In the present study, the effects of L-dopa treatment on cardiovascular variables and peripheral venous tone were assessed in 13 patients with Parkinson's disease (PD) with Hoehn and Yahr stages 1–4. Patients were investigated once with their regular treatment and once after 12 h of interruption of L-dopa treatment. L-Dopa intake significantly reduced systolic and diastolic blood pressure, heart rate and plasma noradrenaline and adrenaline in both the supine and upright (60°) positions. A significant reduction in stroke volume and cardiac output was also seen with L-dopa. The vascular status of the legs was assessed through thigh compression during leg weighing, a new technique developed in our laboratory. Healthy subjects were used to demonstrate that this technique provided reproducible results, consistent with those provided by strain gauge plethysmography of the calf. When using this technique in patients with PD, L-dopa caused a significant lowering of vascular tone in the lower limbs as shown, in particular, by an increase in venous distensibility. Combined with the results of the orthostatic tilting, these findings support that the treatment-linked lowering of plasma noradrenaline in patients with PD was concomitant with a significant reduction in blood pressure, heart rate and vascular tone in the lower limbs. These pharmacological side-effects contributed to reduce venous return and arterial blood pressure which, together with a lowered heart rate, worsened the haemodynamic status.

INTRODUCTION

The frequent arterial hypotension in PD (Parkinson's disease) [14] has been ascribed to the early impairment of autonomic cardiac control, which has been shown clearly through imaging of cardiac sympathetic activity [57]. Clinically, hypotension is frequently observed as orthostatic or postprandial in patients with more advanced disease [3,4,8]. The side-effects of some pharmacological treatments have been shown to contribute to hypotension. The dopaminergic D1 agonist CY 208-243 has been reported to lower BP (blood pressure), HR (heart rate) and noradrenaline (norepinephrine) [9]. We have reported previously [10] that moderate L-dopa regimens lowered both supine and upright BP in long-term treated, as well as in newly diagnosed, PD patients with a moderate degree of disease. The L-dopa-linked decrease in upright BP reflected a dose-dependent lowering of plasma noradrenaline. In addition, the same study [10] showed that L-dopa treatment also lowered HR in both supine and upright positions. This was different from the HR increase concomitant with the BP decrease found in other pathophysiological conditions [11].

Besides the effects on arterial tone and HR control, a lowering of the venous tone would impair circulatory homoeostasis further. Venous compliance contributes to the control of venous return to the heart and, hence, CO (cardiac output) [12]. The effect of L-dopa on veins has not been fully evaluated in human subjects. Therefore, in the present study, we assessed the changes in vascular tone that were associated with L-dopa treatment in addition to the cardiovascular responses to orthostatic tilting. Venous distensibility was assessed in the lower limbs as changes in leg and feet weight during venous TC (thigh compression), using a technique developed in our laboratory and described in the present study. Before using this technique in patients with PD, we assessed the reproducibility of the technique in healthy subjects and compared the results with those provided by SGP (strain gauge plethysmography) of the calf. We also determined whether the technique provided valuable information about pharmacologically induced changes in vascular tone in healthy subjects.

METHODS

Subjects

The study was approved by the Institutional Ethics Committee, and the subjects gave their written informed consent. Two groups of subjects were studied. First, eight healthy subjects (age, 21–48 years; height, 1.68–1.95 m; and weight, 68–91 kg) were involved in the evaluation of the leg-weighing method, i.e. assessment of the reproducibility of the weighing measurements, assessment of the effect of transcutaneous NG (nitroglycerine) and comparison of leg weighing with SGP of the calf.

Secondly, 13 patients with PD and treated with L-dopa (five women; age, 47–76 years; symptom duration, 1–12 years; Hoehn and Yahr stage, 2.0±1.1) were included in the study (Table 1). These patients were investigated twice, once after having taken their usual morning dose of L-dopa and once after 12 h off medication, i.e. without their morning L-dopa intake. In these patients, the two test sessions were completed 1–3 weeks apart in random order [10]. When in use, other medications (for example, dopaminergic agonists) were not discontinued.

Table 1
Clinical characteristics of the patients with PD

UPDRS, Unified PD Rating Scale.

Patient no.AgeSexDisease duration (years)UPDRSHoehn and Yahr stageL-Dopa (mg/day)
65 Female 12 20 2.5 500 
63 Female 15 52 312 
48 Male 11 500 
71 Female 16 19 250 
75 Male 12 375 
74 Male 58 675 
73 Male 16 250 
74 Male 12 187 
72 Female 30 500 
10 73 Male 1.5 27 750 
11 52 Female 1.5 14 187 
12 76 Male 16 250 
13 47 Male 2.5 24 375 
Mean±S.D. 66.4±10.6  5.9±5.3 23.9±15.0 2.0±1.1 393.3±181.3 
Patient no.AgeSexDisease duration (years)UPDRSHoehn and Yahr stageL-Dopa (mg/day)
65 Female 12 20 2.5 500 
63 Female 15 52 312 
48 Male 11 500 
71 Female 16 19 250 
75 Male 12 375 
74 Male 58 675 
73 Male 16 250 
74 Male 12 187 
72 Female 30 500 
10 73 Male 1.5 27 750 
11 52 Female 1.5 14 187 
12 76 Male 16 250 
13 47 Male 2.5 24 375 
Mean±S.D. 66.4±10.6  5.9±5.3 23.9±15.0 2.0±1.1 393.3±181.3 

The patients were included because they had no history or sign of diabetes, heart disease, hypertension, neuropathy or other medical disorders known to alter autonomic function. None of them was treated with drugs known to influence cardiovascular function or the activity of the autonomic nervous system (other than antiparkinsonian drugs). All the test sessions were completed in the morning, at least 2 h after a light breakfast (containing no caffeinated beverages).

