Experimental Physiology

Transcription

1 Exp Physiol (2014) pp Research Paper Research Paper Respiratory pump contributes to increased physiological reserve for compensation during simulated haemorrhage Paula Y. S. Poh 1, Robert Carter III 1, Carmen Hinojosa-Laborde 1, Jane Mulligan 2,GregoryZ.Grudic 2 and Victor A. Convertino 1 1 US Army Institute of Surgical Research, JBSA Fort Sam Houston, TX 78234, USA 2 Flashback Technologies Inc., Boulder, CO 80301, USA Experimental Physiology New Findings What is the central question of this study? The negative intrathoracic pressure created by inspiration (i.e. the respiratory pump) is associated with enhanced venous return, although its contribution as a potential mechanism of compensation during blood loss has not been quantified. What is the main finding and its importance? We demonstrated that optimizing the creation of negative intrathoracic pressure during inspiration represented a mechanism of compensation during haemorrhage that was quantifiably reflected in a reduced rate of diminution of the compensatory reserve, resulting in increased tolerance to progressive reductions in central blood volume. Intrathoracic pressure regulation (IPR) represents a therapy for increasing systemic circulation through the creation of negative intrathoracic pressure. We hypothesized that using this respiratory pump effect would slow the diminution of the physiological reserve to compensate during progressive reductions in central blood volume. The compensatory reserve index (CRI) algorithm was used to measure the proportion (from 100 to 0%) of reserve capacity that remained to compensate for central volume loss before the onset of cardiovascular decompensation. Continuous analog recordings of arterial waveforms were extracted from data files of seven healthy volunteers. Subjects had previously participated in experiments designed to induce haemodynamic decompensation (presyncope) by progressive reduction in central blood volume using graded lower-body negative pressure. The lower-body negative pressure protocol was completed while breathing spontaneously through a standard medical face mask without (placebo) and with a resistance (approximately 7cmH 2 O; active IPR) applied during inspiration. At the onset of presyncope in the placebo conditions, CRI was smaller than the CRI observed at the same time point in the active IPR conditions. The CRI at the onset of presyncope during active IPR (0.08 ± 0.01) was similar to the CRI at presyncope with placebo. Kaplan Meier and log rank tests indicated that CRI survival curves were shifted to the right by active IPR. Optimizing the respiratory pump contributed a small but significant effect of increasing tolerance to progressive reductions in central blood volume by extending the compensatory reserve. (Received 10 June 2014; accepted after revision 1 July 2014; first published online 11 July 2014) Corresponding author V. A. Convertino: US Army Institute of Surgical Research, JBSA Fort Sam Houston, TX 78234, USA. victor.a.convertino.civ@mail.mil DOI: /expphysiol

2 1422 P. Y. S. Poh and others Exp Physiol (2014) pp Introduction Haemorrhage from major trauma is a leading cause of death in both civilian and battlefield settings (Søreide et al. 2007; Eastridge et al. 2011). From the point of injury until delivery to a hospital, maintenance of vital organ perfusion becomes imperative when life-saving interventions and therapies are not readily available (Convertino et al. 2004, 2007). Intrathoracic pressure regulation (IPR) therapy can be used during prehospital care to enhance tissue circulation (Convertino et al. 2011a) by optimizing the reduction in negative intrathoracic pressure during inspiration, resulting in increased venous return, stroke volume and cardiac output (Convertino et al. 2004, 2011b). The impact of IPR therapy was most apparent during severe reductions in central blood volume, when the onset of imminent haemodynamic decompensation and presyncopal symptoms was delayed, with improved systemic haemodynamics (Convertino et al. 2004, 2007, 2011b;Rickardsetal. 