CHANGES IN PLASMA NORADRENALINE AND SEROTONIN LEVELS AND CRAVING DURING ALCOHOL WITHDRAWAL

Ashwin A. Patkar*, Raman Gopalakrishnan, Prakash C. Naik1, Heather W. Murray, Michael J. Vergare and Charles A. Marsden2

Department of Psychiatry, Thomas Jefferson University, Philadelphia, PA 19107, USA,
1 Drug and Alcohol Unit, Lyndon Clinic, Solihull, West Midlands B92 8PW and
2 Department of Physiology and Pharmacology, University of Nottingham Medical School, Nottingham NG7 2UH, UK

Received 11 June 2002; in revised form 4 November 2002; accepted 2 December 2002


    ABSTRACT
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Aims: Despite substantial preclinical evidence that supports the involvement of noradrenergic (NA) and serotonergic (5-HT) mechanisms in alcohol withdrawal, human data remain inconsistent. We examined whether plasma levels of NA and 5-HT were altered during alcohol withdrawal and whether these measures were related to craving. We also explored whether alterations in NA and 5-HT activity differ between type I and type II alcohol-dependent patients during withdrawal. Methods: Plasma measurements of NA and 5-HT and assessments of craving were performed longitudinally in 26 Caucasian alcohol-dependent men who were hospitalized for detoxification, at baseline (day 0), and on the 1st, 7th and 14th days of withdrawal. These measures were compared with NA and 5-HT levels obtained in 28 controls. Results: During withdrawal, NA levels declined significantly from day 1 through day 14, whereas 5-HT levels and craving declined significantly from day 0 through day 14. The NA levels on days 0 and 1 of withdrawal were significantly higher than those in controls; however, by day 7 the NA levels were similar to the control values. In contrast, the 5-HT levels on day 0 of withdrawal resembled control values; however, the 5-HT concentrations on days 1, 7 and 14 were significantly lower than those in controls. There were no significant correlations between NA and 5-HT levels or between craving and the biological measures during withdrawal. Type I and type II patients did not differ in NA or 5-HT levels during withdrawal. Conclusions: These findings indicate that both plasma NA and 5-HT levels change during withdrawal; however, the pattern of change is different for the two measures. Also, while alterations in NA activity appear to normalize by late withdrawal, 5-HT changes seem to be more persistent. Neither craving nor subtypes of alcoholism seem to be related to alterations in NA or 5-HT during withdrawal.


    INTRODUCTION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Several lines of evidence indicate that disturbances in the central noradrenergic (NA) and serotonergic (5-HT) systems may mediate excessive alcohol consumption. For example, animal experiments have shown that acute alcohol exposure increases NA turnover and release, and downregulates central {alpha}2 adrenergic receptors (Wang et al., 1999Go; Thiele et al., 2000Go). Similarly, acute administration of alcohol is known to stimulate 5-HT turnover, while chronic alcohol intake is reported to decrease 5-HT synthesis and release (Tollefson, 1989Go). Reduced central 5-HT has been implicated in mediating alcohol preference in animals (LeMarquand et al., 1994Go).

Compared with the preclinical evidence supporting the effects of alcohol on NA and 5-HT function, human data have been less consistent. A possible reason may be that alcohol’s effects on NA and 5-HT seem to vary with dose and length of exposure or abstinence. Not surprisingly, plasma NA levels have been found to be increased (Smith et al., 1990Go) as well as normal (Bylund et al., 1984Go) in alcoholic patients. Similarly, studies of 5-HT function have found increased (Patkar et al., 1995Go), reduced (Arranz et al., 1999Go) as well as unchanged (Gentsch et al., 1985Go) platelet serotonin uptake sites in alcoholic patients. Biological studies in alcoholic patients are further complicated by the heterogeneity of the clinical population. Studies of problem drinkers and their families have yielded two broad categories termed type I and type II by Cloninger (1987)Go. The type II alcoholism is male-limited, and represented by an earlier age of onset of alcohol problems, stronger familial influences and higher impulsivity and violent behaviour. In contrast, the type I subgroup is more benign with a later onset, higher anxiety levels, fewer complications and a better prognosis. The two subtypes have been found to differ in measures of serotonin and monoamine oxidase activity and in response to serotonergic agents, such as the selective serotonin reuptake inhibitors (Von Knorring et al., 1987Go; Sullivan et al., 1990Go; George et al., 1997Go; Naranjo and Knoke, 2001Go).

The alcohol withdrawal syndrome has been long considered to be a manifestation of neuroadaptive responses to chronic alcohol use and a central feature of alcohol dependence. Since the withdrawal period overlaps with early abstinence, the associated neurobiological changes have attracted interest, due to the possibility that they may contribute to relapse. Alcohol withdrawal has been linked to increased metabolism and release of NA (Fujimoto et al., 1983Go; Hawley et al., 1985Go), reduced {alpha}2 adrenoceptor function (Nutt et al., 1988Go; Berggren et al., 2000Go), reduced 5-HT function (Ballenger et al., 1979Go), and alterations in neuroendocrine responsivity to challenge with NA and 5-HT agents (Krystal et al., 1996Go). Most of the data have come from determinations at one or two time points and very few studies have attempted to longitudinally study the changes in NA and 5-HT during the withdrawal period. Another clinical symptom, craving, has been increasingly recognized to contribute to continued drinking and is believed to play an important role in relapse (Petrakis et al., 1999Go). There is evidence that selective serotonin reuptake inhibitors (SSRI) may reduce craving for alcohol (Naranjo and Knoke, 2001Go), though contrary findings have also been reported (Petrakis et al., 2001Go). Since craving is a prominent symptom in alcohol withdrawal it seems reasonable to explore whether it may be related to neurobiolgical changes.