All measurements were performed in quiet rooms where the ambient temperature was 22–25 °C and the lights dimmed.

Assessment of peripheral vascular tone by TC and leg weighing

TC by means of cuff inflation below diastolic BP reduces the cross-sectional area of the underlying veins and causes an immediate flushing of blood into the abdomen, followed by a slower pooling of blood in the legs and feet. This further filling of veins and capillaries in legs and feet depends on the distensibility of capacitive vessels. Thus changes in blood volume of the legs and feet reflect both the increase in venous BP and compliance of the capacitive vessels. Also, during the early vein filling triggered by TC, the rate of volume changes in the legs directly reflects arterial inflow [13]. All of these changes have often been assessed by calf SGP. Depending on each subject's size, thigh cuffs of 22 or 17 cm long (CC22 or CC17 respectively; Hokanson) were used. They were connected to a source of compressed air through a rapid cuff inflator (models AG-101 and E-20; Hokanson). With this device, inflation to the preset pressure was achieved in 4–5 s. Cuff deflation from any pressure was achieved within 2 s. Thigh cuffs were not inflated beyond 60 mmHg to avoid arterial compression.

Leg weighing relies on the assumption that the continuous measurements of a vertical force sensor on the leaning point under the calf reflect the instantaneous weight of this body segment, i.e. legs, feet and part of the thighs. In a motionless resting subject, the rapid changes (with a time constant of a few seconds) in the vertical force recorded by the sensor reflect a change in blood volume inside the vascular network of the lower limbs. Accordingly, a table was constructed to continuously record the vertical force received by a pan supporting the lower limbs. This device was used to assess the vascular tone in the leg and feet through graded TC manoeuvres. The leg-weighing device consisted of mechanical and electronic components. The subject leaned on a table with calves on one pan (60 cm wide and 37 cm length), which could be slid on two rails, at a chosen distance from the table, to fit the anatomy of each subject. The pan rested on a force transducer (AG type; Scaime) whose force range was 294.2 N (30 kg-force) and sensibility was 59 mN (or 6 g-force). The transducer was connected to an electrical box with a potentiometer to adjust the zero of the transducer signal. This box was connected to an analogue/digital acquisition card inserted into a microcomputer. The ECG signal of an electrocardioscope was also recorded. Software for signals acquisition and processing was developed using Visual C++. The acquisition frequency was set at 650 Hz. Given the high acquisition frequency, smoothed average values of the leaning force (or ‘leg weight’) were computed by averaging 50 consecutive values. A first smoothed value was obtained from the first 50 values. The second smoothed value was obtained by repeating the same process from the second measurement of the first 50 series, i.e. a one-point progression step (see Figure 1). These smoothed weight data were then transferred into Microsoft Excel® files.

Serial changes in leg weight during TC

Figure 1
Serial changes in leg weight during TC

The thick white line drawn through the raw data (650 Hz sampled) is the moving average. Stages 1–5 are described in the Results section. ΔW is the change in weight of the two legs (and feet) caused by TC. ΔV is a reminder of volume change when calculated from calf SGP. AFI is calculated as the slope of weight increase immediately after inflation of the thigh cuffs (tangent to the initial part of the increase). As an example, the Figure was drawn with data from a TC of 3 min. base, resting baseline.

Figure 1
Serial changes in leg weight during TC

The thick white line drawn through the raw data (650 Hz sampled) is the moving average. Stages 1–5 are described in the Results section. ΔW is the change in weight of the two legs (and feet) caused by TC. ΔV is a reminder of volume change when calculated from calf SGP. AFI is calculated as the slope of weight increase immediately after inflation of the thigh cuffs (tangent to the initial part of the increase). As an example, the Figure was drawn with data from a TC of 3 min. base, resting baseline.

During each TC manoeuvre, the rate of vascular filling was quantified as the slope of the time-related weight change at the beginning of the increasing weight curve following application of TC pressure. As it probably tracked the arterial inflow, the value of this slope was called the AFI (arterial flow index) [13], as for SGP, and was expressed as mN/min.

The changes occurring during each TC were calculated from leg weight or calf circumference (SGP), which varied in relation to the amount of blood accumulated in the vascular network of legs and feet. Changes were expressed as mN (leg weighing) or ml (SGP). With weight data, changes were calculated as the difference between baseline before cuff inflation and the plateau reached during cuff inflation (see Figure 1). When SGP was used, the difference between the plateau and the minimum following cuff deflation was usually considered with this measurement [13].

According to SGP procedures, VDI (venous distensibility index) was calculated as the ratio of the difference between the weight changes with TC of 50 and 30 mmHg to the pressure difference, i.e. 20 mmHg [13], and expressed as mN/mmHg. Plotting the weight changes against the different TC pressures displayed the shape of the relationship and allowed the calculation of Cslope (slope of compliance; expressed as mN/mmHg).

