Heart & Lung xxx (2014) 1e7
Contents lists available at ScienceDirect
Heart & Lung journal homepage: www.heartandlung.org
Acute hemodynamic responses to 30 head-down postural drainage in stable, ventilated trauma patients: A randomized crossover trial Guntaragorn Hongrattana, MSc (PT) a, Potipong Reungjui, MD b, Chulee U. Jones, PhD c, * a
Department of Physiotherapy, Khon Kaen Hospital, Thailand Trauma and Critical Care Center, Khon Kaen Hospital, Thailand c School of Physical Therapy, Faculty of Associated Medical Sciences, Khon Kaen University, Thailand b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 September 2013 Received in revised form 14 January 2014 Accepted 16 January 2014 Available online xxx
Objective: To determine whether 30 head-down tilt (HDT) used for secretion clearance is safe for acute trauma patients. Background: There are concerns that HDT may lead to cardiac irregularities in intubated patients in the ICU. Methods: Eleven mechanically ventilated trauma patients (25e42 yrs) without cardiovascular problems received two interventions, one supine HDT for 10 min and a control in the horizontal supine position (HS), in a crossover design. Results: Compared to baseline there were statistically significant (p < 0.05) increases in SBP (6.3 mm Hg; 95% CI 2.5, 12.7) and CVP (7.3 cm H2O; 5.7, 10.0) during 10 min HDT although these were not of clinical concern. Heart rate and oxygen saturation were unchanged. No episodes of arrhythmia or hypoxemia were observed. All values returned close to baseline during 10 min horizontal recovery. There were no significant changes during the control HS intervention. Conclusion: 30 HDT entails minimal risk for trauma patients who have no underlying cardiovascular disease. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Head down tilt Hemodynamic changes Trauma patients ICU Postural drainage
Introduction Most trauma patients admitted to the intensive care unit (ICU) require respiratory interventions such as intubation and mechanical ventilation but in these conditions lung secretions can accumulate1 leading to complications such as pneumonia and atelectasis, especially of the middle and lower lobes.2,3 A variety of chest physical therapy techniques are used to help clear secretions to improve gas exchange and increase lung expansion.4,5 One such technique is postural drainage in which the body is placed in a position that allows gravity to assist mucus drainage from distal bronchi to the proximal airways.6 The standard position for draining the lower lung is 30 head-down tilt (HDT) for 3e15 min, or longer in special situations, depending on the patient’s tolerance to changes in position and the amount of sputum produced.7 Source of support: Funding from Faculty of Associated Medical Sciences and Graduate School, Khon Kaen University. Trial registration: NCT01668875. Conflict of interest: None. * Corresponding author. Tel.: þ66 845164169; fax: þ66 43202085. E-mail addresses: [email protected], [email protected] (C.U. Jones). 0147-9563/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.hrtlng.2014.01.011
Although HDT is a recommended technique for secretion clearance, and patients on mechanical ventilation frequently need such treatment, there is some doubt about the safety of the procedure when used in the ICU. One study examined the effects of side lying in the 10 head-down position for 10 min in 72 critically ill patients (pulmonary infection) and reported that over a third of the patients developed arrhythmias.8 An earlier study of 13 bronchiectasis patients,9 also using the 10 head-down position, reported four patients to have cardiac irregularities. However, a number of patients in both studies may have had pre-existing cardiovascular problems and the combination of the head down position with percussion may also have contributed to the complications. It is also notable in the study of Hammon et al8 that the arrhythmias were seen only in the older subjects. On the other hand, Berney et al10 used head down tilt of up to 45 , together with manual hyperinflation, to improve clearance in 20 ICU patients with mainly chronic conditions and reported no adverse events. This would suggest the technique is safe but the actual changes in cardiovascular parameters were not reported so it is difficult to know quite how safe the procedure is and whether it could be used with patients with a greater degree of cardiovascular risk, such as those suffering acute trauma.
2
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
The present study concerns the use of the head down drainage position in critically ill trauma patients for two reasons. Road traffic accidents are very common in Thailand and the location of the study was a regional center for the care of trauma patients. Secondly, conventional treatments to increase secretion clearance such as percussion are often contra-indicated because of the nature of the injuries so that postural drainage is potentially an important treatment option. We have previously shown that deep breathing with a device that both humidifies and oscillates the inspired air (BreatheMAX) doubles the secretions cleared with intubated patients,11 but this technique can only be used with conscious patients who are able to come off the ventilator for a minute or so. However, it clearly shows the potential benefits that might be expected from an effective intervention that could be used in unconscious patients and those who are wholly dependent on a ventilator. We have briefly reported the respiratory responses to supine HDT in these pateints12 and found no changes of clinical significance. However, the head down position can stress the cardiovascular system13 and this may be particularly important in critical trauma patients. Consequently we report here the effects of HDT on blood pressure, central venous pressure, heart rate, oxygenation and incidence of cardiac irregularities and compare these measures to those when patients were in the horizontal supine position that is commonly used for more conservative physiotherapy procedures.
