- Email: [email protected]
Growth hormone response to the GABA-B agonist baclofen in 3-week abstinent alcoholics
Alcohol 41 (2007) 551e556
Growth hormone response to the GABA-B agonist baclofen in 3-week abstinent alcoholics Saliha Ozsoya,*, Ertugrul Esela, Tayfun Turana, Mustafa Kulab a Department of Psychiatry, Erciyes University School of Medicine, 38039 Kayseri, Turkey Department of Nuclear Medicine, Erciyes University School of Medicine, 38039 Kayseri, Turkey Received 19 May 2007; received in revised form 15 September 2007; accepted 17 September 2007 b
Abstract Gamma-aminobutyric acid (GABA) dysfunction is a known feature of alcoholism. We investigated GABA-B receptor activity in 3-week abstinent alcoholics using the growth hormone (GH) response to baclofen, a GABA-B receptor agonist. The study aimed to investigate the relationship between GABA-B receptor activity and alcohol withdrawal. GH response to baclofen was measured in alcohol-dependent males without depression (n 5 22) who were on day 21 of alcohol abstinence and in healthy control male subjects (n 5 23). After 20 mg baclofen was given orally to the subjects, blood samples for GH assay were obtained every 30 min for the subsequent 150 min. The patients were divided into two subgroups (continuing withdrawal and recovered withdrawal subgroups) according to their withdrawal symptom severity scores on day 21 of alcohol cessation. Baclofen administration significantly altered GH secretion in the controls, but not in the patients. When GH response to baclofen was assessed as DGH, it was lower in the patients with continuing withdrawal symptoms than in the controls and in the recovered withdrawal group. Impaired GH response to baclofen in all patients mainly pertained to the patients whose withdrawal symptoms partly continued. Our results suggest that reduced GABA-B receptor activity might be associated with longerterm alcohol withdrawal symptoms in alcoholic patients. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Gamma-aminobutyric acid; GABA-B receptor; Baclofen; Growth hormone; Alcohol withdrawal; Neuroendocrine
Introduction Gamma-aminobutyric acid (GABA) is an inhibitory amino acid neurotransmitter that plays a major role in central nervous system dysregulation during alcohol withdrawal. Hyperexcitation symptoms in alcohol withdrawal stem, in part, from a decreased central GABAergic inhibitory function probably caused by chronic alcohol consumption that leads to downregulation in GABA receptors (Cagetti et al., 2003). Animal and human studies confirm the decreased activity of GABA during ethanol withdrawal (Faingold et al., 2000; Tsai et al., 1998). The reduction in withdrawal symptoms with GABA agonists and worsening of symptoms with GABA antagonists also support the idea of decreased GABAergic function in alcohol withdrawal (Anton, 2001). Baclofen is a GABA-B receptor agonist, promising in the treatment of alcohol withdrawal and alcohol
* Corresponding author. Department of Psychiatry, Erciyes University School of Medicine, Talas Road, 38039 Kayseri, Turkey. Tel./fax: þ90352-4375702. E-mail address: [email protected] (S. Ozsoy). 0741-8329/07/$ e see front matter Ó 2007 Elsevier Inc. All rights reserved. doi: 10.1016/j.alcohol.2007.09.003
dependence (Addolorato et al., 2006; Colombo et al., 2000; Flannery et al., 2004). GABA-B receptor function in the central nervous system can be assessed using a neuroendocrine-challenge test with baclofen, which is considered to provide an index of GABA-B receptor function in humans. It is suggested that baclofen increases growth hormone (GH) release by inhibition of hypothalamic somatostatin release and/or by dopaminergic control (Davis et al., 1997; Koulu et al., 1979). Another mechanism of GH production stimulated by baclofen has been reported that in GH-producing cells, GABA acts as an autocrine factor via GABA-B receptors to control GH levels (GamelDidelon et al., 2002). Previously, Vescovi and Coiro (2001) and Vescovi et al. (1998) studied GABA-B activity by investigating GABAergic control of GH secretion in alcoholism. They reported that baclofen administration induced serum GH levels in the controls, but not in 2e4-week abstinent alcoholics, and suggested a loss of GABAergic inhibitory control in alcoholism (Vescovi et al., 1998). Another study by these authors investigated whether long-term abstinence restores this stimulatory effect of baclofen on GH secretion, and it was found that GH response to baclofen did not recover
552
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
in 4-year abstinent alcoholics (Vescovi and Coiro, 2001). They suggested that GABAergic dysfunction might be a consequence of chronic alcohol consumption resulting in irreversible damage or might be a trait marker of alcoholism. However, no clear relationship between GABA-B activity and alcohol withdrawal was revealed. It has been reported that withdrawal symptoms such as anxiety, insomnia, and mild autonomic overactivity are likely to continue for 2e6 months after the acute withdrawal has disappeared. These protracted withdrawal symptoms may enhance the probability of relapse (Schuckit, 2005). Under experimental conditions, it has also been shown that long-term behavioral changes and anxiety-like states persist in experimental animals for weeks after withdrawal (Rasmussen et al., 2001). The aim of the present study was to investigate probable alterations in GABA-B activity in abstinent alcoholics, particularly in patients whose withdrawal symptoms partly continue despite they being in the third week of alcohol cessation, measuring GH response to GABA-B agonist, baclofen. The results might increase our understanding of the relationship between GABA-B receptor activity and continuing withdrawal symptoms.
Methods Subjects Twenty-two male inpatients (mean age 6 S.D.: 42.04 6 7.93 years, range: 25e55 years) who met the Diagnostic and Statistical Manual of Mental Disorders,
4th edition, DSM-IV; (American Psychiatric Association, 1994) criteria for alcohol dependence and alcohol withdrawal were recruited for the study. The patients were selected from the inpatient population of the Psychiatry Clinic of Erciyes University Medical School. The DSM-IV diagnosis of alcohol withdrawal was based on information generated from clinical interviews. The patients were in the first day of withdrawal when they were admitted to hospital for the alcohol detoxification program. The patients stayed in hospital throughout the study. Exclusion criteria for patients were (1) a substance use disorder other than alcohol or nicotine, (2) any other DSM-IVaxis I disorder, current or past, (3) any significant medical and endocrine disorder, and/or (4) history of cirrhosis or laboratory evidence of significant liver disease. Twenty-three physically and mentally healthy men (mean age 6 S.D.: 37.00 6 8.97 years, range: 25e55 years) who were recruited from volunteers and hospital staff participated in the study as controls. Exclusion criteria for the controls were the same as those for the patients, plus a history of alcohol or drug use disorders in themselves or their families. All subjects were included in the study after careful evaluation of their laboratory test results and physical and psychiatric examinations and history. All patients and controls were smokers. Detailed characteristics of the subjects are given in Table 1. This study was carried out in accordance with the Helsinki Declaration of the World Medical Association and was approved by the local Ethics Committee. Written informed consent was obtained from each patient after the study was explained to them.
Table 1 Demographical and clinical characteristics of the patients and the controls Demographical and clinical variables
Patients (n 5 22) Mean 6 S.D.
Controls (n 5 23) Mean 6 S.D.