To circumvent a loss of accuracy by a lever effect, the distance of the leg pan to the table, and the distances of anatomical landmarks (e.g., iliac crest, tibial condyle and malleoli) to the edges of the weighing pan, were carefully recorded. These marks were used to ensure that each subject was placed in a similar position during testing. Cuffs were placed loosely around the thighs during an initial 15 min rest period.

After the initial rest period, cuffs were connected to the cuff inflator. The subjects were asked to avoid any movement and to relax during the 8-min period of data acquisition i.e. 1 min control period, 4 min with thigh cuffs inflated at the controlled pressure and 3 min after deflation of cuffs. TC was achieved at successively increasing cuff pressures. Five TC pressures (20, 30, 40, 50 and 60 mmHg) were successively applied in healthy subjects.

Reproducibility of leg weighing measurements

The reproducibility of leg weighing technique was assessed by performing the measurements with TC pressures at 30 and 50 mmHg three times 1 h apart in the eight healthy subjects.

Effects of NG

The effects of NG were assessed in the same subjects. The five sequences (20, 30, 40, 50 and 60 mmHg) of leg weighing were repeated on the next day, at the same time of the day 2 h after application of an NG skin patch (500 mg of Nitriderm®; Novartis Pharma).

Comparison of leg weighing with SGP of the calf

As the calves are resting on the weighing pan in the leg-weighing technique, which might lessen the quality of SGP measurements, two measurements were performed separately in a randomly allocated order for each subject. Thus the five TC pressures were performed once for segmental weighing and once for SGP during which the leg rested on the heel instead of calf. SGP was performed with a Perivein® plethysmograph with automatic resetting to zero and calibration (Jansen Scientific Instruments). The principle and details of the method have been described previously [13,14].

Tilt testing and measurement of haemodynamic variables and vascular tone in patients with PD

The subjects rested initially in the supine position for 20 min after having ECG electrodes attached. An intravenous catheter was then inserted into a vein of the left forearm, and the subject rested for another 15 min after which a venous blood sample was collected whilst supine. The finger cuff of a photoplethysmograph sensor for BP (Finapres®; Ohmeda) was then placed on to the third finger of the right hand and kept at the heart level for the complete test sequence to avoid changes in hydrostatic pressure. HR and BP were collected continuously, as well as during the 10 min of rest in the supine position and 10 min after upright tilting to 60°. Plasma concentrations of noradrenaline, adrenaline and dopamine were determined in the blood sample of each subject at the end of the supine and upright periods by using a sensitive radioenzymatic method [15]. In the supine position SV (stroke volume) and CO were evaluated by using thoracic impedance cardiography (Physioflow; Manatec Biomedical) [16,17].

At 20 min after the postural testing, peripheral vascular tone was assessed during leg weighing. It was more difficult for these patients to remain motionless for a long period of time on the weighing device than for healthy subjects. Therefore, to reduce the occurrence of discomfort and the probability of movements likely to impair the weighing measurements, only TC pressures of 30, 40, and 50 mmHg were performed.

Statistics

Data are means±S.E.M. Changes in BP and HR were evaluated by ANOVA for repeated measurements (numerous serial measurements in the same position). Student's t test and Wilcoxon's paired signed rank were used to detect changes in cardiovascular and hormonal variables. Statistical significance was considered when P<0.05. The reproducibility of weight changes was expressed as individual CVs (coefficients of variation). A simple linear correlation was first used to compare the absolute changes assessed by SGP or leg weighing.

Secondly, volume changes (SGP) and weight changes (leg weighing) were expressed as relative variations (%), which allowed comparisons of quantities of similar magnitude. To evaluate the agreement and bias between the two methods, Bland–Altman plots were used [18], where the differences between relative changes were plotted against the mean of the relative changes for each set of TCs. Limits of agreement were defined as means±two S.D.s. The Spearman rank correlation was used to assess the relationship between the treatment-linked changes in the leg weight effects of TC and plasma noradrenaline in PD patients.

RESULTS

Reproducibility of leg weighing compared with calf SGP in healthy subjects

Changes in leg weight caused by TC

The baseline leg weight was steady after the initial 15 min rest in every subject. Inflation of the thigh cuffs resulted in the following sequence of weight changes: (1) an initial brief decrease, which reflected the flushing of blood from thighs toward the abdomen; (2) a constant rate increase (AFI); (3) a progressive smoothing of the weight increase; (4) a final levelling off at a maximal weight; and (5) a return to baseline level after a few oscillations (Figure 1). The maximal weight increase (ΔW) from the lowest point following the beginning of cuff inflation to the plateau reflected the maximal volume of blood accumulated in the vascular network of legs and feet during TC.

Reproducibility of the changes in leg weight

A lever effect was measured when the leg pan was placed 8 cm closer to, or farther from, the table, i.e. the leaning point of the subject's leg closer or farther from the pelvis. However, for each subject when the distance between the leg pan and the table was reproduced within 2 cm and when the anatomical landmarks were carefully positioned on the table and pan, reproducible results were obtained. When 30 and 50 mmHg were repeated three times, there was no overlap in the increase in the leaning force values at one TC pressure compared with values at the other pressure in any subject. The individual CVs were between 3.5 and 11.5% for TC at 30 mmHg and between 0.7 and 11.7% at 50 mmHg (Table 2b). Similar ranges were obtained for the individual CVs for AFI (Table 2a).

