Abstract
Background: Patients with PD often have signs or symptoms of autonomic failure, including orthostatic hypotension. Cardiac sympathetic denervation occurs frequently in PD, but this has been thought to occur independently of autonomic failure.
Methods: Forty-one patients with PD (18 with and 23 without orthostatic hypotension) and 16 age-matched healthy volunteers underwent PET scanning to visualize sympathetic innervation after injection of 6-[18F]fluorodopamine. Beat-to-beat blood pressure responses to the Valsalva maneuver were used to identify sympathetic neurocirculatory failure and plasma norepinephrine to indicate overall sympathetic innervation.
Results: All patients with PD and orthostatic hypotension had abnormal blood pressure responses to the Valsalva maneuver and septal and lateral ventricular myocardial concentrations of 6-[18F]fluorodopamine-derived radioactivity >2 SD below the normal mean. In contrast, only 6 of the 23 patients without orthostatic hypotension had abnormal Valsalva responses (p < 0.0001 compared with patients with orthostatic hypotension), and only 11 had diffusely decreased 6-[18F]fluorodopamine-derived radioactivity in the left ventricular myocardium (p = 0.0004). Of the 12 remaining patients without orthostatic hypotension, 7 had locally decreased myocardial radioactivity. Supine plasma norepinephrine was lower in patients with than in those without orthostatic hypotension (1.40 ± 0.15 vs 2.32 ± 0.26 nmol/L, p = 0.005). 6-[18F]fluorodopamine-derived radioactivity was less not only in the myocardium but also in the thyroid and renal cortex of patients with PD than in healthy control subjects.
Conclusions: In PD, orthostatic hypotension reflects sympathetic neurocirculatory failure from generalized sympathetic denervation.
Patients with PD frequently have signs or symptoms of autonomic failure, such as constipation, urinary incontinence, orthostatic or postprandial lightheadedness, heat or cold intolerance, and erectile dysfunction.1-3⇓⇓ In particular, orthostatic hypotension occurs commonly in PD.4-7⇓⇓⇓ Because of increased susceptibility to falls and other accidental trauma, orthostatic hypotension in this setting can be not only disabling but also life-threatening.8,9⇓
Orthostatic hypotension in patients with parkinsonism has been attributed to treatment with levodopa10-12⇓⇓ or to another diagnosis, such as striatonigral degeneration13 or the parkinsonian form of multisystem atrophy.14,15⇓ Recent work has suggested an association between orthostatic hypotension and failure of reflexive sympathetically mediated cardiovascular stimulation in response to decreased venous return to the heart—sympathetic neurocirculatory failure.16
All of at least a dozen recent studies have agreed that patients with PD have a high prevalence of loss of sympathetic innervation of the heart, with low myocardial concentrations of radioactivity after injection of the sympathoneural imaging agents, 123I-metaiodobenzylguanidine13,17-27⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓ and 6-[18F]fluoro-dopamine.16 Neurochemical assessments during right-sided heart catheterization have confirmed that the low concentrations of radioactivity result from loss of functional cardiac sympathetic nerve terminals.16
Generalized sympathetic denervation would provide a straightforward explanation for sympathetic neurocirculatory failure and therefore orthostatic hypotension in PD. On this point the available literature is inconsistent. Findings based on 123I-metaiodobenzylguanidine scanning have led to the views that in PD cardiac sympathetic denervation occurs independently of autonomic failure and that the denervation is selective for the heart.19,27⇓ In contrast, patients with PD and orthostatic hypotension have lower mean plasma levels of norepinephrine, the sympathetic neurotransmitter, during supine rest than do patients without orthostatic hypotension,2,28-30⇓⇓⇓ supporting the notion of generalized rather than cardioselective sympathetic denervation.
This study addressed four questions: Does sympathetic neurocirculatory failure attend orthostatic hypotension in PD? Does the frequency of cardiac sympathetic denervation relate to that of orthostatic hypotension? Does plasma norepinephrine in supine patients distinguish PD with from PD without orthostatic hypotension? Do patients with PD have decreased sympathetic innervation in organs other than the heart?
Materials and methods.
The study protocol was approved by the Clinical Research Subpanel of the National Institute of Neurologic Disorders and Stroke. Each subject gave informed, written consent.
Subjects.
Thoracic PET scanning was performed after IV injection of 6-[18F]fluorodopamine in 41 patients with PD and 16 age-matched healthy volunteers (mean age, 62 ± 3 years). Of the 41 patients, 18 had orthostatic hypotension, defined by a decrease in systolic blood pressure of at least 20 mm Hg and decrease in diastolic pressure of at least 5 mm Hg between the supine and upright positions, and 23 did not. Among the patients with orthostatic hypotension, the mean age was 68 ± 2 years and duration of PD averaged 7 ± 2 years. Eleven patients were on levodopa/carbidopa at the time of testing, four had never taken levodopa/carbidopa, and three had discontinued levodopa/carbidopa and had normal plasma levodopa levels at the time of testing. Among the patients without orthostatic hypotension, the mean age was 61 ± 2 years and duration of PD averaged 7 ± 1 years. Seventeen patients were on levodopa/carbidopa at the time of testing, four had never taken levodopa/carbidopa, and two had discontinued levodopa/carbidopa and had normal plasma levodopa levels at the time of testing. All the volunteers had normal medical history, physical examination, electrocardiogram, and laboratory (blood and urine) screening test results.
Caffeine-containing beverages, cigarettes, and alcohol were not allowed for at least 24 hours before the testing. Patients were studied while taking their usual medications, which did not include drugs known to inhibit neuronal uptake of catecholamines but did include levodopa.
Valsalva maneuver.
Sympathetic neurocirculatory failure was defined by the combination of chronic, reproducible orthostatic hypotension and abnormal beat-to-beat blood pressure responses to the Valsalva maneuver—specifically, a progressive decline in blood pressure during phase II and absence of pressure overshoot in phase IV.31 Of the 41 patients, 23 did not have sympathetic neurocirculatory failure and 17 did; Valsalva blood pressure data from one patient with orthostatic hypotension were excluded because of inconsistent results.
PET scanning.
6-[18F]Fluorodopamine, synthesized as described previously,32 was infused IV at a constant rate for 3 minutes. Tomographic images (35 contiguous transaxial slices 4.25 mm apart) were acquired for up to 30 minutes. Three-dimensional PET scans were obtained using an Advance whole-body scanner (General Electric, Milwaukee, WI). Transmission scans of 2 and then 8 minutes in duration, using rotating 68Ge/68Ga pin sources, were obtained for attenuation correction and for confirming proper positioning in the scanner.
Scanning of the head was performed in nine patients with PD and sympathetic neurocirculatory failure, 10 patients with PD who did not have sympathetic neurocirculatory failure, and 10 healthy volunteers. To determine whether 6-[18F]fluorodopamine-derived radioactivity in regions such as the thyroid gland, salivary glands, and nasopharyngeal mucosa reflects neuronal uptake of the sympathoneural image agent, the healthy volunteers underwent repeat scanning of the head at least 1 hour after oral administration of 125 mg of desipramine, which blocks the cell membrane norepinephrine transporter.33
In most patients, tissue perfusion was assessed by PET scanning after IV injection of 13N-ammonia.33
Plasma norepinephrine.
Antecubital venous blood was drawn through an indwelling catheter, after at least 15 minutes of supine rest and after 5 minutes of standing (in one patient during lower body negative pressure at −15 mm Hg). Plasma norepinephrine was assayed by high-pressure liquid chromatography with electrochemical detection after batch alumina extraction, according to the method validated in our laboratory.34
Data analysis and statistics.
PET images were reconstructed after correction for attenuation and for physical decay of 18F. Cardiac images were analyzed as described previously.32 Briefly, circular regions of interest approximately half the ventricular wall thickness were placed on images of the septum, using time-averaged pictures of a single slice. Left ventricular septal radioactivity was averaged from two regions of interest, for the 5-minute scanning interval with midpoint at about 8 minutes after initiation of the infusion. For head scanning, static three-dimensional data were obtained for 10 to 15 minutes. Images of noncardiac structures, including the liver, spleen, renal cortex and renal pelvis, salivary glands, nasopharyngeal mucosa, and thyroid, were reconstructed and analyzed by manually drawing regions of interest outlining the structures. Radioactivity concentrations were normalized by correcting for the radioactivity concentration for the administered dose of radioactive drug per unit body mass of the subject and expressed as nCi-kg/mL-mCi.32
Mean values for 6-[18F]fluorodopamine-derived radioactivity and plasma norepinephrine in patient and control groups and in patient groups with and without orthostatic hypotension were compared using factorial analyses of variance or two-tailed, independent-means t-tests. Differences between groups in trends over time of 6-[18F]fluorodopamine-derived radioactivity were assessed by analyses of variance for repeated measures. Tables of frequency data were examined by calculation of χ2. Relationships between neurochemical values across individual patients were assessed by linear regression. A p value <0.05 defined significance.
Results.
Valsalva maneuver.
All 17 of 17 patients with PD and orthostatic hypotension had abnormal blood pressure responses in both phases II and IV of the Valsalva maneuver (1 patient had technically inadequate results). In contrast, only 6 of the 23 patients without orthostatic hypotension had this combination (χ2 = 21.9, p < 0.0001).
PET scanning.
All 18 patients with PD and orthostatic hypotension had low concentrations of 6-[18F]fluoro-dopamine-derived radioactivity throughout the left ventricular myocardium. The similarity of radioactivity concentrations in the left ventricular myocardium and chamber obviated visualization of the lateral ventricular wall in most patients; however, the interventricular septum could be identified from the negative contrast with the adjacent left and right ventricular chambers during infusion of 6-[18F]fluorodopamine. All 18 patients with PD and orthostatic hypotension had septal myocardial concentrations of 6-[18F]fluorodopamine-derived radioactivity that were > 2 SD below the mean concentration of the healthy control subjects (mean: 2,645 ± 160 vs 9,250 ± 604 nCi-kg/mL-mCi; p < 0.0001; figure 1).
Figure 1. Interventricular septal concentrations of 6-[18F]fluorodopamine-derived radioactivity in patients with PD and orthostatic hypotension (Park w/OH), patients with PD without orthostatic hypotension (Park w/o OH), and healthy control subjects (Normal).
In contrast, of the 23 patients without orthostatic hypotension who underwent thoracic PET scanning after injection of 6-[18F]fluorodopamine, only 11 had the combination of lack of visualization of the lateral ventricular wall and septal 6-[18F]fluorodopamine-derived radioactivity >2 SD below the normal mean. Therefore, whereas all patients with orthostatic hypotension had diffusely decreased 6-[18F]fluorodopamine-derived radioactivity in the left ventricular myocardium, only about half of patients without orthostatic hypotension had diffusely decreased radioactivity (χ2 = 13.3, p = 0.0003).
Of the 12 patients without orthostatic hypotension who did not have diffusely decreased 6-[18F]fluorodopamine-derived radioactivity in the left ventricular myocardium, 5 had septal radioactivity exceeding lateral wall radioactivity by >25% (figure 2). Another two patients had lateral wall radioactivity >2 SD below the normal mean but septal radioactivity within 2 SD of the normal mean. In contrast, among the 16 healthy control subjects, in 15 the septal:lateral wall radioactivity ratio was between 0.75 and 1.25, and the mean septal:lateral wall radioactivity ratio averaged about 1 (1.01 ± 0.03). Therefore, of the 23 patients with PD who did not have orthostatic hypotension, 17 (approximately 3/4) had diffuse or localized decreases in myocardial 6-[18F]fluorodopamine-derived radioactivity.
Figure 2. PET scans of the heart after IV administration of (top) 13N-ammonia and (bottom) 6-[18F]fluorodopamine in (A) a healthy control subject, and (B-D) patients with PD who did not have orthostatic hypotension. Note localized loss of 6-[18F]fluorodopamine-derived radioactivity in the lateral wall in (B and C) and diffuse loss of radioactivity in (D).
Among the 14 patients without orthostatic hypotension who had both the left ventricular free wall and interventricular septum visualized, the mean ratio of septal:free wall radioactivity exceeded 1 (mean: 1.25 ± 0.1, t = 2.6, p = 0.02).
Desipramine treatment in healthy volunteers was associated with decreased 6-[18F]fluorodopamine-derived radioactivity in the nasopharyngeal mucosa (992 ± 48 vs 1,824 ± 340 nCi-kg/mL-mCi) and thyroid gland (1,425 ± 136 vs 3,279 ± 384 nCi-kg/mL-mCi, p = 0.01; figure 3). Proportionately smaller decreases in 6-[18F]fluorodopamine-derived radioactivity in the parotid and submandibular salivary glands were not significant.