Methods Design The study had a randomized crossover design exploring acute hemodynamic responses to the head-down 30 postural drainage position in trauma patients in the intensive care unit of Khon Kaen Hospital, Thailand. The study was approved by the Human Ethics Committee of Khon Kaen Hospital and the participants, or a close relative, gave their written informed consent before data collection began.
Procedure and intervention Patients were transferred to the tilt table and firmly secured with straps around the chest and hip (Fig. 2). Once settled the patients remained in the supine lying position for 10 min. This was followed by either the control intervention in which the patient continued in the same horizontal position (HS) for 10 min, or 10 min in the 30 head down position (HDT), the angle being checked with a protractor. Patients were then returned to the horizontal for a further 10 min recovery. The first intervention of the first patient was randomly assigned by picking a card and thereafter patients alternated between HDT or HS for the first intervention, providing a balanced crossover design (Fig. 1). During the trials the patients were connected to the mechanical ventilator (Savina Bennett 2000Ò) via an endotracheal or tracheostomy tube. Subjects were rested for at least 3 h after the first leg of the study. A preliminary test had shown that blood pressure returned to baseline after HDT within about 10e15 min and it was always established that values were back to baseline before commencing the second leg of the study. Subjects were suctioned via an endotracheal tube to clear the airway 10 min before each leg of the study. Data collection was done by the authors who were not blinded to the procedures. Sample size Blood pressure measurements with a group of young healthy subjects at rest in a standard horizontal supine position and using a manual cuff were 113 5 and 71 9 mm Hg for systolic and diastolic pressures, respectively. Preliminary data for ICU patients using an automated measuring system showed similar variation. In both cases the variation was halved when repeated measurements were made on individuals over a 10 min period. There is no published information about the variability of central venous pressure in trauma patients but preliminary measures in patients showed a standard deviation of 6 cm H2O and a mean value of 13 cm H2O. We anticipated that clinically significant changes would be greater than the standard deviations for both blood and central venous pressures, so the effect size was set as the standard deviation of the blood pressure measurements. For a confidence level of 0.05 and power of 0.8, this indicates a sample size of approximately 10 patients. Criteria for withdrawing patients
Participants Potential participants were patients sequentially admitted to the ICU with traumatic blunt chest and/or abdominal injury between September 1st and October 31st, 2012. The inclusion criteria were (i) mechanical ventilation, (ii) stable cardiopulmonary function with a resting heart rate not fluctuating more than 10 bpm over the course of the day before the study, systolic blood pressure (SBP) 90e140 mm Hg, diastolic blood pressure (DBP) 60e90 mm Hg, respiratory rate 10e20 breaths/min, and oxygen saturation (SpO2) >95%. Patients were excluded if they had spinal injuries such as fracture, herniated nucleus pulposus, whiplash syndrome, spondylolisthesis, fracture of upper and/or lower limbs, acute or chronic cardiac disease (established with 12 lead ECG), craniotomy or craniectomy, intracranial pressure fluctuating >12 mm Hg, or uncontrolled pain. The number of patients included and excluded are shown in Fig. 1 and details of the medical condition of the participants are given in Table 1. Recruitment was sequential from those patients meeting the inclusion criteria. At the time of the study the patients had been in the ICU for between 4 and 7 days and had been mechanically ventilated throughout that time.
Before the study it was determined that the HDT procedure would be stopped immediately if harmful signs or symptoms were detected, largely based on the adverse reactions to physiotherapy procedures identified by Zeppos et al14: BP < 90/60 or >180/ 100 mm Hg, a change in heart rate of more than 20% of resting, any reduction of consciousness, restlessness, SpO2 < 90%, spontaneous respiratory rate >35 beats/min, arrhythmia with premature ventricular contraction (PVC > 5/min) and/or atrial fibrillation, premature atrial contraction, ventricular tachycardia, with hypoxemia and hypotensive signs. Outcome measures Hemodynamic parameters were recorded every 2 min during each of the 10 min periods (Baseline, HS or HDT, Recovery). Recorded variables were heart rate (HR) and electrocardiogram (lead 2); systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), central venous pressure (CVP). Blood pressure was measured with a standard arm cuff and automated bedside monitor (Nihon Kohden-life scopeÒ). CVP was measured with an indwelling subclavian catheter with the transducer standing at the side of the bed and zeroed before each
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
3
Fig. 1. Flow of patients through the study. HDT: head down tilt at 30 in the supine condition; HS: horizontal supine control.
set of measurements. Care was taken to ensure that the pressure cuff and the tip of the CVP catheter were maintained at the level of the heart (rib 4) at all times. Oxygen saturation (SpO2) was monitored using a pulse oximeter applied to a finger. Data analysis Data for each period (Baseline, HS or HDT and Recovery) were averaged over the 10 min to give a single value for each patient.