2007; Ryan et al. 2008). By optimizing the respiratory pump in order to enhance circulation, IPR may represent a therapy for increasing the reserve of physiological mechanisms to compensate for reduced circulating blood volume. However, we are unaware of any data to support such a notion. Compensatory mechanisms (e.g. tachycardia, vasoconstriction and respiration) are activated by a reduction in central blood volume in an effort to maintain perfusion pressure during haemorrhage. We recently introduced the concept of the compensatory reserve as a way to express the proportion of maximal capacity that remains to compensate for central blood volume loss before the onset of decompensation (Convertino et al. 2013). We subsequently developed a novel mathematical algorithm to measure this reserve, called the compensatory reserve index (CRI), which represents the integrated capacity of all compensatory mechanisms (e.g. blood pressure, heart rate) of any individual to maintain adequate systemic tissue perfusion and thus avoid decompensation (Convertino et al. 2013; Mouton et al. 2013; Van Sickle et al. 2013). The CRI algorithm uses feature-extraction and machine-learning techniques capable of real-time analysis and comparison of subtle changes in arterial waveforms obtained from human subjects during progressive reductions in central blood volume that lead to haemodynamic decompensation (Convertino et al. 2013). If the respiratory pump represents a mechanism that contributes to the total compensation in response to blood loss, then optimization of the creation of negative intrathoracic pressure during inspiration should be reflected in a lower rate of diminution of the compensatory reserve during a progressive reduction in central blood volume. In the present study, we estimated the CRI from retrospective analysis of waveforms obtained from published results that demonstrated that the use of active IPR delayed symptoms and increased tolerance to progressively reduced central blood volume (Convertino et al. 2007; Rickards et al. 2008; Ryan et al. 2008). Specifically, we hypothesized that the delayed onset of symptoms and time to intolerance previously reported would be reflected in a slower reduction of the CRI during application of IPR. Methods Calculation of the CRI As described in detail (Convertino et al. 2013), state-of-the-art feature-extraction and machine-learning techniques were used to process arterial waveforms collectively. Briefly, feature extraction is a form of dimensionality reduction that may be used to facilitate pattern recognition in image and signal processing. Machine learning is concerned with the design and development of models that can be used to extract and integrate task-relevant information automatically. The combination of these analytical technologies provided a unique computational tool to make sense rapidly of very large data sets of waveforms. For clinical simplicity, the CRI was normalized on a scale of 1 0 (100 0%, respectively), where 1 reflected the maximal capacity of the sum total of physiological mechanisms (e.g. baroreflexes, respiration) to compensate for relative deficits in central blood volume, and 0 implied imminent cardiovascular instability and decompensation. Values between 1 and 0 indicated the proportion of compensatory reserve remaining. The model calculated the first CRI value estimate after 30 heartbeats of initialization and then, in real time, provided a new CRI estimate after every subsequent heartbeat. Overview of experimental design All experimental procedures were conducted under protocols reviewed and approved by the Brooke Army Medical Center Institutional Review Board and the Institutional Review Board of the Office of Human Research Protection under the US Army Medical Research and Materiel Command, and conducted in accordance with the approved protocols. From a data base of >250 subjects, continuous analog recordings of arterial waveforms were extracted from the data files of only seven healthy, normotensive, non-smoking men and women who repeated exposures of graded lower-body negative pressure (LBNP) while breathing with and without respiratory resistance and had arterial waveform recordings that could be analysed using the CRI algorithm. In an effort to ensure that this sample

3 Exp Physiol (2014) pp Respiration and compensatory reserve 1423 size was adequate, we performed power and sample size calculations using the following parameters: μ(0) (i.e. known mean value) = 0.13; μ(1) expected mean value = 0.08; σ (sample standard deviation) = 0.05; α = 0.05; one-sided test; power = This power analysis revealed that the sample size estimate (seven) was adequate. These experiments were designed to cause progressive reductions in central blood volume similar to those induced by haemorrhage (Ryan et al. 2008). Given that the CRI had not been developed at the time of the experiments on these seven subjects, this investigation represents a retrospective analysis. Subjects were placed in the supine position within an airtight chamber and sealed at the level of the iliac crest by way of a neoprene skirt, resulting in a redistribution of blood