Early stridor onset and stridor treatment predict survival in 136 patients with MSA
Citation Manager Formats
Make Comment
See Comments

Abstract
Objective: To evaluate the predictive value of stridor and its latency of onset and to investigate the role of stridor treatment in a cohort of patients with multiple system atrophy (MSA) referred to a tertiary center.
Methods: We retrospectively identified patients diagnosed with MSA referred to our department beginning in 1991 and evaluated at least yearly during the disease course. Stridor was defined as present when confirmed by a whole night video-polysomnography and as early if presenting within 3 years of disease onset. Survival data, from disease onset to time of death, were calculated with Kaplan-Meier curves. Predictors were identified in univariate and multivariable Cox regression analyses.
Results: We included 136 patients with MSA; 113 were deceased at the time of study. Stridor was diagnosed in 42 patients, and 22 presented early stridor onset. Twelve of the 31 patients treated for stridor received tracheostomy, and 19 received continuous positive airway pressure. Overall survival did not differ between patients with and without stridor, while patients with early stridor onset had a worse prognosis than those developing this symptom later. In the stridor subgroup, early stridor onset was an unfavorable survival predictor. Stridor treatment was significantly associated with survival in our population. The Kaplan-Meier curve did not reveal significant differences in survival between the 2 treatments even though there was a trend toward longer disease duration in patients receiving tracheostomy.
Conclusions: Our results demonstrated that early stridor onset is an independent predictor for shorter survival and that tracheostomy could control stridor, influencing disease duration.
GLOSSARY
- CI=
- confidence interval;
- CPAP=
- continuous positive airway pressure;
- HR=
- hazard ratio;
- IQR=
- interquartile range;
- MSA=
- multiple system atrophy;
- MSA-C=
- cerebellar phenotype of multiple system atrophy;
- MSA-P=
- parkinsonian phenotype of multiple system atrophy;
- VPSG=
- video-polysomnography
Multiple system atrophy (MSA) is a neurodegenerative disorder characterized by an autonomic dysfunction associated with a combination of cerebellar, parkinsonian, or pyramidal signs. The diagnostic criteria define 3 degrees of certainty for diagnosis—possible, probable, and definite—and 2 phenotypes: parkinsonian (MSA-P) or cerebellar (MSA-C), according to the predominant feature at the time of evaluation.1,2 The mean survival in MSA ranges from 6.2 to 10 years,3,–,11 with few patients surviving >15 years.12,13 Various clinical factors have been reported as predictive of shortened survival.5,6,11 Stridor is one of the additional features for the diagnosis of possible MSA showing a high diagnostic positive predicted value14,15; however, its prognostic role is controversial.11,16,17
Stridor is a harsh, high-pitched inspiratory sound due to a laryngeal narrowing, which in MSA typically occurs during sleep, and has been attributed to denervation of laryngeal muscles or to adduction dystonia of vocal cords.15,18
There are 2 main options to treat stridor: tracheostomy or continuous positive airway pressure (CPAP).15,16,19,–,23 Tracheostomy is currently preferred in the advanced disease stage and if wakefulness stridor and immobile vocal cords on laryngoscopy appear.15,22,24 CPAP as a noninvasive therapy can be used for mild stridor.15,19,20,23 However, guidelines for stridor management are lacking, and only a few studies with small sample sizes analyzed the role of stridor treatment in survival.16,17,19,20,25
The aim of the present study was to evaluate the predictive value of stridor and of its latency of onset in a cohort of patients with MSA referred to a tertiary center. Secondarily, we investigated the role of stridor treatment as a predictor of survival in MSA.
METHODS
Patient selection.
We retrospectively selected all patients with a final clinical diagnosis of MSA referred to the Movement Disorders and Autonomic Disorders centers of the Department of Biomedical and Neuromotor Sciences, University of Bologna, Italy, between 1991 and 2014 and evaluated at least once a year during the disease course.
Three neurologists expert in movement disorders independently confirmed the diagnosis of MSA from data available at the last follow-up evaluation according to international criteria.1 Their consensus and the absence of nonsupporting features for MSA were mandatory for inclusion in the study.
The MSA phenotype was defined as predominantly MSA-P or MSA-C on the basis of the predominant motor involvement at the time of the last follow-up visit.
The following medical records were analyzed when available: (1) clinical notes reporting family and patient history, annual neurologic evaluations that included blood pressure measurement in the clinostatic position and within 3 minutes of standing, and information on new symptom onset, treatment changes, response, and adverse effects; (2) brain CT or MRI; (3) neuropsychological evaluation26; (4) all-night video-polysomnography (VPSG)18; (5) tilt test and other cardiovascular reflex tests27; and (6) the effect of levodopa28 assessed by a standardized oral levodopa kinetic-dynamic test or improvement in part III of the Unified Parkinson’s Disease Rating Scale after levodopa was increased to 1 g/d. Orthostatic hypotension had been assessed before and after levodopa administration by standing or tilt test every year.29
The following clinical data were collected from medical records in a standardized fashion by an author and entered into an ad hoc database for statistical analysis: (1) age at disease onset, the age in years at the time of the first reported motor or autonomic symptom or sign that could be related to MSA1; (2) age and cause of death; (3) disease duration, the interval in years from first symptom onset to death or to the end of the study; (4) symptoms at initial presentation; (5) the occurrence of parkinsonian, cerebellar, autonomic, or pyramidal signs; (6) the presence of stridor; and (7) stridor treatment (CPAP or tracheostomy). For each symptom and sign, the latency of occurrence from disease onset was noted. Stridor required confirmation by VPSG and was categorized as early if presenting within 3 years of disease onset.
Finally, we analyzed the timing and latency of the following milestones of disease progression, which usually required additional medical attention: frequent falls (at least 3 falls per year or documentation of frequent or several falls), loss of ambulatory independence, wheelchair dependence, severe dysphagia or percutaneous endoscopic gastrostomy, severe dysarthria, and urinary catheterization.
Survival data were defined on the basis of time to death from the first symptom of disease. Patients or their relatives were contacted by telephone and questioned regarding the clinical course and the time and the cause of death (if applicable) when the patient missed a clinical evaluation within 24 months.
Standard protocol approvals, registrations, and patient consent.
The study was approved by the Department of Biomedical and Neuromotor Sciences institutional review board, and patients gave their written informed consent to personal data processing for research purposes.
Statistical analysis.
The normality of the distribution of the continuous parameters was checked with the Skewness-Kurtosis test, and variables were expressed as the mean ± SD or median and interquartile range (IQR) when appropriate. The t test or Wilcoxon rank-sum test was performed to compare continuous variables as appropriate. Categorical variables were described by their absolute or relative frequencies and compared by use of the χ2 test. Kaplan-Meier curves were used to graphically analyze the overall death survival (or related death survival), and the log-rank test was performed to compare survival between patient subgroups.
To identify variables associated with survival in MSA and in the stridor subgroup, univariate and multivariable Cox regression analyses were performed. Parameters with a value of p < 0.1 on univariate analysis were entered into the multivariable model. The following variables were studied: age at disease onset, sex, predominant clinical phenotype, presence of stridor, and symptoms of disease onset and autonomic onset. The same variables and the latency of stridor onset were considered in the stridor subgroup.
A value of p < 0.05 (2-sided) was considered significant. Statistical analyses were performed with STATA statistical software, version 14.0.
RESULTS
Demographic and clinical features of the study sample are shown in table 1. A total of 136 patients with MSA were included (88 male); 113 were deceased at time of study. Three patients met the consensus criteria for definite MSA, 106 for probable MSA, and 27 for possible MSA. Sixty-eight patients were classified as having MSA-P and the other 68 as having MSA-C. The median disease duration was 7 years (IQR 5–9 years). Kaplan-Meier estimates of death in the overall population are shown in the figure, A. In this analysis, the median duration of illness was 7.84 years.
Demographic and clinical characteristics of the study sample
(A) Survival analysis in all patients with stridor; (B) survival analysis in patients with and without stridor; (C) survival analysis in patients with early and late stridor onset; and (D) survival analysis in patients with stridor without treatment who were treated with tracheostomy and continuous positive airway pressure (CPAP).
Data for VPSG were available for 120 patients. Stridor was diagnosed in 42 patients (25 male; mean ± SD age at stridor onset 59.34 ± 8.38 years; table 2); 36 of these patients were deceased. The median disease duration was 6 years (IQR 5–9 years). There were no differences in sex rate, percentage of deaths, or age at disease onset between patients with and without stridor. Patients with stridor more frequently showed an autonomic onset (76.19% vs 50.00%, p = 0.003) compared to those without. Kaplan-Meier estimates of death in patients with and without stridor showed that the overall survival did not differ between the 2 groups (log-rank test, p = 0.3467; figure, B). Univariate Cox regression analyses of the influence of variables on survival in patients with MSA identified 2 factors associated with short survival: autonomic onset and mode of autonomic onset (table 3).
Demographic and clinical features of the study sample with stridor
Variables associated with survival in patients with multiple system atrophy (MSA) in the univariate Cox regression analysis
Concerning the latency of the stridor onset, available in 41 of 42 patients, there were no differences in sex rate, age at disease onset, percentage of deaths, percentage of treatment, and type of treatment in cases with early stridor onset (n = 22) compared to those with late stridor onset (table 2). Patients with early stridor onset more frequently showed an autonomic onset (90.91% vs 57.89%, p = 0.025) and less frequently a parkinsonian onset (0.00% vs 31.58%, p = 0.003) and had a shorter median disease duration than those with a late stridor onset (6 years [IQR 4–8 years] vs 8 years [IQR 6–12 years]; p = 0.0083). The overall median disease duration after the onset of stridor differed between patients with early stridor onset and those with late onset (5 years [IQR 3–7 years] vs 2 years [IQR 1–5 years], respectively; p = 0.0274). The overall mean ± SD disease duration after treatment for stridor was 3.13 ± 2.02 years with no difference between the 2 groups (p = 0.2992; table 2). The risk of death estimated by Kaplan-Meier analysis (figure, C) was higher in patients developing early stridor (p = 0.0209, log-rank test). The incidence rate of death was 13 per 100 person-years in the first group and 9 per 100 person-years in the second group. In the stridor subgroup, univariate Cox regression analyses identified the following as factors associated with survival: early stridor onset (hazard ratio [HR] = 2.14, 95% confidence interval [CI] = 1.05–4.44, p = 0.040) and mode of autonomic onset (HR = 2.82, 95% CI = 1.12–7.09, p = 0.028). In the multivariable model, early stridor onset remained an independent predictor of mortality after adjustment for mode of autonomic onset (HR = 3.21, 95% CI = 1.41–7.31, p = 0.006).
In regard to the stridor treatment, 12 of the 31 patients treated for stridor received tracheostomy and 19 received CPAP (table 4). The overall mean ± SD treatment latency was 1.93 ± 1.53 years. Patients treated for stridor had a longer median disease duration than those without treatment (8 years [IQR 6–10 years] vs 5 years [IQR 4–6 years], p = 0.0178). Patients treated with tracheostomy showed a longer median disease duration than those treated with CPAP (10 years [IQR 7–13 years] vs 6 years [IQR 5–8 years], p = 0.0131), even though the first group had a longer latency of treatment (2.58 ± 1.38 vs 1.35 ± 1.45 years; p = 0.0300). Patients with tracheostomy showed both a longer disease duration after stridor onset and a longer disease duration after treatment (p = 0.0002 and p = 0.0017, respectively; table 4).
Demographic and disease duration in the study sample treated for stridor
Kaplan-Meier curves showed a difference in survival among patients without treatment, those treated with tracheostomy, and those treated with CPAP (p = 0.0139). This statistical significance was attributable to the difference in mortality between patients without treatment and those treated with tracheostomy (p = 0.0071; figure, D). Incidence rate of death was 16 per 100 person-years in patients without treatment, 9 per 100 person-years in those who underwent tracheostomy, and 11 per 100 person-years in those treated with CPAP.
DISCUSSION
This is one of the few studies investigating the value of stridor and its latency of onset as predictors of survival in a cohort of patients with MSA. The study showed that in our population, in whom stridor prevalence is similar to that reported in the literature,7,11,16,17,19,21,30,–,32 overall survival did not differ between patients with MSA with stridor and without stridor. However, we demonstrated that early stridor onset is an independent risk factor for shorter survival. Furthermore, we showed that treatment is associated with survival in patients with MSA, and concerning treatment type, we observed a trend toward longer disease duration in patients who had undergone tracheostomy compared to those treated with CPAP.
Only 3 retrospective single-center studies included data on the role of stridor as a prognostic indicator, reporting conflicting results.11,16,17 This disagreement may be a consequence of the differences in the design, population characteristics, diagnostic certainty (clinical vs autopsy based), and definition of stridor in the studies. In particular, only one study analyzed survival curves on a large sample,11 and stridor was confirmed by polysomnography only in another study.16 One study conducted on 49 patients with pathologically confirmed MSA showed that the presence of stridor had no significant effect on survival time (HR = 1.21,95% CI = 0.63–2.34, p = 0.57), as in our study, and on the risk of sudden death (HR = 1.84, 95% CI = 0.55–6.16, p = 0.32). In that study, however, the assessment and definition of stridor were not specified in the method section.17 Recently, Coon and colleagues11 investigated in a large population (n = 685) the role on survival time of numerous variables, including stridor, the diagnosis of which was based on clinical history. This symptom was associated with shortened survival (from symptom onset to death, HR = 1.38, 95% CI = 1.14–1.68, p = 0.001) but lost its predictive value in the presence of other variables in the multivariate model.
On the contrary, the study performed by Silber and Levine,16 reporting survival curves of 30 patients with MSA who underwent polysomnography, showed a shorter survival time from polysomnography to death in patients with stridor (n = 11) compared to those without (median survival 3 vs 4 years, respectively, p < 0.05). However, survival curves obtained from disease onset did not significantly differ between groups and diverged after 7 years. This finding suggested, as observed in our study, that the latency of stridor may influence the mortality of patients with MSA. In addition, we found as negative predictors of survival the mode of autonomic onset and autonomic disease onset; the latter also was reported in other studies.5,–,7,11,13,17
Concerning the treatment of stridor, our study demonstrated that this factor is significantly associated with survival in patients with MSA, as previously observed in other studies.19,–,21
Stridor treatment is based mainly on tracheostomy or CPAP,15,16,19,–,22 but guidelines for the use of these treatments in patients with MSA have yet to be established, and few studies with small sample sizes have analyzed the role of treatment type as a predictor of survival.16,17,19,20,23
Our results suggested that patients treated with tracheostomy had longer overall disease duration, longer disease duration after stridor onset, and longer disease duration after treatment compared with those treated with CPAP. However, the Kaplan-Meier curve did not reveal a difference in survival between these 2 treatments even if there was a trend toward longer disease duration in patients who had undergone tracheostomy compared with those treated with CPAP (p = 0.0850), with a difference in the incidence rate of death (9 vs 11 per 100 person-years, respectively). This result could be due to the small sample size and to the fact that, at the end of the study, 100% of the first group had died, while 5 patients (26%) in the second group, enrolled more recently, were still alive. Furthermore, the small number of patients treated for stridor did not allow a multivariate analysis.
Only one previous descriptive study compared these 2 types of stridor treatment and observed survival in a small sample of patients with MSA who underwent polysomnography. In this study, 2 of the 4 patients receiving tracheostomy died 1 year after the sleep evaluation, and the other 2 patients were alive 1.9 and 7 years later, while all 5 patients treated with CPAP died a mean of 2.4 years after the sleep evaluation. However, one patient treated with CPAP had poor compliance, and another patient had audible stridor despite device use. The authors concluded that CPAP was not effective in preventing death in their sample, but increased attention to full compliance and the continued elimination of stridor might result in a more successful outcome.16
Two studies focused on the role of tracheostomy on survival. One study, conducted on 49 patients with definite MSA, showed that tracheostomy reduces the risk of death (HR = 0.21, 95% CI = 0.08–0.56, p < 0.01) and of sudden death (HR = 0.15, 95% CI = 0.02–0.98, p < 0.05) in MSA.17 Another study compared polysomnographic findings in 7 patients with and 11 without tracheostomy, showing that the first group more frequently had central-type sleep-disordered breathing than the second group. Analyzing polysomnographic studies before and after tracheostomy in 3 patients with MSA, the authors reported that this treatment aggravated central sleep apnea, thereby increasing the risk of nocturnal sudden death.33 However, as reported in the comments of that article, these results, obtained from a small sample size, should be interpreted with caution because disease severity was measured only by the Hoehn and Yahr score, which is a reductive scale in MSA; the disease duration of patients with tracheostomy was twice as long as that of the other group; no patients died during the course of the study; and the mean survival time was 11.4 years (in line with a mean life expectancy for patients with MSA).34,35
Other studies with small sample groups of patients with MSA focused only on CPAP treatment demonstrated that this device is effective for eliminating nocturnal stridor19,20,23 and is well-tolerated in the long term, mostly in those with early-stage disease.20 However, the aim of these studies was to assess the long-term tolerance of CPAP in patients with MSA; therefore, the authors did not focus on survival time. Only one of these studies reported data on survival time, showing that the median survival time from disease onset was similar between patients without stridor and those with stridor treated only with CPAP (88 vs 77 months, p = 0.6914).19
Although previous studies and our findings suggest that tracheostomy had a possible effect on survival time, a definitive conclusion remains elusive.
The strengths of our study are that all patients were seen and diagnosed in a single center and that the diagnosis of stridor was confirmed with VPSG. Furthermore, despite the retrospective nature of the study, patient data were updated at every follow-up visit, which was performed at least once a year. The main limitations of the study include a lack of systematic assessment of disease and autonomic failure severity by appropriate scores and of objective data on the number of hours per night of CPAP use. Moreover, treatment selection was not based on a standardized protocol because the study covers a long period of time (1991–2014) and CPAP was introduced as a noninvasive form of stridor treatment only in the last 10 years. Finally, data on causes of death were determined by talking with relatives and were not available for most of our patients.
This study represents one of the largest follow-up studies reported to date focusing on stridor, its treatment, and survival in patients with MSA. Our results demonstrated that early stridor onset has a negative prognostic value for survival and suggested that tracheostomy could better control stridor, influencing the disease duration. However, further prospective studies using a larger sample with systematic assessment of stridor control through VPSG and, if possible, taking into consideration other variables such as latency of stridor onset and latency of treatment are necessary to clarify the role of these treatments in MSA survival.
AUTHOR CONTRIBUTIONS
Dr. Giannini: acquisition, analysis, and interpretation of data, drafting of the manuscript. Dr. Calandra-Buonaura: conception and design of the study, supervision of the study, acquisition and interpretation of data, drafting of the manuscript. Dr. Mastrolilli: acquisition of data, critical revision of the manuscript. Dr. Righini: acquisition of data, analysis and interpretation of data. Dr. Bacchi-Reggiani: analysis and interpretation of data, critical revision of the manuscript. Dr. Cecere: acquisition of data, analysis and interpretation of data. Dr. Barletta: acquisition of data, analysis and interpretation of data. Dr. Guaraldi: conception and design of the study, acquisition and interpretation of data. Dr. Provini: substantial contributions to conception and design of the study, critical revision of the manuscript. Dr. Cortelli: substantial contributions to conception and design of the study, critical revision of the manuscript.
STUDY FUNDING
This study was supported by RFO Alma Mater Studiorum-University of Bologna.
DISCLOSURE
G. Giannini, G. Calandra-Buonaura, F. Mastrolilli, M. Righini, M. Bacchi-Reggiani, A. Cecere, G. Barletta, P. Guaraldi, and F. Provini report no disclosures relevant to the manuscript. P. Cortelli received honoraria for speaking engagements or consulting activities from Allergan Italia, Lundbeck Italy, UCB Pharma S.p.A, Chiesi Farmaceutici, AbbVie srl. Go to Neurology.org for full disclosures.
ACKNOWLEDGMENT
The authors thank Dr. Vicini for his collaboration in the study.
Footnotes
↵* These authors contributed equally to this work.
Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.
- Received March 14, 2016.
- Accepted in final form June 16, 2016.
- © 2016 American Academy of Neurology
REFERENCES
- 1.↵
- Gilman S,
- Wenning GK,
- Low PA, et al.
- 2.↵
- 3.↵
- Ben-Shlomo Y,
- Wenning GK,
- Tison F,
- Quinn NP
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- Watanabe H,
- Saito Y,
- Terao S, et al.
- 9.↵
- O'Sullivan SS,
- Massey LA,
- Williams DR, et al.
- 10.↵
- 11.↵
- Coon EA,
- Sletten DM,
- Suarez MD, et al.
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- Iranzo A,
- Santamaria J,
- Tolosa E, et al.
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- Wenning GK,
- Ben Shlomo Y,
- Magalhaes M,
- Daniel SE,
- Quinn NP
- 31.↵
- 32.↵
- Plazzi G,
- Corsini R,
- Provini F, et al.
- 33.↵
- Jin K,
- Okabe S,
- Chida K, et al.
- 34.↵
- Silber MH
- 35.↵
- Shimohata T,
- Ozawa T,
- Nakayama H,
- Tomita M,
- Nishizawa M
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.