Preserved flow-mediated dilation in the inactive legs of spinal cord-injured individuals

Patricia C. E. de Groot, Fleur Poelkens, Miriam Kooijman, Maria T. E. Hopman

Abstract

The aim of the study was to assess endothelial function, measured by flow-mediated dilation (FMD), in an inactive extremity (leg) and chronically active extremity (arm) within one subject. Eleven male spinal cord-injured (SCI) individuals and eleven male controls (C) were included. Echo Doppler measurements were performed to measure FMD responses after 10 and 5 min of arterial occlusion of the leg (superficial femoral artery, SFA) and the arm (brachial artery, BA), respectively. A nitroglycerine spray was administered to determine the endothelium independent vasodilatation in the SFA. In the SFA, relative changes in FMD were significantly enhanced in SCI compared with C (SCI: 14.1 ± 1.3%; C: 9.2 ± 2.3%), whereas no differences were found in the BA (SCI: 12.5 ± 2.9%; C: 14.2 ± 3.3%). Because the FMD response is directly proportional to the magnitude of the stimulus, the FMD response was also expressed relative to the shear rate. No differences between the groups were found for the FMD-to-shear rate ratio in the SFA (SCI:0.061 ± 0.023%/s−1; C: 0.049 ± 0.024%/s−1), whereas the FMD-to-shear rate ratio was significantly decreased in the BA of SCI individuals (SCI: 0.037 ± 0.01%/s−1; C: 0.061 ± 0.027%/s−1). The relative dilatory response to nitroglycerine did not differ between the groups. (SCI: 15.6 ± 2.0%; C: 13.4 ± 2.3%). In conclusion, our results indicate that SCI individuals have a preserved endothelial function in the inactive legs and possibly an attenuated endothelial function in the active arms compared with controls.

  • vascular endothelial function
  • deconditioning
  • ultrasound

the endothelium plays an essential role in vascular homeostasis and is able to respond to physical and chemical stimuli by the synthesis and release of vasoactive, thromboregulatory, and growth factor substances (37). Impaired endothelial function has been suggested as a key early event in the development of atherosclerosis, and a high correlation between endothelial dysfunction and risk factors for cardiovascular diseases, including hypertension, hypercholesterolemia, cigarette smoking, diabetes, and aging, has been reported (6–8, 12, 29, 33, 47, 48). Besides the above-mentioned traditional risk factors, it is well known that physical inactivity is associated with an increased risk of developing cardiovascular diseases (32). However, at present, the relationship between inactivity and endothelial dysfunction is not clear. In individuals with paraplegia, the part of the body below the lesion level is paralyzed and thus extremely inactive (20, 39). In contrast, the upper limbs are often relatively active because the arms are used for ambulation due to their wheelchair-bound life-style (43). A spinal cord injury (SCI), therefore, offers a unique “human model of nature” to assess peripheral vascular adaptations to inactivity (legs) and activity (arms) on endothelial function within one subject.

Healthy vessels are capable of accommodating to an increase in blood flow by dilating the internal vessel diameter, a phenomenon called flow-mediated dilation (FMD), which is mediated by nitric oxide (NO) release and, as such, indicative for endothelial function (8).

FMD has been shown to be reduced in patients with elevated independent risk factors for cardiovascular diseases such as hypertension, diabetes, smoking, and hypercholesterolemia (7, 12, 29, 47), as well as in the elderly (6, 24). Previous studies, in which the effect of physical activity on FMD was assessed, demonstrate that regular aerobic training enhances FMD in the brachial artery in healthy individuals (3, 10, 13, 17, 21, 45) and in patients with Type 1 diabetes and heart transplantation (17, 22). Hornig et al. (21) investigated the effect of 4 wk of daily handgrip training in patients with chronic heart failure and reported significantly improved FMD in the trained dominant arm but not in the nondominant arm, suggesting a local beneficial mechanism. However, the effect of extreme inactivity of one extremity and chronically increased activity of another on endothelial function has never been studied within one subject.