Table 2
Reproducibility of leg weighing with TC in eight healthy male subjects

AFI and steady increases in leg-leaning force in three successive TC trials. (a) AFI during thigh venous occlusion at 50 mmHg. (b) Changes in leg weight caused by TC.

(a)
AFI (mN/min)
SubjectTrial 1Trial 2Trial 3MeanCV (%)
S1 1755 1638 1667 1697 3.5 
S2 3177 3217 3442 3275 4.3 
S3 2962 3285 3432 3226 7.5 
S4 2864 2991 3138 2991 4.6 
S5 3903 3226 3756 3628 9.8 
S6 3079 2844 3226 3050 6.3 
S7 2108 2373 2442 2305 7.6 
S8 1922 2226 2010 2050 7.6 
(b)
Changes in leg weight (mN)
Thigh cuff pressure…30 mmHg50 mmHg
SubjectTrial 1Trial 2Trial 3MeanCV (%)Trial 1Trial 2Trial 3MeanCV (%)
S1  1687 2078 1746 1863 11.5 3472 3609 3776 3619 4.2 
S2  2550 3040 2775 2785 8.8 6168 5698 5619 5825 5.1 
S3  3295 3854 3648 3599 7.8 7120 7992 8522 7875 8.9 
S4  2138 2452 2452 2344 7.7 4727 4786 4727 4746 0.7 
S5  2079 2432 2481 2334 9.4 4590 4835 4521 4648 3.5 
S6  2471 2726 2805 2667 6.5 4795 5021 5570 5129 7.8 
S7  3472 3648 3746 3619 3.8 4903 5276 6149 5443 11.7 
S8  2275 2432 2471 2393 3.5 3903 4138 4197 4080 3.8 
(a)
AFI (mN/min)
SubjectTrial 1Trial 2Trial 3MeanCV (%)
S1 1755 1638 1667 1697 3.5 
S2 3177 3217 3442 3275 4.3 
S3 2962 3285 3432 3226 7.5 
S4 2864 2991 3138 2991 4.6 
S5 3903 3226 3756 3628 9.8 
S6 3079 2844 3226 3050 6.3 
S7 2108 2373 2442 2305 7.6 
S8 1922 2226 2010 2050 7.6 
(b)
Changes in leg weight (mN)
Thigh cuff pressure…30 mmHg50 mmHg
SubjectTrial 1Trial 2Trial 3MeanCV (%)Trial 1Trial 2Trial 3MeanCV (%)
S1  1687 2078 1746 1863 11.5 3472 3609 3776 3619 4.2 
S2  2550 3040 2775 2785 8.8 6168 5698 5619 5825 5.1 
S3  3295 3854 3648 3599 7.8 7120 7992 8522 7875 8.9 
S4  2138 2452 2452 2344 7.7 4727 4786 4727 4746 0.7 
S5  2079 2432 2481 2334 9.4 4590 4835 4521 4648 3.5 
S6  2471 2726 2805 2667 6.5 4795 5021 5570 5129 7.8 
S7  3472 3648 3746 3619 3.8 4903 5276 6149 5443 11.7 
S8  2275 2432 2471 2393 3.5 3903 4138 4197 4080 3.8 

Changes in leg weight at baseline and during NG stimulation

The average weight changes are shown in Table 3 together with the values measured when the same subjects received a NG skin patch. The weight change increased progressively with TC pressure, and these increases were significantly different step-by-step from 20 to 60 mmHg. NG significantly increased this effect at 30, 40, 50 and 60 mmHg. As a result, VDI and Cslope increased significantly during this pharmacological stimulation [VDI, 136±18 and 169±23 mN/mmHg in control and after NG treatment respectively (P<0.03); Cslope, 134±13 and 152±13 mN/mmHg in control and after NG treatment respectively (P=0.02)]. With NG, a significant decrease in AFI occurred at both 50 mmHg (4634±698 and 3383±388 mN/min in control and after NG treatment respectively; P<0.02) and 60 mmHg (4413±535 and 2677±227 mN/min in control and after NG treatment respectively; P<0.03).

Table 3
Leg weight changes caused by TC in healthy subjects

Changes in leaning force of the legs are expressed as group means±S.E.M. (mN). (a) Leg weight changes. (b) Calf volume changes assessed with SGP during TC. Eight healthy and fit subjects were studied without (control) or with (+NG) an NG skin patch. ΔV, volume change as determined by SGP. *P<0.05 and **P<0.01 compared with controls.