away from the upper body (head and heart) to the abdomen and lower extremities. All subjects were instrumented with an infrared finger plethysmography blood pressure monitor (Finometer R, TNOTPD; Biomedical Instrumentation, Amsterdam, The Netherlands) for the non-invasive, continuous measurement of arterial pulse waves. An appropriately sized cuff was placed on the middle finger of the left hand, which was laid at heart level. The LBNP protocol consisted of a 5 min rest period (baseline) followed by 5 min of chamber decompression at 15, 30, 45 and 60 mmhg, and then additional increments of 10 mmhg every 5 min until the onset of haemodynamic decompensation or the completion of 5minat 100 mmhg. Haemodynamic decompensation was defined by a progressive diminution of systolic blood pressure below 70 mmhg and/or voluntary subject termination due to discomfort from symptoms such as sweating, nausea or dizziness. As previously described (Ryan et al. 2008), subjects were exposed to two LBNP protocols, separated by at least 1 week, while spontaneously breathing through a standard medical face mask with an impedance threshold device (ITD; Advanced Circulatory Systems, Inc., Eden Prairie, MN, USA). One LBNP exposure was without (placebo) and one exposure was with a resistance applied during inspiration (IPR therapy) that was set at approximately 7 cmh 2 O. The ITD devices (placebo or IPR therapy) were placed on the subjects at an identified LBNP level that was required to produce haemodynamic decompensation (Ryan et al. 2008). The order of the placebo or IPR therapy treatments was randomized; all experimental sessions for a given subject were initiated at the same time of day. Statistical analysis The CRI was averaged for the last 3 min of each LBNP stage in both experimental conditions. To determine the rate of change (CRI per minute), the averaged CRI was divided by the length of time (3 min). The 95% confidence interval (CI) was calculated to provide information about the interval computed from the sample data (seven subjects) which would contain the true effect 95% of the time if the study were repeated multiple times using the same experimental material. Pearson correlation and linear regression analyses were used to detect the relationship between systolic blood pressure (SBP) and CRI/min. Kaplan Meier analysis with a log rank test were used to estimate the number of surviving subjects for every LBNP level after IPR therapy was administered. The survival probability at each LBNP stage was calculated by dividing the survival count by the population, and then multiplying it by the previous estimations. The test statistic was calculated as follows where O is defined as observed frequency and E is defined as the expected (theoretical) frequency, asserted by the null hypothesis. χ 2 (log rank) = (O 1 E 1 ) 2 / E 1 + (O 2 E 2 ) 2 /E 2. Unless otherwise stated, all data are expressed as 95% CIs. Results As reported previously (Ryan et al. 2008), breathing with IPR therapy increased LBNP tolerance time by an average of 13% compared with placebo (mean, versus min, respectively). At presyncope, mean (±95% CI) systolic blood pressure (SBP) with placebo (58 ± 11 mmhg) was not statistically indistinguishable from SBP with IPR therapy (71 ± 10 mmhg). Of the seven subjects, LBNP with placebo was terminated at 70 mmhg in one subject, 80 mmhg in three subjects, 90 mmhg in two subjects and 100 mmhg in one subject. As shown in Fig. 1, the mean (±95% CI) CRI before IPR administration was the same between placebo (0.30 ± 0.17) and IPR therapy (0.30 ± 0.16) experimental conditions. At the onset of presyncope with placebo, the mean (±95% CI) CRI (0.07 ± 0.01) was smaller than the CRI observed at the same time point with IPR therapy (0.10 ± 0.01); i.e. 95% of the time, the highest value in the 95% CI at the onset of presyncope (0.08) with the placebo can be expected to be lower than the lowest value in the 95% CI for CRI at the same time point with IPR therapy (0.09). The mean (±95% CI) CRI at the onset of presyncope with IPR therapy (0.07 ± 0.01) was statistically indistinguishable from the CRI at presyncope during the placebo trial. Given that the average time from application of IPR therapy to the time of presyncope was longer, the mean (±95% CI) rate of change (in CRI units per minute) was slower with the IPR therapy (0.008 ± 0.002) than with placebo (without IPR therapy; ± 0.009); i.e. 95% of the time, the highest value in the 95% CI for the CRI rate of change (0.10) with IPR therapy can be expected to be lower than the lowest value in the 95% CI for CRI rate of change (0.11) with the placebo.

