A comparison of passive hindlimb cycling and active upper-limb exercise provides new insights into systolic dysfunction after spinal cord injury

Abstract

Active upper-limb and passive lower-limb exercise are two interventions used in the spinal cord injury (SCI) population. Although the global cardiac responses have been previously studied, it is unclear how either exercise influences contractile cardiac function. Here, the cardiac contractile and volumetric responses to upper-limb (swim) and passive lower-limb exercise were investigated in rodents with a severe high-thoracic SCI. Animals were divided into control (CON), SCI no exercise (NO-EX), SCI passive hindlimb cycling (PHLC), or SCI swim (SWIM) groups. Severe contusion SCI was administered at the T2 level. PHLC and SWIM interventions began on day 8 postinjury and lasted 25 days. Echocardiography and dobutamine stress echocardiography were performed before and after injury. Cardiac contractile indexes were assessed in vivo at study termination via a left ventricular pressure-volume conductance catheter. Stroke volume was reduced after SCI (91 µl in the NO-EX group vs. 188 µl in the CON group, P < 0.05) and was reversed at study termination in the PHLC (167 µl) but not SWIM (90 µl) group. Rates of contraction were reduced in NO-EX versus CON groups (6,079 vs. 9,225 mmHg, respectively, P < 0.05) and were unchanged by PHLC and SWIM training. Similarly, end-systolic elastance was reduced in the NO-EX versus CON groups (0.67 vs. 1.37 mmHg/µl, respectively, P < 0.05) and was unchanged by PHLC or SWIM training. Dobutamine infusion normalized all pressure indexes in each SCI group (all P < 0.05). In conclusion, PHLC improves flow-derived cardiac indexes, whereas SWIM training displayed no cardiobeneficial effect. Pressure-derived deficits were corrected only with dobutamine, suggesting that reduced β-adrenergic stimulation is principally responsible for the impaired cardiac contractile function after SCI.NEW & NOTEWORTHY This is the first direct comparison between the cardiac changes elicited by active upper-limb or passive lower-limb exercise after spinal cord injury. Here, we demonstrate that lower-limb exercise positively influences flow-derived cardiac indexes, whereas upper-limb exercise does not. Furthermore, neither intervention corrects the cardiac contractile dysfunction associated with spinal cord injury.

Keywords: cardiac function; exercise; pressure-volume; spinal cord injury.

Conflict of interest statement

No conflicts of interest, financial or otherwise, are declared by the authors.

Copyright © 2017 the American Physiological Society.

Figures

Fig. 1.

Overview of the study design and schematics of the exercise interventions. A: study design. Animals were assigned to the following groups: control (CON), T2 spinal cord injury (SCI) with no exercise (NO-EX), T2 SCI with passive hindlimb cycling (PHLC), or T2 SCI with swim training (SWIM). B: cycling was performed for 30 min daily for 25 days. C: swimming was also performed daily (6 × 5 min sessions, 20-min break) for 25 days in a custom-made swim tank (60-in. length × 18-in. height × 7-in. width; not drawn to scale). PV, pressure-volume.

Fig. 2.

Cardiovascular function diminishes after T2 400-kDa spinal cord injury (SCI), and passive hindlimb cycling (PHLC) appears to normalize cardiovascular dysfunction. Left ventrular internal diameter at diastole (LVIDd; A) was reduced postinjury but was brought back to control (CON) levels with PHLC and swim training (SWIM) interventions. Stroke volume (SV; B) and cardiac output (CO; C) were only normalized with PHLC. Systolic blood pressure (SBP; D) and mean arterial pressure (MAP; E) were reduced in the T2 SCI with no exercise (NO-EX) group and remained lower after both PHLC and SWIM training. F: heart rate [HR; in beats/min (bpm)]. P values represent post hoc testing after a significant repeated-measures ANOVA main effect for echocardiography-derived indexes (A–C) and one-way ANOVA main effect for pressure-volume analyses (D–F). Data are presented as means ± SD. *P < 0.05, CON vs. NO-EX groups; †P < 0.05, CON vs. PHLC groups; ‡P < 0.05, CON vs. SWIM groups.