(a)
TC pressure (mmHg)
2030405060
Control (mN) 961±38 2745±115 4531±149 5590±212 6335±183 
+NG (mN) 1177±55 3177±111* 5345±180** 6433±235** 7071±197* 
(b)
TC pressure (mmHg)
2030405060
ΔV (ml) 75±2.47 111±3.2 148±4.6 172±4.6 196±5.3 
(a)
TC pressure (mmHg)
2030405060
Control (mN) 961±38 2745±115 4531±149 5590±212 6335±183 
+NG (mN) 1177±55 3177±111* 5345±180** 6433±235** 7071±197* 
(b)
TC pressure (mmHg)
2030405060
ΔV (ml) 75±2.47 111±3.2 148±4.6 172±4.6 196±5.3 

Comparison between leg weighing and SGP

The absolute (mN and ml) and relative (%) changes in weight (ΔW) and leg volume (ΔV) regularly increased with increasing TC pressure (P<0.001; Table 3). Changes in leg weight and calf volume of the eight subjects for the five TC pressures were significantly correlated (r=0.64; P<0.001; Figure 2, upper panel). A Bland–Altman plot for the mean difference of the percentage changes in SGP measurements minus the percentage changes in weight changes was 0.48% and the S.D. was 1.24% (limits of agreement −2%, 2.96%; Figure 2, lower panel). A total of 38 out of 40 measurements took place within the limits of agreement. The confidence interval of the S.E.M. difference ranged from −1.3 to −2.6% for the lower limit of agreement and from 2.3 to 3.6% for the upper limit of agreement. These intervals indicated that the differences between the two techniques were not sizeable and that the changes they revealed were clinically acceptable. AFI values determined at 50 and 60 mmHg from the rate of change in leg weight (2.63±0.18 and 2.36±0.21%/min respectively) were significantly correlated with the AFI values determined from the rate of changes in calf circumference (r=0.50; P<0.05).

Comparison of the changes in leg weight and calf volume measured by SGP caused by TC in healthy subjects
Figure 2
Comparison of the changes in leg weight and calf volume measured by SGP caused by TC in healthy subjects

Upper panel, linear correlation between individual results provided by the two techniques. △V, change in calf volume; △W, change in leg weight. Lower panel, Bland–Altman plot of the difference between leg weighing and calf SGP against the average of the two techniques. The lines of mean difference and limits of agreement (means±2 S.D.) are shown. Individual values at a TC pressure of 20 (▲), 30 (○), 40 (●), 50 (△) and 60 (◆) mmHg are shown.

Figure 2
Comparison of the changes in leg weight and calf volume measured by SGP caused by TC in healthy subjects

Upper panel, linear correlation between individual results provided by the two techniques. △V, change in calf volume; △W, change in leg weight. Lower panel, Bland–Altman plot of the difference between leg weighing and calf SGP against the average of the two techniques. The lines of mean difference and limits of agreement (means±2 S.D.) are shown. Individual values at a TC pressure of 20 (▲), 30 (○), 40 (●), 50 (△) and 60 (◆) mmHg are shown.

Effects of L-dopa on cardiovascular variables with upright tilting and on leg weighing in patients with PD

The usual intake of L-dopa treatment was associated with significantly lower values of systolic (P<0.001) and diastolic (P<0.001) BP compared with when the treatment had been interrupted (Table 4a). The supine SBP (systolic BP) was decreased by an average of 13 mmHg and upright SBP by 15 mmHg. DBP (diastolic BP) decreased by an average of 5 mmHg in both positions. HR was also significantly lowered in both positions with L-dopa, but the orthostatic adaptations of BP and HR were largely maintained. SV and CO were also significantly lowered in the supine position by 7% and 16% respectively, when the patients had taken their usual L-dopa.

Table 4
Effects of L-dopa treatment in patients with PD

Values are group means±S.E.M. (a) Changes in haemodynamic variables and plasma catecholamines during orthostatic tilting in patients with PD without (−) and with (+) L-dopa treatment. Relative change in SV and CO was −7.22% and −16.11% respectively. *P<0.05 compared with without (−) L-dopa (treatment effect); †P<0.05 compared with supine (posture effect). (b) Changes in leg weight and AFI during TC in patients with PD without (−) and with (+) L-dopa. ‡P<0.05 and ‡‡P<0.02 compared with without (−) L-dopa treatment.

(a)
SupineUpright
L-Dopa…++
SBP (mmHg)  136.9±11.4 123±10.2* 135.1±11.2 120.8±10.1* 
DBP (mmHg)  67.7±5.6 62.2±5.1* 74.6±6.1† 70.1±5.8*† 
HR (beats/min)  69.4±5.7 60.7±5.1* 74.6±6.2† 71.3±5.9*† 
SV (ml)  72.4±4.7 67.1±3.9*   
CO (litres/min)  4.7±0.6 3.9±0.5*  
Dopamine (pg/ml)  82.5±7.3 372.9±49.5* 83.7±6.8 397.5±56.1* 
Noradrenaline (pg/ml)  448.8±71.8 223.6±36.5* 605.7±69.4† 352.3±49.8*† 
Adrenaline (pg/ml)  51.4±5.9 31.2±3.7* 89.1±19.2† 48.1±6.5*† 
(b)
TC pressure (mmHg)
304050
Leg weight changes (mN)    
L-Dopa    
  − 1559±276 2.776±373 3128±491 
  + 2187±219‡ 3923±386‡ 4452±556‡‡ 
AFI (mN/min)    
L-Dopa    
  − 1452±324 2256±414 2244±486 
  + 2262±300‡ 3276±414‡ 3606±558‡ 
(a)
SupineUpright
L-Dopa…++
SBP (mmHg)  136.9±11.4 123±10.2* 135.1±11.2 120.8±10.1* 
DBP (mmHg)  67.7±5.6 62.2±5.1* 74.6±6.1† 70.1±5.8*† 
HR (beats/min)  69.4±5.7 60.7±5.1* 74.6±6.2† 71.3±5.9*† 
SV (ml)  72.4±4.7 67.1±3.9*   
CO (litres/min)  4.7±0.6 3.9±0.5*  
Dopamine (pg/ml)  82.5±7.3 372.9±49.5* 83.7±6.8 397.5±56.1* 
Noradrenaline (pg/ml)  448.8±71.8 223.6±36.5* 605.7±69.4† 352.3±49.8*† 
Adrenaline (pg/ml)  51.4±5.9 31.2±3.7* 89.1±19.2† 48.1±6.5*† 
(b)
TC pressure (mmHg)
304050
Leg weight changes (mN)    
L-Dopa    
  − 1559±276 2.776±373 3128±491 
  + 2187±219‡ 3923±386‡ 4452±556‡‡ 
AFI (mN/min)    
L-Dopa    
  − 1452±324 2256±414 2244±486 
  + 2262±300‡ 3276±414‡ 3606±558‡ 