4 1424 P. Y. S. Poh and others Exp Physiol (2014) pp In order to delineate further the relationship of CRI with and without IPR therapy, further statistical analyses were conducted. Based on the Pearson correlation and linear regression analyses, we found that the faster rate of change (in CRI units per minute) with placebo was correlated (r 2 = 0.69) with a greater reduction in SBP during the last 3 min prior to presyncope (Fig. 2). Alternatively, at the same time point when presyncope occurred during the Figure 1. Compensatory reserve index (CRI) before (Pre) and at the time of presyncope (PS) while breathing without (placebo) and with resistance during inspiration (IPR therapy) Values represent means (bars) and 95% confidence intervals ( T lines). denotes differences between the PS Sham IPR and Active IPR at PS Sham IPR (based on 95% CI). Systolic blood pressure (mmhg) Rate of change (CRI units/min -1 ) 0.10 Figure 2. Relationship between the rate of change in the CRI and systolic blood pressure Values represent plots of individual subjects while breathing without resistance during inspiration (placebo; filled circles) and with resistance during inspiration (IPR therapy; open circles). placebo conditions, the correlation coefficient between the rate of degradation in CRI and SBP with IPR therapy was r 2 = Moreover, the Kaplan Meier and log rank tests indicated that CRI survival curves were shifted to the right (P = 0.04) by IPR therapy (Fig. 3). Discussion The use of IPR therapy delayed symptoms and increased tolerance to progressively reduced central blood volume (Convertino et al. 2007; Ryan et al. 2008). The use of arterial waveforms obtained from these subjects provided us with the first opportunity to quantify the impact of optimizing the respiratory pump during progressive simulated haemorrhage. We hypothesized that the delayed onset of symptoms and time to intolerance would be reflected in a slower reduction in the compensatory reserve. Our results supported our hypothesis by demonstrating that the 13% increase in average tolerance to central hypovolaemia was associated with breathing on IPR therapy (Convertino et al. 2004). Thiswasreflectedbyafourfoldreductionintheaverage rate of decrease in the CRI during progressive reductions in central blood volume in the present study. The reserve to compensate for the reduction in central circulating blood volume is defined by the difference between the initial and maximal physiological levels of any specific response (e.g. tachycardia, vasoconstriction) at haemodynamic decompensation (Engelke et al. 1996). The average CRI was 0.30 prior to application of IPR Probability of completion (%) Pre-IPR Time Placebo IPR Therapy Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Stage 6 Figure 3. Probability of completion of lower-body negative pressure stage for subjects treated with IPR therapy (continuous line) and placebo resistance during inspiration (dashed line).

5 Exp Physiol (2014) pp Respiration and compensatory reserve 1425 therapy at the same average level of LBNP in both IPR therapy and placebo experiments (Fig. 1), indicating the reproducibility of the CRI measurement. Given that maximal physiological responses are finite (Convertino et al. 2005), we predicted that the CRI would be equal at the time of decompensation in both IPR therapy and placebo conditions. The results supported this hypothesis and the corollarythatreducingtherateofcridegradationby optimizing the respiratory pump buys time by extending the compensatory reserve and consequently increasing tolerance. As illustrated by the Kaplan Meier survival curves (Fig. 3), IPR therapy increased sustainment of central blood volume loss before collapse. Specifically, the vertical gap between the two curves depicts that at a specific central blood volume loss, IPR therapy had a greater percentage of subjects completing each LBNP stage postintervention (placebo or IPR therapy). For example, at the fourth LBNP stage post-ipr therapy there were 37% completers versus placebo, 8% completers. The horizontal gap between the two curves illustrates that it took longer for subjects with IPR therapy to experience the onset of haemodynamic decompensation. Given that haemorrhage is simulated by LBNP, the results of the