Comparisons
Age (year) BMI (kg/m2) Cigarette (number/day) Duration of smoking (year) Duration of consumption (year) Amount of alcohol consumed (g/day) No. of abstinence attemptsa MADRS score MAST score CIWA-Ar score (first day) CIWA-Ar score (21st day) AST (U/I) ALT (U/I) GGT (first day) (U/I) GGT (21st day) (U/I) T-blb (mg/dl) C-blb (mg/dl)
42.04 6 7.93 24.32 6 4.33 30.90 6 14.36 26.45 6 8.69 23.13 6 7.69 366.71 6 150.06 9.90 6 11.40 6.87 6 4.55 29.63 6 10.56 22.50 6 9.29 4.63 6 5.86 34.05 6 12.34* 36.64 6 18.47 68.50 6 34.25* 60.35 6 31.76* 1.10 6 0.31 0.41 6 0.11
37.00 6 8.97 26.25 6 3.23 24.09 6 11.36 21.77 6 10.69 d d d d d d d 23.36 6 9.31 29.09 6 19.39 26.72 6 10.78 26.72 6 10.78 1.05 6 0.48 0.36 6 0.13
t 5 1.99 t 5 1.70 t 5 1.37 t 5 1.08 d d d d d d d t 5 2.45 t 5 1.03 t 5 4.56 t 5 3.70 t 5 0.22 t 5 1.21
P O.05 P O.05 P O.05 P O.05 d d d d d d d P !.05 P O.05 P !.05 P !.05 P O.05 P O.05
n 5 subject number; BMI 5 body mass index; MADRS 5 MontgomeryeAsberg Depression Rating Scale; MAST 5 Michigan Alcoholism Screen Test; CIWA-Ar 5 revised Clinical Institute Withdrawal Assessment for Alcohol; AST 5 aspartate aminotransferase; ALT 5 alanine aminotransferase; GGT 5 g-glutamyl transpeptidase; T-blb 5 total bilirubin; C-blb 5 conjugated bilirubin; u/l 5 unit/liter. *Significantly higher than those of the controls. a The number of attempts made to abstain with or without treatment.
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
Procedures The patients were admitted to hospital on the first day of cessation. Routine biochemical examinations were performed on the first day of admission. The Michigan Alcoholism Screen Test (Selzer, 1971) was performed to assess the severity of alcoholism within the first week of admission. To measure and follow severity of withdrawal, the revised Clinical Institute Withdrawal Assessment for Alcohol (CIWA-Ar) (Sullivan et al., 1991) was used. The treatment dosage and duration were adjusted according to this scale and clinical symptomatology. The severity of the clinical symptomatology of depression was measured with the MontgomeryeAsberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979) on day 8 of admission. One week after withdrawal, none of the patients fulfilled the DSM-IV criteria of a depressive episode, and MADRS scores were below 15 in all patients. The patients received diazepam (mean dose 6 S.D.: 43.18 6 16.50 mg/day, range: 15e90 mg/day at the beginning of detoxification) and multivitamins for treatment of alcohol withdrawal symptoms. The patients did not receive any additional drug or nondrug therapies. Patients who were on day 21 of abstinence and the controls underwent a baclofen-challenge test. After an overnight fast, a venous catheter was inserted into an antecubital vein at 08:00 hours. A blood sample for basal GH was drawn just after cannula insertion (time 0). The subjects were allowed to rest but not to smoke or eat until the challenge procedure was completed. Blood samples for GH were taken at 30, 60, 90, 120, and 150 min after the administration of baclofen (Lioresal 20 mg p.o., NovartisÒ), which was given just after time 0. Blood samples were centrifuged and the separated serum was stored at 70 C until analysis. GH concentrations were determined by immunoradiometric assay (IRMA) techniques (DSL-1900, USA). For GH, the sensitivity was 0.03 mIU/ml; the intra-assay coefficient of variation was 3.1% for GH pool of 8.73 mIU/ml. The interassay coefficient of variation was 5.9% for GH pool of 1.02 mIU/ml. The patients were divided into two subgroups according to whether their withdrawal symptoms were ongoing on day 21 of abstinence. Patients with ongoing withdrawal symptoms according to their CIWA-Ar score were included in the continuing withdrawal group (n 5 8). Patients who had no withdrawal symptoms were included in the recovered withdrawal group (n 5 14). The threshold for withdrawal symptoms recovery was accepted as 7 in CIWA-Ar score. The mean CIWA-Ar score of the recovered withdrawal patient group was 1.21 6 1.42 (range: 0e4), and that of the continuing withdrawal patient group was 10.62 6 5.92 (range: 7e15).
Data analysis Maximum GH level out of 30e150 min was considered the ‘‘peak GH level’’. Peak GH level for subjects who did
553
not show increases above baseline was determined by using time 120 as a surrogate for peak values because this was the group peak time. GH responses (DGH) were calculated as the difference between baseline concentrations and maximal peak GH response after the challenge. The area under the curve (AUC; the serum concentration vs. time) was also calculated, for evaluating GH response to baclofen in the subjects. The distributions of all variables were checked by KolmogoroveSmirnov test. GH values had an abnormal distribution. To compare hormone levels of the patients and controls, a ManneWhitney U test was performed. A nonparametric Friedman test was preferred to repeated measure ANOVA for investigation of hormone response to baclofen, because GH values had an abnormal distribution. Spearman’s rank correlation test was performed to investigate the relationships between hormonal values and clinical and demographical variables in the patients.
Results There was no significant difference in age, body mass index, number of cigarettes consumed daily, or smoking duration between the patients and controls. The patients had higher serum aspartate aminotransferase and g-glutamyl transpeptidase levels (Table 1). Table 2 shows the comparison of patients’ hormonal values on day 21 of alcohol cessation with those of the controls. Basal GH level in the patients was higher than that in the controls (Z 5 4.727, P !.001). DGH, peak GH, and AUC GH values did not differ significantly between the patients and the controls. When GH response to baclofen was analyzed by Friedman test, a statistically significant increase over time was observed in the controls (c2 5 25.065, P !.001), but not in the patients (c2 5 4.564, P 5 .471) (Fig. 1). The GH response to baclofen measured as DGH in patients within withdrawal syndrome group was lower than that of the controls (Z 5 2.854, P 5.012) and of the recovered withdrawal group (Z 5 1.973, P 5 .042). Nevertheless, DGH of the recovered withdrawal group did not differ from that of the controls (Z 5 0.486, P 5.627). Basal GH levels of both the recovered withdrawal and continuing Table 2 GH values of the patients and the controls
Hormonal values
Patient group (n 5 22) Mean 6 S.D.
Control group (n 5 23) Mean 6 S.D. Comparisons
Basal GH (mIU/ml) 0.18 6 0.15* 0.06 6 0.03 DGH (mIU/ml) 0.55 6 1.45 0.53 6 0.84 Peak GH (mIU/ml) 0.74 6 1.46 0.59 6 0.84 AUC GH 51.07 6 94.10 31.40 6 40.62
Z 5 4.727 Z 5 1.602 Z 5 0.341 Z 5 0.488
GH 5 growth hormone; AUC 5 area under the curve. *Significantly higher than that of the controls.
P !.001 P 5.109 P 5.733 P 5.625
554
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
Fig. 1. Growth hormone response to baclofen in the patients and the controls.
withdrawal groups were higher compared to that of the controls (Z 5 4.168, P !.001; Z 5 3.404, P 5.001) (Table 3, Fig. 2). According to the nonparametric Friedman test, there was no statistically significant GH alteration in the baclofenchallenge test in the recovered withdrawal and continuing withdrawal groups (c2 5 6.882, P 5 .230; c2 5 7.355, P 5.196, respectively). Features of consumption of alcohol (duration, daily amount, and severity of alcoholism) and smoking (duration and daily number) did not differ between the continuing withdrawal and recovered withdrawal groups. CIWA-Ar scores of the continuing withdrawal group were higher than those of the recovered withdrawal group on the first and 21st days (Table 4). There was no significant correlation between hormonal measurements (basal GH, DGH, peak GH, and AUC GH) and features of alcohol consumption (duration, daily amount, and severity of alcohol use) in the patients. There were positive correlations between basal GH and peak GH, and AUC GH values (r 5 0.665, P 5 .001; r 5 0.668, P !.001, respectively), but no correlation between basal GH and DGH (r 5 0.257, P 5 .249).