A major methodological issue in studying peripheral levels of NA and 5-HT and their relationship to central mechanisms, such as craving, is that it is unclear whether plasma NA and 5-HT levels reflect corresponding changes in human brain. Certain metabolic differences are present between periphery and the brain in humans that may need to be considered during interpretation of data [e.g. high-affinity alcohol dehydrogenase enzyme is absent in the brain (Linnoila et al., 1986Go)]. However, human studies have found correlations between brain and plasma levels of 2-methoxy-4-hydroxy-phenylglycol (MHPG) and NA, and plasma NA and its metabolites are believed to be a major determinant of levels of NA in the cerebrospinal fluid (Kopin et al., 1983Go). Also, similarities have been observed in the uptake, storage and release of 5-HT in neurons and blood platelets (Da Prada et al., 1988Go). Since repeated central measurements of neurotransmitters in humans are expensive, invasive and technically difficult, we examined plasma NA and 5-HT levels in our study.

Extending our previous work investigating 5-HT function in alcohol dependence (Patkar et al., 1995Go), we examined the role of NA and 5-HT systems in alcohol withdrawal. The objectives were to examine longitudinal differences in plasma NA and 5-HT levels during alcohol withdrawal, to compare these measures between withdrawing alcoholics and controls and to determine whether craving is associated with NA and 5-HT levels during withdrawal. We also explored whether the type I and type II subgroups of alcoholic patients differ from each other and from controls in measures of NA, 5-HT and craving.


    SUBJECTS AND METHODS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Subjects
The sample consisted of 26 alcohol-dependent Caucasian men, who were hospitalized for detoxification from alcohol, and 28 race- and gender-matched healthy volunteers, who served as the control population. The alcohol-dependent subjects were recruited from a university-affiliated in-patient addiction unit in Nottingham, England. The inclusion criteria were: (1) DSM-IV diagnosis of alcohol dependence (American Psychiatric Association, 1994Go); (2) ability to give informed consent; (3) no interruption of alcohol intake in the previous 72 h as determined by self-reports of alcohol consumption. This permitted us to obtain a baseline blood sample at the beginning of the withdrawal process. Individuals with major depression, bipolar disorders or psychotic disorders, serious medical disorders, current illicit drug misuse, a positive urine drug screen (except benzodiazepines), and those receiving psychotropic medication (except benzodiazepines) were excluded from the study. Smoking status was determined by self-reports (Do you consider yourself as a current smoker, non-smoker or former smoker?) and coded as a dichotomous variable (current smoker or non-smoker/former smoker). Alcohol-dependent subjects were typically recruited within 24 h of admission to the in-patient unit. The detoxification regime consisted of tapering doses of benzodiazepines over a 15-day period along with multivitamins. A fixed dose schedule was used; the starting dose was chlordiazepoxide 20 mg every 6 h, which was reduced by 20% every 3 days. Additional doses were used as clinically needed. The average daily dose of chlordiazepoxide was 44 mg.

Controls were hospital employees or their relatives. They were excluded if they had a positive response to items on the CAGE questionnaire (Ewing, 1984Go), a positive breathalyser reading [reported greater than 21 units of alcohol consumption per week (one unit is equivalent to 10 ml or 8 g of pure alcohol and corresponds to 4 fluid ounces of table wine, 1.5 fluid ounces of spirits or 8 fluid ounces of beer) in the previous month], a major psychiatric or medical disorder, current or past illicit drug misuse, or were receiving psychotropic medication. Family history of a psychiatric illness or substance misuse was not an exclusion criterion.

Behavioural assessments
The study protocol was approved by the ethics committee of the hospital. After a description of the study, written informed consent was obtained. Alcohol-dependent individuals who volunteered for the study underwent a psychiatric interview and had their medical health assessed by medical history and physical examination. Eligible subjects were tested for breath alcohol levels. This was part of the routine admission procedure of the unit. Subjects then completed a visual analogue scale (VAS) for craving (McCormack et al., 1988Go), and a questionnaire to classify subjects into type I and type II subtypes of alcoholism (Cloninger, 1987Go).

The VAS has been frequently used to quantify craving for drugs and alcohol and has been used as a subjective measure of alcohol craving (McCormack et al., 1988Go). It can be completed in 2 min. The VAS consisted of a 100 mm horizontal line anchored on the left with 0 (‘no craving’) and on the right with 100 (‘most craving ever felt for alcohol’). Subjects were asked to place a mark on the VAS based on their perceived intensity of craving at that time. The scores ranged from 0 to 100. A single VAS was administered at four time points during withdrawal. Studies have found a high correlation between craving scores obtained on the VAS and those obtained using a multi-item scale in substance misusers (Mezinskis et al., 1999Go). The type I/type II classification was based on procedures outlined by Gilligan et al.(1988)Go. The questionnaire included 12 variables, which were categorized as representing a type I characteristic (e.g. first dependence symptom after age 25) or a type II characteristic (e.g. family history of alcoholism). All positive responses to type I items (i.e. yes) were weighted positively (+) and all positive responses to type II items were weighted negatively (-). A total score was obtained for each subject by summing scores for all items. Subjects with scores in the positive range were classified as type I; those with negative scores were classified as type II. A score of 0 represented ‘unclassified’ subjects. The questionnaire required 15 min to complete.

Consenting controls were screened using the CAGE questionnaire (Ewing, 1984Go), a clinical interview, a physical examination, and breath alcohol readings. The CAGE is a 4-item questionnaire that takes about 1 min to complete. It has been extensively employed as a screening tool to detect alcoholism in clinical settings (Soderstrom et al., 1997Go).

