Curry, Timothy B.

We tested the hypothesis that erythrocyte count (RBC), hemoglobin (Hgb), and hematocrit (Hct) would be similar during lower body negative pressure (LBNP) and graded blood loss (BL) in humans. Arterial blood pressure, central venous pressure and heart rate were measured in 12 healthy males (88 ± 3 kg) during LBNP and BL. The LBNP protocol consisted of 5 min stages at 0,‐15, ‐30, and ‐45 mmHg. The BL protocol consisted of removal of three 333 mL aliquots of blood (1 L total) which was stored for reinfusion. RBC, Hgb, and Hct were obtained via arterial catheter at baseline and at the end of the final stage for both LBNP and BL. There were no differences (P > 0.05) in baseline values between treatment groups for RBC, Hgb, or Hct. RBC (4.81 ± 0.10 to 4.97 ± 0.10 10 12 /L), Hgb (14.18 ± 0.39 to 14.68 ± 0.36 g/dL), and Hct (40.88 ± 0.82 to 42.36 ± 0.83 %) increased from baseline to final stage (p<0.001 for all) during LBNP, while all values decreased from baseline to final stage (p<0.015 for all) during BL (4.82 ± 0.11 to 4.72 ± 0.11 10 12 /L; 14.33 ± 0.37 to 13.95 ± 0.39 g/dL; 41.09 ± 0.82 to 40.25 ± 0.86 %; respectively). The final stage values were greater in LBNP than BL for RBC, Hgb, and Hct (p< 0.001 for all). LBNP results in relative hemoconcentration while BL results in relative hemodilution. The clinical significance of these findings remains to be elucidated. Grant Funding Source : Supported by U.S. Army MRMC Combat Casualty Care Research Program Grant # W81XWH‐11‐1‐0823.

We tested the hypothesis that the catecholamine response would be similar during lower body negative pressure (LBNP) vs. blood loss (BL). Twelve men underwent randomized LBNP and BL protocols on the same day. LBNP consisted of 5‐minute stages at 0, ‐15, ‐30, and ‐45 mmHg of pressure. The BL protocol included 5 minutes at baseline and then after three stages of 333 ml of whole blood draws (1000 ml total) which were stored for reinfusion. Central venous pressure (CVP) was recorded during each protocol. Blood samples were collected at baseline and at the end of the last stage of each protocol. CVP was greater in LBNP vs. BL at baseline (7.3 ± 0.6 vs. 6.1 ± 0.6 mmHg; P < 0.01) and was lower during LBNP vs. BL (‐0.2 ± 0.6 vs. 1.8 ± 0.8 mmHg; P < 0.01). The increase in epinephrine (EPI) during LBNP (53 ± 7 vs.144 ± 30 pg/ml) and BL (49 ± 7 vs. 103 ± 19 pg/ml; P < 0.01) was similar (P = 0.16). An increase in norepinephrine (NE) during LBNP (148 ± 20 vs. 354 ± 44 pg/ml; P < 0.01) was greater (P < 0.01) than the trend to increase during BL (155 ± 22 vs. 211 ± 29 pg/ml; P = 0.09). Individual slopes of catecholamines vs. CVP were calculated to determine if the responses had similar trajectories between the two protocols. The EPI vs. CVP slope was steeper in LBNP vs. BL (‐13.5 ± 4.6 vs. ‐0.6 ± 0.1 pg/ml/mmHg; P < 0.01). The NE vs. CVP slope was steeper in LBNP vs. BL (‐27.8 ± 5.6 vs. ‐8.7 ± 5.9 pg/ml/mmHg; P = 0.01). LBNP to ‐45 mmHg appears to be a greater sympathetic stressor than 1000 ml of BL. Grant Funding Source : Supported by US Army MRMC Combat Casualty Care Research Program Grant # W81XWH‐11‐1‐0823