Therefore, the main aim of our study is to assess endothelial function, as measured by FMD, in the inactive extremity (below the lesion) and in the chronically active extremity (above the lesion) in SCI individuals. We hypothesize that endothelial function is impaired in the inactive legs and maintained or improved in the active arms.

METHODS

Subjects.

Eleven male SCI individuals with motor-complete spinal cord lesions between T1 and L1 (time since injury 11.6 ± 7.9 yr) and eleven male able-bodied controls (C) participated in the study. Subjects who smoked and were known to have cardiovascular diseases, diabetes, hypercholesterolemia, high blood pressure, or other cardiovascular comorbidity were excluded from the study. None of the subjects received any medication likely to interfere with the cardiovascular system. The Ethical Committee of the University Medical Center Nijmegen approved the study, and all subjects provided written, informed consent before participating. Subject characteristics are presented in Table 1.

View this table:
Table 1.

Subject characteristics of SCI individuals and control subjects

Protocol.

Measurements were carried out in the resting supine position between 9:00 and 11:00 AM after an overnight fast. Subjects were asked to empty their bladder before examination, and they refrained from alcohol, caffeine, nicotine, and exercise at least 12 h before the test. The same investigator performed all the measurements using an echo Doppler device. At first, resting blood cell velocity and diameter of the right common femoral artery (CFA), right superficial femoral artery (SFA), and the right brachial artery (BA) were measured. A large cuff was then placed around the upper thigh, ∼10 cm distally from the greater trochanter. The cuff was inflated to 220 mmHg suprasystolic pressure for 10 min. After cuff deflation, hyperemic flow velocity in the SFA was recorded on videotape for the first 25 s, followed by a continuous registration of the vessel diameter for 5 min. After a resting period of 5 min, the exact same procedure was performed for the BA with the exception that the cuff around the forearm was inflated for 5 min, followed by measurements of hyperemic flow and diameter.

Finally, after a resting period of 5 min, a spray of sublingual nitroglycerine (NTG, 400 μg) was administered in seven SCI individuals and all controls to determine the maximal endothelium independent vasodilatation, which is indicative for smooth muscle function. Vessel diameter of the SFA was continuously recorded between 2 and 6 min after the administration of the spray. Blood pressure was measured manually at the BA before and after the echo Doppler measurements by using a sphygmomanometer.

Reproducibility of the resting measurements and FMD protocol in the SFA and BA was assessed in eight control subjects who where measured twice within 2 wk.

Measurements and data analyses.

Red blood cell velocities and systolic and diastolic vessel diameter of each artery were measured with an echo Doppler device (Megas, ESAOTE; Firenze, Italy) with a 5- to 7.5-MHz broadband linear array transducer. The sample volume was placed in the center of each vessel. For the CFA, images were made just below the inguinal ligament, ∼2 cm proximal of the bifurcation in the deep and SFA. SFA images were made ∼3 cm distal of the bifurcation, and brachial images were obtained ∼3 cm proximal of the olecranon process. The angle of inclination for the velocity measurements was consistently below 60°, and the vessel area was adjusted parallel to the transducer.

From each artery, 4 images with a total of 10–12 velocity profiles were obtained and manually traced afterwards by a single investigator. The average of these 10–12 Doppler spectra waveforms was used to calculate peak velocity (Vpeak) and mean velocity (Vmean).

For resting diameter measurements, two consecutive images in the longitudinal view were frozen at the peak systolic (Ds) and end-diastolic phase (Dd). Off-line, three measurements were performed per diameter image, and the mean diameter (D) was calculated by using the formula: 1/3·systolic diameter + 2/3·diastolic diameter.

Hyperemic velocity was recorded on videotape for the first 25 s after cuff release. A total of six to eight velocity profiles were obtained, and from each velocity profile, the flow velocity integral (FVI) was manually traced by a single investigator. The average of these six to eight velocity profiles was used to calculate peak and mean hyperemic velocity.

