Citation for published version (APA): van Munster, B. C. (2009). Pathophysiological studies in delirium : a focus on genetics.

Transcription

1 UvA-DARE (Digital Academic Repository) Pathophysiological studies in delirium : a focus on genetics van Munster, B.C. Link to publication Citation for published version (APA): van Munster, B. C. (2009). Pathophysiological studies in delirium : a focus on genetics. General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 08 Apr 2019

2 Genetic polymorphisms related to delirium tremens, a systematic review Alcoholism: Clinical and Experimental Research 2007;31(2): Barbara C. van Munster 1,2, Johanna C. Korevaar 1, Sophia E. de Rooij 2, Marcel Levi 2, Aeilko H. Zwinderman Department of Clinical Epidemiology, Biostatistics and Bioinformatics, Academic Medical Center, Amsterdam, the Netherlands 2. Department of Medicine, Academic Medical Center, Amsterdam, the Netherlands

3 Chapter 6 Abstract Background: Delirium tremens (DT) is one of the more severe complications of Alcohol Withdrawal (AW), with a 5 to 10% lifetime risk for alcohol dependent patients. The two most important neurosystems involved in AW are γ aminobutyric acid and glutamate. It is unknown whether these neurosystems are involved in the pathophysiology of DT as well. The candidate gene approach in DT could contribute to this knowledge and demonstrate a possible genetic predisposition for DT. The purpose of this study is to give an overview of all studied genetic polymorphisms in the diverse candidate genes related to DT and to summarize what these studies contribute to insights into the pathophysiology of DT. Methods: The inclusion criteria for this literature study were articles in English analyzing the association between a genetic polymorphism and DT without other AW syndromes. Studies were identified until February 2006 in MEDLINE and EMBASE databases. Results: We found 25 studies dealing with 30 polymorphisms, located in 19 different genes. Positive associations were found in three different candidate genes involved in the dopamine transmission, one gene involved in the glutamate pathway, one neuropeptide gene and one cannabinoid gene. Two candidate genes involved in the dopamine transmission, dopamine receptor D3 and solute carrier family 6, were each associated with DT in two different study populations. The other four positive associations were not replicated in other studies. Conclusions: A total of 8 positive associations out of 30 polymorphisms makes a genetic base for DT plausible. Understanding the pathophysiological process of the development of DT has, indeed, been augmented by the reviewed genetic association studies. These studies suggest that the regulation of dopaminergic neurotransmission may play an important role. 60

4 Genetic polymorphisms related to delirium tremens Introduction Delirium Tremens (DT) is a serious complication of Alcohol Withdrawal (AW) with a mortality rate ranging from 1 to 5% 15. The spectrum of AW varies from minor symptoms as insomnia and tremor to severe complications such as withdrawal seizures and DT 15. The lifetime risk that an alcohol dependent individual will ever have a full blown DT ranges from 5 to 10% 173. The syndrome of DT is characterized by clouding of consciousness, mental confusion or disorientation and is often accompanied by hallucinations and agitation. Risk factors for developing DT include heavy daily alcohol use, concurrent acute medical illness, a history of DT or withdrawal seizures, older age, and abnormal liver function. With long term alcohol consumption, neuroreceptors in the brain affected by alcohol undergo adaptive changes in an attempt to maintain normal function 115. When alcohol consumption ceases, these changes are no longer adaptive and may contribute to the phenomena associated with AW. The two most important neurosystems involved in AW are γ aminobutyric acid (GABA) and glutamate 15,185. GABA is an inhibitory neurotransmitter which acts on the widespread GABA A neuroreceptor. Single doses of alcohol enhance the effect of GABA, resulting in overall decreased brain excitability. Chronic exposure to alcohol results in a compensatory decrease of GABA A receptor mediated response to GABA A. During AW, reduced GABA A receptor sensitivity and reduced brain GABA activity, are thought to contribute to symptoms of nervous system hyperactivity 141. The major excitatory neurotransmitter in the brain is glutamate, which communicates via three major subtypes of glutamate receptors (GR), whereof the N methyl D aspartate (NMDA) neuroreceptors are most sensitive to the effect of alcohol. Acute alcohol exposure inhibits NMDA neuroreceptors, whereas chronic alcohol exposure results in up regulation of these receptors 164. During AW not only the NMDA receptor function but also glutamate release is increased, leading to brain hyperexcitability 188. These alterations in glutamate function manifest clinically as anxiety, irritability, agitation and tremors, as seen in DT. Family, twin, and adoption studies have convincingly demonstrated that genes play an important role in the development of alcohol dependence 38. However, the genetic base of DT is never studied by linkage analysis probably because of the rare co occurrence of DT in the same family. Moreover, the overall genetics of this syndrome is most likely complex, with a relatively minor contribution to the relative risk of each possible genetic risk factor 130. The interaction between genetic and environmental factors leads to the development of the syndrome in genetically susceptible individuals. Many studies have addressed the roles of different polymorphisms in diverse candidate genes in relation to DT: the candidate gene approach. This candidate gene approach can demonstrate a possible genetic predisposition for DT. Apart from this, it provides a good opportunity to increase our knowledge of the pathophysiology of DT, as the pathogenetic mechanism is difficult to study directly in the human brain. Polymorphism, especially with proven 61

