OUP user menu


A. Ammendola , D. Gemini , S. Iannaccone , F. Argenzio , G. Ciccone , E. Ammendola , L. Serio , G. Ugolini , F. Bravaccio
DOI: http://dx.doi.org/10.1093/alcalc/35.4.368 368-371 First published online: 1 July 2000


In some alcohol-related pathologies of chronic alcoholism women are more vulnerable than men. A consecutive sample of 62 chronic alcoholics was studied, 18 females and 44 males, aged between 28 and 69 years to assess the incidence and distribution of peripheral neuropathy with regard to gender. All patients underwent clinical and neurological observations, laboratory tests, and electroneurography. Total lifetime dose of ethanol (TLDE) and other risk factors for neuropathy (disease duration, age, nutritional status) were calculated and correlated to sural nerve sensory-evoked potential (SEP) amplitude. In 42 patients (67.7%), we observed the presence of clinical and/or infraclinical neuropathy, mostly axonal, in 29 males (65.9%) and 13 females (72.2%). In women, compared to men, TLDE and disease duration were significantly inversely correlated to sural nerve SEP amplitude, i.e. in women, SEP amplitude is significantly reduced in relation to TLDE and disease duration increase. These data indicate a higher sensitivity of females towards the toxic effects of ethanol, other than malnutrition, on peripheral nerve fibres.


There are many pathologies associated with chronic alcoholism, which involve the nervous system and other organs. These pathologies generally affect both genders. From the medical literature, it seems that women, when compared to men, are more predisposed to alcohol-induced damage. For example, a more evident female susceptibility to the toxic effects of alcohol has been observed in hepatic, cerebral, cardiac, and muscular alterations (Acker, 1985; Norton et al., 1987; Mann et al., 1992; Urbano-Marquez et al., 1995). We therefore carried out a retrospective, cross-sectional study on chronic alcoholics with the aim of evaluating on the one hand the prevalence and on the other the distribution between the two genders, of a pathology quite frequently observed in the course of chronic alcoholism, namely peripheral neuropathy. The data in the literature concerning the prevalence of alcoholic neuropathy do not always appear to be uniform (Bolton, 1987; Monforte et al., 1995); with regard to the distribution between the two genders. They generally show more or less the same incidence in male and female subjects (Pessione et al., 1995), or sometimes a greater prevalence in males (Mauri et al., 1994) or females (Victor, 1994).


Sixty-two subjects, all chronic alcoholics, admitted and enrolled consecutively over a period of 24 months, were studied. The reason for spontaneous admission for all subjects was to treat their dependence on alcohol. All patients (44 males and 18 females), aged between 28 and 69 years, reported a consumption of at least 100 g of alcohol/day over the 2 years before admission. Alcoholics with other diseases which could cause damage to the peripheral nervous system were excluded from the study. To determine the consumption of alcohol, a detailed anamnesis was taken, and then confirmed by interviews with family members and also by using life events of the subject to facilitate memory of these events, such as military service, marriage, change of job, etc. (‘time-line follow-back method’) (Sobell et al., 1979). Each patient underwent the following assessments. (1) General objective neurological examination: clinical peripheral neuropathy was considered if the patient had two or more of the following clinical abnormalities, muscle weakness, paresthesia, symmetrically depressed or absent tendon reflexes, sensory deficit. (2) Calculation of the total dose of ethanol consumed in their lifetime (TLDE): this was expressed in kg of ethanol/kg of body weight and was estimated by first multiplying the daily consumption of ethanol by the number of days of the periods of exposure to alcohol and then dividing the product by the body weight of the patient when first admitted. (3) Laboratory studies: the following parameters were examined in serum, blood and urine glucose, creatinine, cholesterol, triglycerides, electrolytes, serum aspartate and alanine aminotransferases, γ-glutamyl transpeptidase, protein electrophoresis, lactate dehydrogenase, creatine kinase, aldolase, red blood cell, white blood cell and platelet counts, haematocrit, total protein, prealbumin, albumin, iron, transferrin, prothrombin time, and ethanol levels. (4) Evaluation of nutritional status: this consisted of a comparison between the actual weight and the ideal weight; study of the lean body mass calculated from the circumference of the upper non-dominant arm and the thickness of the tricipital skin fold (expression of the total body fat), and a study of the nutritional proteins evaluating the total protein, albumin, prealbumin, total lymphocytes, transferrin, and of serum folate with the group B vitamins. Caloric or protein malnutrition was considered if body weight was less than 90% of the ideal weight or if the lean body mass was less than 90% of the normal value or at least three of the nutritional parameters examined were altered (Durnin and Womersley, 1974; Burrit and Anderson, 1984; Blackburn et al., 1977). (5) Electroneurographic valuation of the ulnar, median, peroneal, and sural nerves: this involved a study of the maximum motor conduction velocity of the median and peroneal nerves with their compound action potential amplitude, and a study of sensory orthodromic conduction velocity of the ulnar and sural nerves with their sensory-evoked potential (SEP) amplitude. The parameters evaluated were considered pathological when they differed from the average value by ± 2 SD, obtained from a control group of 40 subjects (20 males and 20 females) aged between 30 and 60 years, who did not consume alcoholic drinks. Subclinical peripheral neuropathy was considered if the patient had, in the absence of clinical symptoms, one or more of the above reported electrophysiological parameters altered in at least two of the nerves examined. (6) A comparison of sural nerve SEP amplitude was made between the alcoholic subjects, males and females, and the relative control subjects, to study the behaviour of SEP amplitude, a suitable parameter for the evaluation of a sensory axonal dysfunction.

Statistical analysis was performed with SPSS: software in particular, using Student's t-test the sural nerve SEP amplitude was compared among the alcoholics, males and females, and the relative control subjects, and through Pearson's correlation coefficient analysis the risk factors for alcohol-related neuropathy (TLDE, disease duration, age, nutritional status) were correlated with the sural nerve SEP amplitude in the alcoholic subjects and in both male and female alcoholic groups.


In Table 1, we report the main clinical data for the alcoholics as a single group and as a function of gender. The average duration of alcoholism was less for women than men (11.3 vs 14.8 years respectively). Similarly, the TLDE was lower in women compared to men (16.2 vs 23.6 kg of ethanol/kg body weight). The nutritional status was altered in only a few cases (5 males and 4 females) without a substantial gender difference overall. Peripheral neuropathy was present in 42 cases (67.7%), in 29/44 (65.9%) males and 13/18 (72.2%) females; the difference was not statistically significant (P = 0.63). In the nine cases with altered nutritional status, all patients had caloric malnutrition, seven with protein malnutrition and two with thiamine deficiency.

View this table:
Table 1.

Main clinical data for alcoholics

A cross-tabulation of clinical and electroneurographic signs in the 62 alcoholic patients and the percentages of the 42 subjects with clinical or subclinical neuropathy, with regard to gender, are shown in Tables 2 and 3. The clinical signs are essentially represented by hyposthenia, depressed or absent deep tendon reflexes, and sensory deficit; they were present mostly in the lower limbs.

View this table:
Table 2.

Cross-tabulation of clinical and electroneurographic signs of peripheral neuropathy in the 62 alcoholic patients studied

View this table:
Table 3.

Alcoholic patients with clinical and/or subclinical neuropathy as a function of gender

The electroneurographic alterations (Fig. 1) consisted fundamentally of a proportionately decreased sensory and/or motor conduction velocity, a reduced or absent SEP amplitude, and a low composed muscle action potential amplitude. The most frequent alterations were sensory (Fig. 1) and they predominated in the lower limbs.

Fig. 1.
Fig. 1.

Frequency and type of electroneurographic alterations in 42 cases with peripheral neuropathy.

MCV, motor conduction velocity; MEP, motor-evoked potential; SCV, sensory conduction velocity; SEP, sensory-evoked potential.

The frequency of the neuropathy rose in proportion to the TLDE increase. Thus, by subdividing the male and female alcoholics into three groups, in accordance with levels of TLDE, it was clear (Fig. 2) that the neuropathic patients of both genders showed a TLDE-dependent correlation with neuropathy.

Fig. 2.
Fig. 2.

Correlation between total life dose of ethanol and presence or absence of peripheral neuropathy in the male and female alcoholic groups.

A comparison of the sural nerve SEP amplitude between the alcoholic and the control subjects (Table 4) showed that the SEP average amplitude in the alcoholics was inferior to that in controls, for both males and females, and this difference was statistically significant (P < 0.01) for both genders (Table 4).

View this table:
Table 4.

Comparison and significance of sural nerve sensory-evoked potential (SEP) mean amplitude (μV) between male (n = 20) and female (n = 20) controls and male (n = 44) and female (n = 18) alcoholics

In relation to Pearson's correlation coefficient among sural nerve SEP amplitude and the risk factors considered (TLDE, disease duration, age, and nutritional status), a significant inverse correlation of the SEP amplitude was observed with disease duration and age in the total group and with disease duration and TLDE only in the female group (Table 5).

View this table:
Table 5.

Correlation of sural nerve sensory-evoked potential (SEP) amplitude and risk factors for peripheral neuropathy in all alcoholic patients and as a function of gender


In this retrospective study of 62 chronic alcoholic subjects, the patients seemed to be more frequently affected by peripheral neuropathy, especially when the clinical observation was followed by electrophysiological investigation. In two-thirds of the cases studied (67.7%), we observed clinical and/or instrumental signs of peripheral neuropathy. The most frequent electroneurographic alteration was the sural nerve SEP reduced amplitude or its total absence, confirming the fact that the alcoholic neuropathy is mostly axonal with a major involvement of sensory fibres and is more frequent in the lower limbs (Kimura, 1989).

In our study, the TLDE appears to be an important determinant of noxious action on nerve fibres. In fact, there were only a few cases with altered nutritional status and particularly with thiamine deficiency and the neuropathy was more frequent in alcoholic patients with a greater TLDE. Estruch et al. (1993) suggested that the TLDE is an important risk factor for the development of neuropathy and Palliyath and Schwartz (1993) have demonstrated the role played by abstinence from alcohol in improvement of peripheral nerve function in alcoholic subjects without vitamin supplements.

As regards the distribution of the neuropathy as a function of gender, from the medical literature no data of substantial gender differences emerged, as the disease affects both sexes with possibly males being more affected. Our study, however, shows that women are more affected than men, although the difference was small (72.2% women vs 65.9% of men) and not statistically significant. Also, from our study, it appears that the mean TLDE in women is less than that in men. Furthermore, Pearson's correlations among the sural nerve SEP amplitude and risk factors such as as TLDE, disease duration, age, and nutritional status, indicated an inverse correlation of sural nerve SEP amplitude with TLDE and disease duration for patients of both sexes, but with significant levels only for women. These data could indicate that, in relation to TLDE and disease duration, the SEP amplitude reduction is superior in women. In our opinion, this observation is important if we consider that the SEP amplitude is the parameter which was most frequently altered in our cases, independent of the presence or absence of evident neurological damage. In fact, in our chronic alcoholic subjects, of both genders, the sural nerve SEP average amplitude was significantly inferior to that of control subjects.

Similarly to other organs, such as the liver, brain, heart and muscles, the peripheral nerve fibres seem also to be more susceptible to toxic damage in alcoholic women, when compared to alcoholic men. Possible factors which can be applied to determine the greater frequency of the alcohol-correlated pathologies in female alcoholics have yet to be defined. Certain conditions have been evaluated, such as differences in body composition (higher percentage of fat in women), a difference of the blood levels of alcohol (with the same dose of ethanol these levels are higher in women than in men), or the higher blood levels of acetaldehyde observed in women during ethanol intoxication (Marshall et al., 1983; Arthur et al., 1984; Frezza et al., 1990).

In conclusion, clinical and/or subclinical peripheral neuro-pathy seems to be a frequent pathology in alcoholic subjects of both sexes. Our study suggests a slightly greater, if not statistically significant, prevalence in women, despite our observation of a lower mean TLDE in alcoholic women than in alcoholic men. Also, only in women was the inverse correlation between TLDE and disease duration and sural nerve SEP amplitude significant. These observations suggest a greater sensitivity of females to the toxic effects of alcohol on peripheral nerve fibres unrelated to malnutrition.


  • * Author to whom correspondence should be addressed at: Cattedra e Servizio di Neurofisiopatologia, Policlinico, Piazza L. Miraglia, 2-80138 Napoli, Italia.


View Abstract