A comparison of protocols for simulating hemorrhage in humans: step versus ramp lower body negative pressure

Abstract

Lower body negative pressure (LBNP) elicits central hypovolemia, and it has been used to simulate the cardiovascular and cerebrovascular responses to hemorrhage in humans. LBNP protocols commonly use progressive stepwise reductions in chamber pressure for specific time periods. However, continuous ramp LBNP protocols have also been utilized to simulate the continuous nature of most bleeding injuries. The aim of this study was to compare tolerance and hemodynamic responses between these two LBNP profiles. Healthy human subjects (N = 19; age, 27 ± 4 y; 7 female/12 male) completed a 1) step LBNP protocol (5-min steps) and 2) continuous ramp LBNP protocol (3 mmHg/min), both to presyncope. Heart rate (HR), mean arterial pressure (MAP), stroke volume (SV), middle and posterior cerebral artery velocity (MCAv and PCAv), cerebral oxygen saturation (ScO₂), and end-tidal CO₂ (etCO₂) were measured. LBNP tolerance, via the cumulative stress index (CSI, summation of chamber pressure × time at each pressure), and hemodynamic responses were compared between the two protocols. The CSI (step: 911 ± 97 mmHg/min vs. ramp: 823 ± 83 mmHg/min; P = 0.12) and the magnitude of central hypovolemia (%Δ SV, step: -54.6% ± 2.6% vs. ramp: -52.1% ± 2.8%; P = 0.32) were similar between protocols. Although there were no differences between protocols for the maximal %Δ HR (P = 0.88), the %Δ MAP during the step protocol was attenuated (P = 0.05), and the reductions in MCAv, PCAv, ScO₂, and etCO₂ were greater (P ≤ 0.08) when compared with the ramp protocol at presyncope. These results indicate that when comparing cardiovascular responses to LBNP across different laboratories, the specific pressure profile must be considered as a potential confounding factor.NEW & NOTEWORTHY Ramp lower body negative pressure (LBNP) protocols have been utilized to simulate the continuous nature of bleeding injuries. However, it unknown if tolerance or the physiological responses to ramp LBNP are similar to the more common stepwise LBNP protocol. We report similar tolerance between the two protocols, but the step protocol elicited a greater increase in cerebral oxygen extraction in the presence of reduced blood flow, presumably facilitating the matching of metabolic supply and demand.

Keywords: central hypovolemia; cerebral blood velocity; lower body negative pressure.

Figure 1.

The step (A) and ramp (B) lower body negative pressure (LBNP) protocols. After a 5-min baseline, pressure in the LBNP chamber was decreased every 5-min to −15, −30, −45, −60, −70, −80, −90, and −100 mmHg (step; A), or at a decompression rate of 3 mmHg/min (ramp; B), both until presyncope.

Figure 2.

A: the cumulative stress index (CSI) to a progressive stepwise lower body negative pressure (LBNP) protocol (white bar) and to a continuous ramp LBNP protocol (gray bar). Mean (bars) and individual subject data (circles) are presented. A paired t test was used for analysis. B: percent change from baseline for stroke volume (SV) decreased similarly in both step LBNP (dashed line, open circle) and ramp LBNP (solid line, closed circle). Over the 1 min before presyncope (PS-1), the percent change in SV was not different between step LBNP (open triangle) versus ramp LBNP (closed triangle). Values are shown as means ± SE. A two-factor (protocol, time) linear mixed-model analysis with Holm-corrected multiple comparisons (within and between protocols up to −60 mmHg LBNP) and paired t tests (between protocols at PS-1) were used for analysis. *Difference from baseline within protocol, P ≤ 0.07.

Figure 3.

Heart rate (HR) (A) increased and mean arterial pressure (MAP; B) decreased during both step (dashed line, open circle) and ramp (solid line, closed circle) lower body negative pressure (LBNP) protocols. Over the 1 min before presyncope (PS-1), HR was similar and MAP was different between protocols. Values are shown as means ± SE. A two-factor (protocol, time) linear mixed-model analysis with Holm-corrected multiple comparisons (within and between protocols up to −60 mmHg LBNP) and paired t tests (between protocols at PS-1) were used for analysis. *Difference from baseline within protocol, P ≤ 0.07.

Figure 4.

Middle cerebral artery velocity (MCAv; A) and posterior cerebral artery velocity (PCAv; B) decreased in both step (dashed line, open circle) and ramp (solid line, closed circle) lower body negative pressure (LBNP) protocols. Over the 1 min before presyncope (PS-1), the percent change in MCAv and PCAv was greater with step LBNP (open triangle) versus ramp LBNP (closed triangle). Values are shown as means ± SE. A two-factor (protocol, time) linear mixed-model analysis with Holm-corrected multiple comparisons (within and between protocols up to −60 mmHg LBNP) and paired t tests (between protocols at PS-1) were used for analysis. *Difference from baseline within protocol, P ≤ 0.06.

Figure 5.

Percent change from baseline for cerebral oxygen saturation (ScO₂; A), oxygenated hemoglobin (oxyhemoglobin; B) and deoxygenated hemoglobin (deoxyhemoglobin; C) for step (dashed line, open circle) and ramp (solid line, closed circle) lower body negative pressure (LBNP) protocols. The reduction in ScO₂ was greater during step LBNP versus ramp LBNP, and for the final 1 min before presyncope (PS-1), represented by open (step) and closed (ramp) triangles. Oxyhemoglobin decreased from −15 mmHg LBNP in the step LBNP protocol and was lower versus ramp LBNP at −45 and −60 mmHg LBNP and at the PS-1 time point. Deoxyhemoglobin progressively increased during both step LBNP and ramp LBNP, with no differences at any time point. Values are shown as means ± SE. A two-factor (protocol, time) linear mixed-model analysis with Holm-corrected multiple comparisons (within and between protocols up to −60 mmHg LBNP) and paired t tests (between protocols at PS-1) were used for analysis. *Difference from baseline within protocol, P ≤ 0.08. †Differences between protocols, P ≤ 0.04.