Alcohol-induced autonomic dysfunction: a systematic review

For the most part, the studies identified utilised non-invasive cardiovascular reflex tests in order to measure autonomic dysfunction. In general, the investigations undertaken were as described by Ewing et al. including the heart rate (HR) response to the Valsalva test, standing and deep breathing (DBT), as well as blood pressure (BP) response to standing and sustained handgrip [7]. A minority of studies measured the HR response to atropine administration, HR response to baroreceptor stimulation by external suction, and 24-h heart rate variability (HRV) [5, 8‐11]. Additional non-cardiac measures of autonomic function identified in this review include sympathetic skin response (SSR) and the methacholine test for iridic innervation [12‐17]. The tests identified in this review are summarised in Table 1. These investigations can be dichotomised to represent measures of sympathetic and parasympathetic function. Parasympathetic tests include HR response to DBT, Valsalva test and standing. Sympathetic tests include sweat spot test, SSR, BP response to standing and BP response to sustained hand grip. The studies providing prevalence, symptoms and the split of parasympathetic and sympathetic dysfunction are provided in Table 2. A small number of tests assessed autonomic function using laser Doppler flowmetry, which demonstrated sympathetic dysfunction relative to controls [18‐20].

Amongst the identified studies, there was a significant deal of heterogeneity with respect to the total consumption of alcohol within the cohort in question, definitions of excessive alcohol intake and alcohol dependency, autonomic batteries and definitions of neuropathy utilised. Consequently, the overall prevalence of autonomic neuropathy amongst those who chronically abuse alcohol or are dependent upon alcohol cannot be produced by meta-analysis. Nonetheless, there is a significant body of research available, which supports a high rate of autonomic dysfunction amongst alcohol abusers with cardiovascular reflex measures of neuropathy producing rates ranging between 16% and 73% [12, 21]. This range is clearly quite wide and explained by the heterogeneity between studies, and it is worth noting that the most methodologically sound study with amongst the largest studied populations established that 26% of the study population (patients satisfying DSM IV criteria for alcohol dependence) have abnormal cardiovascular reflexes, defined as two or more abnormal parameters [22]. Interestingly, with respect to the cardiovascular system, parasympathetic impairment occurs with greater frequency than sympathetic, which is less common and in most studies did not occur in isolation—this is illustrated in Table 2 [8, 10, 22‐26]. Currently, there are no relevant longitudinal studies with prolonged follow-up and therefore it cannot be ascertained whether this might represent a natural course in which most individuals begin with parasympathetic dysfunction and later develop sympathetic involvement as their condition progresses. Despite the significant frequency with which cardiovascular reflexes are deranged in alcohol abusers, this does not appear to translate into frequent cardiovascular features with most studies reporting patients to be asymptomatic as detailed in Table 2.

Four studies utilised Holter monitors to measure 24-h heart rate variability (HRV) to detect depressed HRV related to autonomic dysfunction. Malpas et al. found that there were subtle changes in HRV in individuals who had normal cardiovascular reflex tests, and thus proposed that HRV may indicate early ANS derangement amongst alcohol abusers [24]. Malpas et al. found a high rate of 24-h HRV abnormality, affecting 74% of subjects, 30% of which had cardiovascular reflex test abnormality. Similarly, two papers by Weise et al. used mean momentary arrhythmia to measure HRV with a lower prevalence of 16% and 36% [21, 27]. One small study by Melgaard and Somnier failed to show a significant difference in beat-to-beat variation in alcohol abusers relative to controls (n = 14) [28].

Presently, the extensive literature described above in relation to cardiovascular reflex tests represents the most compelling evidence that those who consume large quantities of alcohol over a prolonged period of time are vulnerable to autonomic dysfunction.

Sympathetic skin response (SSR) in chronic alcohol abusers was explored by three studies, two of which demonstrated deranged sudomotor function in alcohol abusers with absent palm SSR in 6–13%, fingertip in 31% and foot SSR in 31–53% [12, 13]. Additionally, the results support that the dysfunction occurs in the most distal segments, an observation that has been shown in patients with alcohol-related large fibre peripheral neuropathy [4]. Navarro et al. investigated the relationship between SSR and the number of pilocarpine-reactive sweat glands, finding that sympathetic involvement is associated with a decrease in their density [12]. However, conflicting these results, a study conducted by Tugnoli et al. failed to find significant differences in SSR between chronic alcohol abusers and healthy controls [14]. As the studies utilised different definitions of excessive alcohol intake, it is possible that the contrasting findings are the result of dissimilar populations. It should also be noted that of the three studies, the population size studied by Tugnoli et al. was the smallest, with 20 subjects compared with a pooled total of 100 individuals in the studies demonstrating differences between chronic alcohol abusers and healthy controls. Therefore, the evidence currently supports that deranged SSR is a feature of alcohol-related autonomic neuropathy. Clinically, however, the symptom of abnormal sweating is not one which is often highlighted as a significant feature in the literature which may indicate that this derangement is generally either subclinical or is not troubling to patients. This is illustrated in Table 2, in which most studies do not identify patients with this symptom with the exception of the article by Myers et al. in which 15% of patients had decreased sweating in the feet [17].

In males, erectile dysfunction appears to be a common feature affecting between 2–29% of chronic alcohol abusers [21, 25, 29, 30]. The relationship between alcohol abuse and erectile dysfunction was explored in greater detail by Wetterling et al. who found that there was increasing frequency of this symptom with increased alcohol consumption [31]. The wide range of the reported prevalence of erectile dysfunction in alcohol abusers was in part due to the various lengths of time between alcohol withdrawal and the commencement of the study [30]. Whilst the frequency of erectile dysfunction in this patient cohort varies significantly between studies, it is consistently the most common symptom reported.

Previously, deranged gastrointestinal (GI) physiology and a number of related clinical features which often occur in chronic alcohol abusers have been attributed to autonomic neuropathy by some authors. Though there is little literature available on the topic, what is presently available does not seem to convincingly support the role of autonomic dysfunction in such derangements. The difficulty in assessing these features arises because it is challenging to disentangle the effects alcohol exerts on gut physiology, as well as direct toxic effects to the gastrointestinal organ system and elements of derangement which are attributed to autonomic dysfunction. Alcoholism has been linked to delayed oro-cecal transit time, and gastric emptying [32‐34]. Further, oesophageal dysmotility and lower oesophageal pressure changes have been identified in this population, although these findings are not consistent [35‐37].

With respect to symptomatology, heavy alcohol consumption has been correlated with dyspepsia, nausea, abdominal pain and diarrhoea [33, 34]. Diarrhoea has been reported as a feature in alcohol abusers, but is not specific to the ANS [25, 29].

It is also worth noting that in studies which identified those with autonomic neuropathy using cardiovascular reflex tests, a relationship between ANS dysfunction and gastrointestinal disorder is inconsistent and consequently it currently remains unclear whether changes to the GI system in the context of alcohol abuse are related to direct toxicity to smooth muscles, nervous dysfunction or some other systemic effect [33, 34, 37].

Total lifetime dose of ethanol (TLDE) is often measured in studies to establish the extent of alcohol abuse. The TLDE is calculated by taking the daily alcohol consumption multiplied by 365 and by the number of years. This measure has been explored as a risk factor for the development of autonomic neuropathy.

Of the six studies to consider TLDE a risk factor, four reported that long-term alcohol abusers with evidence of cardiovascular autonomic dysfunction had a significantly higher TLDE than those without [2, 25, 29, 38]. Two of these only identified a correlation with results from deep breathing tests [25, 29], and one found a correlation only with heart rate and blood pressure responses to standing [2]. The final paper did not identify a correlation with any specific investigation of cardiovascular autonomic dysfunctions [38]. Only one paper by Ferdinandis and De Silva was unable to find any correlation between the TLDE and cardiovascular autonomic dysfunction [39]. However, the TLDE is subject to a degree of recall bias and could be argued is particularly prone to under-reporting given the negative connotation regarding excessive alcohol consumption. Di Ciaula et al. considered gastrointestinal motility as a parameter for autonomic dysfunction and found a correlation between TLDE and gallbladder half-emptying time. They reported a lack of correlation between other measured parameters including gastric motility, dyspeptic scores and various gastrointestinal transit times [34].

Four studies specifically correlated the duration of alcohol abuse with presence of autonomic dysfunction through abnormal findings in cardiovascular parameters; however, the literature is conflicting [21, 22, 29, 38]. A study by Malpas et al. considered groups of chronic alcohol abusers with and without vagal neuropathies, finding no differences between these groups with respect to duration of alcohol abuse or age [24]. Furthermore, two studies showed no correlation between either cardiovascular autonomic dysfunction or abnormal sympathetic skin response and duration of alcohol abuse, although these studies did not appear to adjust for age [40, 41]. Conversely, Agelink et al. showed duration of alcoholism to be one of the most important factors contributing to the development of particularly sympathetic dysfunction, also demonstrating this independent of age [38].

Three papers considered the effect of daily alcohol intake [29, 32, 33]. Papa et al. considered the effect of moderate and heavy alcohol consumption on intestinal transit times, finding that only heavy intake resulted in significantly increased transit times, which might indicate a visceral autonomic neuropathy. However, the authors highlighted the inability to draw a conclusion from this study as the intestinal tract is likely to also experience direct toxic effects from alcohol [32]. Showing similar results, a study considering gastrointestinal transit in chronic alcohol abusers found there was a significant correlation between mean daily consumption and gastric emptying times [33]. A single study evaluated the effects of graduating levels of alcohol consumption on cardiovascular reflex test parameters, and identified no relationship between these variables [29].

There is debate within the literature with respect to the role of hepatic dysfunction in alcohol-related autonomic dysfunction.

Compared to healthy controls, subjects with alcoholic cirrhosis have an increased incidence of autonomic dysfunction [42, 43]. In the largest study (n = 107) which investigated the role of alcohol-related diseases in autonomic dysfunction, no relationship was identified between hepatic disease and autonomic dysfunction [29]. This finding is supported by a further seven smaller studies and provides strong evidence that hepatic disease is not a prerequisite for autonomic dysfunction in the context of chronic alcohol abuse [1, 22, 38, 40, 44‐46]. This is further illustrated by Sebastian and Puranik who performed a study in which they compared alcohol abusers with normal liver function to healthy controls and identified significant sympathetic derangement and a higher resting heart rate but normal parasympathetic tests in those who heavily consumed alcohol, therefore further supporting the hypothesis that hepatic dysfunction is not essential to the disease process [47]. A single study of alcohol abusers with autonomic neuropathy found that these individuals are more likely to have alcoholic liver disease [26].

Five studies assessed cohorts of individuals with hepatic dysfunction of various aetiologies. In these studies, it is clear that those with liver disease have an increased prevalence of autonomic dysfunction, but that this is even higher in the context of alcoholic cause of liver disease [48‐52]. This data implicates ethanol or some other confounding factor such as other health behaviours of those who abuse alcohol or a further substance contained in alcoholic beverages in the pathogenesis of autonomic dysfunction, in addition to any role which hepatic disease plays.

Weighing the available evidence, the most likely conclusion at present is that autonomic dysfunction can occur in the absence of hepatic disease and therefore the latter is not central to the disease process. However, it is feasible that hepatic disease may play a role in the autonomic dysfunction of some alcohol-abusing individuals as a co-occurring pathological factor.

Four studies commented on the role of nutrition, all showing no correlation with alcohol-related autonomic dysfunction [22, 29, 38, 53]. Three of these studies assessed nutritional status through protein and caloric intake profiles, as well as body weight [22, 29, 38]. Two of these studies excluded malnourished patients, and noted that this therefore establishes that malnutrition is not required for the presence of autonomic dysfunction [22, 38]. Monforte et al. included patients with nutritional deficiency but did not demonstrate a relationship between nutritional status and autonomic neuropathy [29]. Similarly, Koike et al. considered the role of B₁₂ deficiency in other neurological disorders such as beriberi postulating that this vitamin may also impact upon the ANS in alcohol abusers, and therefore measured blood B₁₂ levels to exclude B₁₂ deficiency in the study population. In this study, autonomic dysfunction therefore occurred in the absence of B₁₂ deficiency, excluding this as a central requirement to the pathogenesis in patients in this study [53].

At present, there is inadequate exploration of serum vitamin levels in cohorts evaluated for alcohol-related autonomic neuropathy and this prevents satisfactory assessment of this as a risk factor, or pathogenic factor for neuropathy.

Only one study noted self-reported smoking habits in the context of autonomic dysfunction and found no correlation [29].

Four studies, all assessing cardiovascular autonomic outcomes, commented on age as a risk factor [22, 24, 29, 38]. There were no significant differences between the mean ages of patients chronically abusing alcohol who did and did not demonstrate autonomic dysfunction in these studies. Three studies found no relationship; however, one study conducted by Ravaglia et al. identified a significant negative correlation with increasing age and heart rate response to deep breathing, but did not appear to have adjusted for duration of alcohol abuse or TLDE [22]. This study also considered other parameters of cardiovascular, gastrointestinal autonomic dysfunction and impotence.

Only two studies, both assessing only cardiovascular autonomic dysfunction, commented on sex as a risk factor. Neither found any relationship between sex and alcohol-related autonomic dysfunction [38, 40].

Interestingly, one paper commented on the increased incidence of autonomic dysfunction in patients with encephalopathy. However, as a result of the small population sizes and the conflicting results, this correlation remains uncertain [1]. Speculatively, one may consider that this risk factor could reflect a proxy for very high levels of alcohol consumption, hepatic impairment or other particularly sick state.

Only one study commented on the relationship between physical fitness and autonomic dysfunction in alcohol abusers and found an inverse relationship [54].

Ethnicity was not explored as a risk factor for alcohol-related autonomic neuropathy in the available literature. However, it is feasible that this may be a risk factor because of the differences between people of different ethnicities with regards to the metabolism of ethanol. Alcohol metabolism is determined primarily by the enzymes ADH, ALDH and CYP2E1 [55]. These enzymes have numerous isoforms, which are associated with different ethnicities. ADH and ALDH alleles relate to enzymes which catalyse the formation of toxic metabolite acetaldehyde from ethanol thus potentially affecting the risk of alcoholism and alcohol-related tissue damage [55, 56]. Similarly, the CYP2E1 isozyme has ethnic variance and influences the metabolism of alcohol and blood alcohol levels. However, the influence of various isozymes related to alcohol metabolism over one’s tendency to alcoholism or vulnerability to toxicity is an area of research currently considered to have results which are inconclusive [57]. Therefore, ethnicity as a risk factor may represent an interesting area for future research.