For both resting images and hyperemic responses, mean blood flow (in ml/min) was calculated as 1/4·Π(D)2·Vmean (cm/s)·60; peak blood flow (in ml/min) was calculated as 1/4·Π(Ds)2·Vpeak (cm/s)·60; regional peak wall shear rate (PWSR) was calculated as 4·Vpeak/Ds (s−1); and mean wall shear rate (MWSR) was calculated as 4·Vmean/D(s−1).

ΔFlow, ΔPWSR, and ΔMWSR were defined as the differences between rest and hyperemic responses.

Vessel diameters of the SFA and BA after reactive hyperemia were measured off-line from videotape at 45, 60, 90, 120, and 240 s after cuff release and at 2, 3, 4, 5, and 6 min after NTG administration. All diameters were measured at the end-diastolic phase of the cardiac cycle, immediately before the QRS complex (recognized by means of a simultaneous ECG signal). FMD in the SFA and BA and endothelium independent vasodilatation in the SFA were expressed as both the maximal absolute and relative diameter change in end-diastolic baseline diameter. The ratio between the maximal endothelium-dependent vasodilatation and the maximal endothelium independent vasodilatation was expressed as FMD/NTG. Because the FMD response is directly proportional to the magnitude of the stimulus (31), the FMD response was also expressed relative to the Δshear rate. Ratios were calculated for the FMD/ΔPWSR and FMD/ΔMWSR.

Statistical analyses.

Statistical analyses were performed using SPSS 8.0. A Student's t-test for independent groups was used to test differences in subject characteristics and to assess differences between SCI and C in resting arterial characteristics, hyperemic responses, endothelial-dependent vasodilatation, endothelial independent vasodilatation, and the ratios FMD/NTG and FMD/Δshear rates. Reproducibility, expressed as percentage of the indicated parameter, was calculated according to the following formula (1): Math in which x1 represents the result of the first test, x2 the result of the second test, and n the number of paired observations. The level of statistical significance for all tests was set at 5%.

RESULTS

Age, height, body mass, and blood pressure did not differ between the groups (Table 1).

The reproducibility for the resting measurements in eight control subjects in the SFA was 1.5% for diameter, 14% for blood flow, 9% for PWSR, and 13% for MWSR, and in the BA 1.0% for diameter, 18% for blood flow, 11% for PWSR, and 18% for MWSR. The reproducibility for the hyperemic flow and shear rates varied between 12 and 16% for the SFA and 16 and 19% for the BA. The reproducibility for the relative FMD changes was 15% for the SFA and 16% for the BA measurements.

Resting characteristics.

Vessel diameter of the CFA and SFA were significantly reduced, and PWSR and MWSR of the CFA and SFA were significantly elevated in SCI compared with controls (Table 2). No differences in blood flow of the CFA and SFA were found between SCI and controls. Resting arterial characteristics of the BA were not different between the groups (Table 2).

View this table:
Table 2.

Resting arterial characteristics, CFA, SFA, and BA

Hyperemic responses.

In the SFA, hyperemic flow and Δflow after arterial occlusion were significantly higher in the controls than in SCI (P < 0.001). Absolute hyperemic shear rates were significantly higher in SCI (P < 0.05), whereas Δshear rates did not differ between the groups (Tables 3 and 4).

View this table:
Table 3.

Absolute hyperemic responses in SFA and BA

View this table:
Table 4.

Differences between hyperemic responses and resting values in SFA and BA

In the BA, absolute hyperemic flow, Δhyperemic flow, and Δshear rates were significantly higher in SCI than those in controls (P < 0.05; Tables 3 and 4).

Flow-mediated dilation.

In the SFA, absolute changes in FMD were not different between groups (SCI: 0.73 ± 0.01 mm; C: 0.74 ± 0.02 mm), whereas relative changes in FMD were significantly greater in SCI than in controls (SCI: 14.1 ± 1.3%; C: 9.2 ± 2.3%) (P < 0.01) (Fig. 1, A and B). In the BA, absolute and relative changes in FMD were not different between groups (SCI: 0.56 ± 0.1 mm, 12.5 ± 2.9%; C: 0.6 ± 0.1 mm, 14.2 ± 3.3%) (Fig. 1, A and B). Absolute diameters of SFA and BA at rest, posthyperemic (FMD), and after NTG administration are presented in Table 5.