Plasma catecholamines

The intake of L-dopa led to an increased level of plasma dopamine (P<0.001) and reduced significantly plasma levels of noradrenaline (P<0.001) and adrenaline (P<0.01) in both the supine and upright positions (Table 4a).

Indices of vascular tone

Both AFI and the steady increases in leg weight after TC were larger with L-dopa intake (Table 4b). Thus VDI and Cslope were both significantly greater (P<0.05) with than without L-dopa intake (VDI, 109±8.1 compared with 80±6.7 mN/mmHg respectively; Cslope, 113±10.8 compared with 78±5.7 respectively). In addition, the individual treatment-linked increases in venous filling during TC correlated significantly with the treatment-linked decreases in plasma noradrenaline (ρ=−0.418, P<0.01; Figure 3).

Correlation between the treatment-related difference in TC-related increases in leg weight and inhibition of plasma noradrenaline in patients with PD
Figure 3
Correlation between the treatment-related difference in TC-related increases in leg weight and inhibition of plasma noradrenaline in patients with PD

Individual values at a TC pressure of 30 (○), 40 (△) and 50 (●) mmHg are shown. The Spearman's rank correlation is also indicated.

Figure 3
Correlation between the treatment-related difference in TC-related increases in leg weight and inhibition of plasma noradrenaline in patients with PD

Individual values at a TC pressure of 30 (○), 40 (△) and 50 (●) mmHg are shown. The Spearman's rank correlation is also indicated.

DISCUSSION

The present study provides evidence that the L-dopa regimen increased leg venous distensibility and decreased arterial tone in patients with PD. Furthermore, the efficiency of lower limb weighing to assess changes in vascular tone of this peripheral part of the body was also demonstrated. In healthy subjects, the reproducibility of this technique was good when care was taken to reproduce each subject's positioning precisely. In particular, the lever effect did not preclude intrasubject comparisons. When measurements were repeated, the CV of leg weighing remained similar to that of SGP [14,19]. A comparison using the Bland–Altman plot indicated that the two techniques were concordant in evaluating different consequences of change in venous blood volume of the lower limbs. The calf circumference and the blood mass in the legs and feet probably change in the same direction, but larger amounts of blood in the feet (at higher thigh cuff pressure) may have a greater effect on the leaning force value. However, the intrasubject comparison of changes developed in the same direction with the two techniques, as shown by the significant correlations between the indexes of arterial vasomotor tone and venous distensibility.

The relaxant effect of NG on the venous wall was observed, as venous TC caused stepwise higher weight gains in the legs. This increased blood pooling reflected a larger venous compliance.

HR also increased at high TC pressures and this effect was larger with NG (results not shown). These results were consistent with other assessments of the effect of NG on venous compliance [20]. The same study also showed that NG increased sympathetic outflow and systemic vascular resistance [20]. Thus our finding of a NG-linked decrease of the leg-weighing AFI was consistent with an increased arterial tone.

Using the leg-weighing technique, we have found that usual L-dopa regimens increased venous distensibility in patients with PD. The weight increases during TC were smaller in the legs and feet of patients with PD than in normal subjects (Tables 3 and 4). Several factors may have contributed to this difference. The patients were both older and less fit than the control subjects. Venous distensibility of the lower limbs is higher in fit than in sedentary subjects [21]. It has also been assessed, both in vivo and in vitro, that vascular reactivity to drugs changed with aging [22,23]. In healthy humans, the age-related increase in MSNA (muscle sympathetic nerve activity) [24,25] may also reduce venous distensibility in the lower limbs.

Leg weighing with TC revealed a lowered venous vasomotor tone during L-dopa treatment. The accommodation of a larger blood volume in legs and feet during venous TC resulted from an increased venous distensibility. The lower plasma noradrenaline upon L-dopa intake probably resulted from decreased release, which contributed to the lower tone of both venous and arterial vascular smooth muscle. The activation of D2 presynaptic receptors by L-dopa might have caused the lowering of peripheral sympathetic outflow [26]. In addition, low doses of dopamine have been reported to decrease peripheral vascular resistance by activation of vascular dopaminergic receptors, whereas high doses produce vasoconstriction mediated by vascular α-adrenoceptors [27,28]. However, the venoconstricting effect of dopamine is 5–20 times less than that of noradrenaline [29]. In our patients, moderate L-dopa regimens reduced plasma noradrenaline and, in turn, increased venous distensibility in the lower limbs. This venous side-effect of treatment is a drawback to venous return and CO.

We have reported previously [10] that the usual L-dopa regimen contributed to a decrease in BP and HR in patients with PD with moderate degrees of disease. The present study confirmed that in such patients L-dopa lowered diastolic and systolic BP, HR and plasma noradrenaline. These decreases were paralleled by a decreased CO. Consistently, in the present study, leg weighing provided circumstantial evidence that usual L-dopa regimens lower arterial tone in the legs, as reflected by the larger AFI during TC when plasma noradrenaline was lower. All the pharmacological side-effects observed probably contributed to slow down the blood flow in a relaxed peripheral vascular network. Therefore the occurrence of orthostatic or postprandial symptoms of hypotension should not be unexpected in patients with PD treated with L-dopa.

Abbreviations

     
  • AFI

    arterial flow index

  •  
  • BP

    blood pressure

  •  
  • CO

    cardiac output

  •  
  • Cslope

    slope of compliance

  •  
  • CV

    coefficient of variation

  •  
  • DBP

    diastolic BP

  •  
  • HR

    heart rate

  •  
  • NG

    nitroglycerine

  •  
  • PD

    Parkinson's disease

  •  
  • SBP

    systolic BP

  •  
  • SGP

    strain gauge plethysmography

  •  
  • SV

    stroke volume

  •  
  • TC

    thigh compression

  •  
  • VDI

    venous distensibility index

We thank our colleagues D. Chavot, J. Galmiche, C. Portha, P. Severin and T. Moulin for their contribution to this study by referring their patients to us for testing. This work was supported with grants from the French Ministère de l'Education Nationale, de la Recherche et de la Technologie (EA3920) and by the Centre Hospitalier Universitaire de Besançon.

References

References
1
Calne
 
D. B.
Brennan
 
J.
Spiers
 
A. S. D.
Stern
 
G. M.
 
Hypotension caused by L-dopa
Br. Med. J.
1970
, vol. 
1
 (pg. 
474
-
475
)
2
Gross
 
M.
Bannister
 
R.
Godwin-Austen
 
R.
 
Orthostatic hypotension in Parkinson's disease
Lancet
1972
, vol. 
i
 (pg. 
174
-
176
)
3
Micieli
 
G.
Martignoni
 
E.
Cavallini
 
A.
Sandrini
 
G.
Nappi
 
G.
 
Postprandial and orthostatic hypotension in Parkinson's disease
Neurology
1987
, vol. 
37
 (pg. 
386
-
393
)
4
Senard
 
J. M.
Rai
 
S.
Lapeyre-Mestre
 
M.
, et al 
Prevalence of orthostatic hypotension in Parkinson's disease
J. Neurol. Neurosurg. Psychiatry
1997
, vol. 
63
 (pg. 
584
-
589
)
5
Takatsu
 
H.
Nishida
 
H.
Matsuo
 
H.
, et al 
Cardiac sympathetic denervation from the early stage of Parkinson's disease: clinical and experimental studies with radiolabeled MIBG
J. Nucl. Med.
2000
, vol. 
41
 (pg. 
71
-
77
)
6
Goldstein
 
D. S.
Holmes
 
C.
Li
 
S. T.
Bruce
 
S.
Metman
 
L. V.
Cannon
 
R. O.
 
Cardiac sympathetic denervation in Parkinson disease
Ann. Intern. Med.
2000
, vol. 
133
 (pg. 
338
-
347
)
7
Goldstein
 
D. S.
Holmes
 
C. S.
Dendi
 
R.
Bruce
 
S. R.
Li
 
S. T.
 
Orthostatic hypotension from sympathetic denervation in Parkinson's disease
Neurology
2002
, vol. 
58
 (pg. 
1247
-
1255
)
8
Piha
 
S. J.
Rinne
 
J. O.
Rinne
 
U. K.
Seppanen
 
A.
 
Autonomic dysfunction in recent onset and advanced Parkinson's disease
Clin. Neurol. Neurosurg.
1988
, vol. 
90
 (pg. 
221
-
226
)
9
Durieu
 
G.
Senard
 
J. M.
Rascol
 
O.
, et al 
Blood pressure and plasma catecholamine in never-treated parkinsonian patients: effect of a selective D1 agonist (CY 208-243)
Neurology
1990
, vol. 
40
 (pg. 
707
-
709
)
10
Bouhaddi
 
M.
Vuillier
 
F.
Fortrat
 
J. O.
, et al 
Impaired cardiovascular control in newly and long-term treated patients with Parkinson's disease: evidence for involvement of L-dopa therapy
Auton. Neurosci.
2004
, vol. 
116
 (pg. 
30
-
38
)
11
Toussirot
 
E.
Bouhaddi
 
M.
Cappelle
 
S.
, et al 
Abnormal heart rate control in ankylosing spondylitis
Ann. Rheum. Dis.
1999
, vol. 
58
 (pg. 
481
-
487
)
12
Rothe
 
C. F.
 
Shepherd
 
J. T.
Abboud
 
F. M.
 
Physiology of the capacitance vessels. Handbook of Physiology: The Cardiovascular System
Peripheral Circulation and Organ Blood Flow
1983
Bethesda
American Physiological Society
(pg. 
427
-
435
)
13
Louisy
 
F.
Schroiff
 
P.
Guell
 
A.
 