Increased locomotor muscle afferent feedback may be linked to the reduced exercise capacity (peak VO 2 ) and poor ventilatory efficiency (high V E /VCO 2 slope) observed in heart failure (HF). We examined the relationship between peak VO 2 and change in V E /VCO 2 with afferent inhibition during exercise in HF. 9 HF (60±6 yr) and 9 controls (CTL) (63±7 yr) completed 3 visits. Visit 1: cycle exercise to volitional exhaustion. Visits 2‐ 3: 5 min constant‐load cycling (65% of peak work) randomized to lumbar intrathecal fentanyl injection (afferent block) or placebo. VO 2 , ventilation (V E ), carbon dioxide production (VCO 2 ), and end‐tidal CO 2 (P ET CO 2 ) were measured. In HF, peak work (120±34 vs 176±54 W, p <0.05) and VO 2 (18.4±2.7 vs 27.2±4.7 ml/kg/min, p <0.05) were reduced, whereas V E /VCO 2 slope was elevated (39.7±1.8 vs 35.3±0.8, p<0.05) compared with CTL. With afferent inhibition, HF patients had a greater reduction in V E /VCO 2 (‐18.8±8.1 vs ‐5.4±13.3%, p <0.05) and increase in P ET CO 2 (20.3±10.5 vs 6.1±9.5%, p <0.01) during constant‐load cycling. Lower VO 2peak was related to a greater decline in V E /VCO 2 and rise in P ET CO 2 after afferent inhibition in HF (r=0.82, p <0.01 and r=‐0.69, p <0.05, respectively) but not CTL ( p >0.05). Our data demonstrate that patients with the lowest VO 2peak have the greatest improvement in V E /VCO 2 , suggesting a close link between afferent feedback, ventilatory efficiency and exercise capacity in HF. Grant Funding Source : Supported by AHA 12GRNT11630027 and National Center for Advancing Translational Science KL2TR000136.

The relationship between mean arterial pressure (MAP), muscle sympathetic nerve activity (MSNA), and total peripheral resistance (TPR) is altered in young women versus men and postmenopausal women due to endogenous sex hormone differences. Exogenous sex hormones in the form of oral contraceptive pills (OCPs) may further affect this relationship. We performed a retrospective review of 136 premenopausal women in whom MSNA and arterial pressure were measured, including 78 women with natural menstrual cycles in the early follicular phase (26±1 yr) and 58 women in the placebo phase of OCPs (24±1 yr). TPR and cardiac output (CO) were measured in a subset of women in whom continuous intra‐arterial blood pressure was recorded. Compared to those with natural menstrual cycles, women on OCPs had greater MAP (89±1 vs. 84±1 mmHg, p<0.01) and tended to have lower MSNA (15±1 vs. 18±1 bursts/100 heart beats, p=0.06). MAP and MSNA were positively correlated (r=0.22, p<0.05), TPR and MSNA were positively associated (r=0.41, p<0.05) and CO and MSNA (r=‐0.44, p<0.05) were inversely associated in naturally cycling women (n=26) but not in women on OCPs (r=0.15, r=0.11, r=‐0.01, respectively; p>0.05; n=38). In contrast to naturally cycling women, there appears to be no relationship between MSNA and hemodynamics in women on OCPs, suggesting that these women may regulate their blood pressure differently than women. Grant Funding Source : Supported by NIH

Objective: Given recent evidence from rodents supporting a novel role for the carotid bodies in insulin‐mediated sympathoexcitation, we sought to determine whether carotid chemoreceptor desensitization would reduce muscle sympathetic nerve activity (MSNA) during hyperinsulinemia in humans. Methods: Young, healthy subjects (n=4, 30±2 yrs) breathed hyperoxic gas (End‐tidal O2 = 100%, PaO2 505±13 mmHg) for 15 min at baseline and during steady‐state hyperinsulinemia (5±1 vs 53±1 µU/mL). Hyperinsulinemia was achieved using a 90‐min hyperinsulinemic (1 mU/kg FFM/min), euglycemic (108±4 mg/dL) clamp. CO2 was maintained at resting levels (PaCO2 41±1 mmHg). Changes in heart rate, blood pressure, minute ventilation, and MSNA (microneurography of the peroneal nerve) were assessed. Results: Hyperinsulinemia increased MSNA [burst frequency (p<0.01), burst incidence (p=0.02), total activity (p=0.02)]. Hyperoxia decreased heart rate (p=0.01) and minute ventilation (p=0.11), and increased mean blood pressure (p=0.04). Consistent with our hypothesis, the effect of hyperoxia on MSNA (Δ) was different between baseline and hyperinsulinemia (burst frequency, p=0.03; burst incidence, p=0.10). Conclusion: These data are the first to suggest the carotid chemoreceptors may contribute to insulin‐mediated sympathoexcitation in humans. Grant Funding Source : Supported by: NIH R01 DK090541 (MJJ), NIH T32 DK07352 (JKL)