Fig. 1.

A: maximal absolute diameter increase (in mm) of the superficial femoral artery (SFA) during flow-mediated dilation (FMD), SFA after nitroglycerine (NTG) administration, and brachial artery (BA) during FMD in the spinal cord-injured group (SCI) and control group (C). B: maximal relative diameter increase (in %) of the SFA during FMD, SFA after NTG, and BA during FMD in the SCI and C groups. **P < 0.01, significantly different between groups.

View this table:
Table 5.

Absolute diameter at rest, post-FMD, and after NTG administration in SFA and BA in SCI and control individual

Ratio dilation to stimulus.

No significant differences between SCI and controls were found for the FMD/ΔMWSR ratio and the FMD/ΔPWSR ratio in the SFA (Table 6). In the BA, a significantly decreased FMD/ΔMWSR ratio was found for SCI compared with control (P < 0.05), whereas no differences between the groups were found for the FMD/ΔPWSR ratio. Figure 2 illustrates the relationship between %FMD and the Δhyperemic responses of MWSR and PWSR for SCI individuals and controls in SFA (Fig. 2A) and BA (Fig. 2B).

Fig. 2.

FMD response (in %) versus the reactive stimulus delta wall shear rate (hyperemic PWSR and MWSR − resting PWSR and MWSR, respectively) in the SFA (A) and BA (B) in the SCI individuals and the C group.

View this table:
Table 6.

Ratios of FMD/Δ shear rates in SFA and BA in SCI and control individuals

Endothelium independent vasodilatation.

Absolute diameter change after NTG administration was significantly greater in controls than in SCI (SCI: 0.77 ± 0.1 mm; C: 1.06 ± 0.2 mm) (P < 0.01), whereas relative changes were not different between the groups (SCI: 15.6 ± 2%; C: 13.4 ± 2.3%) (Fig. 1, A and B). The ratio between relative changes in FMD/NTG in the SFA was significantly higher in SCI than in controls (SCI: 0.94 ± 0.1; C: 0.70 ± 0.2) (P < 0.01) (Fig. 3).

Fig. 3.

Ratio between maximal diameter increase during FMD and after NTG administration (FMD/NTG) in the SFA in the SCI and the C group. **P < 0.01, significantly different between groups.

DISCUSSION

In contrast to our hypothesis, the results of the present study demonstrate that SCI individuals have a preserved endothelial function in the inactive legs and possibly an attenuated endothelial function in the active arms compared with controls.

Vascular endothelial function, expressed as percentage change in FMD, was enhanced in the femoral artery of SCI individuals compared with controls, whereas no differences between SCI and controls were found in the relative FMD response of the BA. When the stimulus is taken into account (using the FMD/Δshear rate ratio as index for endothelial function), no differences between the groups were found for the superficial femoral artery, whereas the FMD/Δshear rate ratio was significantly lower in the BA of SCI individuals compared with controls.

The reproducibility of the echo Doppler measurements varied from 1 to 1.5% for resting diameters, 9 to 18% for resting and hyperemic flow velocity parameters, and 15 to 16% for the FMD measurements, which is in line with values previously reported (5, 14, 19). Therefore, the reproducibility found in the present study, as well as the inclusion of a control group measured with the same testing procedures, supports the robustness of our observations.

Resting values and hyperemic responses.

In accordance with previous studies, the diameter of the femoral artery was significantly reduced, and resting shear rate levels were significantly elevated in SCI compared with controls (5, 20, 39). Resting common femoral blood flow was not significantly different between SCI and controls, which is probably due to two outliers in the SCI group with relatively high flows (550 and 680 ml/min). The unusual high flow in the two SCI subjects is due to a biphasic velocity profile, which is indicative for a low peripheral resistance. The latter is in contrast with the high leg vascular resistance usually found in SCI individuals (5). Brachial arterial resting characteristics are in agreement with previous studies and did not differ between SCI and controls (14).

The reactive hyperemic blood flow in the femoral artery, a marker for the peak vasodilator capacity, was significantly lower (i.e., 62%) in the legs of SCI compared with controls, which is in agreement with findings by Nash et al. (34). The atrophy of the vascular bed in the legs of SCI probably reflects adaptations to a lower metabolic demand as a consequence of the extreme inactivity of the paralyzed legs. In line with these findings, previous studies demonstrated diminished reactive hyperemic responses after a period of lower limb casting (26) and bed-rest immobilization (15). These changes in reactive hyperemia after a period of inactivity maybe the result of functional as well as structural changes in the downstream vascular bed.

On the other hand, hyperemic responses in the relatively active upper limb in SCI were significantly increased compared with controls (i.e., 54%), indicating that chronic upper extremity activity leads to an enhanced vasodilator capability of the vascular bed. These observations are supported by similar findings in SCI individuals by Shenberger et al. (43) and are in line with earlier studies showing an increased peak vasodilator response in the active arms of tennis players compared with controls (45) and an improved vasodilator capacity in the BA after 4 wk of handgrip training (3, 46). The responsible mechanisms for this training-induced elevation in hyperemic response may again be related to functional as well as structural adaptations in the vascular system.

Endothelial function.

The enhanced relative FMD response in the inactive legs of SCI compared with controls was an unexpected finding, because most training studies show that increased physical activity leads to an enhanced endothelial function (3, 10, 13, 17, 21, 45). Data on the effect of inactivity are controversial. Previous animal studies (23, 40, 51) mimicking physical inactivity by hindlimb unloading report inconsistent results with some studies showing a decreased nitric oxide synthase (NOS) gene expression and concomitant attenuated endothelium-dependent vasodilator responses, whereas other studies (38, 50) report an upregulation of the NOS system. In humans, a recent bed-rest study demonstrated that plasma nitrate and nitrite concentration, an indicator of endogenous NO production, was decreased after 14 days of bed-rest immobilization (25). However, Bonnin et al. (4) reported an enhanced FMD response in the BA after 7 days of bed-rest deconditioning. The authors reported no changes in resting and hyperemic parameters and no changes in endothelium independent vasodilatation from pre- to post-bed rest, which seems to indicate that the enhanced FMD cannot be attributed to differences in the stimulus for NO release or to an increased sensitivity of arterial smooth muscle cell for NO.

Caution should be taken in the interpretation of our FMD results because of differences in baseline vessel diameter of the femoral artery as well as differences in the trigger for FMD between SCI and controls. In previous studies (8, 41), an inverse relationship between vessel size and FMD has been reported. In accordance with this, relative FMD responses can be expected to be higher in SCI subjects, who have smaller diameters of the femoral artery than controls. The absolute responses in FMD, however, were not different between the groups. Because our groups have significantly different baseline diameters, a possible better way to compare the FMD response is to express the diameter change relative to the trigger (i.e., the difference between maximal hyperemic flow velocity and resting flow velocity) (31, 44). When the FMD/Δshear rate as an index for endothelial function is considered, no significant differences between SCI and controls were found in the femoral artery. These findings are illustrated in the plots of Fig. 2A, in which, however, the SCI subjects consistently show a larger relative FMD response per Δshear rate, indicating a preserved endothelial function in the inactive legs of the SCI individuals.

In the BA the SCI group experienced a larger hyperemic stimulus, but exhibited less dilation, resulting in a significantly decreased FMD/ΔMWSR ratio. These findings suggest that SCI individuals may have endothelial dysfunction compared with controls. However, calling a 12.5% FMD response in BA of the SCI individuals “dysfunctional” may be an overstatement, because this is still a large response compared with values from literature in which FMD responses of ∼10% are reported in healthy individuals (8, 11). In addition, in the present study, four subjects in the control group participated in racket sports like tennis and table tennis, which may contribute to the relatively high brachial FMD responses of about 14% in our control subjects.

According to the minimum cost theory, the human arterial system strives to maintain a constant shear stress by adapting the internal vessel diameter to chronic changes in blood flow. This arterial remodeling process has been shown to depend on an intact endothelium (28, 30). In SCI individuals, the vessel diameter of the femoral artery decreases excessively leading to an increase in basal shear stress. The almost doubled resting femoral shear stress levels in SCI individuals may suggest that this process is disturbed after a period of extreme inactivity and or paralyses, which may be indicative for endothelial dysfunction. The possibility of eventual endothelial damage due to chronically elevated shear stress is supported by earlier studies by Fry (16) and more recent by Nomura et al. (35), who present evidence that high shear rates may contribute to the development of atherosclerotic processes as has been reported previously to occur for regions of low shear stress levels (27).

In SCI individuals, it may well be that the enhanced levels of basal shear stress lead to an upregulation of endothelial NOS, because it has been shown that shear stress is a potent physiological stimulus for NO release. Previously, it has been shown that NOS gene expression in endothelial cells is augmented after exposure to increased shear stress levels (49) and similar observations were made in animal training studies, where repeated episodes of increased shear stress seemed to be the basis for a NOS mRNA upregulation (42). Beside the chronically enhanced resting shear stress levels, a lack of variation in shear stress in the vessels supplying the paralyzed and inactive leg muscles of SCI individuals (this in contrast with the great variation in shear stress levels in the active legs of ambulant able-bodied individuals) may contribute to an upregulation in the NO pathway. The present study, however, shows that vascular smooth muscle function was not altered in SCI by demonstrating no differences between SCI and controls in the relative NTG response. The fact that the FMD response and the NTG response were almost similar in the SCI group (FMD/NTG ratio: 0.94) means that the endothelium-dependent FMD response in the inactive legs of the SCI individuals reaches approximately maximal achievable vasodilatation levels (36). Taken into account that the relative dilatory response to NTG was not different between the groups, the explanation for this maximal FMD-induced vasodilation is likely related to a higher NO release in SCI individuals. In controls, the FMD/NTG ratio shows a 30% vasodilator reserve for FMD, which is in agreement with previous studies in the BA reporting a FMD/NTG ratio of 60–70% in healthy controls.

Limitations.

We consider the SCI population as a unique “human model of nature” to assess peripheral vascular adaptations to extreme inactivity. As valuable as information is from this patient population, one should be cautious to extrapolate these results to the general population because of other unique pathologies underlying SCI, such as disturbed sympathetic innervation. Although animal experiments have shown that sympathectomy may affect endothelial function, i.e., long-term sympathectomy in rats causes a decrease in endothelial NOS expression and an increase in endothelin-1 (2), results from several human studies suggest that endothelial function after chronic sympathectomy does not change (9). In addition, previous studies in SCI have shown that most of the adaptations in the circulatory system in SCI are reversible by functional electrostimulation training (5, 18), which suggests that the adaptations in the inactive and paralyzed legs in SCI seem to result primarily from deconditioning.

In conclusion, vascular endothelial function, expressed as percentage change in FMD, was enhanced in the femoral artery of the SCI individual compared with controls, whereas no differences between the groups were found in the relative FMD response of the BA. When the stimulus is taken into account (using the FMD/Δshear rate ratio as an index for endothelial function), the results indicate that SCI individuals have a preserved endothelial function in the inactive legs and possibly an attenuated endothelial function in the active arms compared with able-bodied controls.

GRANTS

This study was part of the research program “Physical strain, work capacity and mechanisms of restoration of mobility in the rehabilitation of individuals with spinal cord injury,” and is made possible with financial support of the Dutch Organization for Health Research and Development (ZON/MW).

Acknowledgments

We acknowledge the participation of all subjects in the study. In addition, we acknowledge Carola van Amerongen and Marc Awater for excellent support with testing procedures and data analyses.

Footnotes

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract