Autonomic dysfunction is associated with disease progression and survival in amyotrophic lateral sclerosis: a prospective longitudinal cohort study

Although amyotrophic lateral sclerosis (ALS) is traditionally viewed as a neurodegenerative disorder affecting the motor neurons, increasing evidence suggests that ALS should be considered as a multisystem disease with 80% of patients reporting at least one non-motor symptom [1]. Psychiatric, cognitive [2], and behavioral symptoms [3] are most common and associated with shorter survival time and reduced quality of life. Among non-motor symptoms, autonomic dysfunction has been also described in ALS and reported in up to 75% of patients, mild in most patients (85%), and moderate in a minority (15%) [4]. The most experienced symptoms include urinary urgency and frequency in 40% of patients [5, 6] and gastrointestinal dysfunction, most typically constipation, in 50% [7, 8]. Although less disabling to patients, cardiovagal abnormalities, including reduced heart rate variability, are also present in 50% of patients [4, 9], though more serious cardiovascular autonomic failure appears rare and only in more advanced disease stages [10]. Since the autonomic nervous system is responsible for the control of vital functions, its impairment has been consistently associated with shorter survival in several neurological diseases, such as multiple system atrophy [11], Parkinson’s disease [12], dementia with Lewy Body [13] and progressive supranuclear palsy [14]. However, autonomic symptoms have not been systematically assessed as prognostic factors in ALS. Therefore, the main objectives of this study were: first, to investigate in a longitudinal study whether the presence of autonomic symptoms at diagnosis can predict disease progression and survival; and second, to instrumentally assess autonomic dysfunction over time using a simple parasympathetic marker, such as the heart rate variability (HRV) [15].

A total of 102 consecutive patients with ALS who fulfilled the entry criteria (63.7% were male, and 83.3% had spinal onset) were included, and their demographic and clinical characteristics are summarized in Table 1. The mean disease severity (ALSFRS-R) was 37 (IQR 34–41). The median follow-up time among censored observations was 12 months (IQR 5–25) and 36.2% of patients had died by January 2022, with a median follow-up time among deceased patients of 13 months (IQR 5–17). The overall median estimated survival from onset was 21 months (IQR 16–35). The King’s 4 stage, used as the main disease milestone, was reached by seventy-one patients (69.6%) with a median time from disease onset of 17 months (IQR 12–28.5). Lastly, ALS patients compared to the HC group did not differ for age (p = 0.11) and sex (p = 0.06).

This is the first prospective and longitudinal study evaluating the autonomic burden at diagnosis and over time in a well-selected and characterized cohort of newly diagnosed ALS patients.

The main finding was that autonomic symptoms were present in most ALS patients at diagnosis, with a higher load in bulbar phenotype. The severity of autonomic symptoms progressed over time, regardless of the clinical phenotype at onset, and was an independent predictor of a more rapid progression of disease disability. Among the autonomic domains, urinary complaints were associated with shorter survival. Parasympathetic dysfunction, indexed by a lower HRV, was present at diagnosis independent of respiratory impairment and of disease onset phenotype and progressed over time.

Autonomic impairment in ALS, analogously to motor neuron dysfunction, is likely resulting from a combination of degeneration within central and peripheral structures. Specifically, central degeneration is supported by pathological studies showing damage in the thoracic [29, 30] and sacral [31] intermediolateral nucleus of the spinal cord, which convey sympathetic and parasympathetic preganglionic neurons, respectively. Peripheral autonomic damage is also plausible since a small fiber pathology, involving somatic and autonomic cutaneous nerves has been described in ALS in the last decade [8, 32-34]. Interestingly, autonomic denervation was associated with vascular remodeling and disease progression rate [8].

Previous literature on autonomic symptoms in ALS was mostly based on retrospective studies, while longitudinal studies in newly diagnosis patients are lacking.

In our cohort, about two thirds of patients at diagnosis complained of autonomic disturbances that were more severe in bulbar onset patients. In this subgroup of patients, the higher autonomic load may also be linked to the spatial contiguity of crucial autonomic nuclei involved in gastrointestinal, urinary, and cardiorespiratory function in the brainstem [35]. Post-mortem studies in SOD 1 patients with autonomic failure have reported neuronal degeneration of autonomic structures in the brainstem [36], and therefore similar studies should be conducted in sporadic ALS patients with bulbar onset.

Regardless of the disease onset phenotype, the most experienced symptoms were gastrointestinal disturbances, mainly constipation or early abdominal fullness, as well as urinary symptoms such as urgency/incontinence or retention observed in more than 40% of patients. Significant gastrointestinal symptoms, such as constipation, abdominal pain, nausea and feeling of fullness, have been previously reported with constipation being the most frequent symptom [4, 7, 8, 37]. Interestingly, two studies of patients with ALS identified a significant proportion with delayed gastric emptying [38] and prolonged colonic transit times [39].

Analogously, urinary symptoms have been reported in the ALS population [4, 8], with urinary urgency described in percentages ranging from 20% [4] to 76% [7] of the cases and retention in 41% of patients [5].

Notably, urodynamic studies typically revealed a neurogenic bladder resulting from detrusor sphincter dyssynergia, and increased post-void residual bladder volumes in patients with greater lower limb spasticity, implicating suprasacral pathology [5, 6]. However, the association between clinical disability or disease phenotype and urinary symptoms was not confirmed in a subsequent study [40], suggesting that increased spasticity and or reduced mobility may not be a major explanation for urinary impairment. Furthermore, previous studies failed to demonstrate an impairment of motor neurons that supply the pelvic floor muscles [41].

The association of autonomic dysfunction with disease progression in ALS patients is in line with previous studies in other neurodegenerative diseases [11‐14]. We also observed that ALS patients who complained of urinary disturbances at diagnosis had a poor survival suggesting that autonomic impairment has a role in blunting adaptive responses reducing life expectancy of ALS patients.

This result agrees with the findings of a recent cross-sectional study that showed that early and severe neurogenic bladder, demonstrated by urodynamic study in motor neuron disease patients was associated with a reduced survival [40], regardless of their phenotype, suggesting a more rapid spread of the disease to extra-motor networks in this patients’ subgroup.

Autonomic dysfunction in our patients was also confirmed by the significant reduction of the SDRR, an index of decreased cardiovagal tone, together with an increase of heart rate indicating sympathetic overactivity. These results are in keeping with previous studies [4, 9, 42‐44] that suggest an imbalance between enhanced sympathetic and reduced parasympathetic drives [44] in the autonomic cardiovascular state of ALS patients. Nevertheless, our results showed that the cardiac sympathetic hyperactivity was driven by pulmonary capacity, whereas the impairment of the parasympathetic tone was independent of respiratory dysfunction as well as of the disease onset phenotype. Moreover, the longitudinal study of autonomic cardiac assessment revealed that only parasympathetic hypofunction progressed over time, whereas sympathetic overactivity tended to remain stable and therefore it appeared unrelated to the reduction of parasympathetic drive. Importantly, the impairment of parasympathetic tone did not correlate with autonomic symptoms, demonstrating that it did not reflect the overall autonomic burden. So, it was not surprising that there was no correlation between the HRV and the aggressiveness of the disease. Furthermore, cardiovascular symptoms were not among the most frequently reported by our patients.

In this study, we could not assess any potential effect of riluzole on the cardiac autonomic impairment, since most of patients were on treatment; however, there is no evidence that riluzole affects the heart rate [45].

In conclusion, our study shows that the presence of autonomic symptoms at diagnosis is an independent factor of disease progression and highlights the concept that early involvement of non-motor networks can significantly predict the clinical picture and prognosis, as it is associated with more rapid rates of motor functional decline and shorter survival. This may imply that early autonomic impairment is the expression of a more aggressive multisystemic degenerative process; however, we cannot exclude that autonomic dysfunction by itself may further affect the disease course.

Lastly, our findings have important clinical implications: clinicians should identify autonomic symptoms as part of the routine clinical evaluation and adopt therapeutic approaches to counteract the effect of autonomic dysfunction on disease, thus improving survival and quality of life of patients with ALS.

Further studies of ALS populations including larger numbers of patients with bulbar phenotype are needed. Moreover, the functional assessment of each autonomic domain, especially the most involved ones such as urinary and gastrointestinal, may help to tailor treatment strategies and objectively monitor autonomic dysfunction.