- Email: [email protected]
The Hemodynamic and Respiratory Effects of Cuirass Ventilation in Healthy Volunteers: Part 1
The Hemodynamic and Respiratory Effects of Cuirass Ventilation in Healthy Volunteers: Part 1 William T. McBride, MD, FCARCSI, FRCA, FFICM,* Giulia Ranaldi, MD,* Mark J. Dougherty, MBBCh, BAO, MRCP, FCARCSI,* Tommaso Siciliano, MSc, MD,* Brian Trethowan, MBBCh, BAO, MRCSI, FRCA, FCARCSI, MRCP,* Peter Elliott, MD, FRCA, FFICM,* Claire Rice,* Sabino Scolletta, MD,† Pierpaolo Giomarelli, MD,† Salvatore Mario Romano, PhD,‡ and David M. Linton, MBChB, FCA, MPhil§ Objective: Negative-pressure ventilation (NPV) by external cuirass (RTX; Deminax Medical Instruments Limited, London, UK) in intubated patients after cardiac surgery improves hemodynamics measured by pulmonary artery catheter (PAC)-based methods, with an increased cardiac output (CO) and stroke volume (SV), without changing the heart rate (HR). The less-invasive pressure recording analytical method (PRAM) (Mostcare; Vytech Health srl, Padova, Italy) allows radial artery– based monitoring of the CO, SV, SV variation, and cardiac cycle efficiency (CCE). The authors investigated the hypothesis that NPV improves PRAMbased hemodynamics and arterial blood gas analysis in spontaneously breathing subjects. Design: A clinical investigation. Setting: A teaching hospital. Participants: Ten healthy volunteers. Interventions: Subjects underwent 5 consecutive experimental ventilation modalities lasting 5 minutes: (1) baseline (no cuirass ventilation), (2) mode 1: cuirass ventilation with a continuous negative pressure of ⴚ20 cmH2O, (3) first rest
period (no cuirass ventilation), (4) mode 2: cuirass ventilation in control mode of 12 breaths/min at ⴚ20 cmH2O, and (5) second rest period. Measurements and Main Results: PRAM parameters were analyzed throughout the final minute of each experimental modality, which concluded with arterial blood gas sampling. Both NPV modes significantly reduced HR without changing CO or systemic vascular resistance. Mode 1 significantly increased CCE and decreased SVV. PO2 decreased in both rest modes compared with baseline. This was prevented by NPV. In 5 smokers, PO2 significantly increased in the control mode compared with first rest period. The control mode NPV improved oxygenation with a reduced PCO2 and reciprocally increased pH. Conclusions: Five minutes of NPV improves hemodynamics and oxygenation in healthy subjects. © 2012 Elsevier Inc. All rights reserved.
T
The paucity of information on the possible cardiorespiratory effects of the negative-pressure cuirass ventilation in spontaneously breathing healthy volunteers may arise from the unacceptably high invasiveness pertaining to the pulmonary artery catheter– based hemodynamic assessment methods previously used. The relatively recent advent of well-validated, less-invasive radial artery– based CO monitoring methods now makes studies involving a hemodynamic evaluation in healthy volunteers possible. It was believed that a volunteer study should have minimal invasiveness and not involve central venous cannulation or the injection of fluid or lithium for calibration and yet have satisfactory validation in unstable or rapidly changing hemodynamic situations. Uniquely fulfilling all these validation7-10 and clinical criteria was the pressure recording analytical method (PRAM) of MOSTCARE (Vytech Health srl, Padova, Italy) in which the analysis of continuous radial artery pressure monitoring allows the continuous recording of heart rate (HR), blood pressure, stroke volume variation (SVV), pulse pressure variation (PPV), stroke volume (SV), CO, cardiac index, and cardiac cycle efficiency (CCE).11-15 CCE describes the cardiocirculatory performance in terms of energy expenditure and reflects the ratio of the power released by the left ventricle with the power dissipated by the cardiovascular system. The maximum (and merely theoretic) value of CCE is ⫹1, with progressively lesser values (eg, ⫹0.2, 0.0, ⫺0.8, or ⫺1.0) representing more and more energy being expended to achieve and maintain cardiovascular homeostasis at a given moment. Recently, CCE has been used clinically in patients undergoing obesity surgery12 and inversely correlates with pro-B natriuretic peptide in decompensated heart failure patients.14 Moreover, the relationship of perioperative CCE and
HE HEMODYNAMIC BENEFITS of negative-pressure ventilation (NPV) have been described in intubated, sedated cardiac surgery patients and include improved cardiac output (CO) and venous return, especially when right ventricular failure occurs.1,2 Moreover, improvements in the oxygenation of recently extubated cardiac surgery patients treated with NPV have been reported.3 However, factors such as sedation, mechanical ventilation, and recent cardiac surgery can singly or in combination alter hemodynamic4 and respiratory functions.5,6 Accordingly, the existing knowledge of cardiorespiratory changes mediated by cuirass ventilation in patients cannot be applied to the unsedated, spontaneously breathing healthy subject.
From the *Department of Cardiac Anaesthesia, Royal Victoria Hospital, Belfast, UK; †Department of Surgery and Bioengineering, University of Siena, Siena, Italy; ‡Unit of Internal Medicine and Cardiology, Department of Critical Care Medicine, University of Florence, Florence, Italy; and §Department of Medicine, Hadassah University Hospital, Ein Karem, Jerusalem, Israel. The authors received unconditional loans of the RTX cuirass ventilator by Deminax Medical Instruments, Ltd, London, UK, and the Mostcare Vytech monitor by Vytech, Ltd, Padova, Italy. S.M.R. is the patent holder and inventor of the pressure recording analytical method (PRAM) technology. Address reprint requests to William T. McBride, MD, FCARCSI, FRCA, FFICM, Department of Cardiac Anaesthesia, Royal Victoria Hospital, Belfast BT12 6BA, Northern Ireland (UK). E-mail: [email protected] © 2012 Elsevier Inc. All rights reserved. 1053-0770/2605-0019$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.05.009 868
KEY WORDS: noninvasive ventilation, pressure recording analytical method, cardiac surgery, hemodynamic monitoring, cuirass ventilation
Journal of Cardiothoracic and Vascular Anesthesia, Vol 26, No 5 (October), 2012: pp 868-872
HEMODYNAMIC AND RESPIRATORY EFFECTS OF CUIRASS VENTILATION
869
Table 1. The Anthropometric Data
Table 3. The Ratio of PPV and SVV
Variable
Volunteer Group (N ⫽ 10)
Age (y) (mean ⫾ SD) Weight (kg) (mean ⫾ SD) Height (cm) (mean ⫾ SD) BSA (mean ⫾ SD) Sex (male/female) Smokers/nonsmokers
34.9 ⫾ 9.0 73.2 ⫾ 14.0 171.9 ⫾ 9.3 1.86 ⫾ 0.21 6/4 5/5
Volunteers Experimental Mode
Baseline Mode 1 Rest 1 Mode 2 Rest 2
PPV (%)
SVV (%)
PPV/SVV
19.5 10.0 15.9 11.9 13.7
17.4 12.0 13.8 13.4 15.4
1.1 0.8 1.2 0.9 0.9
Abbreviation: SD, standard deviation.
control mode of 12 breaths/min at ⫺20 cmH2O, and (5) second rest period (no cuirass ventilation). In addition, before the cessation of each ventilatory mode, arterial blood gas samples were obtained for blood gas analysis using the ABL800 FLEX Blood Gas Analyzer (Radiometer Medical ApS, Brønshøj, Denmark) after completion of the final 1-minute time period of hemodynamic measurements for that ventilatory mode. Healthy volunteers breathed room air throughout their study. The primary endpoints were a significant increase in SV or a decrease in HR. The secondary endpoints were significant changes in PO2, PCO2, and pH. To assess normality for all variables, QQ plots were created using the standardized residuals from the repeatedmeasures analysis of variance. For nonparametric data, within-group changes from the baseline or from selected experimental modalities were analyzed by Friedman comparisons followed by Dunn multiple comparisons. For parametric data, within-group changes from the baseline or from selected experimental modalities were analyzed using analysis of variance with Bonferroni post hoc tests adjusting for multiple comparisons. The Greenhouse-Geisser correction was used when sphericity could not be assumed.
an assay of plasma-mediated in vitro negative inotropy has been investigated.13 This study addresses the question of whether or not 2 modalities of negative-pressure cuirass ventilation (ie, continuous negative pressure at ⫺20 cmH2O and the control mode) in healthy subjects would beneficially alter hemodynamic parameters as measured by noninvasive radial artery⫺based CO and CCE monitoring and ventilation as assessed by arterial blood gas determinations. METHODS The study involved 10 healthy volunteers. Ethical (Office for Research Ethics Committee of Northern Ireland) and institutional (Belfast Trust) research governance approval were obtained before the study (Research Ethics Commitee reference number: 08/NIR02/95). Volunteer inclusion criteria were medical or cardiac surgical intensive care nursing staff with formal or honorary contracts to work in the National Health Service and the ability and willingness to give written informed consent. All participants had continuous radial artery pressure monitoring with MOSTCARE (preset to record the mean values of hemodynamic data over 3 beats), which allowed the continuous recording of HR, blood pressure, SVV, PPV, SV, CO, and CCE. The following mean hemodynamic values were calculated from recordings made over the final minute of each 5-minute experimental ventilation modality: (1) baseline (no cuirass ventilation), (2) mode 1: cuirass ventilation with a continuous negative pressure of ⫺20 cmH2O (Hayek RTX; Deminax Medical Instruments Limited, London, UK), (3) first rest period (no cuirass ventilation), (4) mode 2: cuirass ventilation in
RESULTS
Table 1 shows the anthropometric data, and Tables 2 and 3 show the hemodynamic data in the volunteers. The baseline HR was higher than the HR during mode 1, rest 1, and mode 2. Regarding SV, although analysis of variance just missed significance at the 5% level (p ⫽ 0.058), the estimated marginal mean for mode 1 was higher than during all other time points. Dunnett multiple comparison of modalities with baseline con-
Table 2. Hemodynamic Results Presented as Mean (M) ⴞ Standard Error (SE) or Median (MD) and Lower and Upper Quartile (LQ and UQ) Experimental Modality Variable
Friedmann (F) or Analysis of Variance (A) p Value
B
M1
R1
M2
R2
HR (beat/min) (mean [SE])
71.6 (1.9)
63.6 (2.2)
67.3 (2.2)
65.1 (2.1)
69.0 (2.5)
Stroke volume (L) (mean [SE]) SVV (%) (mean [SE])
0.086 (0.006)
0.097 (0.006)
0.087 (0.006)
0.088 (0.005)
0.085 (0.005)
(A) p ⫽ 0.058
18.65 (1.312)
12.13 (1.265)
15.00 (1.726)
14.85 (1.797)
15.58 (1.427)
21.7 (2.75) 1,283, 1,026, and 1,489 6.19 (0.495)
11.5 (1.82) 1,009, 956, and 1,300 6.26 (0.446)
17.0 (2.26) 1,150, 1,049, and 1,396 5.89 (0.433)
14.2 (2.72) 1,225, 1,054, and 1,604 5.77 (0.386)
14.3 (2.00) 1,345, 1,070, and 1,506 5.72 (0.329)
(A) p ⫽ 0.020 sphericity assumed (A) p ⫽ 0.063 (F) p ⫽ 0.0852
PPV (%) (mean [SE]) SVR (dynes · s · cm⫺5) (MD, LQ, and UQ) Cardiac output (L/min) (mean [SE]) Systolic pressure (mmHg) (mean [SE]) CCE (MD, LQ, and UQ)
132 (4.75) 0.3438, 0.1778 and 0.4339
133 (4.025) 0.5292, 0.4033 and 0.5544
(A) p ⬍ 0.001 sphericity assumed
Pair-wise Comparison B-M1, p ⫽ 0.021 B-R1, p ⫽ 0.043 B-M2, p ⫽ 0.033 (BMC)
B-M1, p ⫽ 0.011 (BMC)
(A) p ⫽ 0.301
133 (3.65)
133 (3.91)
134 (5.49)
(A) p ⫽ 0.763
0.3948, 0.3584 and 0.4883
0.4266, 0.2957 and 0.5138
0.2522, 0.1028 and 0.4238
(F) p ⫽ 0.0051
NOTE. Significant pair-wise comparisons are shown for baseline (B), mode 1 (M1), rest 1(R1), mode 2 (M2), and rest 2 (R2). Abbreviation: BMC, Bonferroni multiple comparison post hoc test.
B-M1, p ⬍ 0.05; M1-R2, p ⬍ 0.01 (Wilcoxon)
870
MCBRIDE ET AL Table 4. Arterial Blood Gas Results Presented as Mean (M) ⴞ Standard Error (SE) or Median (MD) and Lower and Upper Quartile (LQ and UQ) Experimental Modality Variable
PO2 (kPa) (mean [SE])
PO2 (kPa) in 5 smokers (MD, LQ and UQ) PO2 (kPa) in 5 nonsmokers (MD, LQ, and UQ) PCO2 (kPa) (mean [SE])
pH (mean [SE])
Lactate (mmol/L) (mean [SE])
Friedmann (F) or Analysis of Variance (A) p Value
B
M1
R1
M2
R2
13.7 (0.311)
13.2 (0.298)
12.0 (0.394)
14.2 (0.487)
11.8 (0.610)
13.43, 13.13, and 14.00 13.73, 12.55, and 15.25
13.75, 12.92, and 14.49 12.63, 11.97, and 13.46
11.19, 10.18, and 13.10 12.50, 11.79, and 13.30
15.77, 13.88, and 16.00 13.67, 12.14, and 14.41
11.46, 9.77, and 13.00 11.64, 10.36, and 14.16
4.70 (0.134)
4.77 (0.161)
4.98 (0.136)
3.89 (0.194)
4.71 (0.146)
(A) p ⬍ 0.001 (sphericity assumed)
7.43 (0.008)
7.43 (0.013)
7.42 (0.006)
7.49 (0.016)
7.44 (0.011)
(A) p ⬍ 0.001 (sphericity assumed)
0.682 (0.057)
0.690 (0.077)
0.628 (0.068)
0.669 (0.068)
0.646 (0.061)
(A) p ⬍ 0.001 (sphericity assumed)
(F) p ⫽ 0.012
Pair-wise Comparison B-R1, p ⫽ 0.003 B-R2, p ⫽ 0.021 R1-M2, p ⫽ 0.025 M2-R2, p ⫽ 0.043 (BMC) R1-M2, p ⬍ 0.05
(F) p ⫽ 0.029
B-M2, p ⫽ 0.020 M1-M2, p ⫽ 0.001 R1-M2, p ⫽ 0.001 M2-R2, p ⫽ 0.004 (BMC) B-M2, p ⫽ 0.019 M1-M2, p ⫽ 0.035 R1-M2, p ⫽ 0.009 (BMC)
(A) p ⫽ 0.317
NOTE. Significant pair-wise comparisons are shown for baseline (B), mode 1 (M1), rest 1(R1), mode 2 (M2), and rest 2 (R2). Abbreviation: BMC, Bonferroni multiple comparison post hoc test.
trol found SV at mode 1 to be higher than at baseline (p ⬍ 0.05). CCE increased from the baseline to mode 1, and rest 2 was lower than mode 1. There were no between-mode differences with respect to CO, systemic vascular resistance (SVR), PPV, and systolic blood pressure. SVV decreased from baseline to mode 1. PPV mirrored SVV, which decreased with respect to baseline (Table 3). The arterial blood gas changes are shown in Table 4. PO2 (in room air) decreased from baseline at rest 1 and rest 2. PO2 at mode 2 was higher than at rest 1 and rest 2. PCO2 at mode 2 was lower than during all the other modalities. Similarly, the pH level at mode 2 was higher than at baseline, mode 1, and rest 1. Lactate was unchanged throughout. In the 5 volunteers who smoked (3 men and 2 women), PO2 increased from rest 1 to mode 2 (Table 4). In the 5 volunteers who did not smoke (3 men and 2 women), no between-mode differences in PO2 occurred. DISCUSSION
Understanding the mechanism of CO changes associated with negative end-expiratory pressure requires a hemodynamic evaluation of healthy subjects,16 but finding a reliable method of hemodynamic monitoring acceptable to awake healthy subjects already has proven to be difficult for other investigators.16 The CO measurement method chosen in this study was the PRAM (MOSTCARE) system, which allows a noninvasive hemodynamic assessment accurately reflecting rapidly changing hemodynamic conditions but does not require central venous catheter insertion or the injection of fluid, drug, or dye. Because previously reported pulmonary artery catheter– based NPV-related hemodynamic changes in intubated cardiac surgery patients focused on HR, CO, SV, SVV, and PPV,17 it was important that PRAM should at least provide this information and, if possible, additional parameters of cardiac function. One such measure is the CCE.15 It represents the ratio between the hemodynamic work performed by the heart and energy expen-
diture. The underlying mathematic proof recently was reviewed.15 Briefly, an increase in CCE reflects a reduction in energy used by the cardiovascular system to maintain the same hemodynamic balance. Any improvement in CCE can be interpreted as an improvement in ventricular-arterial coupling. The authors found that NPV led to a significant decrease in the HR of volunteers but no decrease in CO. Because CO equals HR ⫻ SV, a decrease in HR accompanied by an unchanged CO suggests an expected increase in SV. However, regarding changes in SV, analysis of variance just missed significance at the 5% level. This means that, although the authors cannot claim statistical significance for the estimated marginal mean for mode 1 being higher than all the other time points, herein may lie a physiologic explanation as to why the significant decrease in HR was not accompanied by a decrease in CO. In defense of this argument, provided the caveat is acknowledged that the SV analysis of variance p value was 0.057, the Dunnett multiple comparison test of SV modalities versus baseline showed that SV at mode 1 was higher than at baseline. In contrast, earlier reports of intubated adult coronary artery bypass graft surgery patients ventilated with intermittent positive pressure ventilation showed that the addition of NPV did not alter HR, blood pressure, or gas exchange but led to significant increases in the SV index and cardiac index.17 These earlier reports of NPV maintaining the HR in intubated patients highlight the unique observation of the healthy volunteers developing a significant decrease in HR upon the commencement of a continuous negative mode, which was seen again during the control mode. The decrease in HR in the volunteers was most likely a response to increased preload and, thus, atrial stretch. The contrasting absence of a decrease in HR in postoperative intubated cardiac surgical patients17 uncovers an asyet-unnoticed difference between healthy and post– cardiac surgery hearts. This difference explains how an arguably
HEMODYNAMIC AND RESPIRATORY EFFECTS OF CUIRASS VENTILATION
important hemodynamic benefit of NPV has been hitherto overlooked in the literature, namely maintaining CO through increasing SV in the presence of a reduced HR, a combination of events that shows conservation of myocardial energy expenditure, which is reflected in the significantly increased CCE. NPV significantly improved CCE in the volunteers. It could be argued that the increased SV and CCE and the decreased HR should be of benefit in nonsurgical patients with ischemic heart disease or in any patient with a critical balance of myocardial oxygen supply and demand. In contrast to the earlier reports of NPV in intubated patients, the healthy volunteers did not have an increase in CO with NPV. However, it could be argued that any increase beyond their already normal baseline CO was unlikely to confer an additional physiologic benefit. The PPV in volunteers mirrored the SVV, which decreased with respect to baseline values (Table 3). The reason why the change in PPV was marginally significant whereas SVV was quite significant may be because PPV is determined by the height of the arterial waveforms, whereas SVV integrates the area under the curve. In fact, the PPV/SVV ratio in volunteers (Table 3) showed a decreasing trend. This could be explained by considering the hemodynamic variables describing arterial elastance studied by Pinsky18 in which the PPV/SVV estimated elastance, which is an indicator of arterial tone, whereas SVR reflected the arteriolar tone. The volunteers’ arterial tone tended to decrease with NPV (in the presence of an unchanged SVR) suggesting adequate fluid hydration before the induced intrathoracic blood accumulation with NPV. Interestingly, the opposite was noted in the present authors’ study of postoperative extubated cardiac surgery patients (part 2). The SVV in volunteers decreased in response to the continuous negative mode. The decreased SVV and the improved CO coupled with the increased SV may suggest an improved venous return. The magnitude of the SVV has been taken as a predictor of fluid responsiveness in mechanically ventilated patients.19-21 In spontaneously breathing subjects, the utility of the related measure of PPV and systolic pressure variation in determining fluid responsiveness has been questioned22 as has the use of SVV and PPV.23,24 The poor detection with pulse contour methods of the response to venous return during spontaneous breathing could be attributable to the poor sensitivity of the available technology in evaluating the differences in SVV and, particularly, in PPV. However, PRAM samples at 1,000 Hz, a frequency 5- to 10-fold higher than other devices, so a higher sensitivity and a better ability to detect small and rapidly occurring variations might be expected provided there is constant vigilance by the attending physician to check the morphology of the arterial wave to prevent over- or underdamping.25,26 This study was not designed to test the hypothesis that NPV-related changes in SVV predict fluid responsiveness. However, the present demonstration of PRAM’s ability to detect rapid changes in PPV measurements and the derived changes in SVV points to the need for such a hypothesis to be tested, which if confirmed would be a more easily applied tool in detecting fluid responsiveness than the inspiratory and expiratory valves suggested by Dahl et al22 in spontaneously breathing subjects. Mode 1 (continuous negative) brought about a significant change from baseline in HR, CCE, and SVV. Moreover, there
871
was a trend toward an increased SV. When looking at mode 2, it effected a significant change from baseline in HR but with trends (albeit not statistically significant) toward the same changes as observed in mode 1. This would suggest that the observed effects were more pronounced in mode 1 than mode 2 but that similar changes were occurring in both modes. Because the decrease in HR in mode 2 was less than in mode 1, the CCE improvements in mode 2 were limited. One explanation may be that continuous NPV in mode 1 provides a constant improvement in venous return, whereas this is interrupted in mode 2, reducing the effect on HR. A second possibility is that because hyperventilation in mode 2 inhibits the known bradycardic effect of the baroreflex in humans,27 the magnitude of the HR reduction in mode 2 is less pronounced than in mode 1. The present authors believe that this is the first time that NPV using a cuirass in healthy subjects has been evaluated for its hemodynamic effects. Overall, it was noted that the cuirass was well tolerated in the volunteers. NPV improved SV (p ⬎ 0.05) and reduced HR with CO staying the same, suggesting an improvement in myocardial oxygen supply/demand, which was reflected by the increased CCE. These findings may suggest a possible new application of NPV as a nonpharmacologic means of supporting the failing or ischemic heart in