Figure 3. PET scans of the head and neck after IV administration of 6-[18F]fluorodopamine in a healthy control subject (top) at baseline without and (bottom) with pretreatment by oral desipramine to block neuronal uptake of catecholamines. Note decreased 6-[18F]fluorodopamine-derived radioactivity in the nasopharyngeal mucosa and thyroid in the setting of desipramine pretreatment.
Thyroid concentrations of 6-[18F]fluorodopamine-derived radioactivity varied as a function of subject group (F = 12.6, p = 0.0002). Among patients with PD, those with orthostatic hypotension tended to have lower thyroid concentrations of 6-[18F]fluorodopamine-derived radioactivity (1,361 ± 257 nCi-kg/mL-mCi) than did patients without orthostatic hypotension (1,908 ± 124 nCi-kg/mL-mCi, p = 0.08; figure 4). Both patient groups had lower thyroid concentrations of 6-[18F]fluorodopamine-derived radioactivity than did healthy volunteers (3,279 ± 367 nCi-kg/mL-mCi, F = 22.9, p = 0.0001).
Figure 4. Tissue concentrations (mean ± SEM) of 6-[18F]fluorodopamine-derived radioactivity in various organs of patients with PD and healthy control subjects. Blue = normal; yellow = PD without orthostatic hypotension, local loss of myocardial 6-[18F]fluorodopamine-derived radioactivity; green = PD without orthostatic hypotension, diffuse loss of myocardial 6-[18F]fluorodopamine-derived radioactivity; red = PD with orthostatic hypotension. **Significantly below control, p < 0.01.
Tissue concentrations of 6-[18F]fluorodopamine-derived radioactivity in the renal cortex increased rapidly during and briefly after infusion of 6-[18F]fluorodopamine (F = 142, p < 0.0001; figure 5). Renal cortical 6-[18F]fluoro-dopamine-derived radioactivity was decreased in patients with PD, compared with that in healthy volunteers (F = 4.5). The patient and control groups also differed in trends of 6-[18F]fluorodopamine-derived radioactivity over time (F = 3.1, p = 0.006). Subgroups of patients with PD with and without orthostatic hypotension did not differ in renal cortical radioactivity.
Figure 5. Renal cortical and renal pelvic mean (± SEM) concentrations of 6-[18F]fluorodopamine-derived radioactivity in healthy control subjects, patients with PD without orthostatic hypotension, and patients with PD and orthostatic hypotension. Open circles = healthy volunteers; filled circles = PD without orthostatic hypotension; filled squares = PD with orthostatic hypotension.
There were no significant differences between patients with PD and healthy control subjects in 6-[18F]fluoro-dopamine-derived radioactivity in the liver, spleen, renal pelvis, salivary glands, or nasopharyngeal mucosa (see figure 4).
Plasma catechols.
Patients with PD and orthostatic hypotension had lower plasma norepinephrine levels during supine rest than did patients without orthostatic hypotension (2.32 ± 0.26 vs 1.40 ± 0.15 nmol/L, t = 3.0, p = 0.005, data for one outlier excluded).
Plasma norepinephrine correlated weakly positively with plasma levodopa levels (r = 0.31, p = 0.08), excluding data from the outlier with extremely high plasma levodopa levels (0.79 including data from the outlier). Below a levodopa concentration of approximately 10 nmol/L, there was no relationship between plasma norepinephrine and plasma levodopa, whereas above a levodopa concentration of approximately 2,000 nmol/L, norepinephrine levels were significantly positively correlated with levodopa levels (r = 0.64; figure 6). The difference in mean norepinephrine concentrations between the group of patients with and the group without orthostatic hypotension remained significant after excluding data from patients with high plasma levodopa levels from levodopa/carbidopa treatment (t = 3.1, p = 0.007).
Figure 6. Plasma concentrations of norepinephrine, l-dopa, and dihydroxyphenylglycol in patients with PD. Filled circles = patients with orthostatic hypotension; open circles = patients without orthostatic hypotension; vertical dashed line separates patients off or never treated with levodopa from patients treated with levodopa at the time of testing; solid line shows line of best fit for the relationship between plasma dihydroxyphenylglycol and plasma norepinephrine in patients with orthostatic hypotension, to highlight the y-intercept near the origin.
Eleven of 18 tested patients without orthostatic hypotension had an increase in plasma norepinephrine levels of ≥60% during orthostasis, whereas none of 11 tested patients with orthostatic hypotension had this finding (χ2 = 10.8, p = 0.001).
Mean plasma dihydroxyphenylglycol levels did not differ between the group of patients with and the group without orthostatic hypotension (see figure 6). Among patients with orthostatic hypotension (excluding data from three patients with extremely high levodopa concentrations), plasma dihydroxyphenylglycol levels were positively correlated with plasma norepinephrine levels (r = 0.62, p = 0.01), with the y-intercept value at about the origin (see figure 6).
Discussion.
More than a dozen recent studies have agreed on the remarkable finding that most patients with PD have cardiac sympathetic denervation. The results of the current study support the view that cardiac sympathetic denervation in PD reflects a pathologic process that, when generalized, produces orthostatic hypotension from sympathetic neurocirculatory failure.
A particular pattern of beat-to-beat blood pressure responses to the Valsalva maneuver permits detection of sympathetic neurocirculatory failure.31 Because of deficient reflexive, sympathetically mediated vasoconstriction in response to decreased cardiac filling, during phase II of the maneuver the blood pressure decreases progressively, and during phase IV the pressure fails to exceed the baseline value. Because all 17 patients with PD who had orthostatic hypotension and were able to perform a technically adequate Valsalva maneuver showed this abnormal response, regardless of levodopa/carbidopa treatment, whereas only 6 of the 23 patients without orthostatic hypotension had the abnormal response, sympathetic neurocirculatory failure attends the orthostatic hypotension in PD.
The cardiac PET scanning findings help to understand the mechanism of sympathetic neurocirculatory failure in PD. Because all 18 patients with PD who had orthostatic hypotension had markedly decreased concentrations of 6-[18F]fluorodopamine-derived radioactivity throughout the left ventricular myocardium; because low myocardial concentrations of 6-[18F]fluorodopamine-derived radioactivity invariably are associated with decreased synthesis, release, reuptake, and turnover of norepinephrine in the heart in PD;16 and because only about half of patients with PD who did not have orthostatic hypotension had decreased 6-[18F]fluorodopamine-derived radioactivity throughout the left ventricular myocardium, orthostatic hypotension in PD is associated with neuroimaging evidence of cardiac sympathetic denervation.
Concentrations of norepinephrine in antecubital venous plasma provide a means, albeit indirect, of detecting sympathetic denervation in the body as a whole.35-37⇓⇓ Normally plasma norepinephrine levels approximately double within 5 minutes of standing from the supine position.38 Because in the supine position patients with PD and orthostatic hypotension had significantly lower plasma norepinephrine concentrations than did patients without orthostatic hypotension, and because all tested patients with PD and orthostatic hypotension had subnormal or absent increments in plasma norepinephrine levels during standing, whereas most patients without orthostatic hypotension had normal increments, the neurochemical findings indicate that orthostatic hypotension in PD is associated with decreased sympathetic innervation in the body as a whole. Moreover, because the sympathetic innervation of the heart contributes little to norepinephrine concentrations in antecubital venous blood,39 the loss of sympathetic innervation in PD with orthostatic hypotension appears to occur in organs besides the heart.
The current results for plasma norepinephrine levels agree with several previous reports describing normal supine plasma norepinephrine in PD without orthostatic hypotension2,28-30,40-43⇓⇓⇓⇓⇓⇓⇓ and decreased plasma norepinephrine in PD with orthostatic hypotension.2,28-30⇓⇓⇓
The extent of loss of sympathetic innervation in PD seems to vary across organs, because the patients had approximately normal tissue concentrations of 6-[18F]fluorodopamine-derived radioactivity in the liver, spleen, salivary glands, and nasopharyngeal mucosa but had decreased concentrations in the thyroid gland and renal cortex.
Thyroid tissue concentrations of 6-[18F]fluoro-dopamine-derived radioactivity at least partly reflect sympathetic innervation, because healthy volunteers treated with desipramine had decreased 6-[18F]fluorodopamine-derived radioactivity in the thyroid. Thyroid radioactivity in patients with PD was decreased to about half of normal—to about the same extent as in desipramine-treated healthy volunteers. Too few patients with or without orthostatic hypotension were tested to determine whether orthostatic hypotension is related to the extent of thyroid or renal cortical sympathetic denervation in PD.
The current evidence for loss of sympathetic innervation in other organs in addition to the heart is consistent with previous studies describing the presence of Lewy bodies in sympathetic ganglia of patients with PD7,44,45⇓⇓ and provides support for the view that sympathetic neurocirculatory failure in PD reflects a form of ganglionic rather than, or in addition to, central neuropathology.
Where would the denervation be that would be missed by 6-[18F]fluorodopamine PET scanning of extra-cardiac organs and cause orthostatic hypotension? The mesenteric organs constitute the site of most of norepinephrine production in the human body.46,47⇓ Because of lack of registration of PET scanning results with those of CT scanning in the current study, we could not assess sympathetic innervation in most mesenteric organs. Concentrations of norepinephrine in arterial and therefore antecubital venous plasma are insensitive to norepinephrine release from the mesenteric organs, because the liver effectively extracts and metabolizes norepinephrine delivered to it via the portal vein.47 Therefore, gastrointestinal sympathetic denervation, if it occurred, would not explain low antecubital venous plasma levels of norepinephrine in patients with PD and orthostatic hypotension.
Several reports have noted decreased skin sympathetic responses in patients with PD.48-54⇓⇓⇓⇓⇓⇓ These responses depend on alterations in sympathetically mediated sweat production, and acetylcholine, rather than norepinephrine, is the main sympathetic neurotransmitter in eccrine sweat glands. The applicability of these reports to relatively low plasma norepinephrine levels in PD and orthostatic hypotension therefore is unclear. Patients with PD and autonomic dysfunction have decreased reflexive vasoconstriction in skin55 and skeletal muscle56; however, other mechanisms might explain these decreases besides attenuated exocytotic release of norepinephrine. Measurement of pulse-synchronous bursts of peroneal nerve traffic by microneurography57,58⇓ or of extracellular fluid levels of norepinephrine by microdialysis58,59⇓ can provide indices of sympathetic outflow to skeletal muscle, but neither approach has been used so far in PD.
The current results reject the long-held notion that orthostatic hypotension in PD results from treatment with levodopa.10 If levodopa therapy caused orthostatic hypotension in PD, then patients with orthostatic hypotension would be expected to be on levodopa, and those without orthostatic hypotension would not. At a minimum, it would be expected that a higher proportion of patients with orthostatic hypotension would be on levodopa therapy than would patients without orthostatic hypotension. The results of the current study disconfirmed these expectations. Patients with orthostatic hypotension did not differ from those without orthostatic hypotension in terms of plasma levodopa concentrations. Perhaps most convincingly, among the 18 patients with orthostatic hypotension, four had never taken levodopa, and three had discontinued levodopa treatment and had normal plasma levodopa levels at the time of testing, yet all 18 had orthostatic hypotension and physiologic and neurochemical evidence of sympathetic neurocirculatory failure and sympathetic denervation.
Does an association between sympathetic denervation and orthostatic hypotension in PD justify inferring that the denervation causes the orthostatic hypotension? Because there is no reason to think that orthostatic hypotension produces sympathetic denervation in PD, the only alternative to inferring a causal role of sympathetic denervation would be a common cause for both orthostatic hypotension and sympathetic denervation in PD. Although it is theoretically possible that a common cause might produce both diffuse sympathetic denervation and orthostatic hypotension in PD, we cannot think of one. (For the reasons discussed here, levodopa treatment cannot be such a common cause.) Meanwhile, because diffuse sympathetic denervation always produces orthostatic hypotension, the finding of such denervation is sufficient to explain orthostatic hypotension in PD. The current results therefore support the inference that diffuse sympathetic denervation not only is associated with but probably causes the orthostatic hypotension. Our findings do not exclude the possibility that vasodilation, elicited by dopamine produced from levodopa, might decrease the blood pressure of some patients with diffuse sympathetic denervation and, because of loss of reflexive sympathetically mediated vasoconstriction, exaggerate orthostatic hypotension.
Because the cardiac sympathetic nerves travel with the coronary arteries, the observation that patients without orthostatic hypotension had more severely decreased myocardial 6-[18F]fluorodopamine-derived radioactivity in the left ventricular free wall than in the interventricular septum suggests a “dying back” process, as opposed to an alternative mechanism whereby the cell bodies would die and then death of the terminals would ensue (as in Wallerian degeneration).
Because about half of the patients without orthostatic hypotension had decreased 6-[18F]fluoro-dopamine-derived radioactivity throughout the left ventricular myocardium, and of the remaining half most had localized decreases, most of the patients with PD in this study had evidence for at least some loss of cardiac sympathetic innervation. This finding agrees with those from centers in several countries based on cardiac sympathetic neuroimaging using 123I-metaiodobenzylguanidine.
In two respects the current results based on 6-[18F]fluorodopamine PET scanning differ from previously reported results based on 123I-metaiodobenzylguanidine scanning. First, the same high prevalence of decreased myocardial 123I-metaiodobenzylguanidine-derived radioactivity has been reported in PD with or without orthostatic hypotension,19 whereas the current results showed more prevalent decreases in myocardial 6-[18F]fluorodopamine-derived radioactivity in patients with orthostatic hypotension. Second, decreased 123I-metaiodobenzylguanidine-derived radioactivity has been reported in the heart of patients with PD but not in other organs,27 whereas the current results showed decreased 6-[18F]fluorodopamine-derived radioactivity also in the thyroid and renal cortex.
Treatment with levodopa produces circulating levodopa concentrations in antecubital venous plasma that typically exceed 5,000 nmol/L, compared with endogenous concentrations <10 nmol/L. Because endogenous levodopa is converted to norepinephrine in sympathetic nerves, it might be expected that such large increases in plasma levodopa levels would increase plasma norepinephrine levels. In the current study, there was no relationship between plasma norepinephrine and plasma levodopa over a >100-fold range of levodopa concentrations above endogenous concentrations. Increased production and turnover of norepinephrine in storage vesicles, without concomitantly increased exocytotic release of norepinephrine, can explain this finding.60,61⇓
Plasma dihydroxyphenylglycol has multiple sources. To relatively small extents, plasma dihydroxyphenylglycol reflects neuronal uptake of circulating norepinephrine and reuptake of released norepinephrine.62,63⇓ In resting, untreated people, however, most of plasma dihydroxyphenylglycol derives from net leakage of norepinephrine from vesicles into the axoplasm in sympathetic nerves.61 This is an ongoing process independent of sympathetic nerve traffic. Because of this continuous formation of dihydroxyphenylglycol in sympathetic nerves, the y-intercept value for the relationship between plasma dihydroxyphenylglycol and plasma norepinephrine levels normally is significantly above the origin.61 The current finding of a y-intercept value near the origin in patients with PD and orthostatic hypotension therefore suggests a generalized decrease in vesicular norepinephrine stores, which would be consistent with generalized sympathetic denervation.
In patients with PD and orthostatic hypotension, plasma norepinephrine levels, although significantly lower than in patients without orthostatic hypotension, were not particularly low for healthy people of similar age64,65⇓ and were higher than in patients with pure autonomic failure.37 It is possible that partial loss of sympathetic terminals leads to augmented traffic to remaining terminals, resulting in increased proportionate release of norepinephrine from the reduced vesicular stores. Moreover, because denervation would produce concurrent decreases in both release and reuptake of norepinephrine, plasma norepinephrine levels might fail to detect a real decrease in norepinephrine release.66
Patients with PD had normal 6-[18F]fluoro-dopamine-derived radioactivity in the parotid and submandibular salivary glands. Some of the radioactivity may have reflected uptake or secretion of metabolites of 6-[18F]fluorodopamine, which would compromise detection of sympathetic denervation by 6-[18F]fluorodopamine scanning. Analogously, in rats, desipramine pretreatment does not significantly reduce salivary gland concentrations of tritium after IV injection of 3H-6-fluorodopamine, although desipramine does reduce concentrations of 3H-6-fluoronorepinephrine.67 The current results therefore do not allow unequivocal inferences about salivary denervation in PD.
Acknowledgments
Acknowledgment
The authors Sandra Brentzel, RN, Patricia Woltz, RN, and the NIH PET Department.
- Received September 10, 2001.
- Accepted December 22, 2001.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