Fig. 3.

Echocardiography-derived responses to the β-agonist dobutamine (DOB) reveals a blunted response at 1 wk postinjury but hyperresponsiveness to DOB at 5 wk postinjury. Pre-spinal cord injury (pre-SCI) responses are presented in A–D, 1-wk responses are presented in E–H, and 5-wk responses are presented in I–L. HR, heart rate [in beats/min (bpm)]; EDV, end-diastolic volume; SV, stroke volume; CO, cardiac output. P values represent post hoc testing after a significant repeated-measures ANOVA main effect for DOB concentration. Data are presented as means ± SD; control (CON) group: n = 6, T2 SCI with no exercise (NO-EX) group: n = 8, T2 SCI with passive hindlimb cycling (PHLC) group: n = 6, T2 SCI with swim-trained (SWIM) group: n = 6. *P < 0.05 for the NO-EX group at that dose vs. 0 μg; ^P < 0.05 for the SWIM group at that dose vs. 0 μg; †P < 0.05 for the PHLC group at that dose vs. 0 μg.

Fig. 4.

Terminal pressure-volume assessments demonstrate reduced baseline pressure-generating capacity and load-independent metrics in all SCI groups groups compared with the control (CON) group. Animals were assigned to the following groups: CON, T2 spinal cord injury (SCI) with no exercise (NO-EX), T2 SCI with passive hindlimb cycling (PHLC), or T2 SCI with swim training (SWIM). A: representative loop from the CON group labeled with the relevant measurements acquired from pressure-volume analysis. ESP, end-systolic pressure; ESV, end-systolic volume; EDV, end-diastolic volume; EDPVR, end-diastolic pressure-volume relationship; SW, stroke work; ESPVR, end-systolic pressure-volume relationship; Ees, slope of the ESPVR. B: representative loop from 1 animal/group demonstrating diminished volumetric and pressure indexes in NO-EX and SWIM group and the recovery of volumetric indexes in the PHLC group. Each loop represents 15 cardiac cycles averaged together. C–F: inferior vena cava occlusions from each group representing the load-independent measures acquired. Note that all SCI groups exhibited a significantly reduced Ees β-coefficient. Load-independent metrics were analyzed with mixed-model regression analysis. The significant reduction in the Ees β-coefficient reflects the average changes in all animals within that group. Data are represented as means ± SE; n = 5 for all groups. *P < 0.05 vs. the CON group.

Fig. 5.

Pressure-volume-derived indexes in response to administration of increasing concentrations of dobutamine (DOB). Animals were assigned to the following groups: control (CON), T2 spinal cord injury (SCI) with no exercise (NO-EX), T2 SCI with passive hindlimb cycling (PHLC), or T2 SCI with swim training (SWIM). A–D: representative pressure-volume loops obtained from 1 animal/group. Each loop contains 15 cardiac cycles averaged together to demonstrate the increase in both volume and pressure indexes in response to DOB. DOB elicited increases in heart rate [HR; in beats/min (bpm); E] and transient increases in end-systolic pressure (Pes; F) and normalized rates of contraction (dP/dtmax; G). Low-frequency systolic blood pressure (LFSBP) was diminished after SCI (NO-EX group) and was not influenced by exercise (H). P values represent significant post hoc testing after a significant repeated-measures ANOVA main effect for DOB (E–G) or group (H). Data are represented as means ± SD; n = 5 for all groups. ‡P < 0.05 for the CON group at that dose vs. 0 μg; *P < 0.05 for the NO-EX group at that dose vs. 0 μg; ^P < 0.05 for the SWIM group at that dose vs. 0 μg; †P < 0.05 for the PHLC group at that dose vs. 0 μg; σP < 0.001 vs. the CON group for LFSBP.