unsedated spontaneously breathing patients. Although the authors earlier documented the use of the cuirass in the continuous negative mode as an effective method of obviating reintubation in extubated cardiac surgery patients developing respiratory dysfunction, they wished to evaluate its ventilatory effects in healthy volunteers. In the volunteers, the baseline PCO2 and pH were in the normal range as would be expected in subjects not under analgesia and sedation. In the healthy volunteers, there was a significant decrease in PO2 at rest 1 and rest 2. This may reflect a degree of V/Q mismatch associated with the recumbent position used in the study. Interestingly, during the application of NPV in the volunteers, PO2 did not decrease significantly below baseline, suggesting that the atelectasis of recumbency was being countered by the NPV. Arguably, consideration of the alveolar gas equation may explain, at least in part, the increased PO2 between rest 1 and mode 2, where reduced PCO2 would allow more space for O2 as happens during accommodation at a high altitude. Of particular interest was the observation in the 5 healthy volunteers who were smokers that the control mode led to a significant increase in oxygenation compared with the preceding rest PO2. This was not the case with the nonsmoking volunteers. With the caveat that numbers were small, there may be a suggestion that the beneficial effects on the ventilation of NPV have particular application in smokers in whom elastic tissue degeneration may allow small airway closure, which becomes clinically significant after a period of recumbency. The control mode led to a marked decrease in PCO2 and a reciprocal increase in pH in the 10 volunteers, suggesting that this mode was causing hyperventilation. Limitations In mode 1, the cuirass rests motionless on the chest with minimal time needed for the subject to become accustomed to
872
MCBRIDE ET AL
the device. In contrast, in mode 2, the device moves with some minutes required for a first-time subject to become accustomed to this ventilation mode. It is possible that the 5 minutes allowed for each ventilation modality (including rests) were not long enough to allow acclimatization to mode 2. A criticism of the study is that the experimental order for modes 1 and 2 was not random. This is because it is standard clinical practice within the authors’ centers to allow a patient to become familiar with mode 1 (in which cuirass is immobile) before progressing to mode 2 (in which cuirass moves). The respiratory rate in the control mode (mode 2) was preset to a rate that led to hyperventilation, which may have limited any HR-reducing effect of this mode. A third ventilatory mode
of the control mode to normocarbia would have been helpful to determine if the significantly reduced HR in mode 1 also occurred in the control mode at normocarbia. Because PRAM was the only method of hemodynamic monitoring used in the study, all the limitations of PRAM apply, including inadvertent arterial pressure damping although care was taken throughout to prevent this. CONCLUSIONS
NPV in volunteers leads to improvement in hemodynamics. NPV decreases PCO2 with a reciprocal increase in pH and in the control mode improves oxygenation in smokers.
REFERENCES 1. Shekerdemian LS, Bush A, Shore DF, et al: Cardiorespiratory responses to negative pressure ventilation after tetralogy of Fallot repair: A hemodynamic tool for patients with a low-output state. J Am Coll Cardiol 33:549-555, 1999 2. Shekerdemian LS, Bush A, Shore DF, et al: Cardiopulmonary interactions after Fontan operations: Augmentation of cardiac output using negative pressure ventilation. Circulation 96:3934-3942, 1997 3. Trethowan B, Elliott P, Varadarajan B, et al: Continuous negative extra thoracic pressure in respiratory complications after cardiac surgery. Br J Anaesth 101:589P-590P, 2008 4. Sorbara C, Armellin G, Bonato R, et al: Postoperative sedation with propofol infusion: Haemodynamics and pharmacokinetics. Clin Drug Invest 16:431-439, 1998 5. Rozanski EA, Bedenice D, Lofgren J, et al: The effect of body position, sedation, and thoracic bandaging on functional residual capacity in healthy deep-chested dogs. Can J Vet Res 74:34-39, 2010 6. Imberger G, McIlroy D, Pace NL, et al: Positive end-expiratory pressure (PEEP) during anaesthesia for the prevention of mortality and postoperative pulmonary complications. Cochrane Database Syst Rev 9:CD007922, 2010 7. Romano SM, Pistolesi M: Assessment of cardiac output from systemic arterial pressure in humans. Crit Care Med 30:1834-1841, 2002 8. Giomarelli P, Biagioli B, Scolletta S: Cardiac output monitoring by pressure recording analytical method in cardiac surgery. Eur J Cardiothorac Surg 26:515-520, 2004 9. Scolletta S, Romano SM, Biagioli B, et al: Pressure recording analytical method (PRAM) for measurement of cardiac output during various haemodynamic states. Br J Anaesth 95:159-165, 2005 10. Romano SM, Scolletta S, Olivotto I, et al: Systemic arterial waveform analysis and assessment of blood flow during extracorporeal circulation. Perfusion 21:109-116, 2006 11. Scolletta S, Carlucci F, Tabucchi A: Cardiac cycle efficiency correlates with pro-B-type natriuretic peptide in cardiac surgery patients. Crit Care 12:249, 2008 12. Balderi T, Forfori F, Marra V, et al: Continuous hemodynamic monitoring during laparoscopic gastric bypass in superobese patients by pressure recording analytical method. Obes Surg 18:1007-1014, 2008 13. Modesti PA, Gamberi T, Bazzini C, et al: Response of serum proteome in patients undergoing infrarenal aortic aneurysm repair. Anesthesiology 111:844-854, 2009
14. Giglioli C, Landi D, Gensini GF, et al: Cardiac efficiency improvement after slow continuous ultrafiltration is assessed by beatto-beat minimally invasive monitoring in congestive heart failure patients: A preliminary report. Blood Purif 29:44-51, 2010 15. Romano SM: Cardiac cycle efficiency: A new parameter able to fully evaluate the dynamic interplay of the cardiovascular system. Int J Cardiol 155:326-327, 2012 16. Fellahi JL, Caille V, Charron C, et al: Noninvasive assessment of cardiac index in healthy volunteers: A comparison between thoracic impedance cardiography and Doppler echocardiography. Anesth Analg 108:1553-1559, 2009 17. Chaturvedi RK, Zidulka AA, Goldberg P, et al: Use of negative extrathoracic pressure to improve hemodynamics after cardiac surgery. Ann Thorac Surg 85:1355-1360, 2008 18. Pinsky MR: The functional haemodynamic monitoring: applied physiology at the bedside, In: Yearbook of emergency and intensive care medicine 537-552, 2001 19. Berkenstadt H, Margalit N, Hadani M, et al: Stroke volume variation as a predictor of fluid responsiveness in patients undergoing brain surgery. Anesth Analg 92:984-989, 2001 20. Michard F: Changes in arterial pressure during mechanical ventilation. Anesthesiology 103:419-428, 2005 21. Reuter DA, Goetz AE, Peter K: Assessment of volume responsiveness in mechanically ventilated patients. Anaesthesist 52:10051007, 1010-1013, 2003 22. Dahl MK, Vistisen ST, Koefoed-Nielsen J, et al: Using an expiratory resistor, arterial pulse pressure variations predict fluid responsiveness during spontaneous breathing: An experimental porcine study. Crit Care 13:R39, 2009 23. Magder S: Predicting volume responsiveness in spontaneously breathing patients: Still a challenging problem. Crit Care 10:165, 2006 24. Soubrier S, Saulnier F, Hubert H, et al: Can dynamic indicators help the prediction of fluid responsiveness in spontaneously breathing critically ill patients? Intensive Care Med 33:1117-1124, 2007 25. Shinozaki T, Deane RS, Mazuzan JE: The dynamic responses of liquid-filled catheter systems for direct measurements of blood pressure. Anesthesiology 53:498-504, 1980 26. Gardner RM: Direct blood pressure measurement—Dynamic response requirements. Anesthesiology 54:227-236, 1981 27. Van de Borne P, Mezzetti S, Montano N, et al: Hyperventilation alters baroreflex control of heart rate and muscle sympathetic nerve activity. Am J Physiol 279:536-541, 2000
period (no cuirass ventilation), (4) mode 2: cuirass ventilation in control mode of 12 breaths/min at ⴚ20 cmH2O, and (5) second rest period. Measurements and Main Results: PRAM parameters were analyzed throughout the final minute of each experimental modality, which concluded with arterial blood gas sampling. Both NPV modes significantly reduced HR without changing CO or systemic vascular resistance. Mode 1 significantly increased CCE and decreased SVV. PO2 decreased in both rest modes compared with baseline. This was prevented by NPV. In 5 smokers, PO2 significantly increased in the control mode compared with first rest period. The control mode NPV improved oxygenation with a reduced PCO2 and reciprocally increased pH. Conclusions: Five minutes of NPV improves hemodynamics and oxygenation in healthy subjects. © 2012 Elsevier Inc. All rights reserved.
