Effect of a single dose of standard levodopa on cardiac autonomic function in Parkinson's disease
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.86536
Source of Support: None, Conflict of Interest: None
Background: Parkinson's disease (PD) is associated with autonomic dysfunction and chronic levodopa therapy has been reported to impair the autonomic control of heart rate. Aim: Our aim was to assess the immediate effect of a single dose of levodopa on heart rate variability (HRV) in idiopathic PD. Materials and Methods: Eleven patients of idiopathic PD (F:M =2:9, mean age 57.3±8.6 years, duration of illness 4.1±2.8 years, Hoehn and Yahr stage 2.1±0.2) on stable levodopa dosage were studied. Motor part of unified Parkinson's disease rating scale and resting Lead II electrocardiogram (ECG) recordings were performed at baseline (12 hours off medication) and after two tablets of 100/10 mg of standard levodopa/ carbidopa. ECG was recorded continuously in the first hour (H1) followed by a 15-min recording in second (H2), third (H3) and fourth (H4) hours. Artifact free 5-min segments of the ECG were analyzed offline to obtain the HRV parameters in time domain (ms) and frequency domains (ms 2 ). Results: Significant increase was observed in standard deviation of normal to normal intervals (23.5±2.7-46.2±6.6, P<0.05), root mean square of successive differences of NN intervals (16.3±2.9-30.7±5.1, P<0.01), total power (568.9±125.7-2739±667.5, P<0.01), low frequency power (146.5±40.8-614.1±206.7, P<0.05) and high frequency power (107.4±33.9-332.7±85.9, P<0.05) in H1. Conclusion: The results are suggestive of an improvement in the overall variability of the heart rate indicating an enhanced vagal tone.
Keywords: Cardiac autonomic dysfunction, levodopa, Parkinson′s disease
Autonomic dysfunction is a common feature of Parkinson's disease (PD). It can be diverse, the most commonly reported being cardiovascular, gastrointestinal and urogenital dysfunctions.  The cardiovascular dysfunction manifests as suppressed heart rate (HR) and blood pressure (BP) responses to various autonomic provocations  and also as disturbances in the circadian rhythms of HR and BP. , The effect of the PD medications on the autonomic nervous system has been a subject of interest. Although there have been several studies, a clear demarcation of obvious modulation of autonomic responses is not yet evident. ,, The associations of these fluctuations with the "ON" and "OFF" periods of motor fluctuations have also been studied. Patients with the "ON-OFF" phenomenon have a higher resting HR, greater orthostatic BP fall and lower responses to Valsalva maneuver and cold pressor stimuli. , These changes in autonomic functions have been attributed to cardiac sympathetic denervation  and also to concurrent parasympathetic dysfunction.  The aim of the present study was to determine the immediate effect of a single dose of levodopa on the heart rate variability (HRV) in PD patients.
Eleven patients of idiopathic PD fulfilling the Parkinson's Disease Society Brain Bank clinical criteria  (2 women and 9 men, mean age=57.3±8.6 years, duration of illness=3.8±2.7 years, mean HandY stage=2.1±0.2), were recruited from the Department of Neurology, National Institute of Mental Health and Neurosciences (NIMHANS). All patients were on a stable dose of levodopa at the time of the study. Two patients had history suggestive of wearing off symptoms. None of the patients had any history of diabetes, hypertension, cardiac disease or other medical disorders and were not being treated with any drugs known to affect the autonomic nervous system. To confirm autonomic dysfunction, the baseline data of the patients was compared with a control group (CTRL) consisting of 11 age and gender matched healthy subjects (2 women and 9 men, mean age 54.7±1.8 years). The study was approved by the Institute Ethics Committee and all subjects gave their written informed consents.
The tests were carried out between 8 AM and 12 noon in the Autonomic Laboratory, Department of Neurophysiology, NIMHANS under standardized conditions.  Motor part of Unified Parkinson's Disease Rating Scale (UPDRS) was recorded in patients at baseline (12 hours off medication). Lead II Electrocardiogram (ECG) was recorded while a normal respiratory rate of 12-15 breaths/min was ensured by recording the respiratory movements. The recorded ECG and breathing signals were conveyed through an analog digital converter (Power lab, 16 channels data acquisition system, AD Instruments, Australia) with a sampling rate of 1024 Hz. MLS 310 Module was used to analyze different HRV measures. Subsequently, the patients were administered 2 tablets of 100/10mg of standard levodopa/carbidopa and ECG was recorded continuously in the first hour (H1) followed by 15 min recordings in the second (H2), third (H3) and fourth (H4) hours. As the HRV fluctuations were hypothesized to occur during the first hour, the recording was performed for the entire hour. However, for analysis, 5 min segments were selected from every quarter i.e., Q1 (0-15 min), Q2 (15-30 min), Q3 (30-45 min) and Q4 (45-60 min). Recorded data was stored in a personal computer and later analyzed offline using an automatic program that allowed visual checking of the raw ECG and breathing signals. UPDRS was also repeated at every hour.
HRV was analyzed as per the guidelines of Taskforce report.  Five minute artifact free segments of the recordings were selected in all time points and analyzed to obtain time domain parameters and frequency domain parameters. The time domain parameters included (i) standard deviation of normal-to-normal (NN) intervals, in millisecond (SDNN, in ms), (ii) square root of the mean of the sum of squares of differences between adjacent NN intervals (RMSSD, in ms), (iii) count of number of pairs of NN intervals differing by >50 ms (NN50), and (iv) percentage of NN50 i.e., NN50 divided by total number of all NN intervals (pNN50). The frequency domain parameters included (i) total power (TP, in ms 2 ), (ii) low frequency power (LF, in ms 2 ), (iii) high frequency power, (HF, in ms 2 ), (iv) LF power expressed in normalized units (LFnu), (v) HF power expressed as normalized units (HFnu) and (vi) LF/HF ratio which represents sympathovagal balance.
Repeated measures of analysis of variance (RMANOVA) were used to analyze the change in the UPDRS with Least significant difference (LSD) post-hoc tests. HRV parameters between patients and controls were analyzed with Mann-Whitney 'U' test and changes over the time points in the patients were analyzed with Friedman's ANOVA and Dunn's multiple comparison post-hoc tests. Analysis of the HRV parameters was done in two stages. First, the parameters at baseline and in Q1, Q2, Q3 and Q4 were compared. As best improvement was seen in Q3, thereafter analysis was performed between the Q3 (represents H1) and H2, H3 and H4. Differences were considered significant at P<0.05. Data are presented as Mean±S.E.M. Spearman R correlation test was used to determine correlation coefficients.
Comparison of UPDRS in H1, H2, H3 and H4 in PD patients
The UPDRS of the patients steadily improved after drug administration. At baseline the mean UPDRS was 37.6±2.0. It reduced significantly to 22.5±2.1 in H1 (P<0.001), 22.8±2.2 in H2 (P<0.001), 26.3±2.4 in H3 (P<0.001) and 31.5±1.9 in H4 (P<0.01).
Comparison of HRV between controls and PD patients
At baseline, the PD patients and controls had comparable mean RR (normal to normal - NN) interval (794.8±32.8 ms, 874.7±25.5ms) and mean heart rate (HR) (76.7±3.1 beats/min, 69.2±2.1 beats/min). However, PD patients had a significantly reduced SDNN (23.5±2.7, 36.3±3.5, P<0.01) and TP (568.9±125.7, 1396.1±284.1, P<0.05) when compared to controls. The rest of the TD and FD parameters were comparable at baseline in both groups [Table 1].
Comparison of HRV in PD patients at baseline and during first hour after drug administration
[Table 2] shows the mean time domain and frequency domain parameters of every quarter of the first hour. The most significant changes were seen in the Q3 i.e., 30-45 min after drug administration.
Comparison of baseline and H1 HRV with H2, H3 and H4
The improvement in HRV seen in the first hour did not sustain from the second hour onwards. [Figure 1] shows the changes in mean UPDRS scores and time domain and frequency domain parameters from baseline to subsequent time points.
The UPDRS motor scores at the different time points did not correlate with any of the time domain and frequency domain parameters of the HRV.
In the present study, the baseline SDNN and TP were significantly reduced in PD patients as compared to controls confirming cardiac autonomic dysfunction in these patients. A significant improvement was observed after taking levodopa in the time domain parameters viz., SDNN and RMSSD and frequency domain parameters viz. TP, LF and HF. The best results were obtained in the first hour after administration of levodopa and it tapered toward baseline values by the fourth hour. The results support a favorable action of levodopa on the cardiac autonomic function. Our study also confirms the findings of an earlier study by Ludwig et al. who reported increased HRV in the ON phase of medication. The improvement in UPDRS motor scores did not correlate with the improvement in the cardiac autonomic parameters suggesting that the mechanisms responsible for these phenomena are probably independent of each other. Although Pursainen et al.  suggested that motor performance and cardiovascular regulation may be inter-related as a consequence of the disease, our findings support the view of Shibata et al.  that dopaminergic neurodegeneration responsible for motor symptoms and autonomic dysfunction are independent.
Normally, around 60% of a standard dose of levodopa-carbidopa is absorbed in 30 minutes and absorption is completed by 2-3 hrs,  and the serum peak level of levodopa has been reported to occur between 30 minutes and 2 hours.  Although serum levels were not measured in the current study, we observed maximum improvement in the HRV parameters in Q3 (30-45 min after drug administration), suggesting peak drug action around 30 th minute.
Mild impairment of autonomic cardiovascular control is known to occur early in the course of PD.  Both central and peripheral mechanisms are hypothesized to be responsible for this. A considerable overlap of degenerative changes in the basal ganglia and also autonomic post-ganglionic neurons in the periphery results in autonomic dysfunction in PD. , Studies have indicated cardiac sympathetic denervation as a result of loss of catecholamine innervations in the nigrostriatal system in the brain and sympathetic nervous system in the heart resulting in autonomic failure in PD.  Bouhaddi et al.  have earlier reported that the side-effects of levodopa aggravated the impairment of autonomic control of blood pressure and heart rate. Levodopa is hypothesized to act on the central as well as peripheral nervous system by stimulating the D1 receptors in the former and D2 receptors in the latter leading to an overall sympatholytic effect.  This sympatholysis may have contributed to tilting the autonomic balance in favor of a vagal tone and hence an enhanced variability. 
In addition to this, parasympathetic dysfunction has also been postulated in PD  and an increase in the parasympathetic dysfunction in addition to the sympathetic denervation leads to impaired cardiovascular responses.  Cardiac ganglia receive major parasympathetic projections from the preganglionic vagal fibers in the ventral nucleus ambiguus (vNA) and some minor projections from the dorsal motor nuclei (DMN) of the vagus. Deposition of Lewy bodies and Lewy neurites in the brainstem nuclei is known to affect the DMN resulting in aberrant cardiovagal activity.  Reports on the effect of levodopa on the parasympathetic system have been few and conflicting. Antonaccio and Robsin  reported that levodopa had no effect on the vagal activity. However, another study demonstrates the possible beneficial effect of levodopa on parasympathetic dysfunction in de-novo PD patients as observed by an improvement in the deep breathing test. 
In the present study, SDNN and TP which are indicators of overall variability were increased after levodopa administration. Concomitant increases were also evident in the LF and HF components of TP. Increase in the LF power is primarily due to heightened sympathetic activity; however, a parasympathetic contribution cannot be ruled out. Enhancement of HF power and RMSSD purely reflects enhanced parasympathetic activity.  Previous studies have indicated that the striatal dopaminergic system is accountable for modulating HR and BP responses. ,, Recently, Yeh and colleagues  demonstrated a negative correlation of striatal dopamine D2/D3 receptor binding with HR, and a positive correlation with cardiac vagal index and LF power in supine resting healthy subjects. Hence, we postulate that increased striatal dopaminergic activity following administration of levodopa may be resulting in enhancement of the cardiac vagal tone as observed in our patients.
To conclude, the results of our study suggest a beneficial effect of levodopa on the cardiac autonomic regulation in PD patients. Our study was limited by a small sample and further studies on larger cohort of PD patients are required to validate these findings.
[Table 1], [Table 2]