Changes in leg vein filling and emptying characteristics and leg volume during long-term head-down bed rest
J. Appl. Physiol.
1997
, vol. 
82
 (pg. 
1726
-
1733
)
14
Louisy
 
F.
Schroiff
 
P.
 
Plethysmography with optoelectronic sensors: reproducibility and correlation with mercury strain gauge plethysmography
Aviat. Space Environ. Med.
1995
, vol. 
66
 (pg. 
1191
-
1197
)
15
Sole
 
M. J.
Hussain
 
M. N.
 
A simple specific radioenzymatic assay for the simultaneous measurement of picogram quantities of norepinephrine, epinephrine and dopamine in plasma and tissues
Biochem. Med.
1977
, vol. 
18
 (pg. 
301
-
307
)
16
Richard
 
R.
Lonsdorfer-Wolf
 
E.
Charloux
 
A.
, et al 
Non invasive cardiac output evaluation during a maximal progressive exercise test, using a new impedance cardiograph device
Eur. J. Appl. Physiol.
2001
, vol. 
85
 (pg. 
202
-
207
)
17
Tordi
 
N.
Mourot
 
L.
Matusheski
 
B.
Hughson
 
R. L.
 
Measurements of cardiac output during constant exercises: comparison of two non-invasive techniques
Int. Sports. Med.
2004
, vol. 
25
 (pg. 
145
-
149
)
18
Bland
 
J. M.
Altman
 
D. G.
 
Statistical methods for assessing agreement between two methods of clinical measurement
Lancet
1986
, vol. 
i
 (pg. 
307
-
310
)
19
Boccalon
 
H.
Ginestet
 
M. C.
Longhi
 
R.
Puel
 
P.
 
Variations de la physiologie veineuse chez le sujet normal. Etude par pléthysmographie posturale à occlusion veineuse
J. Mal. Vascul.
1987
, vol. 
12
 (pg. 
150
-
157
)
20
Gisolf
 
J.
Westerhof
 
B. E.
van Dijk
 
N.
Wesseling
 
K. H.
Wieling
 
W.
Karemaker
 
J. M.
 
Sublingual nitroglycerin used in routine tilt testing provokes a cardiac output-mediated vasovagal response
J. Am. Coll. Cardiol.
2004
, vol. 
44
 (pg. 
588
-
593
)
21
Wecht
 
J. M.
De Meersman
 
R. E.
Weir
 
J. P.
Bauman
 
W. A.
Grimm
 
D. R.
 
Effects of autonomic disruption and inactivity on venous vascular function
Am. J. Physiol. Heart Circ. Physiol.
2000
, vol. 
278
 (pg. 
H515
-
H520
)
22
Egashira
 
K.
Inou
 
T.
Hirooka
 
Y.
, et al 
Effects of age on endothelium-dependent vasodilatation of resistance coronary artery by acetylcholine in humans
Circulation
1993
, vol. 
88
 (pg. 
77
-
81
)
23
Shipley
 
R. D.
Muller-Delp
 
J. M.
 
Aging decreases vasoconstrictor responses of coronary resistance arterioles through endothelium-dependent mechanisms
Cardiovasc. Res.
2005
, vol. 
66
 (pg. 
374
-
383
)
24
Ng
 
A. V.
Callister
 
R.
Johnson
 
D. G.
, et al 
Age and gender influence muscle sympathetic nerve activity at rest in healthy humans
Hypertension
1993
, vol. 
21
 (pg. 
498
-
503
)
25
Dinenno
 
F. A.
Dietz
 
N. M.
Joyner
 
M. J.
 
Aging and forearm postjunctional α-adrenergic vasoconstriction in healthy men
Circulation
2002
, vol. 
106
 (pg. 
1349
-
1354
)
26
Mannelli
 
M.
Ianni
 
L.
Lazzeri
 
C. H.
, et al 
In vivo evidence that endogenous dopamine modulates sympathetic activity in man
Hypertension
1999
, vol. 
34
 (pg. 
398
-
402
)
27
Goldberg
 
L. I.
Rajfer
 
S. O.
 
Dopamine receptors: application in clinical cardiology
Circulation
1985
, vol. 
72
 (pg. 
245
-
248
)
28
Okamura
 
T.
Yamazaki
 
M.
Toda
 
N.
 
Responses to dopamine of isolated human and monkey veins compared with those of the arteries
J. Pharmacol. Exp. Ther.
1991
, vol. 
258
 (pg. 
275
-
279
)
29
Harada
 
K.
Ohmori
 
M.
Kito
 
Y.
Fujimura
 
A.
 
Effects of dopamine on veins in humans: comparison with noradrenaline and influence of age
Eur. J. Clin. Pharmacol.
1998
, vol. 
54
 (pg. 
227
-
230
)