T
The paucity of information on the possible cardiorespiratory effects of the negative-pressure cuirass ventilation in spontaneously breathing healthy volunteers may arise from the unacceptably high invasiveness pertaining to the pulmonary artery catheter– based hemodynamic assessment methods previously used. The relatively recent advent of well-validated, less-invasive radial artery– based CO monitoring methods now makes studies involving a hemodynamic evaluation in healthy volunteers possible. It was believed that a volunteer study should have minimal invasiveness and not involve central venous cannulation or the injection of fluid or lithium for calibration and yet have satisfactory validation in unstable or rapidly changing hemodynamic situations. Uniquely fulfilling all these validation7-10 and clinical criteria was the pressure recording analytical method (PRAM) of MOSTCARE (Vytech Health srl, Padova, Italy) in which the analysis of continuous radial artery pressure monitoring allows the continuous recording of heart rate (HR), blood pressure, stroke volume variation (SVV), pulse pressure variation (PPV), stroke volume (SV), CO, cardiac index, and cardiac cycle efficiency (CCE).11-15 CCE describes the cardiocirculatory performance in terms of energy expenditure and reflects the ratio of the power released by the left ventricle with the power dissipated by the cardiovascular system. The maximum (and merely theoretic) value of CCE is ⫹1, with progressively lesser values (eg, ⫹0.2, 0.0, ⫺0.8, or ⫺1.0) representing more and more energy being expended to achieve and maintain cardiovascular homeostasis at a given moment. Recently, CCE has been used clinically in patients undergoing obesity surgery12 and inversely correlates with pro-B natriuretic peptide in decompensated heart failure patients.14 Moreover, the relationship of perioperative CCE and
HE HEMODYNAMIC BENEFITS of negative-pressure ventilation (NPV) have been described in intubated, sedated cardiac surgery patients and include improved cardiac output (CO) and venous return, especially when right ventricular failure occurs.1,2 Moreover, improvements in the oxygenation of recently extubated cardiac surgery patients treated with NPV have been reported.3 However, factors such as sedation, mechanical ventilation, and recent cardiac surgery can singly or in combination alter hemodynamic4 and respiratory functions.5,6 Accordingly, the existing knowledge of cardiorespiratory changes mediated by cuirass ventilation in patients cannot be applied to the unsedated, spontaneously breathing healthy subject.
From the *Department of Cardiac Anaesthesia, Royal Victoria Hospital, Belfast, UK; †Department of Surgery and Bioengineering, University of Siena, Siena, Italy; ‡Unit of Internal Medicine and Cardiology, Department of Critical Care Medicine, University of Florence, Florence, Italy; and §Department of Medicine, Hadassah University Hospital, Ein Karem, Jerusalem, Israel. The authors received unconditional loans of the RTX cuirass ventilator by Deminax Medical Instruments, Ltd, London, UK, and the Mostcare Vytech monitor by Vytech, Ltd, Padova, Italy. S.M.R. is the patent holder and inventor of the pressure recording analytical method (PRAM) technology. Address reprint requests to William T. McBride, MD, FCARCSI, FRCA, FFICM, Department of Cardiac Anaesthesia, Royal Victoria Hospital, Belfast BT12 6BA, Northern Ireland (UK). E-mail: [email protected] © 2012 Elsevier Inc. All rights reserved. 1053-0770/2605-0019$36.00/0 http://dx.doi.org/10.1053/j.jvca.2012.05.009 868
KEY WORDS: noninvasive ventilation, pressure recording analytical method, cardiac surgery, hemodynamic monitoring, cuirass ventilation
Journal of Cardiothoracic and Vascular Anesthesia, Vol 26, No 5 (October), 2012: pp 868-872
HEMODYNAMIC AND RESPIRATORY EFFECTS OF CUIRASS VENTILATION
869
Table 1. The Anthropometric Data
Table 3. The Ratio of PPV and SVV
Variable
Volunteer Group (N ⫽ 10)
Age (y) (mean ⫾ SD) Weight (kg) (mean ⫾ SD) Height (cm) (mean ⫾ SD) BSA (mean ⫾ SD) Sex (male/female) Smokers/nonsmokers
34.9 ⫾ 9.0 73.2 ⫾ 14.0 171.9 ⫾ 9.3 1.86 ⫾ 0.21 6/4 5/5
Volunteers Experimental Mode
Baseline Mode 1 Rest 1 Mode 2 Rest 2
PPV (%)
SVV (%)
PPV/SVV
19.5 10.0 15.9 11.9 13.7
17.4 12.0 13.8 13.4 15.4
1.1 0.8 1.2 0.9 0.9
Abbreviation: SD, standard deviation.
control mode of 12 breaths/min at ⫺20 cmH2O, and (5) second rest period (no cuirass ventilation). In addition, before the cessation of each ventilatory mode, arterial blood gas samples were obtained for blood gas analysis using the ABL800 FLEX Blood Gas Analyzer (Radiometer Medical ApS, Brønshøj, Denmark) after completion of the final 1-minute time period of hemodynamic measurements for that ventilatory mode. Healthy volunteers breathed room air throughout their study. The primary endpoints were a significant increase in SV or a decrease in HR. The secondary endpoints were significant changes in PO2, PCO2, and pH. To assess normality for all variables, QQ plots were created using the standardized residuals from the repeatedmeasures analysis of variance. For nonparametric data, within-group changes from the baseline or from selected experimental modalities were analyzed by Friedman comparisons followed by Dunn multiple comparisons. For parametric data, within-group changes from the baseline or from selected experimental modalities were analyzed using analysis of variance with Bonferroni post hoc tests adjusting for multiple comparisons. The Greenhouse-Geisser correction was used when sphericity could not be assumed.