Table 3 GH values of the patient subgroups when divided as ones whose withdrawal symptoms continue and ones whose do not on day 21 of alcohol cessation Patient group Mean 6 S.D. Hormonal values
Recovered withdrawal group (n 5 14)
Continuing withdrawal group (n 5 8)
Control group (n 5 23) Mean 6 S.D.
Basal GH (mIU/ml) DGH (mIU/ml) Peak GH (mIU/ml) AUC GH
0.16 6 0.08* 0.80 6 1.78 0.96 6 1.79 65.28 6 115.96
0.22 6 0.22* 0.11 6 0.27** 0.34 6 0.37 26.22 6 22.02
0.06 6 0.03 0.53 6 0.84 0.59 6 0.84 31.40 6 40.62
GH 5 growth hormone; AUC 5 area under the curve. *Higher than those of the controls. **Lower than that of the controls and the recovered withdrawal group.
Fig. 2. Growth hormone response to baclofen in the two patient subgroups and the controls.
Discussion Some patients in our study presented withdrawal symptoms after 3 weeks of abstinence and benzodiazepine treatment. Alcohol withdrawal symptoms are usually expected to diminish within several days after treatment with benzodiazepines (Mayo-Smith, 1997). However, we have been observing for years that some of our alcoholic patients keep presenting at least some withdrawal symptoms until the third or fourth week of alcohol cessation. We think that this might be a racial feature, because protracted alcohol withdrawal has also been frequently reported in Japanese people (Komiyama, 1997). As we expected, some patients in this study presented withdrawal symptoms, such as anxiety, autonomic symptoms, and tactile and visual disturbances, on day 21 of abstinence. It is known that some withdrawal symptoms persist for some months or longer as part of protracted withdrawal. The continuing withdrawal symptoms in this study should not be referred to as protracted withdrawal, because we did not follow-up the patients over a long abstinence period. However, it was interesting that some patients presented withdrawal symptoms, while others did not. The mechanism underlying this individual variation is not clear. The findings of the study might contribute to a better understanding of this matter. Our data showed that while there was a significant GH response to baclofen in the controls, this response was absent in 3-week abstinent alcoholics. This suggests that the GH-stimulating effect of baclofen is impaired in patients in alcoholic withdrawal. Looking at GH response in each patient subgroup separately, that is, as those having continuing withdrawal symptoms and those not, the GH response to baclofen (DGH) in patients who still had withdrawal symptoms in the third week of abstinence was lower than that in patients who had recovered from withdrawal syndrome. This means that impaired GH response to baclofen in all patients mainly pertained to the patients who were still in active withdrawal although they were in the third week of alcohol cessation. In our opinion, this finding is important because it may indicate to a direct
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
555
Table 4 Demographical and clinical characteristics of the patient subgroups Demographical and clinical variables
Recovered withdrawal group (n 5 14)
Continuing withdrawal group (n 5 8)
Comparisons
Age (year) BMI (kg/m2) Cigarette (number/day) Duration of smoking (year) Duration of alcohol consumption (year) Amount of alcohol consumed (g/day) MADRS score MAST score CIWA-Ar score (first day) CIWA-Ar score (21st day)
41.71 6 9.41 24.68 6 4.92 26.42 6 11.33 26.37 6 7.92 22.21 6 7.37 352.28 6 160.37 5.44 6 4.97 27.42 6 10.75 16.57 6 4.91 1.21 6 1.42
42.62 6 4.83 23.68 6 3.24 38.75 6 16.42 26.66 6 12.58 24.75 6 8.48 395.57 6 133.73 8.71 6 3.45 33.50 6 9.66 32.87 6 4.58* 10.62 6 5.92*
Z 5 0.13 Z 5 0.34 Z 5 1.81 Z 5 0.00 Z 5 0.68 Z 5 0.72 Z 5 1.38 Z 5 1.23 Z 5 3.82 Z 5 3.87
P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P !.001 P !.001
BMI 5 body mass index; MADRS 5 MontgomeryeAsberg Depression Rating Scale; MAST 5 Michigan Alcoholism Screen Test; CIWA-Ar 5 revised Clinical Institute Withdrawal Assessment for Alcohol. *Higher than those of the recovered withdrawal group.
association between continuing alcohol withdrawal symptoms and reduced GABA-B receptor activity. A contributing factor to reduced GABA activity in alcohol withdrawal might be downregulated hypothalamic GABA-B receptors (Cagetti et al., 2003). Thus, one can consider that GABA-B receptor dysfunction may be one of the causes of the clinical picture of alcohol withdrawal, because responselessness to baclofen seems to be peculiar to the patients who had protracted withdrawal symptoms in the present study. Supporting this idea, some studies have implicated GABA-B receptor dysfunction in the pathophysiology of alcohol withdrawal. For instance, gammahydroxy-butyric acid (GHB), which interacts with GABAB receptor via a different mechanism from baclofen (Koek et al., 2007), has been reported to reduce substantially the symptoms of acute alcohol withdrawal (Korninger et al., 2003). Another clinical study showed the lack of effect of GHB on GH release in alcoholic patients, contrary to the increase in the release of GH observed in healthy subjects (Addolorato et al., 1999). Moreover, recent preclinical and clinical studies have suggested that baclofen suppresses symptoms of alcohol withdrawal syndrome (Addolorato et al., 2006; Colombo et al., 2000; Walker and Koob, 2007). On the other hand, impaired GABA-B receptor activity may be a reason for protracted withdrawal, rather than a consequence of chronic alcohol consumption or alcohol withdrawal. Patients who had more reduced GABA-B activity might be susceptible to the development of protracted withdrawal symptoms. Furthermore, protracted withdrawal may be a predictor for postedetoxification relapse (O’Connor et al., 1991). The finding that baclofen is an effective treatment in the prevention of relapse due to its ability to reduce craving in alcoholic patients also supports this consideration (Addolorato et al., 2002). Impaired GABAergic activity is probably responsible, in part, for alcohol craving and for withdrawal symptoms (De Witte, 2004). Considering all these findings together with our present finding, one may speculate that protracted withdrawal in alcoholic patients is associated with decreased GABA-B activity.
From a different perspective, blunted GH response to baclofen may reflect a heritable trait that may be a risk factor for alcoholism and it may not recover over time as some prior reports supported (Vescovi and Coiro, 2001; Vescovi et al., 1998). The finding of the present study that the patients had blunted GH response to baclofen is partly consistent with this idea. It has also been suggested that blunted GH response to baclofen may be related with comorbid anxiety disorders in alcohol-dependent patients (Condren et al., 2003; Gerra et al., 1998). Therefore, it may be considered that persistently elevated CIWA-Ar scores in the present study might be associated with a comorbid anxiety disorder in alcoholic patients rather than continuing alcohol withdrawal symptoms. However, we tried to eliminate confounding effect of anxiety by excluding the patients complicated by axis I psychopathologies such as anxiety disorders. One of the limitations of the present study was that all patients had received diazepam for treatment of alcohol withdrawal. The clinical effects of classical benzodiazepines such as diazepam are mediated by specific GABA-A receptor subtypes (Rudolph et al., 1999). For that reason, we do not consider that diazepam treatment directly affects GABA-B receptor activity, and this undermines our conclusions. The other limitation is the insufficient sample sizes of the patient subgroups. This might have caused a type II statistical error. Subjects were not able to match for smoking. However, features of smoking (duration and daily number) did not differ between the patients and the controls, and between the continuing withdrawal and recovered withdrawal subgroups. Finally, the lack of other clinical assessments such as anxiety was the other limitation. In conclusion, the present work suggests that GABA deficit during alcohol withdrawal may result from a dysfunction of GABA-B receptors and GABA-A receptors. GABA-B receptor dysfunction in alcoholic patients might be related to alcohol withdrawal symptoms per se, as they continue. However, this conclusion should be evaluated very cautiously because our results are preliminary and the study has important limitations as stated above.
556
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
References Addolorato, G., Capristo, E., Gessa, G. L., Caputo, F., Stefanini, G. F., and Gasbarrini, G. (1999). Long-term administration of GHB does not affect muscular mass in alcoholics. Life Sci. 65(14), PL191–PL196. Addolorato, G., Caputo, F., Capristo, E., Domenicali, M., Bernardi, M., Janiri, L., Agabio, R., Colombo, G., Gessa, G. L., and Gasbarrini, G. (2002). Baclofen efficacy in reducing alcohol craving and intake: a preliminary double-blind randomized controlled study. Alcohol Alcohol. 37, 504–508. Addolorato, G., Leggio, L., Abenavoli, L., Agabio, R., Caputo, F., Capristo, E., Colombo, G., Gessa, G. L., and Gasbarrini, G. (2006). Baclofen in the treatment of alcohol withdrawal syndrome: a comparative study vs diazepam. Am. J. Med. 119, e13–e18. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Press. Anton, R. F. (2001). Pharmacologic approaches to the management of alcoholism. J. Clin. Psychiatry 62, 11–17. Cagetti, E., Liang, J., Spigelman, I., and Olsen, R. W. (2003). Withdrawal from chronic intermittent ethanol treatment changes subunit composition, reduces synaptic function, and decreases behavioral responses to positive allosteric modulators of GABA-A receptors. Mol. Pharmacol. 63, 53–64. Colombo, G., Agabio, R., Carai, M. A., Lobina, C., Pani, M., Reali, R., Addolorato, G., and Gessa, G. L. (2000). Ability of baclofen in reducing alcohol intake and withdrawal severity: IdPreclinical evidence. Alcohol. Clin. Exp. Res. 24, 58–66. Condren, R. M., Lucey, J. V., and Thakore, J. H. (2003). A preliminary study of baclofen-induced growth hormone release in generalised social phobia. Hum. Psychopharmacol. 18, 125–130. Davis, L. L., Trivedi, M., Choate, A., Kramer, G. L., and Petty, F. (1997). Growth hormone response to the GABAB agonist baclofen in major depressive disorder. Psychoneuroendocrinology 22, 129–140. De Witte, P. (2004). Imbalance between neuroexcitatory and neuroinhibitory amino acids causes craving for ethanol. Addict. Behav. 24, 1325–1339. Faingold, C., Li, Y., and Evans, M. S. (2000). Decreased GABA and increased glutamate receptor-mediated activity on inferior colliculus neurons in vitro are associated with susceptibility to ethanol withdrawal seizures. Brain Res. 868, 287–295. Flannery, B. A., Garbutt, J. C., Cody, M. W., Renn, W., Grace, K., Osborne, M., Crosby, K., Morreale, M., and Trivette, A. (2004). Baclofen for alcohol dependence: a preliminary open-label study. Alcohol. Clin. Exp. Res. 28, 1517–1523. Gamel-Didelon, K., Corsi, C., Pepeu, G., Jung, H., Gratzl, M., and Mayerhofer, A. (2002). An autocrine role for pituitary GABA: activation of GABA-B receptors and regulation of growth hormone levels. Neuroendocrinology 76, 170–177. Gerra, G., Ferri, M., Zaimovic, A., Giucastro, G., Palladino, M., Sartori, R., Delsignore, R., Maestri, D., Marzocchi, G., and Brambilla, F. (1998). GABAergic function in detoxified heroin addicts: relationship to anxiety disorders. Psychiatry Res. 77, 89–96. Koek, W., Mercer, S. L., and Coop, A. (2007). Cataleptic effects of gamma-hydroxybutyrate (GHB), its precursor gamma-butyrolactone (GBL), and GABA(B) receptor agonists in mice: differential
antagonism by the GABA (B) receptor antagonist CGP35348. Psychopharmacology (Berl.) 192(3), 407–414. Komiyama, T. (1997). Clinical features of alcoholics after alcohol withdrawaldwithdrawal syndrome, protracted withdrawal syndrome and prolonged disturbance of alcohol dependence. Nippon Rinsho 55, 393–405. Korninger, C., Roller, R. E., and Lesh, O. M. (2003). Gamma-hydroxybutyric acid in the treatment of alcohol withdrawal syndrome in patients admitted to hospital. Acta Med. Austriaca 30, 83–86. Koulu, M., Lammintausta, R., and Dahlstrom, S. (1979). Stimulatory effect of acute baclofen administration on human growth hormone and prolactin secretion in man. J. Clin. Endocrinol. Metab. 48, 1038–1040. Mayo-Smith, M. F. (1997). Pharmacological management of alcohol withdrawal. A meta-analysis and evidence-based practice guideline. American Society of Addiction Medicine Working Group on Pharmacological Management of Alcohol Withdrawal. JAMA 278(2), 144– 151. Montgomery, S. A., and Asberg, M. (1979). A new depression scale designed to be sensitive to change. Am. J. Psychiatry 134, 382–389. O’Connor, P. G., Gottlieb, L. D., Kraus, M. L., Segal, S. R., and Horwitz, R. I. (1991). Social and clinical features as predictors of outcome in outpatient alcohol withdrawal. J. Gen. Intern. Med. 6, 312–316. Rasmussen, D. D., Mitton, D. R., Green, J., and Puchalski, S. (2001). Chronic daily ethanol and withdrawal: 2. Behavioral changes during prolonged abstinence. Alcohol. Clin. Exp. Res. 25(7), 999–1005. Rudolph, U., Crestani, F., Benke, D., Brunig, I., Benson, J. A., Fritschy, J. M., Martin, J. R., Bluethmann, H., and Mohler, H. (1999). Benzodiazepine actions mediated by specific gamma-aminobutyric acid A receptor subtypes. Nature 401, 796–800. Schuckit, M. A. (2005). Alcohol-related disorders. In B. J. Sadock, & V. A. Sadock (Eds.), Kaplan and Sadock’s comprehensive textbook of psychiatry (pp. 1168–1188). New York, NY: Lippincott Williams & Wilkins. Selzer, M. L. (1971). The Michigan Alcoholism Screening Test: the quest for a new diagnostic instrument. Am. J. Psychiatry 127, 1653–1658. Sullivan, J. T., Swift, R. M., and Lewis, D. C. (1991). Benzodiazepine requirements during alcohol withdrawal syndrome: clinical implications of using a standardized withdrawal scale. J. Clin. Psychopharmacol. 11, 291–295. Tsai, G. E., Ragan, P., Chang, R., Chen, S., Linnoila, V. M., and Coyle, J. T. (1998). Increased glutamatergic neurotransmission and oxidative stress after alcohol withdrawal. Am. J. Psychiatry 155, 726–732. Vescovi, P. P., and Coiro, V. (2001). Different control of GH secretion by gamma-amino- and gamma-hydroxy-butyric acid in 4-year abstinent alcoholics. Drug. Alcohol Depend. 61, 217–221. Vescovi, P. P., Volpi, R., and Coiro, V. (1998). Alcoholism abolishes the gamma-aminobutyric acid (GABA)ergic control of GH secretion in humans. Alcohol 16, 325–328. Walker, B. M., and Koob, G. F. (2007). The gamma-aminobutyric acid-B receptor agonist baclofen attenuates responding for ethanol in ethanoldependent rats. Alcohol. Clin. Exp. Res. 31(1), 11–18.
