ± 0.05) in comparison to PC-ACV, PSV & spontaneous breathing (0.18 ± 0.04 Vs 0.18 ± 0.04 Vs 0.18 ± 0.04) in impaired systolic function group The higher mean values of TAPSE & Tricuspid “S” waves signifies improved RV systolic function during PAV mode.

On the other hand, Khalaf Eldehily et al. (2016), reported no significant change in mean TAPSE & Tricuspid “S' wave values between ACV & PSV in successful patients (2.48 ± 0.36 & 19.28 ± 6.93 Vs 2.55 ± 0.54 18.4 ± 2.65) [31]. Different results were shown in our study where mean TAPSE value was higher in PC-ACV than PSV (22.1 + 5.7 Vs 21.4 + 4.3). The contradiction can be attributed to different methodology as he used ACV with volume controlled ventilation while we used ACV with pressure control ventilation. All pressure modes are associated with a “decelerating” flow pattern during inspiration. This decelerating flow pattern represents the speed of the gas, which is initially very high but gradually lowers as the chest fills. This characteristic flow pattern is considered more physiologic than that associated with volume-based ventilation and may contribute to better gas distribution and altered hemodynamic response [31].

We showed that mean Tricuspid E/E′ ratio was higher during spontaneous breathing (4.3 ± 1.9) in comparison to PC-ACV, PAV+ &, PSV 4.1 ± 1.4 Vs 3.9 ± 1.2 Vs 4 ± 1.6 denoting elevation of RV filling pressures following liberation from mechanical ventilation (P 0.002).

6. Conclusion & recommendations

PAV+ may have favorable effects on hemodynamics, left & right ventricular functions in patients ready for weaning. PAV+ favourable effects were apparent in patients with normal or impaired left ventricular systolic functions.

7. Study limitations

  • The study was a single center trial on a limited number of patients and with a short follow up time only 48 h. In addition, Hemodynamic variables were performed using a Non invasive device which is less accurate than pulmonary artery catheter. Also, Echocardiography was done to all patients by single operator. Finally, the study excluded patients who failed SBT, so hemodynamic changes in weaning failure patients were not studied.

Conflict of interest

Authors states that there is no conflict of interest in this article.