Data were analyzed using SPSS version 13, checked for normality by the ShapiroeWilk test and reported as means and standard deviations or as mean and 95% confidence intervals for the changes between baseline and both intervention and recovery. Significance of changes between baseline, intervention and recovery were tested using the General Linear Model repeated measures ANOVA separately for the HDT and HS conditions. Differences between the conditions were assessed by comparing changes from baseline using the General Linear Model univariate ANOVA. In both cases
Table 1 Participant characteristics. No.
Gender
Age (yrs)
Medical diagnosis
Mode of mechanical ventilation
Sedation/analgesia
Apache II
Modified Ramsay sedation score
1
Male
38
12
2
Male
30
Propofol
18
4
3
Male
25
SIMV; VT 550, Ti 1.8, RR 12, PS 25, FiO2 0.4, PEEP 5 BIPAP; PiP 26, Ps 14, RR 16, Ti 1.1, FiO2 0.4, PEEP 5 CPAP; PS 12, FiO2 0.4, PEEP 5
Morphine
2
Morphine
18
2
4
Female
42
SIMV; VT 520, Ti 1.4, RR 12, PS 12, FiO2 0.3, PEEP 5
Propofol
20
5
5
Female
40
18
3
Male
36
Fentanyl, Morphine
14
3
7
Male
26
Propofol
19
5
8
Male
32
Propofol
18
4
9
Female
30
Propofol
14
5
10
Female
25
Fentanyl
16
2
11
Male
26
BIPAP; PiP 24, Ps 12, RR 16, Ti 1.1, FiO2 0.5, PEEP 5 SIMV; VT 500, Ti 1.3, RR 14, PS 25, FiO2 0.5, PEEP 5 CMV; VT 450, Ti 1.3, RR 12, PS 22, FiO2 0.5, PEEP 5 BIPAP; PiP 26, Ps 14, RR 16, Ti 1.2, FiO2 0.4, PEEP 5 CMV; VT 450, Ti 1.3, RR 12, PS 22, FiO2 0.5, PEEP 5 SIMV; VT 500, Ti 1.3, RR 14, PS 25, FiO2 0.5, PEEP 5 BIPAP; PiP 24, Ps 12, RR 16, Ti 1.1, FiO2 0.5, PEEP 5
Fentanyl, Morphine
6
Blunt abdominal with respiratory failure with bowel injury grade I Bilateral hemopneumothorax s/p ICD with respiratory failure Blunt abdominal with Fx. Rib 8, 9 with respiratory failure Blunt abdominal with small bowel injury grade II with fracture rib 7e8 with Rt. hemopneumothorax s/p EL with repaired small bowel Hemopnuemothorax s/p ICD with respiratory failure with Fx. Right Rib 3, 4 Blunt abdominal with respiratory failure with Fx. Left Rib 5, 6, Mild head injury with respiratory failure with Fx. Right rib 6 Bilateral hemopneumothorax s/p ICD with respiratory failure Mild head injury with respiratory failure
Propofol
17
5
Blunt abdominal with small bowel injury grade II s/p EL with repaired small bowel Blunt abdominal with respiratory failure with bowel injury grade I
s/p ¼ status post, ICD ¼ intercostal drainage, EL ¼ exploratory laparotomy, CMV ¼ controlled mechanical ventilation, SIMV ¼ synchronized intermittent mandatory ventilation, BIPAP ¼ biphasic positive airway pressure, CPAP ¼ continuous positive airway pressure, FiO2 ¼ fraction of inspired oxygen, PEEP ¼ positive end expiratory pressure, VT ¼ tidal volume, Ti ¼ inspiratory time, PS ¼ pressure support, RR ¼ respiratory rate, Fx ¼ fracture.
4
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
Fig. 2. Patient positioning. A) Patient in the horizontal supine (HS) position. B) Patient in the 30 head down (HDT) position.
where the F statistic achieved a significance of p < 0.05, main effects were identified by post-hoc pairwise comparisons, with Bonferroni adjustment for multiple comparisons. The main aim of this study was to determine the safety of the HDT procedure and while means and variances might be within safe limits the averaging process can hide the occasional outlier who has a response that is outside the safe limits. To allow for this possibility we have shown the responses of individual patients in Fig. 3, as well as the mean data in Tables 2 and 3.
period, together with heart rate, are presented in Table 2 while in Table 3, the changes from baseline are presented for both HDT and HS. Heart rate Heart rates were high during the baseline and recovery periods, on average just over 100 bpm. There was a tendency to increase during HDT but this was not statistically or clinically significant (Table 3). There was no change in heart rate during the control HS intervention.