Growth hormone response to the GABA-B agonist baclofen in 3-week abstinent alcoholics Saliha Ozsoya,*, Ertugrul Esela, Tayfun Turana, Mustafa Kulab a Department of Psychiatry, Erciyes University School of Medicine, 38039 Kayseri, Turkey Department of Nuclear Medicine, Erciyes University School of Medicine, 38039 Kayseri, Turkey Received 19 May 2007; received in revised form 15 September 2007; accepted 17 September 2007 b
Abstract Gamma-aminobutyric acid (GABA) dysfunction is a known feature of alcoholism. We investigated GABA-B receptor activity in 3-week abstinent alcoholics using the growth hormone (GH) response to baclofen, a GABA-B receptor agonist. The study aimed to investigate the relationship between GABA-B receptor activity and alcohol withdrawal. GH response to baclofen was measured in alcohol-dependent males without depression (n 5 22) who were on day 21 of alcohol abstinence and in healthy control male subjects (n 5 23). After 20 mg baclofen was given orally to the subjects, blood samples for GH assay were obtained every 30 min for the subsequent 150 min. The patients were divided into two subgroups (continuing withdrawal and recovered withdrawal subgroups) according to their withdrawal symptom severity scores on day 21 of alcohol cessation. Baclofen administration significantly altered GH secretion in the controls, but not in the patients. When GH response to baclofen was assessed as DGH, it was lower in the patients with continuing withdrawal symptoms than in the controls and in the recovered withdrawal group. Impaired GH response to baclofen in all patients mainly pertained to the patients whose withdrawal symptoms partly continued. Our results suggest that reduced GABA-B receptor activity might be associated with longerterm alcohol withdrawal symptoms in alcoholic patients. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Gamma-aminobutyric acid; GABA-B receptor; Baclofen; Growth hormone; Alcohol withdrawal; Neuroendocrine
Introduction Gamma-aminobutyric acid (GABA) is an inhibitory amino acid neurotransmitter that plays a major role in central nervous system dysregulation during alcohol withdrawal. Hyperexcitation symptoms in alcohol withdrawal stem, in part, from a decreased central GABAergic inhibitory function probably caused by chronic alcohol consumption that leads to downregulation in GABA receptors (Cagetti et al., 2003). Animal and human studies confirm the decreased activity of GABA during ethanol withdrawal (Faingold et al., 2000; Tsai et al., 1998). The reduction in withdrawal symptoms with GABA agonists and worsening of symptoms with GABA antagonists also support the idea of decreased GABAergic function in alcohol withdrawal (Anton, 2001). Baclofen is a GABA-B receptor agonist, promising in the treatment of alcohol withdrawal and alcohol
* Corresponding author. Department of Psychiatry, Erciyes University School of Medicine, Talas Road, 38039 Kayseri, Turkey. Tel./fax: þ90352-4375702. E-mail address: [email protected] (S. Ozsoy). 0741-8329/07/$ e see front matter Ó 2007 Elsevier Inc. All rights reserved. doi: 10.1016/j.alcohol.2007.09.003
dependence (Addolorato et al., 2006; Colombo et al., 2000; Flannery et al., 2004). GABA-B receptor function in the central nervous system can be assessed using a neuroendocrine-challenge test with baclofen, which is considered to provide an index of GABA-B receptor function in humans. It is suggested that baclofen increases growth hormone (GH) release by inhibition of hypothalamic somatostatin release and/or by dopaminergic control (Davis et al., 1997; Koulu et al., 1979). Another mechanism of GH production stimulated by baclofen has been reported that in GH-producing cells, GABA acts as an autocrine factor via GABA-B receptors to control GH levels (GamelDidelon et al., 2002). Previously, Vescovi and Coiro (2001) and Vescovi et al. (1998) studied GABA-B activity by investigating GABAergic control of GH secretion in alcoholism. They reported that baclofen administration induced serum GH levels in the controls, but not in 2e4-week abstinent alcoholics, and suggested a loss of GABAergic inhibitory control in alcoholism (Vescovi et al., 1998). Another study by these authors investigated whether long-term abstinence restores this stimulatory effect of baclofen on GH secretion, and it was found that GH response to baclofen did not recover
552
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
in 4-year abstinent alcoholics (Vescovi and Coiro, 2001). They suggested that GABAergic dysfunction might be a consequence of chronic alcohol consumption resulting in irreversible damage or might be a trait marker of alcoholism. However, no clear relationship between GABA-B activity and alcohol withdrawal was revealed. It has been reported that withdrawal symptoms such as anxiety, insomnia, and mild autonomic overactivity are likely to continue for 2e6 months after the acute withdrawal has disappeared. These protracted withdrawal symptoms may enhance the probability of relapse (Schuckit, 2005). Under experimental conditions, it has also been shown that long-term behavioral changes and anxiety-like states persist in experimental animals for weeks after withdrawal (Rasmussen et al., 2001). The aim of the present study was to investigate probable alterations in GABA-B activity in abstinent alcoholics, particularly in patients whose withdrawal symptoms partly continue despite they being in the third week of alcohol cessation, measuring GH response to GABA-B agonist, baclofen. The results might increase our understanding of the relationship between GABA-B receptor activity and continuing withdrawal symptoms.
Methods Subjects Twenty-two male inpatients (mean age 6 S.D.: 42.04 6 7.93 years, range: 25e55 years) who met the Diagnostic and Statistical Manual of Mental Disorders,
4th edition, DSM-IV; (American Psychiatric Association, 1994) criteria for alcohol dependence and alcohol withdrawal were recruited for the study. The patients were selected from the inpatient population of the Psychiatry Clinic of Erciyes University Medical School. The DSM-IV diagnosis of alcohol withdrawal was based on information generated from clinical interviews. The patients were in the first day of withdrawal when they were admitted to hospital for the alcohol detoxification program. The patients stayed in hospital throughout the study. Exclusion criteria for patients were (1) a substance use disorder other than alcohol or nicotine, (2) any other DSM-IVaxis I disorder, current or past, (3) any significant medical and endocrine disorder, and/or (4) history of cirrhosis or laboratory evidence of significant liver disease. Twenty-three physically and mentally healthy men (mean age 6 S.D.: 37.00 6 8.97 years, range: 25e55 years) who were recruited from volunteers and hospital staff participated in the study as controls. Exclusion criteria for the controls were the same as those for the patients, plus a history of alcohol or drug use disorders in themselves or their families. All subjects were included in the study after careful evaluation of their laboratory test results and physical and psychiatric examinations and history. All patients and controls were smokers. Detailed characteristics of the subjects are given in Table 1. This study was carried out in accordance with the Helsinki Declaration of the World Medical Association and was approved by the local Ethics Committee. Written informed consent was obtained from each patient after the study was explained to them.
Table 1 Demographical and clinical characteristics of the patients and the controls Demographical and clinical variables
Patients (n 5 22) Mean 6 S.D.
Controls (n 5 23) Mean 6 S.D.
