Multiple system atrophy (MSA) is a heterogeneous neurodegenerative disorder, with a clinical presentation combining extrapyramidal, cerebellar, autonomic or pyramidal symptoms. There are two major subtypes: MSA‐P, with a clinical predominance of parkinsonism, and MSA‐C, with a clinical predominance of cerebellar symptoms. Although various factors have been proposed to predict survival in MSA, including age at onset and several phenotypic features,1 the terminal/end of life events have never been systematically studied. We present our results from a study on the causes of death in a series of pathologically confirmed, definite MSA cases.
All patients registered with the University of Miami/NPF Brain Endowment Bank (UM/BEB) donation programme with a diagnosis of neuropathologically confirmed, definite multiple system atrophy (MSA; n = 21) were included in this study. Pertinent information was gathered by two prospectively filled questionnaires used as part of the UM/BEB's recruitment process: (a) the UM/BEB Parkinson's disease registry form, a 128‐item, self‐administered questionnaire on demographics, environmental exposures, personal and family history, comorbid conditions, activities of daily living, clinical and treatment details; and (b) the “agonal state” form, a 25‐item questionnaire on events covering the 48 h before death completed by the treating doctor/nurse. For comparisons, each MSA case was closely matched for age at disease onset (±2 years) and sex with a Parkinson's disease brain donor by an investigator blinded to the disease status and clinical information. Medical, hospital and hospice records of brain donors were also collected on an annual basis and all disease‐related information was extracted by two independent clinical investigators (blinded to the aims of the study), and entered into separate databases that were checked for consistency. Brain removal, autopsy, fixation and sectioning were performed according to standard protocols. Postmortem diagnosis of MSA and Parkinson's disease were based on well‐accepted criteria.2,3 For statistical analysis, Mann–Whitney U test for two samples was used in non‐parametric comparisons, and χ2 tests with Yates' corrected p value and two‐tailed Fisher exact p values in the comparison of proportions, as appropriate. The study was approved by the local institutional review board.
Table 1 shows the demographics and primary causes of death of all patients. Patients with MSA had significantly shorter disease duration (p = 0.02) than matched patients with Parkinson's disease, and most presented with parkinsonian‐predominant symptoms. None of the patients had a predominantly autonomic presentation. In all, 15 of 21 (71.4%) patients with end‐stage MSA had permanent in‐dwelling balloon (Foley) catheters because of symptoms of urinary incontinence for at least 6 months before death; 13 (61.3%) had recurrent lower urinary tract infections (LUTIs). The recurrence of infections did not correlate with the presence of permanent Foley catheters; 4 of 13 (30.8%) patients with LUTIs did not have permanent Foley catheters. Two patients with MSA used clean intermittent self‐urinary bladder catheterisation. Of the 13 patients with recurrent LUTIs, 5 (38.5%) died as a result of their infections. In addition, 7 (33.3%) patients had recurrent (⩾2) episodes of aspiration and 8 (38.1%) had percutaneous gastrostomy (PEG) feeding tubes inserted because of swallowing/feeding difficulties and aspiration. The recurrence of aspirations was independent of PEG tubes, as 4 of 7 (57.1%) patients with PEG continued having episodes of aspiration after PEG. Of patients with recurrent aspirations, one died as a result of acute aspiration and two as a result of aspiration pneumonia.
Table 1 Characteristics and primary causes of death.
| MSA (n = 21) | PD (n = 21) | |
|---|---|---|
| Sex | 10M/11F | 10M/11F |
| Mean (SD) age at disease onset (years) | 59.4 (10.1) | 59.9 (8.9) |
| Mean (SD) disease duration (years)* | 8.5 (4.7) | 13.4 (8.2) |
| MSA type at presentation (%) | ||
| MSA‐P | 16 (76.2) | |
| MSA‐C | 5 (23.8) | |
| Permanent balloon (Foley) catheters (%)† | 15 (71.4) | 7 (33.3) |
| Percutaneous gastrostomy (%) | 8 (38.1) | 3 (14.3) |
| Tracheostomy (%) | 2 (9.5) | None |
| Sudden death (%)‡ | 8 (38.1) | 3 (14.3) |
| Cause of death (%) | ||
| Cardiopulmonary arrest | 7 (33.3) | 1 (4.8) |
| Urinary tract infection | 5 (23.8) | None |
| Aspiration pneumonia | 2 (9.5) | None |
| Infectious pneumonia | 2 (9.5) | 7 (33.3) |
| Acute aspiration | 1 (4.8) | 1 (4.8) |
| Wasting syndrome | 3 (14.3) | 1 (4.8) |
| Cerebrovascular accident | None | 3 (14.3) |
| Myocardial infarction | None | 2 (9.5) |
| Cancer | None | 2 (9.5) |
| Other | 1 (4.8) | 1 (4.8) |
F, female; M, male; MSA, multiple system atrophy; MSA‐C, MSA with a clinical predominance of cerebellar symptoms; MSA‐P; MSA with a clinical predominance of parkinsonism; PD, Parkinson's disease.
Results in the comparisons between MSA and PD groups were significant for *p = 0.02, †p = 0.03 and ‡p = 0.02.
Sudden death related to MSA was reported in 8 (38.1%) patients. In seven patients, sudden death was characterised as cardiopulmonary arrest of otherwise unknown aetiology, and in one as acute aspiration during sleep. In all, 5 of 6 (23.8%) patients with reported symptoms of laryngeal stridor as part of their disease died suddenly. Two of the patients with stridor had a permanent tracheostomy. Sudden death was reported in one of them. Skin infections in the form of complicated pressure ulcers were present in 6 (28.6%) patients. However, skin infections were not associated with death in any case. Marked weight loss (⩾10% of premorbid weight) was reported in 16 (76.2%) patients. Weight loss was considerably less common in patients with PEG (5–31.2%) compared with those without PEG (11–68.8%). Three patients died as a result of weight loss and wasting. In contrast with Parkinson's disease, all patients with MSA died as a result of events related to their disease. One patient with MSA died from intestinal perforation after PEG tube misplacement.
More than one third of patients with MSA in this study died suddenly. Several mechanisms for sudden death have been proposed in MSA. The combination of passive glottic narrowing by selective paralysis of the vocal cord abductors and active narrowing by adductor activation during inspiration4 have been associated with stridor and acute airway obstruction. Furthermore, patients with MSA show minimal to no chemosensitivity to hypoxia (especially during sleep) possibly owing to the degeneration of brain stem respiratory centres. These may explain sudden death in patients with MSA even after tracheostomy.5
Dysphagia caused by delays in the oral and pharyngeal phases of swallowing, in combination with laryngeal (airway and sensory) and oesophageal sphincter disturbances may lead to both aspiration pneumonia and acute aspiration.6 PEG tube feeding prevented considerable weight loss and wasting, but not the recurrence of aspirations and aspiration pneumonia. Additional measures, such as improvement of oral/dental hygiene and proper patient postprandial and sleep positioning,7 may be considered to decrease mortality from aspiration. An additional finding of this study is the high prevalence of weight loss among patients with end‐stage MSA. Weight loss is a risk factor for mortality in chronically ill patients.8 Dietary adjustments, early swallowing studies and PEG tube feeding may reduce mortality in patients with MSA.
Neurogenic lower urinary tract dysfunction is considered a valuable diagnostic tool for MSA.9 Urinary urgency or incontinence (storage disorder) and incomplete emptying or urinary retention (voiding disorder) may occur simultaneously and lead to intractable LUTIs, which are major causes of morbidity in this disorder.10 More than half of our patients reported recurrent LUTIs and a large number died from complications related to LUTIs. Frequent urological monitoring and treatment of complications, in addition to the use of clean intermittent self‐urinary bladder catheterization, may reduce the risk of LUTI‐associated mortality.
In summary, all patients with MSA died from disease‐related events, with sudden death and infections being the most common. We propose that careful screening for laryngeal stridor, neurogenic bladder dysfunction and dysphagia with aggressive treatment may increase total survival time in patients with MSA. More studies on the patient are warranted. Research efforts should be directed towards the development of more efficient identification and prevention strategies for the major complications of MSA.
Acknowledgements
We thank the donors and their families for making these studies possible.
Footnotes
Funding: The Brain Endowment Bank is funded in part by the National Parkinson Foundation. This work was supported in part by a grant from the National Parkinson Foundation (Miami, Florida, USA).
Competing interests: None declared.
References
- 1.Watanabe H, Saito Y, Terao S.et al Progression and prognosis in multiple system atrophy: an analysis of 230 Japanese patients. Brain 2002125(Pt 5)1070–1083. [DOI] [PubMed] [Google Scholar]
- 2.Gilman S, Low P A, Quinn N.et al Consensus statement on the diagnosis of multiple system atrophy. J Neurol Sci 199916394–98. [DOI] [PubMed] [Google Scholar]
- 3.Hughes A J, Daniel S E, Kilford L.et al Accuracy of clinical diagnosis of idiopathic Parkinson's disease: a clinico‐pathological study of 100 cases. J Neurol Neurosurg Psychiatry 199255181–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Isozaki E, Osanai R, Horiguchi S.et al Laryngeal electromyography with separated surface electrodes in patients with multiple system atrophy presenting with vocal cord paralysis. J Neurol 1994241551–556. [DOI] [PubMed] [Google Scholar]
- 5.Silber M H, Levine S. Stridor and death in multiple system atrophy. Mov Disord 200015699–704. [DOI] [PubMed] [Google Scholar]
- 6.Higo R, Nito T, Tayama N. Swallowing function in patients with multiple‐system atrophy with a clinical predominance of cerebellar symptoms (MSA‐C). Eur Arch Otorhinolaryngol 2005262646–650. [DOI] [PubMed] [Google Scholar]
- 7.Matsui T, Yamaya M, Ohrui T.et al Sitting position to prevent aspiration in bed‐bound patients. Gerontology 200248194–195. [DOI] [PubMed] [Google Scholar]
- 8.Sauerwein H P, Romijn J A. More consideration to dietary protein in the nutrition of chronically ill adults with tendency to weight loss. Ned Tijdschr Geneeskd 1999143886–889. [PubMed] [Google Scholar]
- 9.Gilman S, Low P A, Quinn N.et al Consensus statement on the diagnosis of multiple system atrophy. J Auton Nerv Syst 199874189–192. [PubMed] [Google Scholar]
- 10.Ito T, Sakakibara R, Yasuda K.et al Incomplete emptying and urinary retention in multiple‐system atrophy: when does it occur and how do we manage it? Mov Disord 200621816–823. [DOI] [PubMed] [Google Scholar]