Results Patient characteristics Eleven intubated patients with mechanical ventilation were enrolled; seven male and five female patients, median age 30 years (IQR 26e37). Details of the medical condition, mode of ventilation, the severity of their condition (Apache II score15), sedation and analgesia and Ramsey sedation score16 are given in Table 1. Hemodynamic responses to the interventions Blood pressure and central venous pressure during HDT are shown for individual patients in Fig. 3. The data averaged over each
Systolic blood pressure In the majority of patients, SBP rose by about 12 mm Hg in the first 2 min of HDT and remained at a similar level (F(2,20) ¼ 5.69, p ¼ 0.026) until the patient returned to horizontal supine lying and recovered to values not significantly different to the baseline within 4e6 min (Table 2, Fig. 3A). There were significant differences in changes from baseline between HDT and HS (F(3,40) ¼ 4.67, p ¼ 0.007, Table 3). For patient 5 there were no systematic changes in pressure while for patient 7 there was a decrease in pressure during HDT although SBP for this patient was rather variable during both the baseline and recovery periods. There were no changes in systolic pressure during the control HS interventions.
Fig. 3. Hemodynamic changes with the head down position. A) Changes in systolic blood pressure (sBP). B) Changes in diastolic blood pressure (dBP). C) Changes in mean arterial pressure (MAP). D) Changes in central venous pressure (CVP). Data are expressed relative to the average value for each subject over the first 10 min baseline period as indicated by the dashed horizontal line. The solid horizontal bar between 10 and 20 min indicates the duration of the head down position.
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7 Table 2 Mean values for hemodynamic variables. Baseline
HR (bpm) SBP (mm Hg) DBP (mm Hg) MAP (mm Hg) CVP (cm H2O)
Mean SD Mean SD Mean SD Mean SD Mean SD
Intervention
Recovery
HDT
HS
HDT
HS
HDT
HS
104 11 114 8.6 72 5.5 85 4.8 13.7 6.4
104 13 113 6.9 71 7.2 84 6.1 12.5 6.2
108 7 120* 13.1 77 7.9 91 7.9 21.0# 6.6
106 10 114 7.8 75 6.4 88 5.1 15.2 7.0
106 10 114 7.8 75 6.4 88 5.1 15.2 7.0
104 13 112 7.5 72 6.2 85 5.7 12.5 6.4
Mean and SD of the hemodynamic variables for the three periods, Baseline, Intervention and Recovery and the two conditions, head down tilt (HDT) or the control horizontal supine lying position (HS). HR, heart rate; sBP, systolic blood pressure; dBP, diastolic blood pressure; MAP, mean arterial pressure; CVP, central venous pressure. *differs significantly from HDT baseline, p ¼ 0.026; #differs significantly from HDT baseline and recovery, p < 0.001, pair-wise comparisons with Bonferroni correction.
Diastolic blood pressure Changes in DBP (Table 2, Fig. 3B) were more variable than SBP. There was very little change, or a slight decrease, in diastolic pressure in three subjects while the others showed an increase over the 10 min period. For patient 1, however, there was a large increase in DBP of 25 mm Hg during HDT. The same patient also showed an increase of 10 mm Hg during HS although there were no changes in DBP during the control intervention for any of the other patients. The changes did not reach statistical significance in either the HDT or HS condition (Table 3). Mean arterial pressure The maximum rise in MAP during HDT was between 11 and 12 mm Hg but did not reach statistical significance (Table 2, Fig. 3C). In most patients MAP was back to baseline within the 10 min recovery period but in four it remained slightly elevated as a consequence of the slower recovery of diastolic blood pressure (Fig. 3B). Central venous pressure Central venous pressure increased by 6e12 cm H2O in all patients in the first 2 min of HDT (F(2,20) ¼ 59.8, p < 0.001), remaining elevated and roughly constant for 10 min in the majority Table 3 Changes from baseline of hemodynamic variables. Change from baseline Intervention
HR (bpm)
Change 95% CI
SBP (mm Hg)
Change 95% CI p Change 95% CI p Change 95% CI p Change 95% CI p
DBP (mm Hg)
MAP (mm Hg)
CVP (cm H2O)
Recovery
HDT
HS
HDT
HS
6.9 0.65, 7.70 ns 6.3 2.5, 10.1 0.037 5.4 1.68, 9.15 ns 5.7 2.78, 8.60 ns 7.3 5.71, 8.87
Contents lists available at ScienceDirect
Heart & Lung journal homepage: www.heartandlung.org
Acute hemodynamic responses to 30 head-down postural drainage in stable, ventilated trauma patients: A randomized crossover trial Guntaragorn Hongrattana, MSc (PT) a, Potipong Reungjui, MD b, Chulee U. Jones, PhD c, * a
Department of Physiotherapy, Khon Kaen Hospital, Thailand Trauma and Critical Care Center, Khon Kaen Hospital, Thailand c School of Physical Therapy, Faculty of Associated Medical Sciences, Khon Kaen University, Thailand b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 8 September 2013 Received in revised form 14 January 2014 Accepted 16 January 2014 Available online xxx