Comparisons
Age (year) BMI (kg/m2) Cigarette (number/day) Duration of smoking (year) Duration of consumption (year) Amount of alcohol consumed (g/day) No. of abstinence attemptsa MADRS score MAST score CIWA-Ar score (first day) CIWA-Ar score (21st day) AST (U/I) ALT (U/I) GGT (first day) (U/I) GGT (21st day) (U/I) T-blb (mg/dl) C-blb (mg/dl)
42.04 6 7.93 24.32 6 4.33 30.90 6 14.36 26.45 6 8.69 23.13 6 7.69 366.71 6 150.06 9.90 6 11.40 6.87 6 4.55 29.63 6 10.56 22.50 6 9.29 4.63 6 5.86 34.05 6 12.34* 36.64 6 18.47 68.50 6 34.25* 60.35 6 31.76* 1.10 6 0.31 0.41 6 0.11
37.00 6 8.97 26.25 6 3.23 24.09 6 11.36 21.77 6 10.69 d d d d d d d 23.36 6 9.31 29.09 6 19.39 26.72 6 10.78 26.72 6 10.78 1.05 6 0.48 0.36 6 0.13
t 5 1.99 t 5 1.70 t 5 1.37 t 5 1.08 d d d d d d d t 5 2.45 t 5 1.03 t 5 4.56 t 5 3.70 t 5 0.22 t 5 1.21
P O.05 P O.05 P O.05 P O.05 d d d d d d d P !.05 P O.05 P !.05 P !.05 P O.05 P O.05
n 5 subject number; BMI 5 body mass index; MADRS 5 MontgomeryeAsberg Depression Rating Scale; MAST 5 Michigan Alcoholism Screen Test; CIWA-Ar 5 revised Clinical Institute Withdrawal Assessment for Alcohol; AST 5 aspartate aminotransferase; ALT 5 alanine aminotransferase; GGT 5 g-glutamyl transpeptidase; T-blb 5 total bilirubin; C-blb 5 conjugated bilirubin; u/l 5 unit/liter. *Significantly higher than those of the controls. a The number of attempts made to abstain with or without treatment.
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
Procedures The patients were admitted to hospital on the first day of cessation. Routine biochemical examinations were performed on the first day of admission. The Michigan Alcoholism Screen Test (Selzer, 1971) was performed to assess the severity of alcoholism within the first week of admission. To measure and follow severity of withdrawal, the revised Clinical Institute Withdrawal Assessment for Alcohol (CIWA-Ar) (Sullivan et al., 1991) was used. The treatment dosage and duration were adjusted according to this scale and clinical symptomatology. The severity of the clinical symptomatology of depression was measured with the MontgomeryeAsberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979) on day 8 of admission. One week after withdrawal, none of the patients fulfilled the DSM-IV criteria of a depressive episode, and MADRS scores were below 15 in all patients. The patients received diazepam (mean dose 6 S.D.: 43.18 6 16.50 mg/day, range: 15e90 mg/day at the beginning of detoxification) and multivitamins for treatment of alcohol withdrawal symptoms. The patients did not receive any additional drug or nondrug therapies. Patients who were on day 21 of abstinence and the controls underwent a baclofen-challenge test. After an overnight fast, a venous catheter was inserted into an antecubital vein at 08:00 hours. A blood sample for basal GH was drawn just after cannula insertion (time 0). The subjects were allowed to rest but not to smoke or eat until the challenge procedure was completed. Blood samples for GH were taken at 30, 60, 90, 120, and 150 min after the administration of baclofen (Lioresal 20 mg p.o., NovartisÒ), which was given just after time 0. Blood samples were centrifuged and the separated serum was stored at 70 C until analysis. GH concentrations were determined by immunoradiometric assay (IRMA) techniques (DSL-1900, USA). For GH, the sensitivity was 0.03 mIU/ml; the intra-assay coefficient of variation was 3.1% for GH pool of 8.73 mIU/ml. The interassay coefficient of variation was 5.9% for GH pool of 1.02 mIU/ml. The patients were divided into two subgroups according to whether their withdrawal symptoms were ongoing on day 21 of abstinence. Patients with ongoing withdrawal symptoms according to their CIWA-Ar score were included in the continuing withdrawal group (n 5 8). Patients who had no withdrawal symptoms were included in the recovered withdrawal group (n 5 14). The threshold for withdrawal symptoms recovery was accepted as 7 in CIWA-Ar score. The mean CIWA-Ar score of the recovered withdrawal patient group was 1.21 6 1.42 (range: 0e4), and that of the continuing withdrawal patient group was 10.62 6 5.92 (range: 7e15).
Data analysis Maximum GH level out of 30e150 min was considered the ‘‘peak GH level’’. Peak GH level for subjects who did
553
not show increases above baseline was determined by using time 120 as a surrogate for peak values because this was the group peak time. GH responses (DGH) were calculated as the difference between baseline concentrations and maximal peak GH response after the challenge. The area under the curve (AUC; the serum concentration vs. time) was also calculated, for evaluating GH response to baclofen in the subjects. The distributions of all variables were checked by KolmogoroveSmirnov test. GH values had an abnormal distribution. To compare hormone levels of the patients and controls, a ManneWhitney U test was performed. A nonparametric Friedman test was preferred to repeated measure ANOVA for investigation of hormone response to baclofen, because GH values had an abnormal distribution. Spearman’s rank correlation test was performed to investigate the relationships between hormonal values and clinical and demographical variables in the patients.
Results There was no significant difference in age, body mass index, number of cigarettes consumed daily, or smoking duration between the patients and controls. The patients had higher serum aspartate aminotransferase and g-glutamyl transpeptidase levels (Table 1). Table 2 shows the comparison of patients’ hormonal values on day 21 of alcohol cessation with those of the controls. Basal GH level in the patients was higher than that in the controls (Z 5 4.727, P !.001). DGH, peak GH, and AUC GH values did not differ significantly between the patients and the controls. When GH response to baclofen was analyzed by Friedman test, a statistically significant increase over time was observed in the controls (c2 5 25.065, P !.001), but not in the patients (c2 5 4.564, P 5 .471) (Fig. 1). The GH response to baclofen measured as DGH in patients within withdrawal syndrome group was lower than that of the controls (Z 5 2.854, P 5.012) and of the recovered withdrawal group (Z 5 1.973, P 5 .042). Nevertheless, DGH of the recovered withdrawal group did not differ from that of the controls (Z 5 0.486, P 5.627). Basal GH levels of both the recovered withdrawal and continuing Table 2 GH values of the patients and the controls
Hormonal values
Patient group (n 5 22) Mean 6 S.D.
Control group (n 5 23) Mean 6 S.D. Comparisons
Basal GH (mIU/ml) 0.18 6 0.15* 0.06 6 0.03 DGH (mIU/ml) 0.55 6 1.45 0.53 6 0.84 Peak GH (mIU/ml) 0.74 6 1.46 0.59 6 0.84 AUC GH 51.07 6 94.10 31.40 6 40.62
Z 5 4.727 Z 5 1.602 Z 5 0.341 Z 5 0.488
GH 5 growth hormone; AUC 5 area under the curve. *Significantly higher than that of the controls.
P !.001 P 5.109 P 5.733 P 5.625
554
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
Fig. 1. Growth hormone response to baclofen in the patients and the controls.
withdrawal groups were higher compared to that of the controls (Z 5 4.168, P !.001; Z 5 3.404, P 5.001) (Table 3, Fig. 2). According to the nonparametric Friedman test, there was no statistically significant GH alteration in the baclofenchallenge test in the recovered withdrawal and continuing withdrawal groups (c2 5 6.882, P 5 .230; c2 5 7.355, P 5.196, respectively). Features of consumption of alcohol (duration, daily amount, and severity of alcoholism) and smoking (duration and daily number) did not differ between the continuing withdrawal and recovered withdrawal groups. CIWA-Ar scores of the continuing withdrawal group were higher than those of the recovered withdrawal group on the first and 21st days (Table 4). There was no significant correlation between hormonal measurements (basal GH, DGH, peak GH, and AUC GH) and features of alcohol consumption (duration, daily amount, and severity of alcohol use) in the patients. There were positive correlations between basal GH and peak GH, and AUC GH values (r 5 0.665, P 5 .001; r 5 0.668, P !.001, respectively), but no correlation between basal GH and DGH (r 5 0.257, P 5 .249).
Table 3 GH values of the patient subgroups when divided as ones whose withdrawal symptoms continue and ones whose do not on day 21 of alcohol cessation Patient group Mean 6 S.D. Hormonal values
Recovered withdrawal group (n 5 14)
Continuing withdrawal group (n 5 8)
Control group (n 5 23) Mean 6 S.D.
Basal GH (mIU/ml) DGH (mIU/ml) Peak GH (mIU/ml) AUC GH
0.16 6 0.08* 0.80 6 1.78 0.96 6 1.79 65.28 6 115.96
0.22 6 0.22* 0.11 6 0.27** 0.34 6 0.37 26.22 6 22.02
0.06 6 0.03 0.53 6 0.84 0.59 6 0.84 31.40 6 40.62
GH 5 growth hormone; AUC 5 area under the curve. *Higher than those of the controls. **Lower than that of the controls and the recovered withdrawal group.
Fig. 2. Growth hormone response to baclofen in the two patient subgroups and the controls.
Discussion Some patients in our study presented withdrawal symptoms after 3 weeks of abstinence and benzodiazepine treatment. Alcohol withdrawal symptoms are usually expected to diminish within several days after treatment with benzodiazepines (Mayo-Smith, 1997). However, we have been observing for years that some of our alcoholic patients keep presenting at least some withdrawal symptoms until the third or fourth week of alcohol cessation. We think that this might be a racial feature, because protracted alcohol withdrawal has also been frequently reported in Japanese people (Komiyama, 1997). As we expected, some patients in this study presented withdrawal symptoms, such as anxiety, autonomic symptoms, and tactile and visual disturbances, on day 21 of abstinence. It is known that some withdrawal symptoms persist for some months or longer as part of protracted withdrawal. The continuing withdrawal symptoms in this study should not be referred to as protracted withdrawal, because we did not follow-up the patients over a long abstinence period. However, it was interesting that some patients presented withdrawal symptoms, while others did not. The mechanism underlying this individual variation is not clear. The findings of the study might contribute to a better understanding of this matter. Our data showed that while there was a significant GH response to baclofen in the controls, this response was absent in 3-week abstinent alcoholics. This suggests that the GH-stimulating effect of baclofen is impaired in patients in alcoholic withdrawal. Looking at GH response in each patient subgroup separately, that is, as those having continuing withdrawal symptoms and those not, the GH response to baclofen (DGH) in patients who still had withdrawal symptoms in the third week of abstinence was lower than that in patients who had recovered from withdrawal syndrome. This means that impaired GH response to baclofen in all patients mainly pertained to the patients who were still in active withdrawal although they were in the third week of alcohol cessation. In our opinion, this finding is important because it may indicate to a direct
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
555
Table 4 Demographical and clinical characteristics of the patient subgroups Demographical and clinical variables
Recovered withdrawal group (n 5 14)
Continuing withdrawal group (n 5 8)
Comparisons
Age (year) BMI (kg/m2) Cigarette (number/day) Duration of smoking (year) Duration of alcohol consumption (year) Amount of alcohol consumed (g/day) MADRS score MAST score CIWA-Ar score (first day) CIWA-Ar score (21st day)
41.71 6 9.41 24.68 6 4.92 26.42 6 11.33 26.37 6 7.92 22.21 6 7.37 352.28 6 160.37 5.44 6 4.97 27.42 6 10.75 16.57 6 4.91 1.21 6 1.42
42.62 6 4.83 23.68 6 3.24 38.75 6 16.42 26.66 6 12.58 24.75 6 8.48 395.57 6 133.73 8.71 6 3.45 33.50 6 9.66 32.87 6 4.58* 10.62 6 5.92*
Z 5 0.13 Z 5 0.34 Z 5 1.81 Z 5 0.00 Z 5 0.68 Z 5 0.72 Z 5 1.38 Z 5 1.23 Z 5 3.82 Z 5 3.87
P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P O.05 P !.001 P !.001
BMI 5 body mass index; MADRS 5 MontgomeryeAsberg Depression Rating Scale; MAST 5 Michigan Alcoholism Screen Test; CIWA-Ar 5 revised Clinical Institute Withdrawal Assessment for Alcohol. *Higher than those of the recovered withdrawal group.
association between continuing alcohol withdrawal symptoms and reduced GABA-B receptor activity. A contributing factor to reduced GABA activity in alcohol withdrawal might be downregulated hypothalamic GABA-B receptors (Cagetti et al., 2003). Thus, one can consider that GABA-B receptor dysfunction may be one of the causes of the clinical picture of alcohol withdrawal, because responselessness to baclofen seems to be peculiar to the patients who had protracted withdrawal symptoms in the present study. Supporting this idea, some studies have implicated GABA-B receptor dysfunction in the pathophysiology of alcohol withdrawal. For instance, gammahydroxy-butyric acid (GHB), which interacts with GABAB receptor via a different mechanism from baclofen (Koek et al., 2007), has been reported to reduce substantially the symptoms of acute alcohol withdrawal (Korninger et al., 2003). Another clinical study showed the lack of effect of GHB on GH release in alcoholic patients, contrary to the increase in the release of GH observed in healthy subjects (Addolorato et al., 1999). Moreover, recent preclinical and clinical studies have suggested that baclofen suppresses symptoms of alcohol withdrawal syndrome (Addolorato et al., 2006; Colombo et al., 2000; Walker and Koob, 2007). On the other hand, impaired GABA-B receptor activity may be a reason for protracted withdrawal, rather than a consequence of chronic alcohol consumption or alcohol withdrawal. Patients who had more reduced GABA-B activity might be susceptible to the development of protracted withdrawal symptoms. Furthermore, protracted withdrawal may be a predictor for postedetoxification relapse (O’Connor et al., 1991). The finding that baclofen is an effective treatment in the prevention of relapse due to its ability to reduce craving in alcoholic patients also supports this consideration (Addolorato et al., 2002). Impaired GABAergic activity is probably responsible, in part, for alcohol craving and for withdrawal symptoms (De Witte, 2004). Considering all these findings together with our present finding, one may speculate that protracted withdrawal in alcoholic patients is associated with decreased GABA-B activity.
From a different perspective, blunted GH response to baclofen may reflect a heritable trait that may be a risk factor for alcoholism and it may not recover over time as some prior reports supported (Vescovi and Coiro, 2001; Vescovi et al., 1998). The finding of the present study that the patients had blunted GH response to baclofen is partly consistent with this idea. It has also been suggested that blunted GH response to baclofen may be related with comorbid anxiety disorders in alcohol-dependent patients (Condren et al., 2003; Gerra et al., 1998). Therefore, it may be considered that persistently elevated CIWA-Ar scores in the present study might be associated with a comorbid anxiety disorder in alcoholic patients rather than continuing alcohol withdrawal symptoms. However, we tried to eliminate confounding effect of anxiety by excluding the patients complicated by axis I psychopathologies such as anxiety disorders. One of the limitations of the present study was that all patients had received diazepam for treatment of alcohol withdrawal. The clinical effects of classical benzodiazepines such as diazepam are mediated by specific GABA-A receptor subtypes (Rudolph et al., 1999). For that reason, we do not consider that diazepam treatment directly affects GABA-B receptor activity, and this undermines our conclusions. The other limitation is the insufficient sample sizes of the patient subgroups. This might have caused a type II statistical error. Subjects were not able to match for smoking. However, features of smoking (duration and daily number) did not differ between the patients and the controls, and between the continuing withdrawal and recovered withdrawal subgroups. Finally, the lack of other clinical assessments such as anxiety was the other limitation. In conclusion, the present work suggests that GABA deficit during alcohol withdrawal may result from a dysfunction of GABA-B receptors and GABA-A receptors. GABA-B receptor dysfunction in alcoholic patients might be related to alcohol withdrawal symptoms per se, as they continue. However, this conclusion should be evaluated very cautiously because our results are preliminary and the study has important limitations as stated above.