an assay of plasma-mediated in vitro negative inotropy has been investigated.13 This study addresses the question of whether or not 2 modalities of negative-pressure cuirass ventilation (ie, continuous negative pressure at ⫺20 cmH2O and the control mode) in healthy subjects would beneficially alter hemodynamic parameters as measured by noninvasive radial artery⫺based CO and CCE monitoring and ventilation as assessed by arterial blood gas determinations. METHODS The study involved 10 healthy volunteers. Ethical (Office for Research Ethics Committee of Northern Ireland) and institutional (Belfast Trust) research governance approval were obtained before the study (Research Ethics Commitee reference number: 08/NIR02/95). Volunteer inclusion criteria were medical or cardiac surgical intensive care nursing staff with formal or honorary contracts to work in the National Health Service and the ability and willingness to give written informed consent. All participants had continuous radial artery pressure monitoring with MOSTCARE (preset to record the mean values of hemodynamic data over 3 beats), which allowed the continuous recording of HR, blood pressure, SVV, PPV, SV, CO, and CCE. The following mean hemodynamic values were calculated from recordings made over the final minute of each 5-minute experimental ventilation modality: (1) baseline (no cuirass ventilation), (2) mode 1: cuirass ventilation with a continuous negative pressure of ⫺20 cmH2O (Hayek RTX; Deminax Medical Instruments Limited, London, UK), (3) first rest period (no cuirass ventilation), (4) mode 2: cuirass ventilation in
RESULTS
Table 1 shows the anthropometric data, and Tables 2 and 3 show the hemodynamic data in the volunteers. The baseline HR was higher than the HR during mode 1, rest 1, and mode 2. Regarding SV, although analysis of variance just missed significance at the 5% level (p ⫽ 0.058), the estimated marginal mean for mode 1 was higher than during all other time points. Dunnett multiple comparison of modalities with baseline con-
Table 2. Hemodynamic Results Presented as Mean (M) ⴞ Standard Error (SE) or Median (MD) and Lower and Upper Quartile (LQ and UQ) Experimental Modality Variable
Friedmann (F) or Analysis of Variance (A) p Value
B
M1
R1
M2
R2
HR (beat/min) (mean [SE])
71.6 (1.9)
63.6 (2.2)
67.3 (2.2)
65.1 (2.1)
69.0 (2.5)
Stroke volume (L) (mean [SE]) SVV (%) (mean [SE])
0.086 (0.006)
0.097 (0.006)
0.087 (0.006)
0.088 (0.005)
0.085 (0.005)
(A) p ⫽ 0.058
18.65 (1.312)
12.13 (1.265)
15.00 (1.726)
14.85 (1.797)
15.58 (1.427)
21.7 (2.75) 1,283, 1,026, and 1,489 6.19 (0.495)
11.5 (1.82) 1,009, 956, and 1,300 6.26 (0.446)
17.0 (2.26) 1,150, 1,049, and 1,396 5.89 (0.433)
14.2 (2.72) 1,225, 1,054, and 1,604 5.77 (0.386)
14.3 (2.00) 1,345, 1,070, and 1,506 5.72 (0.329)
(A) p ⫽ 0.020 sphericity assumed (A) p ⫽ 0.063 (F) p ⫽ 0.0852
PPV (%) (mean [SE]) SVR (dynes · s · cm⫺5) (MD, LQ, and UQ) Cardiac output (L/min) (mean [SE]) Systolic pressure (mmHg) (mean [SE]) CCE (MD, LQ, and UQ)
132 (4.75) 0.3438, 0.1778 and 0.4339
133 (4.025) 0.5292, 0.4033 and 0.5544
(A) p ⬍ 0.001 sphericity assumed
Pair-wise Comparison B-M1, p ⫽ 0.021 B-R1, p ⫽ 0.043 B-M2, p ⫽ 0.033 (BMC)
B-M1, p ⫽ 0.011 (BMC)
(A) p ⫽ 0.301
133 (3.65)
133 (3.91)
134 (5.49)
(A) p ⫽ 0.763
0.3948, 0.3584 and 0.4883
0.4266, 0.2957 and 0.5138
0.2522, 0.1028 and 0.4238
(F) p ⫽ 0.0051
NOTE. Significant pair-wise comparisons are shown for baseline (B), mode 1 (M1), rest 1(R1), mode 2 (M2), and rest 2 (R2). Abbreviation: BMC, Bonferroni multiple comparison post hoc test.
B-M1, p ⬍ 0.05; M1-R2, p ⬍ 0.01 (Wilcoxon)
870
MCBRIDE ET AL Table 4. Arterial Blood Gas Results Presented as Mean (M) ⴞ Standard Error (SE) or Median (MD) and Lower and Upper Quartile (LQ and UQ) Experimental Modality Variable
PO2 (kPa) (mean [SE])
PO2 (kPa) in 5 smokers (MD, LQ and UQ) PO2 (kPa) in 5 nonsmokers (MD, LQ, and UQ) PCO2 (kPa) (mean [SE])
pH (mean [SE])
Lactate (mmol/L) (mean [SE])
Friedmann (F) or Analysis of Variance (A) p Value
B
M1
R1
M2
R2
13.7 (0.311)
13.2 (0.298)
12.0 (0.394)
14.2 (0.487)
11.8 (0.610)
13.43, 13.13, and 14.00 13.73, 12.55, and 15.25
13.75, 12.92, and 14.49 12.63, 11.97, and 13.46
11.19, 10.18, and 13.10 12.50, 11.79, and 13.30
15.77, 13.88, and 16.00 13.67, 12.14, and 14.41
11.46, 9.77, and 13.00 11.64, 10.36, and 14.16
4.70 (0.134)
4.77 (0.161)
4.98 (0.136)
3.89 (0.194)
4.71 (0.146)
(A) p ⬍ 0.001 (sphericity assumed)
7.43 (0.008)
7.43 (0.013)
7.42 (0.006)
7.49 (0.016)
7.44 (0.011)
(A) p ⬍ 0.001 (sphericity assumed)
0.682 (0.057)
0.690 (0.077)
0.628 (0.068)
0.669 (0.068)
0.646 (0.061)
(A) p ⬍ 0.001 (sphericity assumed)
(F) p ⫽ 0.012
Pair-wise Comparison B-R1, p ⫽ 0.003 B-R2, p ⫽ 0.021 R1-M2, p ⫽ 0.025 M2-R2, p ⫽ 0.043 (BMC) R1-M2, p ⬍ 0.05
(F) p ⫽ 0.029
B-M2, p ⫽ 0.020 M1-M2, p ⫽ 0.001 R1-M2, p ⫽ 0.001 M2-R2, p ⫽ 0.004 (BMC) B-M2, p ⫽ 0.019 M1-M2, p ⫽ 0.035 R1-M2, p ⫽ 0.009 (BMC)
(A) p ⫽ 0.317
NOTE. Significant pair-wise comparisons are shown for baseline (B), mode 1 (M1), rest 1(R1), mode 2 (M2), and rest 2 (R2). Abbreviation: BMC, Bonferroni multiple comparison post hoc test.