Objective: To determine whether 30 head-down tilt (HDT) used for secretion clearance is safe for acute trauma patients. Background: There are concerns that HDT may lead to cardiac irregularities in intubated patients in the ICU. Methods: Eleven mechanically ventilated trauma patients (25e42 yrs) without cardiovascular problems received two interventions, one supine HDT for 10 min and a control in the horizontal supine position (HS), in a crossover design. Results: Compared to baseline there were statistically significant (p < 0.05) increases in SBP (6.3 mm Hg; 95% CI 2.5, 12.7) and CVP (7.3 cm H2O; 5.7, 10.0) during 10 min HDT although these were not of clinical concern. Heart rate and oxygen saturation were unchanged. No episodes of arrhythmia or hypoxemia were observed. All values returned close to baseline during 10 min horizontal recovery. There were no significant changes during the control HS intervention. Conclusion: 30 HDT entails minimal risk for trauma patients who have no underlying cardiovascular disease. Ó 2014 Elsevier Inc. All rights reserved.
Keywords: Head down tilt Hemodynamic changes Trauma patients ICU Postural drainage
Introduction Most trauma patients admitted to the intensive care unit (ICU) require respiratory interventions such as intubation and mechanical ventilation but in these conditions lung secretions can accumulate1 leading to complications such as pneumonia and atelectasis, especially of the middle and lower lobes.2,3 A variety of chest physical therapy techniques are used to help clear secretions to improve gas exchange and increase lung expansion.4,5 One such technique is postural drainage in which the body is placed in a position that allows gravity to assist mucus drainage from distal bronchi to the proximal airways.6 The standard position for draining the lower lung is 30 head-down tilt (HDT) for 3e15 min, or longer in special situations, depending on the patient’s tolerance to changes in position and the amount of sputum produced.7 Source of support: Funding from Faculty of Associated Medical Sciences and Graduate School, Khon Kaen University. Trial registration: NCT01668875. Conflict of interest: None. * Corresponding author. Tel.: þ66 845164169; fax: þ66 43202085. E-mail addresses: [email protected], [email protected] (C.U. Jones). 0147-9563/$ e see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.hrtlng.2014.01.011
Although HDT is a recommended technique for secretion clearance, and patients on mechanical ventilation frequently need such treatment, there is some doubt about the safety of the procedure when used in the ICU. One study examined the effects of side lying in the 10 head-down position for 10 min in 72 critically ill patients (pulmonary infection) and reported that over a third of the patients developed arrhythmias.8 An earlier study of 13 bronchiectasis patients,9 also using the 10 head-down position, reported four patients to have cardiac irregularities. However, a number of patients in both studies may have had pre-existing cardiovascular problems and the combination of the head down position with percussion may also have contributed to the complications. It is also notable in the study of Hammon et al8 that the arrhythmias were seen only in the older subjects. On the other hand, Berney et al10 used head down tilt of up to 45 , together with manual hyperinflation, to improve clearance in 20 ICU patients with mainly chronic conditions and reported no adverse events. This would suggest the technique is safe but the actual changes in cardiovascular parameters were not reported so it is difficult to know quite how safe the procedure is and whether it could be used with patients with a greater degree of cardiovascular risk, such as those suffering acute trauma.
2
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
The present study concerns the use of the head down drainage position in critically ill trauma patients for two reasons. Road traffic accidents are very common in Thailand and the location of the study was a regional center for the care of trauma patients. Secondly, conventional treatments to increase secretion clearance such as percussion are often contra-indicated because of the nature of the injuries so that postural drainage is potentially an important treatment option. We have previously shown that deep breathing with a device that both humidifies and oscillates the inspired air (BreatheMAX) doubles the secretions cleared with intubated patients,11 but this technique can only be used with conscious patients who are able to come off the ventilator for a minute or so. However, it clearly shows the potential benefits that might be expected from an effective intervention that could be used in unconscious patients and those who are wholly dependent on a ventilator. We have briefly reported the respiratory responses to supine HDT in these pateints12 and found no changes of clinical significance. However, the head down position can stress the cardiovascular system13 and this may be particularly important in critical trauma patients. Consequently we report here the effects of HDT on blood pressure, central venous pressure, heart rate, oxygenation and incidence of cardiac irregularities and compare these measures to those when patients were in the horizontal supine position that is commonly used for more conservative physiotherapy procedures.
Methods Design The study had a randomized crossover design exploring acute hemodynamic responses to the head-down 30 postural drainage position in trauma patients in the intensive care unit of Khon Kaen Hospital, Thailand. The study was approved by the Human Ethics Committee of Khon Kaen Hospital and the participants, or a close relative, gave their written informed consent before data collection began.
Procedure and intervention Patients were transferred to the tilt table and firmly secured with straps around the chest and hip (Fig. 2). Once settled the patients remained in the supine lying position for 10 min. This was followed by either the control intervention in which the patient continued in the same horizontal position (HS) for 10 min, or 10 min in the 30 head down position (HDT), the angle being checked with a protractor. Patients were then returned to the horizontal for a further 10 min recovery. The first intervention of the first patient was randomly assigned by picking a card and thereafter patients alternated between HDT or HS for the first intervention, providing a balanced crossover design (Fig. 1). During the trials the patients were connected to the mechanical ventilator (Savina Bennett 2000Ò) via an endotracheal or tracheostomy tube. Subjects were rested for at least 3 h after the first leg of the study. A preliminary test had shown that blood pressure returned to baseline after HDT within about 10e15 min and it was always established that values were back to baseline before commencing the second leg of the study. Subjects were suctioned via an endotracheal tube to clear the airway 10 min before each leg of the study. Data collection was done by the authors who were not blinded to the procedures. Sample size Blood pressure measurements with a group of young healthy subjects at rest in a standard horizontal supine position and using a manual cuff were 113 5 and 71 9 mm Hg for systolic and diastolic pressures, respectively. Preliminary data for ICU patients using an automated measuring system showed similar variation. In both cases the variation was halved when repeated measurements were made on individuals over a 10 min period. There is no published information about the variability of central venous pressure in trauma patients but preliminary measures in patients showed a standard deviation of 6 cm H2O and a mean value of 13 cm H2O. We anticipated that clinically significant changes would be greater than the standard deviations for both blood and central venous pressures, so the effect size was set as the standard deviation of the blood pressure measurements. For a confidence level of 0.05 and power of 0.8, this indicates a sample size of approximately 10 patients. Criteria for withdrawing patients
Participants Potential participants were patients sequentially admitted to the ICU with traumatic blunt chest and/or abdominal injury between September 1st and October 31st, 2012. The inclusion criteria were (i) mechanical ventilation, (ii) stable cardiopulmonary function with a resting heart rate not fluctuating more than 10 bpm over the course of the day before the study, systolic blood pressure (SBP) 90e140 mm Hg, diastolic blood pressure (DBP) 60e90 mm Hg, respiratory rate 10e20 breaths/min, and oxygen saturation (SpO2) >95%. Patients were excluded if they had spinal injuries such as fracture, herniated nucleus pulposus, whiplash syndrome, spondylolisthesis, fracture of upper and/or lower limbs, acute or chronic cardiac disease (established with 12 lead ECG), craniotomy or craniectomy, intracranial pressure fluctuating >12 mm Hg, or uncontrolled pain. The number of patients included and excluded are shown in Fig. 1 and details of the medical condition of the participants are given in Table 1. Recruitment was sequential from those patients meeting the inclusion criteria. At the time of the study the patients had been in the ICU for between 4 and 7 days and had been mechanically ventilated throughout that time.
Before the study it was determined that the HDT procedure would be stopped immediately if harmful signs or symptoms were detected, largely based on the adverse reactions to physiotherapy procedures identified by Zeppos et al14: BP < 90/60 or >180/ 100 mm Hg, a change in heart rate of more than 20% of resting, any reduction of consciousness, restlessness, SpO2 < 90%, spontaneous respiratory rate >35 beats/min, arrhythmia with premature ventricular contraction (PVC > 5/min) and/or atrial fibrillation, premature atrial contraction, ventricular tachycardia, with hypoxemia and hypotensive signs. Outcome measures Hemodynamic parameters were recorded every 2 min during each of the 10 min periods (Baseline, HS or HDT, Recovery). Recorded variables were heart rate (HR) and electrocardiogram (lead 2); systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), central venous pressure (CVP). Blood pressure was measured with a standard arm cuff and automated bedside monitor (Nihon Kohden-life scopeÒ). CVP was measured with an indwelling subclavian catheter with the transducer standing at the side of the bed and zeroed before each
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
3
Fig. 1. Flow of patients through the study. HDT: head down tilt at 30 in the supine condition; HS: horizontal supine control.
set of measurements. Care was taken to ensure that the pressure cuff and the tip of the CVP catheter were maintained at the level of the heart (rib 4) at all times. Oxygen saturation (SpO2) was monitored using a pulse oximeter applied to a finger. Data analysis Data for each period (Baseline, HS or HDT and Recovery) were averaged over the 10 min to give a single value for each patient.
Data were analyzed using SPSS version 13, checked for normality by the ShapiroeWilk test and reported as means and standard deviations or as mean and 95% confidence intervals for the changes between baseline and both intervention and recovery. Significance of changes between baseline, intervention and recovery were tested using the General Linear Model repeated measures ANOVA separately for the HDT and HS conditions. Differences between the conditions were assessed by comparing changes from baseline using the General Linear Model univariate ANOVA. In both cases
Table 1 Participant characteristics. No.
Gender
Age (yrs)
Medical diagnosis
Mode of mechanical ventilation
Sedation/analgesia
Apache II
Modified Ramsay sedation score
1
Male
38
12
2
Male
30
Propofol
18
4
3
Male
25
SIMV; VT 550, Ti 1.8, RR 12, PS 25, FiO2 0.4, PEEP 5 BIPAP; PiP 26, Ps 14, RR 16, Ti 1.1, FiO2 0.4, PEEP 5 CPAP; PS 12, FiO2 0.4, PEEP 5
Morphine
2
Morphine
18
2
4
Female
42
SIMV; VT 520, Ti 1.4, RR 12, PS 12, FiO2 0.3, PEEP 5
Propofol
20
5
5
Female
40
18
3
Male
36
Fentanyl, Morphine
14
3
7
Male
26
Propofol
19
5
8
Male
32
Propofol
18
4
9
Female
30
Propofol
14
5
10
Female
25
Fentanyl
16
2
11
Male
26
BIPAP; PiP 24, Ps 12, RR 16, Ti 1.1, FiO2 0.5, PEEP 5 SIMV; VT 500, Ti 1.3, RR 14, PS 25, FiO2 0.5, PEEP 5 CMV; VT 450, Ti 1.3, RR 12, PS 22, FiO2 0.5, PEEP 5 BIPAP; PiP 26, Ps 14, RR 16, Ti 1.2, FiO2 0.4, PEEP 5 CMV; VT 450, Ti 1.3, RR 12, PS 22, FiO2 0.5, PEEP 5 SIMV; VT 500, Ti 1.3, RR 14, PS 25, FiO2 0.5, PEEP 5 BIPAP; PiP 24, Ps 12, RR 16, Ti 1.1, FiO2 0.5, PEEP 5
Fentanyl, Morphine
6
Blunt abdominal with respiratory failure with bowel injury grade I Bilateral hemopneumothorax s/p ICD with respiratory failure Blunt abdominal with Fx. Rib 8, 9 with respiratory failure Blunt abdominal with small bowel injury grade II with fracture rib 7e8 with Rt. hemopneumothorax s/p EL with repaired small bowel Hemopnuemothorax s/p ICD with respiratory failure with Fx. Right Rib 3, 4 Blunt abdominal with respiratory failure with Fx. Left Rib 5, 6, Mild head injury with respiratory failure with Fx. Right rib 6 Bilateral hemopneumothorax s/p ICD with respiratory failure Mild head injury with respiratory failure
Propofol
17
5
Blunt abdominal with small bowel injury grade II s/p EL with repaired small bowel Blunt abdominal with respiratory failure with bowel injury grade I
s/p ¼ status post, ICD ¼ intercostal drainage, EL ¼ exploratory laparotomy, CMV ¼ controlled mechanical ventilation, SIMV ¼ synchronized intermittent mandatory ventilation, BIPAP ¼ biphasic positive airway pressure, CPAP ¼ continuous positive airway pressure, FiO2 ¼ fraction of inspired oxygen, PEEP ¼ positive end expiratory pressure, VT ¼ tidal volume, Ti ¼ inspiratory time, PS ¼ pressure support, RR ¼ respiratory rate, Fx ¼ fracture.
4
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7
Fig. 2. Patient positioning. A) Patient in the horizontal supine (HS) position. B) Patient in the 30 head down (HDT) position.
where the F statistic achieved a significance of p < 0.05, main effects were identified by post-hoc pairwise comparisons, with Bonferroni adjustment for multiple comparisons. The main aim of this study was to determine the safety of the HDT procedure and while means and variances might be within safe limits the averaging process can hide the occasional outlier who has a response that is outside the safe limits. To allow for this possibility we have shown the responses of individual patients in Fig. 3, as well as the mean data in Tables 2 and 3.
period, together with heart rate, are presented in Table 2 while in Table 3, the changes from baseline are presented for both HDT and HS. Heart rate Heart rates were high during the baseline and recovery periods, on average just over 100 bpm. There was a tendency to increase during HDT but this was not statistically or clinically significant (Table 3). There was no change in heart rate during the control HS intervention.
