Introduction

Individuals with spinal cord injury (SCI) represent one of the most physically inactive populations because of their profound immobility. Accordingly, such immobility results in an elevated risk of a wide array of cardiovascular complications, including coronary artery disease,1 peripheral artery disease,2 atherosclerosis3 and accelerated cardiovascular aging.4 Although severe physical inactivity has been shown to be associated with impaired diastolic function in able-bodied individuals,5, 6, 7 findings regarding diastolic function after SCI remain equivocal. Left ventricular diastolic function has been shown by some to be preserved in those with SCI,8, 9, 10 whereas more recent reports convey diastolic impairments after SCI.11, 12, 13, 14 Therefore, it is unknown whether diastolic dysfunction actually manifests after SCI.

It is important to note that echocardiographic parameters of diastolic function are highly influenced by preload. Specifically, an increase in preload augments diastolic values whereas a decrease in preload reduces them.15, 16, 17, 18 Individuals with SCI demonstrate reduced ventricular filling and preload, partly due to a reduction in blood volume secondary to the chronic physical inactivity.19 Therefore, it is not clear whether the reported diastolic impairments following SCI are a result of reduced blood volume or due to pathological alterations in the left ventricle, per se. In addition, diastolic function in individuals with SCI has always been examined while participants were in a relaxed state, and therefore it is unclear how ventricular filling after SCI functions when the heart is exposed to stress. Rapid saline infusion is a diagnostic method that increases preload, and may reveal left ventricular impairments that may otherwise not manifest during resting conditions.20 In addition, the normal diastolic responses to rapid saline infusion are well documented,17, 18 and therefore this method may be a useful tool to elucidate any left ventricular diastolic impairments in individuals with SCI.

Accordingly, the purpose of this study was to determine whether individuals with SCI demonstrate diastolic impairments at rest and/or during a volume stress induced by rapid saline infusion. We also aimed to determine whether reduction in blood volume following SCI contributes to the putative diastolic impairments in this population.

Materials and methods

Participants

Thirteen individuals with SCI (C4–T6; AIS A-D; 16.5±9.7 years post injury) and 12 able-bodied individuals matched for age, sex and body mass participated in this study. Participants were nonsmokers, nonhypertensive, did not have history of heart disease and had a normal resting and exercise electrocardiogram. As this is the first study to perform rapid saline infusion in individuals with SCI, those with unstable autonomic dysreflexia were excluded. Participants from both groups included a range of sedentary to physically active individuals but none were competitive or highly trained athletes. Individuals with SCI were recruited from an outpatient community-based exercise center for SCI and other neuromuscular conditions (Power Cord, St Catharines, ON, USA). All the participants provided informed consent to participate in the study, and we certify that all applicable institutional regulations concerning the ethical use of human volunteers were followed during the course of this research.

General study protocol

All participants were required to void their bladder before testing, especially those with SCI to reduce any risk of autonomic dysreflexia. After a 15 min resting period in the left lateral decubitus position, resting blood pressure was obtained manually from the brachial artery, and a continuous electrocardiogram recording (lead II) was taken throughout the testing session. Following the resting period, assessment of left ventricular structure and function was performed via two-dimensional and M-mode echocardiography. Participants then performed the carbon monoxide rebreathe test for the measurement of total blood volume. This was followed by rapid saline infusion. After 5 min of the saline infusion, left ventricular diastolic function was assessed once more.

Echocardiography

All echocardiographic images were obtained using a commercially available ultrasound system (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway) with a 1.5 MHz phased-array transducer. All images were acquired by a single experienced sonographer and were stored for later offline analysis using a commercially available software (EchoPac version 6.0; GE Vingmed Ultrasound AS). Images were obtained while participants were in the left lateral decubitus position at end expiration and three consecutive cycles for each image were stored and averaged for offline analysis.

Left ventricular wall dimensions were assessed at end diastole and end systole via M-mode recordings (and values were confirmed by two-dimensional recordings) in accordance to the recommendations of the American Society of Echocardiography.21 Left ventricular wall dimensions included intraventricular septal thickness at diastole and systole, posterior left ventricular internal diameter at diastole (LVIDd) and systole (LVIDs) as well as posterior wall thickness in diastole and systole. Hemodynamic measures that included end-diastolic volume, end-systolic volume, ejection fraction, as well as stroke volume were measured from the long parasternal axis view through M-mode, as apical chamber views from individuals with SCI were suboptimal for the use of Simpson’s Biplane. Cardiac output was calculated as the product of stroke volume and heart rate. Left ventricular mass (LVM) was calculated using the Devereux formula22 and was indexed for body surface area (LVMi).23