The NA and 5-HT assay
Twenty millilitres of venous blood were collected in glass tubes from subjects and controls. For controls a single blood sample was obtained. For alcoholic subjects, blood was drawn on the day of admission (day 0), the next day (day 1), 7 days after admission (day 7) and 14 days after admission (day 14). Patients who could not produce at least two blood samples for any reason were not included in the study. Blood samples were collected in the morning and transported to the laboratory on ice within 2 h. The samples were centrifuged within 4 h of arrival at the laboratory. The plasma was separated into platelet-rich and platelet-poor fractions, platelet counts were performed on both fractions to ensure optimal separation, and the plasma was stored at -80°C. For the 5-HT assay, a good blood sampling technique, adequate platelet stabilization in the test tubes and rapid processing after collection reduced the possibility of platelet aggregation and activation that may stimulate 5-HT release and inflate the measured values (Beck et al., 1993Go). The NA and 5-HT assay was performed on the platelet-poor fraction using high-performance liquid chromatography (HPLC) (Beckman 110 system) with electrochemical detection (Warwick Instruments) optimized for the detection of NA and 5-HT. Details of the techniques for the NA and 5-HT assay have been previously described in published studies from our laboratory (Marsden, 1987Go; Fulford and Marsden, 1997Go). Briefly, after preparation of test samples, a solvent delivery pump was used to circulate the mobile phase. The samples were injected onto a column via a Rheodyne 7125 injector. NA and 5-HT were measured electrochemically using a dual glassy carbon electrode maintained by a silver/silver chloride reference electrode and a platinum auxiliary electrode. Changes in potential were measured by an amperometric detector linked to an integrator for the display of chromatograms. The HPLC system was calibrated using standard solutions to give a linear response over a range of NA and 5-HT concentrations. The protein assay was performed by the methods described by Lowry et al. (1951)Go. The intra-assay and inter-assay coefficients of variation for NA were 1.2 and 3.9%, respectively, and for 5-HT were 1.8 and 3.7%, respectively. The NA and 5-HT concentrations were expressed in picomoles per millilitre (pmol/ml) and nanomoles per litre (nmol/l) of plasma, respectively. The mean and standard deviation values for plasma NA and 5-HT levels in healthy volunteers measured using techniques similar to the present study have been reported to be 2.34 ± 0.18 pmol/ml (Forster et al., 1991Go) and 2.9 ± 1.8 nmol/l (Hindberg and Naesh, 1992Go), respectively. The laboratory personnel were blind to the clinical data.

Blood samples from alcoholic subjects were also analysed for serum {gamma}-glutamyl transpeptidase (GGT) and mean corpuscular volume (MCV) values. Elevated GGT and MCV are both indicative of excessive alcohol consumption and reported to be elevated in 50–75% of alcoholic individuals (Allemann, 2000Go).

Statistical analysis
Two-tailed t-tests and chi-squared tests were employed to compare subjects and controls on continuous and dichotomous variables respectively. Changes in NA and 5-HT levels and craving during withdrawal were evaluated using repeated measures analysis of variance (ANOVA). For this purpose, a Huynh-Feldt correction was used if the Mauchley test indicated that the data did not meet the assumption of sphericity (Gagnon et al., 1989Go). Relationships between clinical variables and NA or 5-HT levels were examined using Pearson or point biserial correlations. When appropriate, the data were analysed using analyses of covariance (ANCOVA) to control for the effects of other dependent variables on the biological measures. Differences in group sizes for biological measures reflect data lost due to missed blood draws or laboratory-related problems. Power calculations were based on our previous study that found a significant difference in platelet tritiated imipramine binding between 29 alcoholic patients and 29 controls (P < 0.005); moreover, platelet imipramine binding was significantly increased during alcohol withdrawal (P < 0.01) (Patkar et al., 1995Go). Assuming a similar effect size (f = 0.34), the sample in the current study had a power of 0.85 to detect differences between alcoholics and controls and a power of 0.87 to detect changes in biological measures over the four time points during withdrawal.


    RESULTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Subjects
A total of 54 male, Caucasian, individuals were studied: 26 were alcohol-dependent and 28 were controls. Table 1Go summarizes the demographic and clinical characteristics of the sample.


View this table:
[in this window]
[in a new window]
 
Table 1. Clinical characteristics of alcoholic subjects and controls
 
As seen in Table 1Go, the two groups did not differ significantly in age; however alcohol-dependent individuals were significantly more likely to be single, unemployed and tobacco smokers, than controls. No individuals tested positive for HIV; however, eight alcohol-dependent individuals had a history of medical diseases or were taking medications for their medical conditions. Five alcohol-dependent patients had a history of illicit drug misuse; however, none of these was currently misusing drugs. Five patients were taking benzodiazepines at the time of admission. Thirteen alcohol-dependent patients were classified as type I, 12 as type II and one individual was unclassified. The two subtypes did not differ in age, employment or marital status, duration of drinking, breathalyser readings of alcohol, or MCV values (all t < 1.72, df = 23, P > 0.05; {chi}2 < 0.84, df = 1, P > 0.05 in each case). However, type II patients had higher GGT values (231.15 ± 238.76 IU/l) than type I alcoholics (64.16 ± 44.03 IU/l) (t = 2.38, df = 23, P < 0.05). Four of the 12 (33.3%) type II patients had a history of illicit drug use, compared with one of the 13 (7.7%) type I patients. The two subtypes did not differ in tobacco use ({chi}2 = 1.35, P > 0.05).

Plasma NA levels during withdrawal
The plasma NA levels in controls were 3.03 ± 1.30 pmol/ml. During alcohol withdrawal the NA levels were: 5.19 ± 2.37 pmol/ml on day 0, 5.38 ± 2.67 pmol/ml on day 1, 2.51 ± 1.22 pmol/ml on day 7, and 2.46 ± 1.05 pmol/ml on day 14. As summarized in Fig. 1Go, a repeated measure ANOVA that compared all four test days revealed a significant effect of time on plasma NA levels during alcohol withdrawal [F(3,75) = 25.55, P < 0.001).



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 1. Changes in noradrenaline levels during alcohol withdrawal (n = 26). F(3,75) = 25.55, P < 0.01; vertical bars, ± 1 SD; —{blacksquare}, mean NA concentration.

 
The ANOVA analyses comparing two test days at a time (i.e. day 0 with day 1; day 7 with day 1) showed that NA levels were similar on days 0 and 1 [F(1,25) = 0.41, P = 0.52]; however, there was a significant reduction in NA concentrations on day 7 compared with day 1 [F(1,23) = 30.43, P < 0.001]. The NA levels on days 7 and 14 were not significantly different [F(1,21) = 0.23, P = 0.89]. When NA levels in controls were compared with those during alcohol withdrawal, the control levels were significantly lower than day 0 (t = 4.25, P < 0.001) and day 1 (t = 4.22, P < 0.001) values; however, control values were similar to day 7 (t = -1.51, P = 0.14) and day 14 (t = -1.73, P = 0.09) values. We also re-analysed the data controlling for benzodiazepines prescribed for detoxification. For this purpose, the mean daily total benzodiazepine dose in diazepam equivalents was used as covariate in an ANCOVA analysis. This showed that the effect of NA during withdrawal continued to remain significant after adjusting for benzodiazepine use [F(3,75) = 22.70, P < 0.001]. We then examined whether there were differences in NA levels during withdrawal between type I and type II alcoholics. No significant interaction was observed between the two subtypes and NA levels during withdrawal [F(3,72) = 0.49, P = 0.68].

Plasma 5-HT levels during withdrawal
The plasma 5-HT levels in controls were 3.73 ± 1.81 nmol/l. During withdrawal the 5-HT values were: 2.29 ± 1.88 nmol/l on day 0, 1.96 ± 1.58 nmol/l on day 1, 1.35 ± 0.87 nmol/l on day 7, and 1.34 ± 0.85 nmol/l on day 14. Similar to NA, a significant effect of time on plasma 5-HT levels was observed during alcohol withdrawal [F(3,75) = 5.47, P < 0.05]. The 5-HT levels during withdrawal are summarized in Fig. 2Go.



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 2. Changes in 5-HT levels during alcohol withdrawal (n = 26). F(3,75) = 5.47, P < 0.05; —{blacksquare}, mean 5-HT concentration; vertical bars, ± 1 SD.

 
ANOVA comparing two test days at a time showed that there was a significant decrease in 5-HT levels on day 1 compared with day 0 [F(1,25) = 6.41, P < 0.05], and day 7 compared with day 1 [F(1,23) = 5.69, P < 0.05], whereas the day 7 and day 14 values were not significantly different [F(1,21) = 0.31, P = 0.83]. The plasma 5-HT levels in controls were significantly higher than the day 1 (t = -2.12, P < 0.05), day 7 (t = -2.86, P < 0.01) and day 14 (t = -2.28, P < 0.05) values in alcoholic patients; however, the control 5-HT levels were similar to the day 0 values during withdrawal (t = -0.85, P = 0.40).

ANCOVA analysis using the benzodiazepine dose as a covariate did not change the direction or significance of findings [F(3,75) = 5.56 and F(3,75) = 5.47, respectively, P < 0.05]. Type I and type II alcoholic patients did not significantly differ in plasma 5-HT levels during withdrawal [F(3,72) = 1.43, P = 0.31].

Craving during withdrawal
As summarized in Fig. 3Go, a significant effect of time on craving was found during withdrawal [F(3,75) = 25.11, P < 0.001].



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 3. Changes in VAS craving scores during alcohol withdrawal (n = 26). F(3,25) = 25.11, P < 0.001; vertical bars, ± 1 SD; —{blacksquare}, mean VAS scores.

 
The VAS craving scores on day 0 (71 ± 30) were significantly higher than those on day 1 (49 ± 28) [F(1,25) = 18.08, P < 0.001], whereas the day 1 scores were higher than the day 7 scores (31 ± 28) [F(1, 23) = 11.09, P < 0.01] and the day 7 scores were similar to day 14 (29 ± 28) scores [F(1,21) = 0.46, P = 0.50].

Relationship between NA, 5-HT and craving during withdrawal
No significant correlations were found between NA, 5-HT and VAS craving scores on day 0, day 1, day 7 and day 14 (r ranged from 0.005 to 0.37, P > 0.05 in each case). To examine the relationship between changes in plasma 5-HT and NA levels and craving scores, we calculated the absolute and percentage change in the various measures between the four time points. No significant correlations were observed between changes in NA and 5-HT levels from day 0 to 1 (r = -0.28), day 1 to 7 (r = 0.11), day 7 to 14 (r = -0.22) and day 0 to 14 (r = -0.16). Similarly, the change in craving scores were not correlated to changes in NA and 5-HT from day 0 to 1 (r = 0.18 and 0.10, respectively), day 1 to 7 (r = 0.27 and -0.17, respectively), day 7 to 14 (r = -0.16 and 0.21, respectively) and day 0 to 14 (r = 0.18 and 0.13, respectively) during withdrawal (P > 0.05 in each case) (all correlations reported for absolute changes).

Plasma NA and 5-HT during alcohol withdrawal and in controls
As both NA and 5-HT levels declined significantly during alcohol withdrawal, we examined whether these measures differed from control values. The NA levels on days 0 and 1 of withdrawal were significantly higher than the NA levels in controls; however, the day 7 and day 14 NA levels were similar to the control values. In contrast, the 5-HT levels on day 0 of withdrawal resembled control values; however, the 5-HT concentrations on days 1, 7 and 14 were significantly lower than those in controls.

Due to demographic and clinical differences between alcoholic subjects and controls, we examined whether these variables were related to plasma NA or 5-HT levels. We found that NA or 5-HT levels were not significantly correlated with employment, marital status or smoking status ({chi}2 ranged from 0.48 to 1.87, df = 1, P > 0.05 in each case), or with GGT values or breath alcohol levels. Smoking status was also not related to craving scores ({chi}2 = 1.64, df = 1, P > 0.05).


    DISCUSSION
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
There were four main findings of interest from this study. First, both plasma NA and 5-HT levels changed significantly during the withdrawal period. Secondly, NA levels were higher than controls at the beginning of withdrawal, but subsequently declined and resembled those in controls within 7 days of abstinence. In contrast, 5-HT levels were similar to controls at the beginning of withdrawal, then declined and remained reduced compared with controls even after 14 days of abstinence. Thirdly, the alterations in NA or 5-HT activity were not specific for type I or type II alcoholism. Finally, craving was not related to alterations in NA or 5-HT during withdrawal.

Plasma NA levels during withdrawal
Our findings that plasma NA levels were significantly elevated in early withdrawal are consistent with reports of changes in plasma NA (Smith et al., 1990Go) and CSF 2-methoxy-4-hydroxyphenylglycol (MHPG) (Hawley et al., 1985Go). Elevated levels of circulating catecholamines can lead to downregulation of {alpha}2-adrenergic receptors. Consistent with our results, Nutt et al.(1988)Go documented blunting of neuroendocrine responses to intravenous clonidine, an {alpha}2 agonist in withdrawing alcoholics, compared with controls. In contrast, other studies have yielded negative results (Bylund et al., 1984Go). A possible reason for these discrepant findings could be that most studies have examined NA levels at one or two time points during withdrawal. Since we found that NA levels declined from early to late withdrawal, the stage of withdrawal during which data were obtained could have affected the results. It is also noteworthy that clinical signs and symptoms of alcohol withdrawal peak by 48–72 h and then diminish progressively over the next 10–14 days (Stockwell, 1994Go). This parallels the change in NA levels found in our study, suggesting that at least a part of the alcohol withdrawal syndrome may be mediated by the NA system (Linnoila et al., 1986Go; Lubman et al., 1983Go; Uzbay et al., 1998Go). An excellent review of the possible mechanisms for increased NA activity during early withdrawal has been published by Linnoila et al.(1987)Go. Since the NA levels resembled control values by 7 days of abstinence, it appears that NA changes during withdrawal may be reversible and temporary. Data in this area remain inconsistent with both normal and disturbed NA function having been reported in the literature (Muller et al., 1989Go; Berggren et al., 2000Go). Further studies, incorporating longitudinal designs, are needed to clarify whether changes in NA activity during various stages of alcohol dependence are state-dependent phenomena or represent long-term consequences of alcohol consumption.

Plasma 5-HT levels during withdrawal
There could be two possible explanations for the decline in plasma 5-HT levels throughout the withdrawal period. Since alcohol consumption has been known to stimulate serotonin turnover (Tollefson, 1989Go), it is possible that alcohol consumption may be an attempt to compensate for an underlying serotonin-deficit state that is unmasked during withdrawal. Alternatively, the withdrawal process may lead to reduction in the firing rates of serotonergic neurons resulting in decreased 5-HT release (Bailey et al., 2000Go). Consistent with several reports (Bailly et al., 1990; Arranz et al., 1999Go), 5-HT levels did not return to normal values at the end of withdrawal. It was beyond the scope of this study to determine whether the reduced 5-HT function would recover with extended abstinence; however, there is increasing evidence that central serotonergic function may be decreased in alcoholics (Krystal et al., 1996Go; Heinz et al., 1998Go). Considering the variability in response to treatment with 5-HT agents in alcoholism (Naranjo and Knoke, 2001Go), it may be timely to examine whether biological variables such as low 5-HT levels may help to predict response to treatment with selective serotonin reuptake inhibitors.

NA, 5-HT and craving in alcohol withdrawal
The patterns of changes in NA and 5-HT during withdrawal were different and there were no significant relationship between the two systems. This suggests that the disturbances in the NA and 5-HT systems during alcohol withdrawal may be independent of each other and possibly mediate distinct clinical symptoms. Consistent with studies of the course of withdrawal symptoms (Skinner and Allen, 1982Go; Stuppaeck et al., 1994Go), craving declined over the withdrawal period; however, it was not associated with changes in NA or 5-HT levels. This suggests that self-reports of urges to drink during withdrawal may not be extensively modulated by changes in plasma NA or 5-HT turnover. Also, craving is a multidimensional phenomenon having cognitive, affective and motivational components; a VAS does not reflect all these attributes. Because we did not measure withdrawal symptoms, it was not possible to determine whether symptoms such as anxiety that overlap with craving, could have been related to plasma NA or 5-HT. Our results support findings from studies that found no relationship between NA measures (Petrakis et al., 1999Go) or serotonergic manipulations (Petrakis et al., 2001Go) and cue-induced craving in alcohol misusers. However, some researchers have suggested that craving studies in alcoholics may require larger sample sizes than studies of other substance misusers (Tiffany et al., 2000Go). Although the current study was sufficiently powered to detect changes in NA (effect size = 0.58) and craving (effect size = 0.55), the relationship between the two may be less strong and larger sample sizes may be necessary to detect such small effects.

Type I and type II alcoholism
Contrary to our earlier findings that type I and II alcohol-dependent individuals differ in platelet tritiated imipramine binding (Patkar et al., 1995Go), the two subtypes did not differ in plasma 5-HT or NA levels during withdrawal. While the data on differences in NA function between the two subtypes are not robust, substantial evidence indicates that type I and type II alcoholics may differ in 5-HT measures (Roy et al., 1987Go; Virkunnen et al., 1994Go). The discrepant findings may be related to studying alcoholics during a state of withdrawal, a period characterized by significant perturbations in neurobiological systems, as opposed to chronic drinking or prolonged abstinence. Also, it must be noted that stratifying alcoholics in the two subtypes almost halved the group sizes and reduced the power to detect between-subject differences. With advances in neuroimaging, it seems feasible to examine central measures of NA and 5-HT, which may clarify the neurobiological differences between the two subtypes.

Limitations
Our data should be interpreted in the light of certain inherent limitations, given the clinical nature of the sample. First, patients received a regular dose of benzodiazepines, which may suppress NA or 5-HT activity (Finlay et al., 1995Go). While we statistically controlled for this influence, we cannot completely exclude the confounding effects of medications. Secondly, we considered day 0 as the day of admission, which may not have corresponded to beginning of withdrawal for all patients. Nevertheless, based on breathalyser values and clinical evaluations, we felt that the four time points represented early as well as late withdrawal. Thirdly, we used a single blood sample for the NA and 5-HT measurements in controls as opposed to four samples in alcohol-dependent patients. Fourthly, because we excluded patients with current substance misuse, this may have disproportionately excluded severe type II alcoholic individuals, who frequently misuse illicit substances. Also, it is unclear the extent to which blood levels of NA and 5-HT reflect changes in function in the brain, although central and peripheral measures of NA and 5-HT have been correlated in some studies (Kopin et al., 1983Go; Da Prada et al., 1988Go). Finally, since the sample was restricted to Caucasian men, our findings may not be generalizable to individuals from other ethnic backgrounds and women.

We conclude that the present study demonstrates that plasma NA and 5-HT levels change significantly during alcohol withdrawal in humans. The normalization of NA activity with brief abstinence suggests that the withdrawal-related NA changes may be a state-dependent phenomena. In contrast, the clinical significance of low 5-HT levels at the end of withdrawal need to be explored. Although NA or 5-HT measures do not appear to influence craving for alcohol, it is possible they may affect other factors such as negative mood states that may be important contributors to relapse. Examining the relationships between NA and 5-HT mechanisms during different stages of alcohol dependence may clarify the importance of these systems in mediating the addictive properties of alcohol.


    ACKNOWLEDGEMENTS
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
This work was carried out at the Drug and Alcohol Unit, Mapperley Hospital, and Department of Physiology and Pharmacology, University of Nottingham Medical School, UK. Drs Patkar and Naik were formerly associated with Mapperley Hospital. Dr Patkar was also associated with the Department of Physiology and Pharmacology, University of Nottingham. We thank Dr Philip McLean for medical input, David Forster for technical assistance and Stephen Weinstein for review of the manuscript. This research was supported by the Trent Regional Health Authority and Upjohn-Duphar Pharmaceuticals.


    FOOTNOTES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
* Author to whom correspondence should be addressed. Back


    REFERENCES
 TOP
 FOOTNOTES
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Allemann, P. I. (2000) Markers for excessive alcohol use (screening). Therapeutische Umschau. Revue thérapeutique 57, 185–190.[Medline]

American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders, 4th edn. American Psychiatric Press, Washington, DC.

Arranz, B., Rosel, P., Sarro, S., Zaldivar, E., Cano, R., Navarro, M. and San, L. (1999) Platelet serotonergic binding sites in alcohol-dependent patients. Alcohol and Alcoholism 34, 726–732.[Abstract/Free Full Text]

Bailey, C., Andrews, N., McKnight, A., Hughes, J. and Little, H. (2000) Prolonged changes in neurochemistry of dopamine neurons after chronic ethanol consumption. Pharmacology Biochemistry and Behavior 66, 153–161.[CrossRef][ISI][Medline]

Bailly, D., Vignua, J., Racadot, N., Beuscart, R., Servant, D. and Parquet, P. J. (1993) Platelet serotonin levels in alcoholic patients; changes related to physiological and pathological factors. Psychiatry Research 47, 57–88.[CrossRef][ISI][Medline]

Ballenger, J., Goodwin, F., Major, L. and Brown, G. (1979) Alcohol and central serotonin metabolism in man. Archives of General Psychiatry 36, 224–227.[Abstract]

Beck, O., Wallen, N. H., Broijersen, A., Larsson, P. T. and Hjemdahl, P. (1993) On the accurate determination of serotonin in human plasma. Biochemical and Biophysical Research Communications 196, 260–266.[CrossRef][ISI][Medline]

Berggren, U., Fahlke, C., Norrby, A., Zachrisson, O. and Balldin, J. (2000) Subsensitive alpha-2 adrenoceptor function in male alcohol-dependent individuals during 6 months of abstinence. Drug and Alcohol Dependence 57, 255–260.[CrossRef][ISI][Medline]

Bylund, D., Solomon, N., Ray-Prenger, C., Jones, S. and Harris, R. (1984) Comparison of adrenergic receptor binding in blood cells from alcoholics and controls. Alcohol 1, 369–372.[CrossRef][Medline]

Cloninger, C. R. (1987) Neurogenetic adaptive mechanisms in alcoholism. Science 236, 410–416.[ISI][Medline]

Da Prada, M., Cesura, A. M., Launay, J. M. and Richards, J. G. (1988) Platelets as models for neurons? Experimental Research 44, 115–126.[ISI]

Ewing, J. A. (1984) Detecting alcoholism: the CAGE questionnaire. Journal of the American Medical Association 252, 1905–1907.[Abstract]

Finlay, J. M., Zigmond, M. J. and Abercombie, E. D. (1995). Increased dopamine and norpepinephrine release in medial prefrontal cortex induced by acute and chronic stress: effects of diazepam. Neuroscience 64, 619–628.[CrossRef][ISI][Medline]

Forster, C. D., Taylor, J. Y. and Macdonald, I. A. (1991) The assay of the catecholamine content of small volumes of human plasma. In Monitoring Molecules in Neuroscience: Proceedings of the 5th International Conference on In Vivo Methods, Rollema, H., Westerink, B. H. C., Drijfhout, W. J. eds. University Centre for Pharmacy, Groningen, The Netherlands.

Fujimoto, A., Nagao, T., Ebara, T., Sato, M. and Otsuki, S. (1983) Cerebrospinal fluid monoamine metabolites during alcohol withdrawal syndrome and recovered state. Biological Psychiatry 18, 1141–1152.[ISI][Medline]

Fulford, A. J. and Marsden, C. A. (1997) Effect of isolation-rearing on noradrenaline release in rat hypothalamus and hippocampus in vitro. Brain Research 748, 93–99.[CrossRef][ISI][Medline]

Gagnon, J., Roth, J., Carroll, M., Hofmann, R., Haycock, K., Plamondon, J., Feldman, D. J. and Simpson, J. S. (1989) SuperANOVA: Accessible General Linear Modeling. Abacus Concepts, Berkeley, CA.

Gentsch, C., Lichtsteiner, M., Gastpar, M., Gastpar, G. and Feer, H. (1985) 3H-imipramine binding sites in platelets of hospitalized psychiatric patients. Psychiatry Research 14, 177–187.[CrossRef][ISI][Medline]

George, D. T., Benkelfat, C., Rawlings, R. R., Eckardt, M. J., Phillips, M. J., Nutt, D. J., Wynne, D., Murphy, D. L. and Linnoila, M. (1997) Behavioral and neuroendocrine responses to m-chlorophenylpiperazine in subtypes of alcoholics and in healthy comparison subjects. American Journal of Psychiatry 154, 81–87.[Abstract]

Gilligan, S. B., Reich, T. and Cloninger, R. C. (1988) Alcohol related symptoms in heterogeneous families of hospitalized alcoholics. Alcoholism: Clinical and Experimental Research 12, 671–678.[ISI][Medline]

Hawley, J. R., Major, L. F., Schulman, E. A. and Linnoila, M. (1985) Cerebrospinal fluid 2-methoxy-4-hydroxyphenylglycol and norepinephrine levels in alcohol withdrawal: correlations with clinical signs. Archives of General Psychiatry 42, 1056–1062.[Abstract]

Heinz, A., Ragan, P., Jones, D., Hommer, D., Williams, W., Knable, M., Gorey, J., Doty, L., Geyer, C., Lee, K., Coppola, R., Weinberger, D. and Linnoila, M. (1998) Reduced central serotonin transporters in alcoholism. American Journal of Psychiatry 155, 1544–1549.[Abstract/Free Full Text]

Hindberg, I. and Naesh, O. (1992) Serotonin concentrations in plasma and variations during menstrual cycle. Clinical Chemistry 38, 2087–2089.[Abstract/Free Full Text]

Kopin, I. J., Gordon, E. K., Jimerson, D. C. and Polinski, R. J. (1983) Relationship between plasma and cerebrospinal fluid levels of 3- methoxy-4-hydroxy-phenyl glycol. Science 219, 73–75.[ISI][Medline]

Krystal, J., Webb, E., Cooney, N., Kranzler, H., Southwick, S., Heninger, G. and Charnery, D. (1996) Serotonergic and noradrenergic dysregulation in alcoholism: m-chlorophenylpiperazine and yohimbine effects in recently detoxified alcoholics and healthy comparison subjects. American Journal of Psychiatry 153, 83–92.[Abstract]

LeMarquand, D., Pihl, R. O. and Benkelfat, C. (1994) Serotonin and alcohol intake, abuse, and dependence: findings of animal studies. Biological Psychiatry 15, 395–421.[CrossRef]

Linnoila, M., Guthrie, S., Lane, E. A., Karoum, F., Rudorfer, M. and Potter, W. Z. (1986) Clinical studies on norepinephrine metabolism: how to interpret the numbers. Psychiatry Research 17, 229–239.[CrossRef][ISI][Medline]

Linnoila, M., Mefford, I., Nutt, D. and Adinoff, B. (1987) Alcohol withdrawal and noradenergic function. Annals of Internal Medicine 107, 875–889.[ISI][Medline]

Lowry, O., Rosenbough, N., Farr, L. and Randall, R. (1951) Protein measurement with the Follin phenol reagent. Journal of Biological Chemistry 193, 265–275.[Free Full Text]

Lubman, A., Emrick, C., Mosiman, W. and Freedman, R. (1983) Altered mood and norepinephrine metabolism following withdrawal from ethanol. Drug and Alcohol Dependence 12, 3–13.[CrossRef][ISI][Medline]

Marsden, C. (1987) Changes in human whole blood 5-hydroxytrptamine (5-HT) and platelet 5-HT uptake during treatment with paroxetine, a selective 5-HT uptake inhibitor. Journal of Psychopharmacology 14, 244–250.

McCormack, H. M., Horne, D. J. and Sheather, S. (1988) Clinical applications of visual analog scales. A critical review. Psychological Medicine 18,1007–1019.[ISI][Medline]

Mezinskis, J., Somoza, E., Wishnick, H., Kidwell, C., Dyrenforth, S., Winhusen, T. and Kropp, F. (1999) Intercorrelations between several cocaine craving scales and their relationship to recent cocaine use. In Problems of Drug Dependence, 1998: Proceedings of the 60th Annual Scientific Meeting, College on Problems of Drug Dependence, Harris, K. ed. US Government Printing Office, Washington, DC.

Muller, N., Hoehe, M., Klein, H., Nieberle, G., Kapfhammer, H., May, F., Muller, O. and Fichter, M. (1989) Endocrinological studies in alcoholics during withdrawal and after abstinence. Psychoneuroendocrinology 14, 113–123.[CrossRef][ISI][Medline]

Naranjo, C. A. and Knoke, D. (2001) The role of selective serotonin reuptake inhibitors in reducing alcohol consumption. Journal of Clinical Psychiatry 62, 18–25.

Nutt, D., Glue, P., Molyneux, S. and Clark, E. (1988) Alpha-2-adrenoceptor function in alcohol withdrawal: a pilot study of the effects of i.v. clonidine in alcoholics and normals. Alcoholism: Clinical and Experimental Research 12, 14–18.[ISI][Medline]

Patkar, A. A., Naik, P., Marsden, C. A., Mclean, P. C., Kendall, D. A. and Al-Chalabi, T. (1995) Platelet 3H imipramine binding in type I and type II alcoholism. Human Psychopharmacology 10, 1–6.[ISI]

Petrakis, I., Trevisan, L., D’Souza, D., Gil, R., Krasnicki, S., Boutros, N., Cooney, N. and Krystal, J. (1999) CSF monoamine metabolite and beta-endorphin levels in recently detoxified alcoholics and healthy controls: Prediction of alcohol cue-induced craving. Alcoholism: Clinical and Experimental Research 23, 1336–1341.[ISI][Medline]

Petrakis, I., Trevisan, L., Nashaat, N., Krasnicki, S., Boutros, N., Limoncelli, D., Cooney, N. L. and Krystal, J. H. (2001) Effect of tryptophan depletion on alcohol cue-induced craving in abstinent alcoholic patients. Alcoholism: Clinical and Experimental Research 25, 1151–1155.[CrossRef][ISI][Medline]

Roy, A., Virkkunen, M. and Linnoila, M. (1987) Reduced central 5-HT turnover in a subgroup of alcoholics? Progress in Neuropsychopharmacology and Biological Psychiatry 11, 173–177.[CrossRef][ISI][Medline]

Skinner, H. A. and Allen, B. A. (1982) Alcohol dependence syndrome: measurement and validation. Journal of Abnormal Psychology 91, 199–209.[CrossRef][ISI][Medline]

Smith, A. J., Brent, P. J., Henry, D. A. and Foy, A. (1990) Plasma noradrenaline, platelet {alpha}2-adrenoceptors, and functional scores during ethanol withdrawal. Alcoholism: Clinical and Experimental Research 14, 497–502.[ISI][Medline]

Soderstrom, C. A., Smith, G. S., Kufera, J. A., Dischinger, P. C., Hebel, J. R., McDuff, D. R., Gorelick, D. A., Ho, S. M., Kerns, T. J. and Read, K. M. (1997) The accuracy of the CAGE, the Brief Michigan Alcoholism Screening Test, and the Alcohol Use Disorders Identification Test in screening trauma center patients for alcoholism. Journal of Trauma 43, 962–969.[ISI][Medline]

Stockwell, T. (1994) Alcohol withdrawal: an adaptation to heavy drinking of no practical significance? Addiction 89, 1447–1453.[ISI][Medline]

Stuppaeck, C. H., Barnas, C., Falk, M., Guenther, V., Hummer, M., Oberbauer, H., Pycha, R., Whitworth, A. B. and Fleischhacker, W. W. (1994). Assessment of the alcohol withdrawal syndrome — validity and reliability of the translated and modified Clinical Institute Withdrawal Assessment for Alcohol scale (CIWA-A). Addiction 89, 1287–1292.[ISI][Medline]

Sullivan, J. L., Baeziger, J. C. and Wagner, D. L (1990) Platelet MAO in subtypes of alcoholism. Biological Psychiatry 27, 911–922.[CrossRef][ISI][Medline]

Thiele, T. E., Cubero, I., van Dijk, G., Mediavilla, C. and Bernstein, I. L. (2000) Ethanol-induced c-fos expression in catecholamine- and neuropeptide Y-producing neurons in rat brainstem. Alcoholism: Clinical and Experimental Research 24, 802–809.[CrossRef][ISI][Medline]

Tiffany, S., Carter, B. and Singleton, E. (2000) Challenges in the manipulation assessment and interpretation of craving relevant variables. Addiction 95, 177S–187S.

Tollefson, G. (1989) Serotonin and alcohol: interrelationships. Psychopathology 22, 37–48.[ISI][Medline]

Uzbay, I., Usanmaz, S., Tapanyigit, E., Aynacioglu, S. and Akarsu, E. (1998) Dopaminergic and serotonergic alterations in the rat brain during ethanol withdrawal: association with behavioral signs. Drug and Alcohol Dependence 53, 39–47.[CrossRef][ISI][Medline]

Virkkunen, M., Rawlings, R., Tokola, R., Poland, R., Giudotti, A., Nemeroff, C., Bissette, G., Kalogeras, K., Kavonon, S. L. and Linnoila, M. (1994) CSF biochemistries, glucose metabolism and diurnal activity and rhythms in alcoholic, violent offenders, fire setters and healthy volunteers. Archives of General Psychiatry 51, 20–27.[Abstract]

Von Knorring, L., Oreland, L. and Von Knorring, A. L. (1987) Personality traits and platelet MAO activity in alcohol and drug abusing teenage boys. Acta Psychiatrica Scandanavica 75, 307–314.[ISI][Medline]

Wang, Y., Freund, R. K. and Palmer, M. R. (1999) Potentiation of ethanol effects in cerebellum by activation of endogenous noradrenergic inputs. Journal of Pharmacology and Experimental Therapeutics 288, 211–220.[Abstract/Free Full Text]





This Article
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Patkar, A. A.
Articles by Marsden, C. A.
PubMed
PubMed Citation
Articles by Patkar, A. A.
Articles by Marsden, C. A.