trol found SV at mode 1 to be higher than at baseline (p ⬍ 0.05). CCE increased from the baseline to mode 1, and rest 2 was lower than mode 1. There were no between-mode differences with respect to CO, systemic vascular resistance (SVR), PPV, and systolic blood pressure. SVV decreased from baseline to mode 1. PPV mirrored SVV, which decreased with respect to baseline (Table 3). The arterial blood gas changes are shown in Table 4. PO2 (in room air) decreased from baseline at rest 1 and rest 2. PO2 at mode 2 was higher than at rest 1 and rest 2. PCO2 at mode 2 was lower than during all the other modalities. Similarly, the pH level at mode 2 was higher than at baseline, mode 1, and rest 1. Lactate was unchanged throughout. In the 5 volunteers who smoked (3 men and 2 women), PO2 increased from rest 1 to mode 2 (Table 4). In the 5 volunteers who did not smoke (3 men and 2 women), no between-mode differences in PO2 occurred. DISCUSSION
Understanding the mechanism of CO changes associated with negative end-expiratory pressure requires a hemodynamic evaluation of healthy subjects,16 but finding a reliable method of hemodynamic monitoring acceptable to awake healthy subjects already has proven to be difficult for other investigators.16 The CO measurement method chosen in this study was the PRAM (MOSTCARE) system, which allows a noninvasive hemodynamic assessment accurately reflecting rapidly changing hemodynamic conditions but does not require central venous catheter insertion or the injection of fluid, drug, or dye. Because previously reported pulmonary artery catheter– based NPV-related hemodynamic changes in intubated cardiac surgery patients focused on HR, CO, SV, SVV, and PPV,17 it was important that PRAM should at least provide this information and, if possible, additional parameters of cardiac function. One such measure is the CCE.15 It represents the ratio between the hemodynamic work performed by the heart and energy expen-
diture. The underlying mathematic proof recently was reviewed.15 Briefly, an increase in CCE reflects a reduction in energy used by the cardiovascular system to maintain the same hemodynamic balance. Any improvement in CCE can be interpreted as an improvement in ventricular-arterial coupling. The authors found that NPV led to a significant decrease in the HR of volunteers but no decrease in CO. Because CO equals HR ⫻ SV, a decrease in HR accompanied by an unchanged CO suggests an expected increase in SV. However, regarding changes in SV, analysis of variance just missed significance at the 5% level. This means that, although the authors cannot claim statistical significance for the estimated marginal mean for mode 1 being higher than all the other time points, herein may lie a physiologic explanation as to why the significant decrease in HR was not accompanied by a decrease in CO. In defense of this argument, provided the caveat is acknowledged that the SV analysis of variance p value was 0.057, the Dunnett multiple comparison test of SV modalities versus baseline showed that SV at mode 1 was higher than at baseline. In contrast, earlier reports of intubated adult coronary artery bypass graft surgery patients ventilated with intermittent positive pressure ventilation showed that the addition of NPV did not alter HR, blood pressure, or gas exchange but led to significant increases in the SV index and cardiac index.17 These earlier reports of NPV maintaining the HR in intubated patients highlight the unique observation of the healthy volunteers developing a significant decrease in HR upon the commencement of a continuous negative mode, which was seen again during the control mode. The decrease in HR in the volunteers was most likely a response to increased preload and, thus, atrial stretch. The contrasting absence of a decrease in HR in postoperative intubated cardiac surgical patients17 uncovers an asyet-unnoticed difference between healthy and post– cardiac surgery hearts. This difference explains how an arguably
HEMODYNAMIC AND RESPIRATORY EFFECTS OF CUIRASS VENTILATION
important hemodynamic benefit of NPV has been hitherto overlooked in the literature, namely maintaining CO through increasing SV in the presence of a reduced HR, a combination of events that shows conservation of myocardial energy expenditure, which is reflected in the significantly increased CCE. NPV significantly improved CCE in the volunteers. It could be argued that the increased SV and CCE and the decreased HR should be of benefit in nonsurgical patients with ischemic heart disease or in any patient with a critical balance of myocardial oxygen supply and demand. In contrast to the earlier reports of NPV in intubated patients, the healthy volunteers did not have an increase in CO with NPV. However, it could be argued that any increase beyond their already normal baseline CO was unlikely to confer an additional physiologic benefit. The PPV in volunteers mirrored the SVV, which decreased with respect to baseline values (Table 3). The reason why the change in PPV was marginally significant whereas SVV was quite significant may be because PPV is determined by the height of the arterial waveforms, whereas SVV integrates the area under the curve. In fact, the PPV/SVV ratio in volunteers (Table 3) showed a decreasing trend. This could be explained by considering the hemodynamic variables describing arterial elastance studied by Pinsky18 in which the PPV/SVV estimated elastance, which is an indicator of arterial tone, whereas SVR reflected the arteriolar tone. The volunteers’ arterial tone tended to decrease with NPV (in the presence of an unchanged SVR) suggesting adequate fluid hydration before the induced intrathoracic blood accumulation with NPV. Interestingly, the opposite was noted in the present authors’ study of postoperative extubated cardiac surgery patients (part 2). The SVV in volunteers decreased in response to the continuous negative mode. The decreased SVV and the improved CO coupled with the increased SV may suggest an improved venous return. The magnitude of the SVV has been taken as a predictor of fluid responsiveness in mechanically ventilated patients.19-21 In spontaneously breathing subjects, the utility of the related measure of PPV and systolic pressure variation in determining fluid responsiveness has been questioned22 as has the use of SVV and PPV.23,24 The poor detection with pulse contour methods of the response to venous return during spontaneous breathing could be attributable to the poor sensitivity of the available technology in evaluating the differences in SVV and, particularly, in PPV. However, PRAM samples at 1,000 Hz, a frequency 5- to 10-fold higher than other devices, so a higher sensitivity and a better ability to detect small and rapidly occurring variations might be expected provided there is constant vigilance by the attending physician to check the morphology of the arterial wave to prevent over- or underdamping.25,26 This study was not designed to test the hypothesis that NPV-related changes in SVV predict fluid responsiveness. However, the present demonstration of PRAM’s ability to detect rapid changes in PPV measurements and the derived changes in SVV points to the need for such a hypothesis to be tested, which if confirmed would be a more easily applied tool in detecting fluid responsiveness than the inspiratory and expiratory valves suggested by Dahl et al22 in spontaneously breathing subjects. Mode 1 (continuous negative) brought about a significant change from baseline in HR, CCE, and SVV. Moreover, there
871