Pulsed-wave Doppler and tissue Doppler were used to assess diastolic function according to the recommendation of the American Society of Echocardiography.24 For mitral inflow velocity, from an apical four-chamber view, a 4-mm sample volume was placed between the mitral valve leaflet tips during diastole. The Doppler velocity curves were digitized to obtain peak early (E) and late (A) mitral flow velocities, and their ratio (E:A) was calculated. The descending limb of the E wave was used to measure deceleration time, and isovolumetric relaxation time was measured as the time between aortic valve closure and mitral valve opening. Mitral annular velocities were measured via tissue Doppler imaging from the four-chamber apical view. A 2-mm sample volume was placed along the longitudinal movement of the basal septal and basal lateral walls. Peak early (E′) and late (A′) diastolic myocardial tissue velocities were recorded, and E:E′ was calculated as a noninvasive measure of left ventricular filling pressure.25

Carbon monoxide rebreathe test

The carbon monoxide rebreathe test was used for the assessment of total blood volume in accordance to the Burge and Skinner method.26 Briefly, a 20-gauge catheter was placed in the antecubital vein for instant blood draws. A baseline blood sample was taken for measurement of hematocrit. After a 4 min period of breathing 100% oxygen (Praxair, Hamilton, ON, Canada), participants inhaled a 15–20 ml priming dose of 100% CO gas (Praxair) and a 1 ml blood sample was obtained in a heparinized vacutainer for assessment of carboxy-hemoglobin and total hemoglobin. A second dose of carbon monoxide gas (40–50 ml) was inhaled, and after 10 min of breathing, a final blood sample was obtained for carboxy-hemoglobin assessment. All blood samples were analyzed by a blood–gas analyzer via photospectromery (Phox-Cooximeter; NovaBiomedical, Mississauga, ON, Canada). The carbon monoxide rebreathe test yields a low reproducibility error in both able-bodied27 and SCI individuals,19 and total blood volume was reported in absolute values and indexed for body mass.

Volume loading

Preload was increased via rapid infusion of warm isotonic saline (0.9% NaCl) through the antecubital vein by an 18–20-gauge catheter. The saline dose consisted of 15 ml kg−1 at a rate of 100 ml min−1.

Statistical analysis

A Shapiro–Wilk test was used to check for normalcy in all parameters, and as a result, parametric statistics were justified. Unpaired t-tests were used to compare all baseline left ventricular structural, diastolic and hemodynamic parameters between groups. An analysis of covariance was used to assess differences in baseline diastolic parameters while adjusting for LVIDd. The LVIDd was used as a covariate because a reduction in left ventricular size following SCI has been proposed as a mechanism to maintain normal diastolic function despite a reduction in preload,8, 28 and LVIDd had equal regression coefficient associated with both groups (assumption of homogeneity of regression). A two-way repeated-measures analysis of variance was used to detect group by condition (IV infusion) interactions as well as main effects for diastolic parameters. Statistical analyses were performed using Microsoft Excel 2013 and Statistical Package for Social Sciences (SPSS 21.0, IBM, Chicago, IL, USA) software. Data were reported as means±s.d., and the level of statistical significance was set at P⩽0.05.

Results

Participants

All participant characteristics are provided in Table 1. There were no between-group differences in age, height, body mass, heart rate or blood pressure (able-bodied: 105±13 mm Hg vs SCI: 101±16 mm Hg; P>0.05). All participants tolerated the rapid saline infusion well, and no adverse events occurred.

Table 1 Participant characteristics
Full size table

Baseline left ventricular structure and hemodynamics

Compared with able-bodied participants, individuals with SCI exhibited cardiac atrophy as evidenced by smaller LVIDd. In addition, the SCI group demonstrated lower end-diastolic volume, stroke volume and cardiac output; however, ejection fraction was not different between groups (Table 2). There were no between-group differences in LVM, LVMi or any of the other structural parameters. Absolute total blood volume and blood volume indexed for body mass were significantly lower in the SCI group (Table 2). In addition, there were no between-group differences in any of the diastolic parameters (Table 3); however, after controlling for LVIDd, the SCI group demonstrated a significantly lower E:A ratio (Table 3).