present study are translatable to effective non-invasive treatment of bleeding trauma patients and combat casualties in the prehospital setting (Hinojosa-Laborde et al. 2013). The Kaplan Meier survival curve for placebo conditions may be viewed as the current standard of care (no IPR therapy). In contrast, the IPR therapy curve may represent a novel form of a stand-in or adjunct therapy designed to optimize the respiratory pump in a way that could potentially increase the likelihood of survival by extending the time until emergency personnel arrive. Breathing with IPR therapy was able to delay cardiovascular collapse and increase the number of survivors at each LBNP stage postintervention. Clinically, the use of IPR therapy and CRI together could provide a novel combination of a therapeutic intervention designed to optimize the compensatory mechanism of breathing and a diagnostic assessment of the effectiveness of the therapy to delay the onset of system collapse in haemorrhaging patients. The total integrated reserve required to compensate for circulatory compromise involved various mechanisms that were activated by reductions in central blood volume in an effort to maintain adequate perfusion to vital organs during haemorrhage. As illustrated by Fig. 2, a faster rate of compensatory reserve diminution leads to greater development of hypotension. When the respiratory pump was optimized by breathing with IPR therapy, the CRI diminution led to consumption of the physiological reserve at a slower rate and consequently led to maintenance of SBP at the same time point as placebo presyncope. The extended CRI and longer LBNP tolerance induced by IPR therapy in the present investigation indicated that the respiratory pump contributes a small but significant part of the total compensatory reserve. Perhaps more revealing is that a hyperventilatory response reported in the very late stages of compensation to progressive simulated haemorrhage does not become evident until >80% of tolerable central blood volume reduction and the presence of severe hypotension (Convertino et al. 2009). Taken together, the evidence provided by the present and previous investigations may reflect a requirement for a hypotensive stimulus that partly explains the late effect of IPR therapy on restoring compensatory reserve. In summary, the findings of the present study were consistent with the notion that optimizing the respiratory pump vacuum during simulated haemorrhage can represent an important, albeit late, compensatory mechanism to defend cerebral perfusion pressure and blood flow. Spontaneous gasping has been associated with improved cardiovascular function and increased cerebral blood flow in a fatal haemorrhagic shock rodent model (Suzuki et al. 2009), and is reported to occur frequently in cardiac arrest patients, where it has been associated with successful clinical outcomes (Kentsch et al. 1990; Manole & Hickey, 2006; Bobrow et al. 2008). As such, the use of the CRI with IPR therapy in the present investigation provided unique evidence that the respiratory pump acts as an inherent mechanism for increasing the physiological reserve to compensate for reduced circulating blood volume. References BobrowBJ,ZuercherM,EwyGA,ClarkL,ChikaniV,Donahue D, Sanders AB, Hilwig RW, Berg RA & Kern KB (2008). Gasping during cardiac arrest in humans is frequent and associated with improved survival. Circulation 118, Convertino VA, Cooke WH & Lurie KG (2005). Inspiratory resistance as a potential treatment for orthostatic intolerance and hemorrhagic shock. Aviat Space Environ Med 76, Convertino VA, Grudic GZ, Mulligan J & Moulton SL (2013). Estimation of individual-specific progression to impending cardiovascular instability using arterial waveforms. JAppl Physiol 115, Convertino VA, Moulton SL, Grudic GZ, Rickards CA, Hinojosa-Laborde C, Gerhardt RT, Blackbourne LH & Ryan KL (2011a). Use of advanced machine-learning techniques for noninvasive monitoring of hemorrhage. J Trauma 71, S25 S32. Convertino VA, Ratliff DA, Ryan KL, Doerr DF, Ludwig DA, Muniz GW, Britton DL, Clah SD, Fernald KB, Ruiz AF, Lurie KG & Idris AH (2004). Hemodynamics associated with breathing through an inspiratory impedance threshold device in human volunteers. Crit Care Med 32, S381 S386.

6 1426 P. Y. S. Poh and others Exp Physiol (2014) pp Convertino VA, Rickards CA, Lurie KG & Ryan KL (2009). Hyperventilation in response to progressive reduction in central blood volume to near syncope. Aviat Space Environ Med 80, Convertino VA, Ryan KL, Rickards CA, Cooke WH, Idris AH, Metzger A, Holcomb JB, Adams BD & Lurie KG (2007). Inspiratory resistance maintains arterial pressure during central hypovolemia: implications for treatment of patients with severe hemorrhage. Crit Care Med 35, Convertino VA, Ryan KL, Rickards CA, Glorsky SL, Idris AH, Yannopoulos D, Metzger A & Lurie KG (2011b). Optimizing the respiratory pump: harnessing inspiratory resistance to treat systemic hypotension. Respir Care 56, Eastridge BJ, Hardin M, Cantrell J, Oetjen-Gerdes L, Zubko T, Mallak C, Wade CE, Simmons J, Mace J, Mabry R, Bolenbaucher R & Blackbourne LH (2011). Died of wounds on the battlefield: casuation and implications for improving combat casualty care. J Trauma 71, S4 S8. Engelke KA, Halliwill JR, Proctor DN, Dietz NM & Joyner MJ (1996). Contribution of nitric oxide and prostiglandins to reactive hyperemia in human forearm. JApplPhysiol81, Hinojosa-Laborde C, Shade RE, Muniz GW, Bauer C, Goei KA, Pidcoke HF, Chung KK, Cap AP & Convertino VA (2013). Validation of lower body negative pressure as an experimental model of hemorrhage. JApplPhysiol116, Kentsch M, Stendel M & Mueller-Esch G (1990). Early prediction of prognosis in out-of-hospital cardiac arrest. Intensive Care Med 16, Manole SL & Hickey RW (2006). Preterminal gasping and effects on the cardiac function. Crit Care Med 34, S438-S441. Mouton SL, Mulligan J, Grudic GZ & Convertino VA (2013). Running on empty? The compensatory reserve index. J Trauma 75, Rickards CA, Cohen KD, Bergeron LL, Burton L, Khatri PJ, Lee CT, Ryan KL, Cooke WH, Doerr DF, Lurie KG & Convertino VA (2008). Inspiratory resistance, cerebral blood flow velocity, and symptoms of acute hypotension. Aviat Space Environ Med 79, RickardsCA,RyanKL,CookeWH,LurieKG&ConvertinoVA (2007). Inspiratory resistance delays the reporting of symptoms with central hypovolemia: association with cerebral blood flow. Am J Physiol Regul Integr Comp Physiol 293, R243 R250. Ryan KL, Cooke WH, Rickards CA, Lurie KG & Convertino VA (2008). Breathing through an inspiratory threshold device improves stroke volume during central hypovolemia in humans. JApplPhysiol104, Søreide K, Krüger AJ, Vårdal AL, Ellingsen CL, Søreide E & Lossius HM (2007). Epidemiology and contemporary patterns of trauma deaths: changing place, similar pace, older face. World J Surgery 31, Suzuki M, Funabiki T, Hori S & Aikawa N (2009). Spontaneous gasping increases cerebral blood flow during untreated fatal hemorrhagic shock. Resuscitation 80, Van Sickle C, Schafer K, Mulligan J, Grudic GZ, Moulton SL & Convertino VA (2013). A sensitive shock index for real-time patient assessment during simulated hemorrhage. Aviat Space Environ Med 84, Additional Information Competing interests Drs Grudic and Mulligan are co-founders of Flashback Technologies Inc., developers of the CRI model used in this study. Author contributions All authors contributed to study implementation and writing of the manuscript. Paula Y.S. Poh conducted the literature search, and data analysis. She wrote the manuscript and produced the figures. Gregory Grudic and Jane Mulligan contributed to discussions about study design and data interpretation and developed the software-based algorithm. Carmen Hinojosa-Laborde, Robert Carter III, and Victor Convertino conducted the LBNP experiments, produced the database of results, and were responsible for data interpretation. Drs. Hinojosa-Laborde and Carter supervised the data analysis and contributed to the writing of the manuscript. Dr. Convertino designed the study, conducted a critical review and produced the final revision of the manuscript. Funding Funding support was provided by the United States Army Medical Research and Materiel Command Combat Casualty Research Program. Disclaimer The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense.