Gastric emptying in Parkinson's disease

Gastrointestinal dysfunction in Parkinson's disease (PD) is gaining increasing interest, particularly concerning its involvement in oral levodopa absorption in patients with response fluctuations. Several studies showed a high frequency of gastric atony and reduced gastric motility in PD patients, ^[1-5] but most were based on indirect evidence. Symptoms such as early postprandial satiety, abdominal fullness, nausea, and vomiting are common in PD. ^[6] Patients who have undergone gastrectomy absorb levodopa more rapidly and efficiently. ^[7,8] Bypassing the stomach and administering levodopa as a suspension via nasoduodenal or gastrojejunostomy tubes produce rapid elevations of plasma levodopa levels. ^[9-12] In some parkinsonian patients, problems such as ``delayed-on'' (prolonged latencies from levodopa dose ingestion to ``start up'') and ``no-on'' (total dose failure) may be caused mainly by stagnation of levodopa in the atonic stomach. <sup>[13]</sup> These phenomena may first emerge after vagotomy, which attenuates gastric motility. <sup>[14]</sup> Prokinetic drugs that enhance gastric emptying (e.g., cisapride and domperidone) improve some subtypes of response fluctuations, especially ``delayed-on'' and ``no-on'' phenomena, and cause rapid elevations of plasma levodopa levels. ^[15] It is still unknown whether all PD patients or only those with advanced disease have impaired gastric emptying with or without response fluctuations. We therefore investigated gastric motility in patients with PD with a radionuclide gastric emptying technique and correlated the findings with disease severity and response fluctuations.

Methods.

Thirty patients (19 men and 11 women) with idiopathic PD were included in the study. Mean age of the patients was 63 plus minus 9 years (range 44 to 80), and disease duration was 6.4 plus minus 4.6 years (range 0 to 18). None of the patients received anticholinergic drugs or other medications that might delay gut motility. Also, no patient suffered from gastrointestinal disease or had a history of gastric surgery.

The patients were divided into two groups: 15 patients with fluctuating symptoms and 15 patients without response fluctuations. All patients with fluctuating symptoms had ``wearing-off'' and ``delayed-on'' phenomena. ``Delayed on'' was manifested by latencies longer than 30 minutes from ingestion of levodopa to turning ``on'' for almost all daily doses. Twelve patients also had the ``no-on'' phenomenon (i.e., episodic total unresponsiveness to single doses of levodopa). Mean age of patients with fluctuating symptoms was 62 plus minus 9 years, mean disease duration was 8.8 plus minus 4.6 years, and Hoehn and Yahr stages 2 to 4 were seen. ^[16] Mean age of patients without fluctuating symptoms was 64.8 plus minus 9.9 years, mean disease duration was 4.0 plus minus 3.3 years, and Hoehn and Yahr stages 1 to 4 were seen Table 1.

Results.

All 30 PD patients had prolonged gastric emptying manifested by higher stomach retention after 1 hour: 70.7 plus minus 16.1% (mean plus minus SD) compared with less than 60% obtained for normal subjects. They also had longer lag time and emptying time after the lag: 728 plus minus 562 seconds compared with less than 600 seconds for control subjects, and 158 plus minus 153 minutes compared with less than 60 minutes for normal subjects.

Gastric retention measured after 60 minutes was higher in patients with fluctuations than in patients without fluctuations (77.4 plus minus 15.5% versus 64.0 plus minus 14.3%; p less than 0.05). t₁/2 of emptying after the lag phase was significantly delayed in patients with fluctuations (221 plus minus 202 minutes) compared with patients without fluctuations (85 plus minus 31 minutes; p less than 0.05). There was no significant difference in lag time between patients with and without response fluctuations.

Among the patients with response fluctuations, only 2 of 15 had normal values of retention after 60 minutes and 5 of 15 had a severe impairment of gastric emptying (more than 90%). In the group without response fluctuations, 8 of 15 patients had normal retention values and 7 of 15 patients had a moderate impairment of gastric emptying (60 to 90%).

There was no correlation between gastric emptying and disease duration. For instance, four patients with recently diagnosed disease had impairment of gastric emptying, whereas six patients with long-standing disease had normal retention.

Discussion.

Our results demonstrate that impairment of gastric motility is frequent in PD and occurs in about 70% of patients, especially in those with response fluctuations. However, this percentage might be somewhat overestimated because the mean age of the healthy volunteers was lower than the mean age of the patients, and gastric emptying might be impaired even in healthy control subjects over 65 years old. These findings are in accordance with other studies that found that delayed gastric emptying is a pathogenic factor in nonresponding PD patients. ^[1,3,13] We measured gastric emptying by a radionuclide method. In radionuclide gastric emptying studies, there may be a problem of changing tissue attenuation as the meal moves anteriorily from fundus to antrum. This can be corrected using several procedures, but all have many limitations. ^[17] Data obtained from left anterior oblique acquisition may be a proper substitute because it approximates results of other procedures in normal subjects and in symptomatic patients with normal gastric anatomy. ^[18,19] Patients with the ``delayed-on'' and ``no-on'' subtypes of response fluctuations had significantly longer retention times of food in the stomach than patients without response fluctuations. These findings serve as strong support for the argument that the ``delayed-on'' and ``no-on'' phenomena are caused by peripheral pharmacokinetic mechanisms, particularly delayed gastric emptying, which leads to retarded or absent levodopa absorption. ^[4,20]

Several mechanisms may be responsible for the observed changes in gastric motility. Emptying of the stomach is partly controlled by local gastric factors and also depends on feedback signals from the duodenum. Nervous reflexes are mediated mainly by way of afferent nerve fibers in the vagus nerve to the brainstem and then back through efferent neuronal projections to the stomach, also via the vagus. Some of the signals are probably transmitted directly through the myenteric plexus as well. ^[21] Impaired gastric emptying might be caused by the primary disease process itself. In this study, five patients with recent onset of illness (four not yet treated with levodopa before the examination) also had an impairment of gastric emptying. PD is a neurodegenerative disease affecting mainly the pars compacta of the substantia nigra and other pigmented brainstem nuclei, but peripheral organs possibly are also affected. Esophageal Lewy bodies may be present in PD patients with dysphagia. ^[22] Wakabayashi et al. ^[23] studied the enteric nervous system of the alimentary tract in seven patients with PD and found Lewy bodies in Auerbach's and Meissner's plexuses of the lower esophagus and stomach in all patients. Cytoplasmic inclusions similar to Lewy bodies were present in ganglion cells of colonic myenteric plexuses in a patient with PD. ^[24] These observations might indicate that the primary disease process itself plays a role in gastric atony. Neuronal degeneration occurring in the dorsal nucleus of the vagus, in addition to the nigrostriatal tract, may be responsible for the degeneration of the gastrointestinal myenteric plexuses.

An alternative cause of reduced stomach motility in PD may be linked to the effect of levodopa itself on dopaminergic receptors present in the gastrointestinal tract and particularly in the stomach. ^[25] Levodopa may be decarboxylated to dopamine by aromatic amino acid decarboxylase in the mucosa of the stomach. Dopamine itself may cause decreased gastrointestinal motility. ^[26-28] Levodopa administration inhibits gastric emptying in normal volunteers. ^[29] In vivo studies also demonstrated delayed gastric emptying after dopamine infusions. Symptoms of epigastric discomfort are aggravated by treatment with levodopa that relaxes the lower esophageal sphincter and inhibits its stimulation by metoclopramide. ^[27] Therefore, chronic use of levodopa itself may decrease gastric motility, prolong the duration of the drug in the stomach, delay its own absorption from the gut, and thus postpone the production of a critical plasma peak level necessary for entry of levodopa from the circulation into the brain and consequently the start-up of a single-dose response.

There was no correlation between disease severity or duration and the presence of gastric impairment in our patients. Six patients with long-standing disease had normal gastric emptying. They may represent a subset of PD patients with no degenerative involvement of the neuronal plexuses within the gastrointestinal tract. Another explanation is that gastric motility is erratic in these patients (i.e., there could be times of effective contractions, strong enough to induce rapid transit of food bulk into the duodenum, with resultant normal gastric motility values). The dopaminergic receptors in their stomachs may also have been downregulated after long-term treatment with large doses of levodopa, with reduced response to endogenous dopamine that would normally operate to reduce motility. Meunter and Tyce ^[30] concluded that the absorption rate was more rapid during long-term levodopa therapy than at initiation of treatment, implying that tolerance to levodopa, or an adaptive alteration in the intestinal handling of levodopa, may occur as a result of prolonged exposure to the drug. We will follow our patients longitudinally and attempt to determine whether there is a correlation between other gastrointestinal symptoms such as constipation and dysphagia with gastric emptying impairment.

We conclude that gastric motility is impaired in PD, especially in fluctuating patients with ``delayed-on'' and ``no-on'' phenomena, and it may have a role in causing these complications. Such problems may be improved by avoiding drug retention in the stomach by administering levodopa directly into the duodenum via nasogastric tubes or gastrostomies, ^[9-12] administering levodopa methylester intravenously, or administering soluble dopamine agonists subcutaneously. ^[31-33] Partial amelioration may be achieved by combining oral levodopa with prokinetic drugs that accelerate gastric motility. ^[15] Administration of crushed levodopa taken with liquids occasionally overcomes the problem and induces relief ^[34]; this suggests that soluble levodopa prodrugs taken orally may be even more effective than crushed levodopa solutions.