Table 2 Baseline LV structural and hemodynamic parameters
Full size table
Table 3 Baseline diastolic function before and after controlling for LV internal diameter
Full size table

Diastolic response to rapid saline infusion

There was no difference in the average amount of saline given to both groups (SCI: 1184±214 ml vs able-bodied: 1208±252 ml; P=0.80). There were no group by condition interactions for any of the diastolic measures (P>0.05), suggesting that both groups responded similarly to the rapid saline infusion. However, there was a significant main effect for condition, specifically when collapsing across groups, the saline infusion resulted in a significant increase in E (0.78±0.10 m s−1 to 0.90±0.11 m s−1; P<0.01; Figure 1a) and the E:A ratio (2.06±0.61 to 2.82±1.05; P<001; Figure 1b), as well as a significant decrease in A (0.40±0.09 m s−1 to 0.35±0.11 m s−1; P=0.05; Figure 1c), deceleration time (208.37±50.04 ms to 183.16±37.21ms; P=0.01; Figure 1d) and isovolumetric relaxation time (102.04±17.54ms to 88.42±14.11ms; P<0.01; Figure 1e). The saline infusion also resulted in a significant increase in E′Sept (0.13±0.02 m s−1 to 0.14±0.03 m s−1; P=0.01) and E′Lat (0.15±0.04 m s−1 to 0.17±0.04 m s−1; P<0.001) velocities. Rapid saline infusion resulted in no changes in A′Sept velocity (0.09±0.02 m s−1 to 0.10±0.03; P=0.20), A′Lat velocity (0.08±0.03 m s−1 to 0.08±0.03 m s−1; P=0.40) or left ventricular filling pressure as indicated by EE′ (5.8±1.3 to 6.0±1.2; P=0.40). In addition, Table 4 shows that the percent change in diastolic values in response to the saline infusion was similar between groups.

Figure 1
figure 1

(a) Early transmitral diastolic velocity after rapid saline infusion. (b) Ratio of early to late transmitral filling velocity after rapid saline infusion. (c) Late transmitral diastolic velocity after rapid saline infusion. (d) Transmitral deceleration time after rapid saline infusion. (e) Isovolumetric relaxation time after rapid saline infusion. A, late transmitral diastolic velocity; AB, able-bodied; DT, deceleration time; E, early transmitral diastolic velocity; E:A, early to late diastolic velocity ratio; IVRT, isovolumentric relaxation time; SCI, spinal cord injury. Error bars are s.d. * denotes significant main effect for condition (P<0.05).

Full size image
Table 4 Percent change in diastolic function after rapid saline infusion
Full size table

Discussion

The main findings of this investigation are that (i) individuals with SCI have preserved left ventricular diastolic function at rest despite having a lower blood volume, (ii) cardiac atrophy following SCI may have a role in preserving diastolic function in response to lower preload and (iii) rapid saline infusion resulted in similar left ventricular filling patterns between spinal cord injured and able-bodied individuals.

Baseline diastolic function

The current investigation shows that individuals with SCI have preserved left ventricular diastolic function at rest. This has been corroborated by previous studies where normal diastolic function was reported in individuals with complete tetraplegia8 and paraplegia10 regardless of being sedentary9 or active.28 However, other investigators have reported reduced diastolic function following SCI.11, 12, 13, 14 It has been proposed that the chronic physical inactivity after SCI may have a role in such diastolic impairments, as active spinal cord injured individuals have been shown to present better diastolic patterns compared with sedentary ones.11, 14 However, Maggioni et al.29 reported no difference in diastolic function between trained and untrained spinal cord injured individuals and De Groot et al.8 reported no difference in diastole between able-bodied individuals and sedentary individuals with severe tetraplegia. In addition, Driussi et al.12 reported lower diastolic function in those with SCI compared with able-bodied individuals even after controlling for levels of physical activity, suggesting that physical activity did not have a major role in the diastolic impairments. It is interesting to note, however, that in two of the studies that reported diastolic impairments after SCI,12, 13 the spinal cord injured groups exhibited left ventricular concentric hypertrophy, which is a pattern often seen in individuals with diastolic heart failure.30 Therefore, these individuals may have had other concomitant disorders that resulted in unfavorable left ventricular structural alterations, which in turn caused diastolic impairments.

Diastolic function after saline infusion

This is the first study to perform rapid saline infusion in individuals with SCI. Rapid saline infusion is an excellent diagnostic method for unmasking cardiac dysfunctions that otherwise may not manifest during resting conditions. If a ventricle is compliant, then diastolic function should be augmented in response to increased loading without significant changes in left ventricular filling pressures.17, 18 In contrast, a noncompliant ventricle will demonstrate significant elevations in filling pressure after increased volume loading.20 Both groups in the current investigation demonstrated similar changes in all diastolic parameters, and both the groups showed no changes in the left ventricular filling pressures, as shown by EE′. This is evidence that the left ventricle of the current SCI cohort is compliant and does not have intrinsic pathological alterations. In addition, both groups received the same average dose of saline despite the SCI group having a smaller left ventricle. This means that the left ventricle in those with SCI likely expanded relatively more during the infusion with no significant changes in the filling pressure, which further supports the notion that left ventricular elasticity is maintained after SCI.

Positive morphological adaptation

Owing to the lower blood volume and preload (end-diastolic volume) demonstrated by the spinal cord injured group, a reduction in left ventricular diastolic function was expected. However, both groups showed similar values for all diastolic parameters. This was unexpected, as hypovolemia has been shown to reduce ventricular diastolic function5, 15 and distensibility.6, 7 However, despite the spinal cord injured participants being hypovolemic with lower preload, diastolic function was still preserved. This may be mediated by the ventricular atrophy experienced after SCI, as shown by a smaller LVIDd, which is likely a morphological adaptation to the chronic reduction in preload. A smaller chamber size may preserve diastolic velocities despite a lower preload by preserving ventricular wall stress. Although this has been hypothesized to occur after SCI,8, 28 this is the first study to show evidence of such a mechanism, as the SCI group demonstrated reduced diastolic function (lower E:A ratio) only when LVID was controlled for. It is interesting to note that the current SCI cohort did not show a reduction in LVM. This could possibly be because of the preservation of normal systolic blood pressure (afterload), which maintains a certain level of resistance to preserve LVM. This has been previously shown, as individuals with SCI who show reduced LVM also demonstrate reduced blood pressure,8, 28 whereas those who present normal LVM exhibit normal blood pressure.12 This suggests that LVID is more important than LVM in maintaining diastolic velocities during hypovolemia. Furthermore, SCI is associated with severe skeletal muscle atrophy,32 peripheral vasculature atrophy33 and a decrease in capillary density,34 all of which have a role in reduced overall oxygen demand.33, 35 As a result, the heart is not required to maintain a normal size and likely also atrophies in accordance to the muscular and vascular atrophy after paralysis. Therefore, the cardiac atrophy that occurs after SCI may be a positive morphological adaptation, which preserves normal cardiac function and allows the heart to operate in accordance to lower systemic demands.

Structural and hemodynamic findings

Although stroke volume and cardiac output were lower in the SCI group, this was likely because of the reduced preload and not because of impaired ventricular contractility, as evidenced by similar ejection fractions between the groups. Furthermore, the observed reduction in LVIDd demonstrated by the spinal cord injured participants is in agreement with several previous reports.10, 11, 12, 28 However, the commonly reported reduction in LVM in this population8, 9, 10, 28, 29 was not observed in the current participants. As mentioned, the smaller LVIDd can be explained by the chronic reduction in preload, as demonstrated by a lower end-diastolic volume and blood volume. The reduction in preload decreases ventricular mass to volume ratio, thus decreasing ventricular wall stress and consequently resulting in adaptive cardiac atrophy.36 In addition, the left ventricle is highly adaptive to changes in loading conditions, as cardiac atrophy has been shown to occur in as little as 2 weeks of reduced preload.36

The blood volume values obtained in our study are comparable to those reported by Houtman et al.,19 whose participants had a similar distribution of injury levels compared with those in the current study (C4–T6). Physical inactivity is a known stimulus for hypovolemia,6, 7, 37 thus the smaller blood volume in SCI individuals is likely because of the chronically reduced physical activity levels. However, it is interesting to note that the reduction in blood volume in response to physical inactivity is typically demonstrated from models of severe immobilization, such as bed rest31, 37 and spaceflight. The spinal cord injured individuals in the current study were not completely immobilized, as they use manual wheelchairs for their main method of transportation, they perform daily transfers and the majority of them participate in regular aerobic and resistance exercise with the upper limbs. However, although being relatively active, they still demonstrated a lower blood volume compared with the able-bodied group. This suggests that moderate upper limb activity is insufficient to maintain blood volume after SCI and perhaps lower limb exercise is required in this regard. This hypothesis is supported by an exercise study that reported no change in blood volume following 6 weeks of arm ergometry in individuals with SCI despite an increase in exercise performance.38 In contrast, Houtman et al.19 demonstrated significant elevations in blood volume in individuals with SCI following only 2 weeks of electrically stimulated leg cycling exercise.

Conclusion

Although individuals with SCI were hypovolemic, which results in reduced preload, diastolic function is still preserved at rest. The cardiac atrophy that occurs after SCI may have a role in maintaining normal diastolic function. In addition, individuals with SCI demonstrated normal ventricular filling responses to rapid saline infusion, suggesting that the left ventricle is compliant in these individuals. However, further and more invasive examination of human SCI ventricular compliance is required to confirm our results. These investigations can use methods such as pulmonary wedge capillary pressure or analysis of human myocardial tissue.

Limitations

The primary limitation to this study was the small number of participants given the difficulty of recruiting a large number of individuals with SCI. Another limitation is that we are basing our conclusion of compliant ventricles following SCI from noninvasive measures; thus, further studies using more direct measures of ventricular compliance are warranted. Finally, all individuals with SCI were grouped together due to the small sample size; therefore, larger studies are required to compare the diastolic function between different severities and levels of injuries.

DATA ARCHIVING

There were no data to deposit.