was a trend toward an increased SV. When looking at mode 2, it effected a significant change from baseline in HR but with trends (albeit not statistically significant) toward the same changes as observed in mode 1. This would suggest that the observed effects were more pronounced in mode 1 than mode 2 but that similar changes were occurring in both modes. Because the decrease in HR in mode 2 was less than in mode 1, the CCE improvements in mode 2 were limited. One explanation may be that continuous NPV in mode 1 provides a constant improvement in venous return, whereas this is interrupted in mode 2, reducing the effect on HR. A second possibility is that because hyperventilation in mode 2 inhibits the known bradycardic effect of the baroreflex in humans,27 the magnitude of the HR reduction in mode 2 is less pronounced than in mode 1. The present authors believe that this is the first time that NPV using a cuirass in healthy subjects has been evaluated for its hemodynamic effects. Overall, it was noted that the cuirass was well tolerated in the volunteers. NPV improved SV (p ⬎ 0.05) and reduced HR with CO staying the same, suggesting an improvement in myocardial oxygen supply/demand, which was reflected by the increased CCE. These findings may suggest a possible new application of NPV as a nonpharmacologic means of supporting the failing or ischemic heart in unsedated spontaneously breathing patients. Although the authors earlier documented the use of the cuirass in the continuous negative mode as an effective method of obviating reintubation in extubated cardiac surgery patients developing respiratory dysfunction, they wished to evaluate its ventilatory effects in healthy volunteers. In the volunteers, the baseline PCO2 and pH were in the normal range as would be expected in subjects not under analgesia and sedation. In the healthy volunteers, there was a significant decrease in PO2 at rest 1 and rest 2. This may reflect a degree of V/Q mismatch associated with the recumbent position used in the study. Interestingly, during the application of NPV in the volunteers, PO2 did not decrease significantly below baseline, suggesting that the atelectasis of recumbency was being countered by the NPV. Arguably, consideration of the alveolar gas equation may explain, at least in part, the increased PO2 between rest 1 and mode 2, where reduced PCO2 would allow more space for O2 as happens during accommodation at a high altitude. Of particular interest was the observation in the 5 healthy volunteers who were smokers that the control mode led to a significant increase in oxygenation compared with the preceding rest PO2. This was not the case with the nonsmoking volunteers. With the caveat that numbers were small, there may be a suggestion that the beneficial effects on the ventilation of NPV have particular application in smokers in whom elastic tissue degeneration may allow small airway closure, which becomes clinically significant after a period of recumbency. The control mode led to a marked decrease in PCO2 and a reciprocal increase in pH in the 10 volunteers, suggesting that this mode was causing hyperventilation. Limitations In mode 1, the cuirass rests motionless on the chest with minimal time needed for the subject to become accustomed to
872
MCBRIDE ET AL
the device. In contrast, in mode 2, the device moves with some minutes required for a first-time subject to become accustomed to this ventilation mode. It is possible that the 5 minutes allowed for each ventilation modality (including rests) were not long enough to allow acclimatization to mode 2. A criticism of the study is that the experimental order for modes 1 and 2 was not random. This is because it is standard clinical practice within the authors’ centers to allow a patient to become familiar with mode 1 (in which cuirass is immobile) before progressing to mode 2 (in which cuirass moves). The respiratory rate in the control mode (mode 2) was preset to a rate that led to hyperventilation, which may have limited any HR-reducing effect of this mode. A third ventilatory mode
of the control mode to normocarbia would have been helpful to determine if the significantly reduced HR in mode 1 also occurred in the control mode at normocarbia. Because PRAM was the only method of hemodynamic monitoring used in the study, all the limitations of PRAM apply, including inadvertent arterial pressure damping although care was taken throughout to prevent this. CONCLUSIONS
NPV in volunteers leads to improvement in hemodynamics. NPV decreases PCO2 with a reciprocal increase in pH and in the control mode improves oxygenation in smokers.
REFERENCES 1. Shekerdemian LS, Bush A, Shore DF, et al: Cardiorespiratory responses to negative pressure ventilation after tetralogy of Fallot repair: A hemodynamic tool for patients with a low-output state. J Am Coll Cardiol 33:549-555, 1999 2. Shekerdemian LS, Bush A, Shore DF, et al: Cardiopulmonary interactions after Fontan operations: Augmentation of cardiac output using negative pressure ventilation. Circulation 96:3934-3942, 1997 3. Trethowan B, Elliott P, Varadarajan B, et al: Continuous negative extra thoracic pressure in respiratory complications after cardiac surgery. Br J Anaesth 101:589P-590P, 2008 4. Sorbara C, Armellin G, Bonato R, et al: Postoperative sedation with propofol infusion: Haemodynamics and pharmacokinetics. Clin Drug Invest 16:431-439, 1998 5. Rozanski EA, Bedenice D, Lofgren J, et al: The effect of body position, sedation, and thoracic bandaging on functional residual capacity in healthy deep-chested dogs. Can J Vet Res 74:34-39, 2010 6. Imberger G, McIlroy D, Pace NL, et al: Positive end-expiratory pressure (PEEP) during anaesthesia for the prevention of mortality and postoperative pulmonary complications. Cochrane Database Syst Rev 9:CD007922, 2010 7. Romano SM, Pistolesi M: Assessment of cardiac output from systemic arterial pressure in humans. Crit Care Med 30:1834-1841, 2002 8. Giomarelli P, Biagioli B, Scolletta S: Cardiac output monitoring by pressure recording analytical method in cardiac surgery. Eur J Cardiothorac Surg 26:515-520, 2004 9. Scolletta S, Romano SM, Biagioli B, et al: Pressure recording analytical method (PRAM) for measurement of cardiac output during various haemodynamic states. Br J Anaesth 95:159-165, 2005 10. Romano SM, Scolletta S, Olivotto I, et al: Systemic arterial waveform analysis and assessment of blood flow during extracorporeal circulation. Perfusion 21:109-116, 2006 11. Scolletta S, Carlucci F, Tabucchi A: Cardiac cycle efficiency correlates with pro-B-type natriuretic peptide in cardiac surgery patients. Crit Care 12:249, 2008 12. Balderi T, Forfori F, Marra V, et al: Continuous hemodynamic monitoring during laparoscopic gastric bypass in superobese patients by pressure recording analytical method. Obes Surg 18:1007-1014, 2008 13. Modesti PA, Gamberi T, Bazzini C, et al: Response of serum proteome in patients undergoing infrarenal aortic aneurysm repair. Anesthesiology 111:844-854, 2009
14. Giglioli C, Landi D, Gensini GF, et al: Cardiac efficiency improvement after slow continuous ultrafiltration is assessed by beatto-beat minimally invasive monitoring in congestive heart failure patients: A preliminary report. Blood Purif 29:44-51, 2010 15. Romano SM: Cardiac cycle efficiency: A new parameter able to fully evaluate the dynamic interplay of the cardiovascular system. Int J Cardiol 155:326-327, 2012 16. Fellahi JL, Caille V, Charron C, et al: Noninvasive assessment of cardiac index in healthy volunteers: A comparison between thoracic impedance cardiography and Doppler echocardiography. Anesth Analg 108:1553-1559, 2009 17. Chaturvedi RK, Zidulka AA, Goldberg P, et al: Use of negative extrathoracic pressure to improve hemodynamics after cardiac surgery. Ann Thorac Surg 85:1355-1360, 2008 18. Pinsky MR: The functional haemodynamic monitoring: applied physiology at the bedside, In: Yearbook of emergency and intensive care medicine 537-552, 2001 19. Berkenstadt H, Margalit N, Hadani M, et al: Stroke volume variation as a predictor of fluid responsiveness in patients undergoing brain surgery. Anesth Analg 92:984-989, 2001 20. Michard F: Changes in arterial pressure during mechanical ventilation. Anesthesiology 103:419-428, 2005 21. Reuter DA, Goetz AE, Peter K: Assessment of volume responsiveness in mechanically ventilated patients. Anaesthesist 52:10051007, 1010-1013, 2003 22. Dahl MK, Vistisen ST, Koefoed-Nielsen J, et al: Using an expiratory resistor, arterial pulse pressure variations predict fluid responsiveness during spontaneous breathing: An experimental porcine study. Crit Care 13:R39, 2009 23. Magder S: Predicting volume responsiveness in spontaneously breathing patients: Still a challenging problem. Crit Care 10:165, 2006 24. Soubrier S, Saulnier F, Hubert H, et al: Can dynamic indicators help the prediction of fluid responsiveness in spontaneously breathing critically ill patients? Intensive Care Med 33:1117-1124, 2007 25. Shinozaki T, Deane RS, Mazuzan JE: The dynamic responses of liquid-filled catheter systems for direct measurements of blood pressure. Anesthesiology 53:498-504, 1980 26. Gardner RM: Direct blood pressure measurement—Dynamic response requirements. Anesthesiology 54:227-236, 1981 27. Van de Borne P, Mezzetti S, Montano N, et al: Hyperventilation alters baroreflex control of heart rate and muscle sympathetic nerve activity. Am J Physiol 279:536-541, 2000