Results Patient characteristics Eleven intubated patients with mechanical ventilation were enrolled; seven male and five female patients, median age 30 years (IQR 26e37). Details of the medical condition, mode of ventilation, the severity of their condition (Apache II score15), sedation and analgesia and Ramsey sedation score16 are given in Table 1. Hemodynamic responses to the interventions Blood pressure and central venous pressure during HDT are shown for individual patients in Fig. 3. The data averaged over each
Systolic blood pressure In the majority of patients, SBP rose by about 12 mm Hg in the first 2 min of HDT and remained at a similar level (F(2,20) ¼ 5.69, p ¼ 0.026) until the patient returned to horizontal supine lying and recovered to values not significantly different to the baseline within 4e6 min (Table 2, Fig. 3A). There were significant differences in changes from baseline between HDT and HS (F(3,40) ¼ 4.67, p ¼ 0.007, Table 3). For patient 5 there were no systematic changes in pressure while for patient 7 there was a decrease in pressure during HDT although SBP for this patient was rather variable during both the baseline and recovery periods. There were no changes in systolic pressure during the control HS interventions.
Fig. 3. Hemodynamic changes with the head down position. A) Changes in systolic blood pressure (sBP). B) Changes in diastolic blood pressure (dBP). C) Changes in mean arterial pressure (MAP). D) Changes in central venous pressure (CVP). Data are expressed relative to the average value for each subject over the first 10 min baseline period as indicated by the dashed horizontal line. The solid horizontal bar between 10 and 20 min indicates the duration of the head down position.
G. Hongrattana et al. / Heart & Lung xxx (2014) 1e7 Table 2 Mean values for hemodynamic variables. Baseline
HR (bpm) SBP (mm Hg) DBP (mm Hg) MAP (mm Hg) CVP (cm H2O)
Mean SD Mean SD Mean SD Mean SD Mean SD
Intervention
Recovery
HDT
HS
HDT
HS
HDT
HS
104 11 114 8.6 72 5.5 85 4.8 13.7 6.4
104 13 113 6.9 71 7.2 84 6.1 12.5 6.2
108 7 120* 13.1 77 7.9 91 7.9 21.0# 6.6
106 10 114 7.8 75 6.4 88 5.1 15.2 7.0
106 10 114 7.8 75 6.4 88 5.1 15.2 7.0
104 13 112 7.5 72 6.2 85 5.7 12.5 6.4
Mean and SD of the hemodynamic variables for the three periods, Baseline, Intervention and Recovery and the two conditions, head down tilt (HDT) or the control horizontal supine lying position (HS). HR, heart rate; sBP, systolic blood pressure; dBP, diastolic blood pressure; MAP, mean arterial pressure; CVP, central venous pressure. *differs significantly from HDT baseline, p ¼ 0.026; #differs significantly from HDT baseline and recovery, p < 0.001, pair-wise comparisons with Bonferroni correction.
Diastolic blood pressure Changes in DBP (Table 2, Fig. 3B) were more variable than SBP. There was very little change, or a slight decrease, in diastolic pressure in three subjects while the others showed an increase over the 10 min period. For patient 1, however, there was a large increase in DBP of 25 mm Hg during HDT. The same patient also showed an increase of 10 mm Hg during HS although there were no changes in DBP during the control intervention for any of the other patients. The changes did not reach statistical significance in either the HDT or HS condition (Table 3). Mean arterial pressure The maximum rise in MAP during HDT was between 11 and 12 mm Hg but did not reach statistical significance (Table 2, Fig. 3C). In most patients MAP was back to baseline within the 10 min recovery period but in four it remained slightly elevated as a consequence of the slower recovery of diastolic blood pressure (Fig. 3B). Central venous pressure Central venous pressure increased by 6e12 cm H2O in all patients in the first 2 min of HDT (F(2,20) ¼ 59.8, p < 0.001), remaining elevated and roughly constant for 10 min in the majority Table 3 Changes from baseline of hemodynamic variables. Change from baseline Intervention
HR (bpm)
Change 95% CI
SBP (mm Hg)
Change 95% CI p Change 95% CI p Change 95% CI p Change 95% CI p
DBP (mm Hg)
MAP (mm Hg)
CVP (cm H2O)
Recovery
HDT
HS
HDT
HS
6.9 0.65, 7.70 ns 6.3 2.5, 10.1 0.037 5.4 1.68, 9.15 ns 5.7 2.78, 8.60 ns 7.3 5.71, 8.87