We examined the relationship between total peripheral resistance (TPR) and central artery pressure and indices of wave reflection in 16 subjects (10F/6M; 26 ± 1 yr). Non‐invasive aortic pressure waveforms were synthesized from high‐fidelity radial pressure waveforms obtained from applanation tonometry of the radial artery. Radial blood pressure (BP) was calibrated from brachial artery BP (catheter). Central artery pressures, aortic augmentation index (AI a ), roundtrip travel time of the reflected wave to and from the periphery (Δt p ; inverse of pulse wave velocity), and wasted left ventricular energy (E w ) were calculated from the generated aortic pressure waveform. Cardiac output (CO) was measured using an open‐circuit acetylene uptake technique and used to calculate total peripheral resistance (mean arterial pressure/CO). TPR was positively related to aortic pulse pressure (r = 0.49), augmented pressure of the central artery wave (r = 0.60), pulse pressure amplification (r =0.54), AI a (r = 0.53), and E w (r = 0.66), (P < 0.05 for all). Additionally, TPR was inversely related to Δt p (r = − 0.51, P < 0.05). Our data suggests that increased levels of TPR are related to measures of central artery pressure, wave reflection characteristics and indices of central artery stiffness. Supported by NIH K23 DK‐082424 (TBC), AHA 2170087(ECH) and CTSA RR‐024150

We examined the effects of acute β‐adrenergic blockade on central artery hemodynamics and wave reflection characteristics in young healthy humans. High‐fidelity radial artery waveforms were recorded non‐invasively by applanation tonometry before and after intravenous propranolol (0.25 mg/kg bolus for 5 min followed by a maintenance infusion of 0.004/mg/kg/min for 20–30 min) in fourteen subjects (9 F/5M; 27 ± 2 years). Radial blood pressure (BP) was calibrated from the brachial BP (arterial catheter). Aortic augmentation index (AI a ; a measure of amplitude and timing of arterial pressure wave reflection) and roundtrip travel time of the reflected wave to and from the periphery (Δt p ; inverse of pulse wave velocity) were calculated from the generated aortic pressure waveform and are indices of central arterial stiffness. Systemic β‐blockade reduced heart rate (HR; 60 ± 2 vs. 53 ± 2 bpm, P < 0.001) and increased AI a (9.9 ± 4.3% vs. 4.3 ± 4.4%, P < 0.05), but did not alter Δt p (156 ± 4 msec vs. 152 ± 2 msec, P = 0.34). The AI a normalized to a HR of 75bpm (AI a @75) was not increased following β‐blockade (P = 0.67). Taken together, our data indicate that the increase in AI a following systemic β‐blockade is a result of a slower HR and not due to changes in β‐ mediated vascular tone. Supported by NIH K23 DK‐082424 (TBC), AHA 2170087(ECH) and CTSA RR‐024150

Insulin plays an important role in post‐prandial blood pressure (BP) regulation. Within the central nervous system insulin acts to stimulate sympathetic outflow to skeletal muscle. Concurrently, insulin acts locally in the periphery to promote vasodilation and is suggested to also attenuate sympathetically‐mediated vasoconstriction. Given these divergent actions, the ‘net effect’ of insulin on the sympathetic control of BP remains incompletely understood. Therefore, the purpose of this study was to examine the effect of hyperinsulinemia on sympathetic transduction to BP. To accomplish this, in 10 young healthy men, heart rate (ECG), BP (finometer or brachial arterial line) and muscle sympathetic nerve activity (MSNA) were measured at rest and during euglycemic‐hyperinsulinemia. Signal‐averaging was used to characterize the changes in mean arterial pressure (MAP) and total vascular conductance (TVC; derived from ModelFlow) over 10 cardiac cycles following spontaneous bursts of MSNA at rest and during hyperinsulinemia. No significant differences were found in the peak MAP (Rest: ∆2.8 ± 0.3; Insulin: ∆2.4 ± 0.2 mmHg, P=0.15) or TVC (Rest: ∆‐3.0 ± 0.4; Insulin: ∆‐2.7 ± 0.5 mL/min/mmHg, P=0.38) responses following bursts of MSNA between rest and hyperinsulinemia (i.e., ‘net’ sympathetic transduction was unchanged). Notably, the average size of normalized MSNA bursts was increased during hyperinsulinemia (Rest: 45.6 ± 1.8; Insulin 65.3 ± 5.0 Au, P<0.01). To take into account changes in burst height, MSNA bursts were divided into amplitude‐matched bins. When MSNA burst amplitude bins were matched between rest and hyperinsulinemia, sympathetic transduction for a given burst amplitude was blunted during insulin infusion (e.g., largest resting bursts: MAP, Rest: ∆4.4 ± 0.5; Insulin: ∆3.0 ± 0.3 mmHg, P=0.02; TVC, Rest: ∆‐5.2 ± 0.8; Insulin: ∆‐3.1 ± 0.7 mL/min/mmHg, P=0.05). However, it is important to note that 15 ± 5% of MSNA bursts exhibited amplitudes greater than that which existed at rest and overall during hyperinsulinemia ~50% of all MSNA bursts were comparable or greater in amplitude to the largest bursts present at rest. Collectively, these preliminary findings suggest that during hyperinsulinemia, sympathetic transduction for a given burst amplitude is attenuated while ‘net’ sympathetic transduction remains unchanged due to the generation of larger MSNA bursts.