References

[1] Brochard L, Rauss A, Benito S, Conti G, Mancebo J, Rekik N, Gasparetto A, Lemaire F. Comparison of three methods of gradual withdrawal from ventilatory support during weaning from mechanical ventilation. PubMed: 7921460.
[2] Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, et al. Weaning from mechanical ventilation. Eur Respir J 2007;29:1033-1056. doi: https://doi.org/10.1183/09031936.00010206.
[3] Mosteller RD. Simplified calculation of body-surface area. N Engl J Med 1987;317:1098.
[4] Appleton CP, Jensen JL, Hatle LK, Oh JK. Doppler evaluation of left and right ventricular diastolic function: a technical guide for obtaining optimal flow velocity recordings. J Am Soc Echocardiogr 1997;10:271-292. doi: https://doi.org/10.1016/S0894-7317(97)70063-4.
[5] De Sutter J, De Backer J, Van de Veire N, Velghe A, De Buyzere M, Gillebert TC. Effects of age, gender, and left ventricular mass on septal mitral annulus velocity (E') and the ratio of transmitral early peak velocity to E' (E/E'). Am J Cardiol 2005;95:1020-1023.
[6] Tamborinin G, Pepi M, Galli CA, Maltagliati A, Celesti F, Muratori M, et al. Feasibility and accuracy of a routine echocardiographic assessment of right ventricular function. Int J Cardiol 2007;115:86-89.
[7] Lindqvist P, Walendstrom A, Henien M, Morner S, Kazzam E. Regional and global right ventricular function in healthy individuals aged 20-90 years: a pulsed Doppler tissue imaging study: Umea general population heart study. Echocardiography 2005;22:305-314.
[8] Berman GO, Reichek N, Brownson D, Douglas PS. Effect of sample volume location, imaging view, heart rate and age on tricuspid velocimetry in normal subjects. Am J Cardiol 1990;65:1026-1030.
[9] Chan YH. Biostatistics102: quantitative data - parametric & non-parametric tests. Singapore Med J 2003;44(8):391-396.
[10] Chan YH. Biostatistics 103: qualitative data - tests of independence. Singapore Med J 2003;44(10):498-503.
[11] Chan YH. Biostatistics301: repeated measurement analysis. Singapore Med J 2004;45(8):354-369.
[12] Wise RA, Robotham JL, Summer WR. Effects of spontaneous ventilation on the circulation [review]. Lung 1981;159:175-186.
[13] Grasso S, et al. Am J Respir. Crit Care Med 2000;161(3 Pt 1):819-826.
[14] Varelmann Dirk, Wrigge Hermann, Zinserling Jörg, Muders Thomas, Hering Rudolf, Putensen Christian. Crit Care Med 2005;33:1968-1975.
[15] Papaioannou, Vasilios E. et al.Changes of heart and respiratory rate dynamics during weaning from mechanical ventilation: A study of physiologic complexity in surgical critically ill patients. J Crit Care, 26(3), 262-272.
[16] Mols G, Rohr E, Benzing A, et al. Breathing pattern associated with respiratory comfort during automatic tube compensation and pressure support ventilation in normal subjects. Acta Anaesthesiol Scand 2000;44:223-230.
[17] Anwar Ahmed Elganady, Bassem Nashaat Beshey, Amr Abdelaziz Hassan Abdelaziz. Proportional assist ventilation versus pressure support ventilation in the weaning of patients with acute exacerbation of chronic obstructive pulmonary disease Original. http://dx.doi.org/10.1016/j.ejcdt.2014.04.001.
[18] Giannouli E, Webster K, Roberts D, Younes M. Am J Respir Crit Care Med 1999;159(6):1716-1725. doi: https://doi.org/10.1164/ajrccm.159.6.9704025.
[19] Wrigge H, Golisch W, Zinserling J, Sydow M, Almeling G, Burchardi H. Proportional assist versus pressure support ventilation: effects on breathing pattern and respiratory work of patients with chronic obstructive pulmonary disease. Intensive Care Med 1999;25:790-798.
[20] Eumorfia Kondili, Nectaria Xirouchaki, Katerina Vaporidi, Maria Klimathianaki, Dimitris Georgopoulos, Short-term cardiorespiratory effects of proportional assist and pressure-support ventilation in patients with acute lung injury/acute respiratory distress syndrome.
[21] Banner MJ, Kirby RR, Kirton OC, DeHaven CB, Blanch PB. Breathing frequency and pattern are poor predictors of work of breathing in patients receiving pressure support ventilation. Chest 1995;108:1338-1344.
[22] Marantz S, Patrick W, Webster K, Roberts D, Oppenheimer L, Younes M. Response of ventilator-dependent patients to different levels of proportional assist. J Appl Physiol 1996;80:397-403.
[23] Chadda Karim, Annane Djillali, Hart Nicholas, Gajdos Philippe, Raphaël Jean Claude, Lofaso Frédéric. Cardiac and respiratory effects of continuous positive airway pressure and noninvasive ventilation in acute cardiac pulmonary edema. Crit Care Med 2002;30(11).
[24] Xu L1, Zhang NX, Qin YZ, Zhan C, Wang SP. Effect on hemodynamics during different modes of low assist ventilation. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2006;18(6):363-366.
[25] Saad A, Rahman M, Marques NR, Funston J, Pacheco L, Whitehead W, et al. Cardiac hemodynamic changes during weaning from mechanical ventilation using NICOM. Crit Care Med 2015;43(12 Suppl 1):19.
[26] Qin Gu, Ning Liu, Ying Xu. The influence of different respiratory frequencies on hemodynamics in patients on artificial ventilation. Zhongguo weizhongbing jijiuyixue 19(9):525-7.
[27] Vetrugno L, Costa M, Centonze C, Langiano N, Rojatti M, Della Rocca G. Trans-thoracic echo evaluation before and during noninvasive ventilation. Crit Care 2012;16(Suppl 1):P88.
[28] Ait-Oufella H, Tharaux PL, Baudel JL, Vandermeersch S, Meyer P, Tonnellier M, Dussaule JC, Guidet B, Offenstadt G, Maury E. Variation in natriuretic peptides and mitral flow indexes during successful ventilatory, 2007.
[29] Caille Vincent, Amiel Jean-Bernard, Charron Cyril, Belliard Guillaume, Vieillard-Baron Antoine, Vignon Philippe. Echocardiography: a help in the weaning process. Crit Care 2010;14:R120.
[30] Gerbaud E, Erickson M, Grenouillet-Delacre M, Beauvieux MC, Coste P, Durrieu-Jaïs C, et al. Echocardiographic evaluation and N-terminal pro-brain natriuretic peptide measurement of patients hospitalized for heart failure during weaning from mechanical ventilation. Minerva Anestesiol 2012;78:415-425.
[31] Eldehily Khalaf, Nassar Yasser, Soliman Randa, Elmaraghy Sameh, Khalid Hassan. Tricuspid annular plane systolic excursion and right ventricular tissue Doppler imaging as predictors of weaning failure in mechanically ventilated patients. EJICT 2017;7(1).
© 2018 by Lippincott Williams & Wilkins, Inc.