Parkinsonian and Cerebellar Phenotypes of Probable MSA: An Insight in to Prognostic Factors Based on Autonomic Functions.

BACKGROUND: Multiple system atrophy is an adult-onset, sporadic, neurodegenerative disorder with parkinsonian (MSA-P) and cerebellar (MSA-C) subtypes.
OBJECTIVE: We aimed to elucidate differences in prognostic factors between MSA subtypes.
METHODS: The study population comprised 45 probable MSA patients (MSA-P = 22; MSA-C = 23) and 45 healthy controls. Clinical parameters, heart rate variability (HRV), and conventional cardiac autonomic function testing (AFT) were the study tools.
RESULTS: Mean age of onset of MSA was 54.7 ± 9 years for MSA-P and 51.9 ± 7 years for MSA-C subgroups. Median disease duration was 2 years in both subgroups. A greater percentage of MSA-P patients (45.5%) had beneficial response to levodopa (P < 0.01). Patients in both subgroups reported significant autonomic disturbances, such as postural symptoms, bladder disturbances, and erectile dysfunction. MSA-P patients had a trend for a greater number of falls and bladder disturbances than MSA-C patients (P = 0.05). Cardiac AFT showed that in MSA-P, 22.2% had definitive and 77.7% had severe autonomic dysfunction, whereas in MSA-C, 9.5% had early, 28.5% had definitive, and 57.1% had severe autonomic dysfunction. HRV analysis showed significant reduction in overall HRV, sympathetic activity, and parasympathetic activity in MSA patients as compared with controls (P < 0.0001). The sympathetic limb was more severely affected in MSA-P patients as compared with MSA-C patients (P < 0.05).
CONCLUSION: Autonomic dysfunction and postural instability, negative prognostic markers, were relatively more common in the MSA-P than in the MSA-C patients. This implies that MSA-P patients have poorer prognosis as compared with MSA-C. Dopaminergic medications can be beneficial in MSA-P patients. Copyright:

INTRODUCTION

Multiple system atrophy (MSA) is an adult onset, sporadic, neurodegenerative disorder. It is characterized clinically by a variable combination of parkinsonism, cerebellar dysfunction, autonomic dysfunction, and pyramidal tract involvement, usually with a poor response to dopaminergic medications. Pathologically, MSA has features of cell loss, gliosis, and abnormal alpha synuclein cytoplasmic inclusions in oligodendroglia of several structures of the central nervous system. MSA is divided in to two subgroups based on the predominant motor system involved: the parkinsonian variant (MSA-P) and the cerebellar variant (MSA-C).[1234]

Autonomic dysfunction is one of the cardinal features of MSA. Orthostatic hypotension, increased urinary frequency, urinary urgency, urge incontinence or incomplete bladder emptying, and erectile dysfunction (in men) are observed in individuals affected with MSA.[1234] Autonomic dysfunction may result from neurodegeneration in autonomic regulatory regions of the brain, spinal cord, or peripheral autonomic ganglia.[25] Various studies across different ethnic groups (American, European, and Japanese) have been carried out to describe the clinical characteristics and autonomic dysfunction in MSA.[678] There are limited data in this perspective from India.[9] Also, the differences in cardiac autonomic dysfunction between MSA-P and MSA-C groups have not been assessed in most of the studies. Assessing whether there are potential differences in the severity of cardiac autonomic dysfunction between subgroups may give insight to the differences in survival.

We hypothesized that there is significant cardiac autonomic dysfunction in Indians affected with MSA and there are differences in cardiac autonomic function and certain nonmotor and demographic features between probable MSA-P and MSA-C subgroups. In this study, we aimed to assess the clinical characteristics and nature of cardiac autonomic dysfunction in MSA. We also aimed to determine whether there are any significant differences in prognostic factors between MSA-P and MSA-C subgroups.

MATERIALS AND METHODS

We carried out a retrospective descriptive study (chart review) at the National Institute of Mental Health and Neurosciences (NIMHANS), Bangalore, India, after obtaining ethical clearance from Institutional Human Ethics Committee vide letter no. NIMHANS/2nd IEC (BS and NS DIV.)/2016 dated December 3, 2016. Individuals affected with probable MSA who had attended the Out-Patient Department or were admitted in the In-Patient Department, Department of Neurology, NIMHANS, during the years May 2011 to May 2016, constituted the study population. These patients were evaluated by movement disorder specialists and had undergone cardiac autonomic function testing. Inclusion criteria were a diagnosis of probable MSA-P or MSA-C that fulfils the criteria given by second consensus statement on the diagnosis of MSA,[10] and age between 30 and 70 years. Exclusion criteria were a history of stroke or other neurological disorders, brain surgery, comorbid cardiovascular disorders, diabetes mellitus, dementia, major psychiatric disorders, and intake of medications that affect cardiac autonomic function.

Review of hospital charts of patients suspected of probable MSA was carried out in the Department of Neurology, NIMHANS. Clinical data of the probable MSA patients, who fulfilled the study criteria, were obtained from the charts. Age, gender, MSA subtype, the age of onset of disease, disease duration, parkinsonian features, cerebellar features, orthostatic symptoms, bladder disturbances, nonmotor symptoms, levodopa equivalent drug dosage (LEDD), and levodopa responsiveness were the demographic and clinical parameters of interest. Total LEDD per day was calculated based on the protocol for calculating total LEDD.[11] MSA patients had undergone assessment of short-term heart rate variability (HRV) and conventional cardiac autonomic function testing in our autonomic laboratory. HRV data of age- and gender-matched healthy volunteers were obtained from the normative database of autonomic laboratory.

Short-term HRV

Subjects were advised to avoid coffee, tea, smoking, and alcohol for 8 h before the test and have light food 3 h before the test. Resting lead II electrocardiogram (ECG) was recorded for 15 min. They were allowed to rest in the supine position for about 15 min to obtain steady-state hemodynamics before the commencement of the ECG recording. The laboratory has an analog digital converter (Power lab, 16 channels data acquisition system, AD Instruments, Sydney, Australia) with a sampling rate of 1,024 Hz, through which the ECG signals are conveyed. The data acquired were stored in a personal computer and analyzed offline with the help of HRV analysis software V1.1 (Power lab, AD Instruments). An ectopic-free 5-min segment was selected from the 15 min of ECG recording and analyzed to get time and frequency domain parameters (linear methods) of the short-term HRV according to the standards established by the Task Force of the European Society of Cardiology and The North American Society of Pacing and Electrophysiology.[12] The components of time domain measurement of short-term HRV are heart rate, standard deviation of all NN intervals (SDNN), square root of the mean of the sum of squares of differences between adjacent NN intervals (RMSSD), count of number of pairs of adjacent NN intervals differing by >50 ms (NN50), and percentage of NN50 count of all NN intervals (pNN50). SDNN is sensitive to all sources of heart rate variation and represents overall HRV, whereas RMSSD, NN50, and pNN50 are most sensitive to parasympathetic activity/vagal tone. The components of frequency domain of short-term HRV are total power (sum of the constituent frequencies); high-frequency power (HF), the power of frequency band ranging from 0.15 to 0.4 Hz; low-frequency power (LF), the power of frequency band ranging from 0.04 to 0.15 Hz; and LF/HF ratio, the ratio of the power of LF component to the power of HF component. Total power is sensitive to all sources of heart rate variation reflecting overall HRV. High-frequency power is sensitive to parasympathetic/vagal tone, whereas low-frequency power is sensitive to both sympathetic and parasympathetic tone but predominantly sympathetic tone for practical purposes. LF/HF ratio signifies sympathovagal balance. An increase in LF/HF ratio signifies sympathetic dominance, whereas a reduction in LF/HF ratio signifies parasympathetic dominance.[121314]

Conventional cardiac autonomic function testing

Conventional cardiac autonomic function tests include a standard battery of tests based on the protocol of Ewing and Clarke.[15] The subjects were asked to perform certain physiological maneuvers and the changes in heart rate and blood pressure during these maneuvers were recorded. The five parameters of interest were deep breathing difference (HR based), Valsalva ratio (HR based), increase in diastolic BP during isometric handgrip maneuver (BP based), maximal: minimal ratio (HR based), and drop in systolic BP (BP based) during the orthostatic test. Deep breathing difference was calculated from the average of the differences between maximum heart rate during inspiration and minimum heart rate during expiration of six deep breathing cycles. Valsalva ratio was calculated as the ratio of longest RR interval after strain (phase IV) to the shortest RR interval during strain (phase II). An increase in diastolic BP at the end of isometric handgrip maneuver as compared with baseline was calculated. The maximum: minimum ratio was calculated as the ratio of the longest RR interval around the 30th beat after standing up to the shortest RR interval around the 15th beat. Fall in systolic BP, if any, during orthostasis as compared with baseline was noted.[1315] Based on these five tests, cardiac autonomic dysfunction was graded as follows: Normal – all five tests are normal or one is borderline; Early involvement – one of three HR/BP-based tests are abnormal or two tests are borderline; Definitive involvement – two or more HR-based tests are abnormal; and Severe involvement – two or more HR-based tests are abnormal with one or both BP-based tests abnormal.[16]

Statistical analysis

Statistical analysis of all the study variables was carried out using IBM SPSS software (IBM SPSS Statistics for Windows, Version 22.0; IBM Corp., Armonk, NY). Categorical variables were summarized using frequency and percentage. Quantitative variables were summarized using mean and standard deviation for normally distributed variables, median, and interquartile range for non-normal data. Test of association between the categorical variables was done using Pearson's Chi-square/Fisher's exact test. Independent samples “t”-test was used to compare the means of quantitative variables, between MSA-P and MSA-C groups that are normally distributed, whereas Mann–Whitney U-test was used to compare the medians of non-normal quantitative variables between MSA-P and MSA-C groups. Further, we used one way analysis of variance followed by Bonferroni post hoc test to compare the normal quantitative variables, between MSA-P, MSA-C, and control groups. Finally, Kruskal–Wallis test followed by Mann–Whitney U-test was used to compare the non-normal quantitative variables, between MSA-P, MSA-C, and control groups. All tests were two sided and level of significance was set at P < 0.05.

RESULTS

In total, 112 hospital charts of all suspected cases of probable MSA were reviewed. Out of 112 cases, 80 fulfilled the criteria given by second consensus statement for the diagnosis of probable MSA.[10] Out of these 80 probable MSA patients, 35 patients were excluded. These 35 excluded patients comprised five patients whose age was >70 years, 10 patients with heart disease, 17 patients with diabetes mellitus, and 3 patients whose magnetic resonance imaging (MRI) brain showed infarcts. Only 45 probable MSA patients fulfilled the study criteria. Out of these 45 patients, 23 were probable MSA-C and 22 were probable MSA-P. Heart rate variability data of 45 age- and gender-matched healthy volunteers obtained from the normative data base of the autonomic laboratory were compared with MSA patients.

Demographic and clinical characteristics

We have summarized the demographic and clinical features in Table 1. There were 12 males in MSA-P group, 19 males in MSA-C group, and 31 males in control group. The number of females in MSA-P, MSA-C, and control groups was 10, 4, and 14, respectively. Mean age was 57.2 ± 9 years for MSA-P group, 54.2 ± 7 years for MSA-C group, and 55.6 ± 9 years for the control group. Mean age of onset of MSA was 54.6 ± 9 years in MSA-P group and 51.9 ± 7 years in MSA-C group. Median disease duration was 2 (1.6) years in MSA-P and 2 (1.5) years in MSA-C groups. There was no significant difference between the groups in the demographic features. MSA-P and MSA-C patients had a variable combination of parkinsonian, cerebellar, and autonomic features. Dysarthria was similar between subgroups. MSA-P patients had a trend toward an increased number of falls and bladder disturbances as compared with MSA-C (P = 0.05). Postural dizziness, orthostatic hypotension, and erectile dysfunction were similar between the MSA subgroups. Presence of REM behavior disorder (RBD) and duration of RBD were also similar between the subgroups (MSA-P – 36.4%, 4.4 ± 3 years; MSA-C – 47.8%, 2.3 ± 1.7 years).

Table 1

Demographic and clinical features of probable MSA-P and MSA-C patients

Demographic and clinical parameters	Overall (n=45)	MSA-P (n1=22)	MSA-C (n2=23)	Test statistic	P
Age in years# (mean±SD)	55.69±8.09	57.23±8.67	54.22±7.39	0.71	0.494
Age of onset of disease in years# (mean±SD)	53.27±8.25	54.68±8.98	51.91±7.42	1.12	0.265
Disease duration in years$ [median (IQR)]	2 (1.5)	2 (1.6)	2 (1.5)	153	0.78
Gender
Male [n (%)]@	31 (68.9)	12 (54.5)	19 (82.6)	4.13	0.127
Female [n (%)]@	14 (13.11)	10 (45.5)	4 (17.4)
Parkinsonian features
Tremor [n (%)]@	25 (55.6)	15 (68.2)	10 (43.5)	2.78	0.095
Rigidity [n (%)]@	38 (84.4)	21 (95.5)	17 (73.9)	3.97	0.046*
Bradykinesia [n (%)]@	35 (77.8)	21 (95.5)	14 (60.9)	7.78	0.005**
Stooped posture [n (%)]@	8 (17.8)	7 (31.8)	1 (4.3)	5.8	0.02*
Postural Instability [n (%)]@	8 (17.8)	5 (22.7)	3 (13)	0.72	0.39
Freezing of gait [n (%)]@	7 (15.6)	7 (31.8)	0 (0)	8.66	0.003**
Cerebellar features
Ataxic gait [n (%)]@	14 (31.1)	3 (13.6)	11 (47.8)	6.13	0.013*
Finger nose incoordination [n (%)]@	24 (53.3)	3 (13.6)	21 (91.3)	27.25	<0.001***
Knee-Heel-Shin Incoordination [n (%)]@	22 (48.9)	2 (9.1)	20 (87)	27.28	<0.001***
Dysdiadochokinesia [n (%)]@	20 (44.4)	3 (13.6)	17 (73.9)	16.54	<0.001***
Impaired tandem gait [n (%)]@	21 (46.7)	3 (13.6)	18 (78.3)	18.86	<0.001***
Dysarthria [n (%)]@	28 (62.2)	12 (54.5)	16 (69.6)	1.08	0.299
Up-gaze restriction [n (%)]@	16 (35.6)	11 (50)	5 (21.7)	3.92	0.048*
Down-gaze restriction [n (%)]@	3 (6.7)	2 (9.1)	1 (4.3)	0.41	0.608
Lateral gaze restriction [n (%)]@	1 (2.2)	0 (0)	1 (4.3)	0.98	0.323
Slow saccades [n (%)]@	7 (15.6)	4 (18.2)	3 (13)	0.23	0.634
Jerky pursuit [n (%)]@	3 (6.7)	2 (9.1)	1 (4.3)	0.41	0.608
Vertical gaze evoked nystagmus [n (%)]@	1 (2.2)	1 (4.5)	0 (0)	1.07	0.489
Horizontal gaze evoked nystagmus [n (%)]@	6 (13.3)	2 (9.1)	4 (17.4)	0.67	0.665
Falls [n (%)]@	24 (53.3)	15 (68.2)	9 (39.1)	3.81	0.051
Cardiac dysautonomia features
Postural dizziness/orthostatism [n (%)]@	25 (55.6)	10 (45.5)	15 (65.2)	1.78	0.236
Orthostatic hypotension [n (%)] @	22 (48.9)	10 (45.5)	12 (52.2)	0.20	0.768
Δ↓Systolic BP in mmHg [Median (IQR)]$	23 (25)	25 (30)	23 (17)	112	0.752
Δ↓Diastolic BP in mmHg [Median (IQR)]$	12 (19)	12 (21)	14 (21)	97	0.519
Urogenital features
Bladder disturbances [n (%)]@	35 (77.7)	19 (86.36)	16 (69.56)	3.74	0.053
Erectile dysfunction [n (%)]@	9 (20)	2 (9.1)	7 (30.4)	3.20	0.074
REM behavior disorder (RBD)
RBD present [n (%)]@	19 (42.2)	8 (36.4)	11 (47.8)	0.61	0.436
Duration of RBD in years (Mean±SD)#	3.11±2.58	4.4±3.36	2.31±1.75	1.48	0.165

MSA=multiple system atrophy; MSA-P=parkinsonian variant; MSA-C=cerebellar variant; SD=standard deviation; IQR=interquartile range. #Independent samples “t” test, $Mann-Whitney U-test,@Pearson’s Chi-square/Fisher’s exact test, *P<0.05, **P<0.01, ***P<0.001

All the MSA-P patients were on dopaminergic medications. Fifteen MSA-P patients were on levodopa alone. Seven MSA-P patients were on a combination of levodopa and other dopaminergic medications, such as pramipexole, ropinirole, rasagiline, and amantadine. Out of 23 MSA-C patients, 21 MSA-C patients were on dopaminergic medications. Four MSA-C patients were on levodopa alone, 13 were on amantadine alone, and 4 were on a combination of levodopa and amantadine. Total levodopa equivalent drug dosage (LEDD) was calculated based on the protocol given by Tomlinson et al.[11] Total LEDD and levodopa responsiveness are summarized in Table 2. Median LEDD was 400 (413) mg/day in MSA-P group and 200 (100) mg/day in MSA-C group. About 45.5% of MSA-P and 4.3% of MSA-C patients had minimal benefit from levodopa, whereas 40.9% MSA-P and 91.3% MSA-C patients did not have any benefit from levodopa. One MSA-P (4.5%) patient had minimal benefit from levodopa initially for about 6 months but later no response was observed. About 9% patients with MSA-P and 4% MSA-C patients initially had a good response to levodopa for about 5–6 months. but later no response was observed. MSA-P patients had relatively more benefit from levodopa as compared with MSA-C patients (P = 0.004).

Table 2

LEDD and levodopa responsiveness in probable MSA subgroups

	Overall (n=43)	MSA-P (n1=22)	MSA-C (n2=21)	Test statistic	P
Total LEDD in mg/day [Median (IQR)]$	300 (300)	400 (413)	200 (100)	141.5	0.028*
Levodopa responsiveness@ [n (%)]
No response	30 (66.7)	9 (40.9)	21 (91.3)	13.481	0.004**
Minimal response (1-30%)	11 (24.4)	10 (45.5)	1 (4.3)
Good response (31-60%)	0 (0)	0 (0)	0 (0)
Very good response (>60%)	0 (0)	0 (0)	0 (0)
Initially minimal response but later no response	1 (2.2)	1 (4.5)	0 (0)
Initially good response but later no response	3 (6.7)	2 (9.1)	1 (4.3)

MSA=multiple system atrophy; MSA-P=parkinsonian variant; MSA-C=cerebellar variant; LEDD=levodopa equivalent drug dose; IQR=interquartile range, $Mann-Whitney U-test, @Pearson’s Chi-square/Fisher’s exact test, *P<0.05, **P<0.01

Cardiac autonomic dysfunction

We have summarized the results of conventional cardiac autonomic function test in Figure 1. In MSA-P group, 22.2% had definitive and 77.7% had severe cardiac autonomic dysfunction. In MSA-C group, 9.5% had early, 28.5% had definitive, and 57.1% had severe cardiac autonomic dysfunction. Four MSA-P patients and two MSA-C patients were unable to perform conventional cardiac autonomic function tests. MSA-P patients had a nonsignificant trend toward severe cardiac autonomic dysfunction. We have summarized the results of short-term heart rate variability of MSA patients and age- and gender-matched healthy controls in [Table 3]. HRV analysis showed that SDNN, RMSSD, NN50, pNN50, total power, low-frequency power, and high-frequency power were significantly reduced in both MSA subgroups as compared with controls (P < 0.001). Average heart rate was significantly higher in MSA subgroups as compared with controls (P < 0.001). LF/HF ratio was similar between the MSA patients and controls. We have summarized the results of short-term HRV of MSA subgroups in Table 4. Post hoc analysis showed that low-frequency power (sympathetic activity) was significantly lower in MSA-P patients as compared with MSA-C patients (P = 0.016). SDNN, RMSSD, and total power were also relatively lower in MSA-P patients as compared with MSA-C patients, though not statistically significant.

Figure 1

Clustered column graph showing the severity of cardiac autonomic dysfunction (conventional AFT) in probable MSA-P and MSA-C subgroups (AFT = autonomic function test; MSA = multiple system atrophy; MSA-P = parkinsonian variant; MSA-C = cerebellar variant; Pearson's Chi-square test; values expressed as a percentage)

Table 3

Time and frequency domain parameters of short-term heart rate variability in probable MSA subgroups and healthy controls

Short-term heart rate variability	MSA-P (n1=22)	MSA-C (n2=23)	Controls (n3=45)	Test statistic	P
Time domain parameters
SDNN (ms)$	14.09 (8.54)	16.59 (13.03)	41.26 (17.36)	55.42	<0.001***
RMSSD (ms)$	8.7 (9.12)	11.9 (10.07)	30.29 (18.12)	44.41	<0.001***
NN50 (count)$	0.001 (2)	0.001 (1)	26 (60)	46.64	<0.001***
pNN50 (%)$	0.001 (0.58)	0.001 (0.29)	8.53 (15.05)	44.07	<0.001***
Average HR#	82.67±14.48	77.7±12.14	68±11.18	10.26	<0.001***
Frequency domain parameters
Total power (ms2)$	183.78 (250)	255.61 (384.1)	1639.74 (1500)	55.53	<0.001***
HF power (ms2)$	26.03 (38.7)	29.67 (123.93)	352.23 (496.9)	47.89	<0.001***
LF power (ms2)$	26.59 (46.77)	48.49 (79.72)	421.28 (390)	54.75	<0.001***
LF/HF ratio$	1.25 (1.53)	1.76 (2.44)	0.87 (0.88)	4.31	0.116

MSA=multiple system atrophy; MSA-P=parkinsonian variant; MSA-C=cerebellar variant; SDNN=standard deviation of RR Intervals; RMSSD=square root of the mean of the sum of squares of differences between adjacent RR intervals; NN50=count of number of pairs of RR intervals differing by >50 ms; pNN50 (%) = percentage of count of number of pairs of adjacent RR intervals differing by more than 50 ms; average HR=average heart rate; HF power=high-frequency power; LF power=low-frequency power; LF/HF ratio=sympathovagal balance; IQR=interquartile range. #One way analysis of variance, values expressed as Mean±SD. $Kruskal-Wallis test, values expressed as median (IQR). ***P<0.001

Table 4

Time and frequency domain parameters of short-term heart rate variability in MSA-P and MSA-C subgroups

Short-term heart rate variability	MSA-P (n1=22)	MSA-C (n2=23)	Test statistic	P
SDNN (ms)$	14.09 (8.54)	16.59 (13.03)	195	0.188
RMSSD (ms)$	8.7 (9.12)	11.9 (10.07)	180	0.097
NN50 (count)$	0.001 (2)	0.001 (1)	243	0.78
pNN50 (%)$	0.001 (0.58)	0.001 (0.29)	246	0.85
Total power (ms2)$	183.78 (250)	255.61 (384.1)	202	0.24
HF power (ms2)$	26.03 (38.7)	29.67 (123.93)	203	0.25
LF power (ms2)$	26.59 (46.77)	48.49 (79.72)	147	0.016*
LF/HF ratio$	1.25 (1.53)	1.76 (2.44)	218	0.42

MSA=multiple system atrophy; MSA-P=parkinsonian variant; MSA-C=cerebellar variant; SDNN=standard deviation of RR Intervals; RMSSD=square root of the mean of the sum of squares of differences between adjacent RR intervals; NN50=Count of number of pairs of RR intervals differing by >50 ms; pNN50 (%) = percentage of count of number of pairs of adjacent RR intervals differing by more than 50 ms; HF power=high-frequency power; LF power=low-frequency power; LF/HF ratio=sympathovagal balance; IQR=interquartile range. $Mann-Whitney U-test, values expressed as median (IQR). *P<0.05

DISCUSSION

The MSA patients in this single-center retrospective study had a variable combination of parkinsonian, cerebellar, and autonomic symptoms with poor levodopa responsiveness. Mean age of onset of MSA-P was 54.7 ± 9 years in our study, whereas it was 56 ± 10 years in the U.S. cohort,[6] 56.8 ± 9 years in a European cohort,[7] 56.7 ± 7.2 years in a Japanese cohort,[8] and 56.1 ± 10 years in another Indian cohort.[9] Mean age of onset of MSA-C was 51.9 ± 7 years in our study, 56 ± 8 years in the U.S. cohort, 55.4 ± 7.4 years in the European cohort, 54.8 ± 8.7 years in the Japanese cohort, and 53.1 ± 9 years in the other Indian cohort. Mean age of onset of MSA subtypes in our study was lower than in the other cohorts.[6789] A mild male preponderance was seen in our study as well as U.S., European, and Japanese cohorts. Demographic variables such as mean age of onset of MSA, disease duration, and gender distribution were similar between the MSA subgroups in our study as well as these cohorts.[6789]

Parkinsonian features and cerebellar features tended to overlap in both groups, although parkinsonism was more likely to occur in MSA-C than ataxia in MSA-P. Similar results were seen in a retrospective study in U.S. population.[17] Among the motor features, dysarthria was similar between MSA phenotypes. MSA-P patients had a trend toward an increased number of falls as compared with MSA-C. This finding is in contrast to U.S. retrospective study,[17] wherein MSA-C patients had an increased number of falls. A greater proportion of MSA-P patients responded to dopaminergic medications as compared with MSA-C patients (P = 0.004). Our study revealed that a greater percentage of MSA-P than MSA-C patients had a beneficial response to levodopa, such as in the U.S.[18] and European studies:[7] 45.5% in our study, 42.5% in European study, and 57% in U.S. study.

MSA-P patients had a trend toward an increase in bladder disturbances as compared with MSA-C patients. Postural symptoms were similar between the subgroups. Cardiac autonomic dysfunction assessed with conventional cardiac autonomic function tests also showed that a relatively greater percentage of MSA-P patients had severe autonomic dysfunction but was not statistically significant. Similar results were obtained in studies carried out by Schmidt et al.[19] and Tandon et al.[9] However, our results are in contradiction to the results obtained by studies carried by Coon et al.[17] and Wenning et al.[20] Coon et al. found that postural symptoms were more predominant in MSA-P, whereas Wenning et al. found that severity of orthostatic hypotension was more in MSA-C. Analysis of short-term heart rate variability showed that overall HRV (total power, SDNN), sympathetic activity (low-frequency power), and parasympathetic activity (high-frequency power, RMSSD, NN50, and pNN50) were significantly reduced in both MSA-P and MSA-C patients as compared with controls (P < 0.001). The LF/HF ratio, which signifies sympathovagal balance, was similar between MSA and control groups. This implies that both sympathetic and parasympathetic limbs are affected in MSA. Studies carried by Furushima et al.,[21] Brisinda et al.,[22] and Kiyono et al.[23] have also shown that all the time and frequency domain parameters of HRV, except LF/HF ratio, are significantly reduced in MSA patients as compared with controls. In our study, we found that the sympathetic limb was more severely affected in MSA-P as compared with MSA-C patients (P = 0.016). During a literature search, we did not come across any studies which have assessed the differences in HRV between MSA-P and MSA-C patients. Hence, we could not compare this result with other studies.

Our study results demonstrate that there is significant autonomic dysfunction in MSA patients, as evidenced by symptomatology, conventional cardiac autonomic function testing, and short-term HRV. Postmortem pathological examination of MSA patients has revealed cell loss in the dorsal motor nucleus of the vagus nerve, locus coeruleus, catecholaminergic neurons of the ventrolateral medulla, and pontomedullary reticular formation in the brain; and degeneration of sympathetic and parasympathetic preganglionic nuclei in spinal cord regions.[5] Degeneration of autonomic regulatory areas is likely responsible for the cardiac autonomic dysfunction observed in our study. Autonomic function testing is important for assessing the degree of severity of dysautonomia. This is of clinical utility for predicting the prognosis and for treatment planning to prevent potential life-threatening complications.

The sympathetic limb was more severely affected in MSA-P as compared with MSA-C patients. Functional integrity of the sympathetic limb is vital to prevent orthostatic hypotension and syncopal attacks.[24] Bladder disturbances and postural instability were also relatively more common in MSA-P as compared with MSA-C patients. Cardiac autonomic dysfunction increases the risk of syncopal attacks, recurrent falls, and ventricular arrhythmias.[2] Various studies have shown that MSA patients with early and more severe autonomic dysfunction have shorter survival.[7] They carry a sevenfold higher risk for sudden death.[25] Postural instability, bladder disturbances and cardiac autonomic dysfunction are negative prognostic factors, which were more severe in MSA-P patients in this study. This implies that MSA-P patients might have relatively worse prognosis as compared with MSA-C patients. We did not assess survival in this study. Previous survival studies carried out by Wenning et al.[7] and Saito et al.[26] have shown that the parkinsonian phenotype is a negative prognostic factor in MSA.

Our study has focused on characterization of clinical features and cardiac autonomic dysfunction including heart rate variability, in Indians affected with probable MSA-P and MSA-C. Also, unique to this study is that all MSA patients were seen and diagnosed at a single tertiary care center by movement disorder specialists and underwent comprehensive standardized autonomic testing. There is a lesser chance of misdiagnosis because we included only probable and not possible MSA patients, and they had a mean follow-up of 1 ± 0.5 years. A major limitation of this study is that it is retrospective in nature and certain variables of our study interest, such as RBD might not be documented. To avoid introducing false positive symptoms and signs, we treated missing data on specific clinical features as “not present.” A prospective follow-up study with a larger cohort of MSA might give more robust results.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.