Background The baroreflex plays an integral role in blood pressure maintenance by altering heart rate and sympathetic activity in response to changes in arterial pressure. Baroreflex sensitivity (BRS) has been shown to decrease linearly during progressive exposure to central hypovolemia as induced by lower body negative pressure (LBNP), and can be used as an index of decompensation status during hemorrhage. Circulating epinephrine is significantly elevated during high stress situations such as hemorrhage; however, it is unclear whether epinephrine influences BRS during progressive exposure to central hypovolemia. Based on previous work, we hypothesized epinephrine would attenuate any reduction in spontaneous cardiac baroreflex sensitivity during LBNP. Methods Ten young (35±3 yrs), healthy (BMI 24±1 kg/m 2 ) men and women (6M/4F) underwent LBNP protocols on 2 separate days consisting of a 10‐min pre‐infusion baseline, 15‐min infusion period (randomized to saline or epinephrine, 0.1 μg/kg/min), 10‐min post‐infusion baseline, and 5 minutes of LBNP at −15, −30, −45, −60, −70, and −80 mmHg until the onset of presyncope (range= −30 to −80 mmHg). Presyncope was defined as a rapid decrease in systolic blood pressure greater than 15 mmHg, systolic blood pressure less than 80 mmHg, bradycardia, or voluntary termination due to presyncopal symptoms. Heart rate (HR; electrocardiogram) and blood pressure (BP; radial arterial catheter) were recorded continuously. Spontaneous cardiac BRS was determined using the sequence method (WinCPRS). Data are reported from both 10‐min baselines and from 1 minute at 20, 40, 60, 80, and 100 percent of each individual's total LBNP exposure time. Results We observed no significant differences in total exposure time (1475±150 vs. 1415±79s; p=0.69) or cumulative stress index (820±135 vs. 699±69 min*mmHg; p=0.37) between saline and epinephrine visits, respectively. Baroreflex sensitivity did not change during post‐infusion baselines (ΔDown‐Down =−2±1 vs. −2±2 ms/mmHg; saline vs. epinephrine, respectively; p=0.09). Baroreflex sensitivity was significantly reduced with LBNP (p<0.01 main effect of LBNP). There was no difference between the change in baroreflex sensitivity during LBNP with saline or epinephrine infusion (ΔDown‐Down =−14±4 vs. −8±4 ms/mmHg, p=0.17). Conclusions Consistent with previous work, spontaneous cardiac baroreflex sensitivity fell during LBNP. Contrary to our hypothesis, epinephrine infusion did not attenuate the fall in spontaneous cardiac baroreflex sensitivity observed during simulated blood loss. Because epinephrine did not alter the effect of LBNP on BRS, these data suggest that BRS may accurately reflect decompensation status during traumatic hemorrhage independent of elevated circulating epinephrine. Support or Funding Information US Army Medical Research and Materiel Command Contract W81XWH‐13‐2‐0038

Due to the relationship between stroke volume (SV) and pulse pressure (PP) during lower body negative pressure (LBNP), we hypothesized that PP reflects SV during hemorrhage (HEM). After a 5‐minute baseline, 6 men were exposed to LBNP at ‐15, ‐30, and ‐45 mmHg for 5 minutes at each pressure. After 45 minutes of recovery, three progressive 333‐ml stages of HEM (1 L total) were performed. SV was assessed using a CC Nexfin and PP was calculated using an arterial catheter. SV (111 ± 2.1 vs. 77 ± 6.7 mL; p = 0.04) and PP (58±2.7 vs. 41 ± 4.3 mm Hg; p = 0.03) decreased from baseline to ‐45 mmHg LBNP. SV (118 ± 4.4 to 99 ± 6.8 mL; p = 0.06) and PP decreased (58 ± 2.9 to 48 ± 4.9 mmHg; p = 0.06) from baseline to ‐1 L of HEM. PP correlated with SV during HEM (amalgamated r 2 = 0.98) and LBNP (amalgamated r 2 = 0.99). There was limited inter‐subject variability in PP vs. SV during LBNP ( r 2 range 0.57 to 0.99, 5/6 had an r 2 >; 0.90), but substantial inter‐subject variability during HEM ( r 2 range 0.21 to 0.99, 2/6 had an r 2 >; 0.90). Reductions in PP induced by LBNP or HEM are strongly correlated with SV. However large inter‐individual differences in this correlation during HEM, which may indicate a significantly variable compensatory response in human arterial distensibility, may limit the use of PP as a surrogate for SV during clinical blood loss. Funded by U.S. Army MRMC Combat Casualty Care Research Program Grant # W81XWH‐11–1‐0823.