556
S. Ozsoy et al. / Alcohol 41 (2007) 551e556
References Addolorato, G., Capristo, E., Gessa, G. L., Caputo, F., Stefanini, G. F., and Gasbarrini, G. (1999). Long-term administration of GHB does not affect muscular mass in alcoholics. Life Sci. 65(14), PL191–PL196. Addolorato, G., Caputo, F., Capristo, E., Domenicali, M., Bernardi, M., Janiri, L., Agabio, R., Colombo, G., Gessa, G. L., and Gasbarrini, G. (2002). Baclofen efficacy in reducing alcohol craving and intake: a preliminary double-blind randomized controlled study. Alcohol Alcohol. 37, 504–508. Addolorato, G., Leggio, L., Abenavoli, L., Agabio, R., Caputo, F., Capristo, E., Colombo, G., Gessa, G. L., and Gasbarrini, G. (2006). Baclofen in the treatment of alcohol withdrawal syndrome: a comparative study vs diazepam. Am. J. Med. 119, e13–e18. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Press. Anton, R. F. (2001). Pharmacologic approaches to the management of alcoholism. J. Clin. Psychiatry 62, 11–17. Cagetti, E., Liang, J., Spigelman, I., and Olsen, R. W. (2003). Withdrawal from chronic intermittent ethanol treatment changes subunit composition, reduces synaptic function, and decreases behavioral responses to positive allosteric modulators of GABA-A receptors. Mol. Pharmacol. 63, 53–64. Colombo, G., Agabio, R., Carai, M. A., Lobina, C., Pani, M., Reali, R., Addolorato, G., and Gessa, G. L. (2000). Ability of baclofen in reducing alcohol intake and withdrawal severity: IdPreclinical evidence. Alcohol. Clin. Exp. Res. 24, 58–66. Condren, R. M., Lucey, J. V., and Thakore, J. H. (2003). A preliminary study of baclofen-induced growth hormone release in generalised social phobia. Hum. Psychopharmacol. 18, 125–130. Davis, L. L., Trivedi, M., Choate, A., Kramer, G. L., and Petty, F. (1997). Growth hormone response to the GABAB agonist baclofen in major depressive disorder. Psychoneuroendocrinology 22, 129–140. De Witte, P. (2004). Imbalance between neuroexcitatory and neuroinhibitory amino acids causes craving for ethanol. Addict. Behav. 24, 1325–1339. Faingold, C., Li, Y., and Evans, M. S. (2000). Decreased GABA and increased glutamate receptor-mediated activity on inferior colliculus neurons in vitro are associated with susceptibility to ethanol withdrawal seizures. Brain Res. 868, 287–295. Flannery, B. A., Garbutt, J. C., Cody, M. W., Renn, W., Grace, K., Osborne, M., Crosby, K., Morreale, M., and Trivette, A. (2004). Baclofen for alcohol dependence: a preliminary open-label study. Alcohol. Clin. Exp. Res. 28, 1517–1523. Gamel-Didelon, K., Corsi, C., Pepeu, G., Jung, H., Gratzl, M., and Mayerhofer, A. (2002). An autocrine role for pituitary GABA: activation of GABA-B receptors and regulation of growth hormone levels. Neuroendocrinology 76, 170–177. Gerra, G., Ferri, M., Zaimovic, A., Giucastro, G., Palladino, M., Sartori, R., Delsignore, R., Maestri, D., Marzocchi, G., and Brambilla, F. (1998). GABAergic function in detoxified heroin addicts: relationship to anxiety disorders. Psychiatry Res. 77, 89–96. Koek, W., Mercer, S. L., and Coop, A. (2007). Cataleptic effects of gamma-hydroxybutyrate (GHB), its precursor gamma-butyrolactone (GBL), and GABA(B) receptor agonists in mice: differential
antagonism by the GABA (B) receptor antagonist CGP35348. Psychopharmacology (Berl.) 192(3), 407–414. Komiyama, T. (1997). Clinical features of alcoholics after alcohol withdrawaldwithdrawal syndrome, protracted withdrawal syndrome and prolonged disturbance of alcohol dependence. Nippon Rinsho 55, 393–405. Korninger, C., Roller, R. E., and Lesh, O. M. (2003). Gamma-hydroxybutyric acid in the treatment of alcohol withdrawal syndrome in patients admitted to hospital. Acta Med. Austriaca 30, 83–86. Koulu, M., Lammintausta, R., and Dahlstrom, S. (1979). Stimulatory effect of acute baclofen administration on human growth hormone and prolactin secretion in man. J. Clin. Endocrinol. Metab. 48, 1038–1040. Mayo-Smith, M. F. (1997). Pharmacological management of alcohol withdrawal. A meta-analysis and evidence-based practice guideline. American Society of Addiction Medicine Working Group on Pharmacological Management of Alcohol Withdrawal. JAMA 278(2), 144– 151. Montgomery, S. A., and Asberg, M. (1979). A new depression scale designed to be sensitive to change. Am. J. Psychiatry 134, 382–389. O’Connor, P. G., Gottlieb, L. D., Kraus, M. L., Segal, S. R., and Horwitz, R. I. (1991). Social and clinical features as predictors of outcome in outpatient alcohol withdrawal. J. Gen. Intern. Med. 6, 312–316. Rasmussen, D. D., Mitton, D. R., Green, J., and Puchalski, S. (2001). Chronic daily ethanol and withdrawal: 2. Behavioral changes during prolonged abstinence. Alcohol. Clin. Exp. Res. 25(7), 999–1005. Rudolph, U., Crestani, F., Benke, D., Brunig, I., Benson, J. A., Fritschy, J. M., Martin, J. R., Bluethmann, H., and Mohler, H. (1999). Benzodiazepine actions mediated by specific gamma-aminobutyric acid A receptor subtypes. Nature 401, 796–800. Schuckit, M. A. (2005). Alcohol-related disorders. In B. J. Sadock, & V. A. Sadock (Eds.), Kaplan and Sadock’s comprehensive textbook of psychiatry (pp. 1168–1188). New York, NY: Lippincott Williams & Wilkins. Selzer, M. L. (1971). The Michigan Alcoholism Screening Test: the quest for a new diagnostic instrument. Am. J. Psychiatry 127, 1653–1658. Sullivan, J. T., Swift, R. M., and Lewis, D. C. (1991). Benzodiazepine requirements during alcohol withdrawal syndrome: clinical implications of using a standardized withdrawal scale. J. Clin. Psychopharmacol. 11, 291–295. Tsai, G. E., Ragan, P., Chang, R., Chen, S., Linnoila, V. M., and Coyle, J. T. (1998). Increased glutamatergic neurotransmission and oxidative stress after alcohol withdrawal. Am. J. Psychiatry 155, 726–732. Vescovi, P. P., and Coiro, V. (2001). Different control of GH secretion by gamma-amino- and gamma-hydroxy-butyric acid in 4-year abstinent alcoholics. Drug. Alcohol Depend. 61, 217–221. Vescovi, P. P., Volpi, R., and Coiro, V. (1998). Alcoholism abolishes the gamma-aminobutyric acid (GABA)ergic control of GH secretion in humans. Alcohol 16, 325–328. Walker, B. M., and Koob, G. F. (2007). The gamma-aminobutyric acid-B receptor agonist baclofen attenuates responding for ethanol in ethanoldependent rats. Alcohol. Clin. Exp. Res. 31(1), 11–18.