Extended Data Fig. 2: Rodent model develops haemodynamic instability across the natural history of spinal cord injury.
From: Neuroprosthetic baroreflex controls haemodynamics after spinal cord injury
a, We recorded haemodynamics and sympathetic nerve activity during and after the spinal cord contusion (n = 6). b, We observed an immediate increase in blood pressure and sympathetic nerve activity following the onset of the contusion. c, Quantifications revealed a significant increase in systolic blood pressure (paired one-tailed t-test; 91 mmHg vs 128 mmHg; t = 5.40; P = 0.001) and mean arterial pressure (paired one-tailed t-test; 63 mmHg vs 96 mmHg; t = 4.50; P = 0.003), a decrease in heart rate (paired one-tailed t-test; 309 bpm vs 100 bpm; t = −10.26; P = 7.56 × 10−5), and an increase in sympathetic nerve activity (paired one-tailed t-test; t = 2.26; P = 0.037) during contusion. These were followed by complete reversal after the contusion, where systolic blood pressure (paired one-tailed t-test; 116 mmHg vs 79 mmHg; t = −5.97; P = 0.0009), mean arterial pressure (paired one-tailed t-test; 85 mmHg vs 53 mmHg; t = −5.14; P = 0.002), sympathetic nerve activity (paired one-tailed t-test; t = −3.29; P = 0.011), and heart rate (paired one-tailed t-test; 350 bpm vs 313 bpm; t = −2.91; P = 0.017) decreased compared to pre-injury. d, We next established the natural history of haemodynamics using 24/7 recordings. We found that throughout the recording period animals with spinal cord injury deviated outside key thresholds representing ‘normal’ values (points scaled by size and transparency based on their deviation outside our set thresholds (dotted lines)). e, We found a left-shift in the distribution of haemodynamic values, and a right-ward shift in heart rate values, indicating generally lower blood pressure, higher heart rate, and more aberrant sympathetic nerve activity (Kolmogorov–Smirnov test; all P < 2.2 × 10−16). f, Quantifications revealed an increase in the number of deviations for systolic blood pressure (independent samples one-tailed t-test; t = 5.92; P = 0.0005), diastolic blood pressure (independent samples one-tailed t-test; t = 3.68; P = 0.007), mean arterial pressure (independent samples one-tailed t-test; t = 3.23; P = 0.011), heart rate (independent samples one-tailed t-test; t = 2.0; P = 0.0499), and sympathetic nerve activity (independent samples one-tailed t-test; t = 3.20; P = 0.006). We also found an increase in the variance of systolic blood pressure (independent samples one-tailed t-test; t = 2.70; P = 0.011), diastolic blood pressure (independent samples one-tailed t-test; t = 2.01; P = 0.036), mean arterial pressure (independent samples one-tailed t-test; t = 2.85; P = 0.009), and sympathetic nerve activity (independent samples one-tailed t-test; t = 3.20; P = 0.006), fitting the criteria for haemodynamic instability after SCI. g, Formal baseline recordings revealed that baseline systolic blood pressure (two-way repeated measures ANOVA; interaction effect F6,1 = 7.05; P = 1.10 × 10−5; all Tukey post hoc P < 0.001) and mean arterial pressure (two-way repeated measures ANOVA; interaction effect F6,1 = 4.93; P = 0.0004; all Tukey post hoc P < 0.001) were reduced, in agreement with spontaneous 24/7 data recordings. Data are mean ± s.e.m. h, We devised a closed-loop negative-pressure system to mimic an orthostatic challenge in rats. Animals with spinal cord injury could not respond to decreasing pressures, whereas uninjured animals responded and slowly recovered. This response was consistent across all six weeks post injury and is contrasted against week 0 (no injury for both groups). i, Quantification of this response revealed increased negative deltas for the spinal cord injury group for systolic blood pressure (two-way repeated measures ANOVA; interaction effect F6,1 = 3.71; P = 0.003; all Tukey post hoc P < 0.05) and mean arterial pressure (two-way repeated measures ANOVA; interaction effect F6,1 = 4.20; P = 0.001; all Tukey post hoc P < 0.05). j, After SCI we observed a time-dependent increase in the linear relationship between chamber pressure and blood pressure (likelihood ratio test of nested models; P < 0.001), indicating that haemodynamics cannot be stabilized during orthostatic challenge without a functioning baroreflex. In all panels, percentage change is presented for clarity as needed, while all statistics are calculated from raw values. Bar charts represent the mean with raw data overlaid. *P < 0.05; **P < 0.01; ***P < 0.001. HR, heart rate; iSNA, integrated sympathetic nerve activity; MAP, mean arterial pressure.