5 Chapter 6 functionality, can make a patient more susceptible to the development of DT when they are present in a gene region coding for a protein involved in the pathophysiology. An article by Schmidt et al., reviewed the candidate gene approach and their association studies in relation to AW 172. So far, no review has been published with a specific focus on DT. The purpose of the current study was to give an overview of all studied genetic polymorphisms in the diverse candidate genes in relation to DT. Our main research question was what these association studies contributed to the knowledge of the pathophysiology of DT. Materials and Methods Literature selection Studies were identified in February 2006 using the search strategy: 1. delirium or alcohol withdrawal in combination with 2. genetics or genetic or polymorphism This search was performed in MEDLINE ( ) and EMBASE databases ( ). EMBASE search expressions were slightly different from MEDLINE but the search terms were identical. The reference list of relevant papers was checked. In addition we searched for citations of the relevant studies and other articles of the first author. All titles were screened and a selection was made based on the abstract. Study selection The inclusion criteria for this literature study were articles in English analyzing the association between a previously known genetic polymorphism and DT without other AW syndromes. When duplicate articles were identified, considering the same polymorphism in the same study population, the article including the highest number of patients was selected. We excluded sequencing studies, screening for all possible polymorphisms in a candidate gene. All studies were assessed by two independent researchers (BM and JK). Results We identified 1022 studies, whereof 51 studies were retrieved for this review based on title and abstract. The excluded studies did not describe a genetic polymorphism in relation with AW. 29 studies were identified in which DT was analyzed separately from AW syndrome. We excluded four studies: one study was a duplicate article 139, two studies sequenced the whole candidate gene 135,137, and one study did not specify the exact data 163. This resulted in 25 studies included in the final analyses. The study of Okubo et al., determined an association in three different genes with DT 138. However, we excluded the association analysis with the NAD(P)H dehydrogenase 62

6 Genetic polymorphisms related to delirium tremens quinine 2 (NQO2) gene because of the multiple testing with 11 identified polymorphisms. For the genes, glutathione S transferase M1 (GSTM1) and NQO1, an association study was performed with one specific polymorphism, so these polymorphisms were included in our analysis. Wodarz et al. 207, studying a polymorphism in the dopamine receptor D2 (DRD2) gene and DRD3 gene, probably included a sample of the same study population studied by Johann et al. 83. In the latter study, the association between DT and the same polymorphism in the DRD2 gene was determined. We decided to exclude the results from the first study concerning the DRD2 polymorphism to prevent considering the same results twice. The variable number of tandem repeats (VNTR) polymorphism in the dopamine transporter gene (solute carrier family 6, member 3 (SLC6A3)) described by Sander et al. 167, was also determined in the study by Wernicke et al. 204, in the same patients. Although there were more patients in the latter study, we only included the first because of the more detailed description in this study. Another previously found polymorphism in the SLC6A3 gene, the Dral restriction site, was not polymorphic in the study of Wernicke et al. 204, and was left out of consideration. Table 1 shows the study characteristics of the 25 included studies dealing with, in total, 30 polymorphisms. All these studies had a case control design comparing alcoholic subjects with nonalcoholic subjects. The study population was mostly of mixed gender; one included only women and five included only men. Except for seven studies, the included patients had a Caucasian background. The frequency of DT varied between 0.11 and 0.60, with an absolute count of patients with DT between 9 and 165 (median 66). In ten studies, people without alcohol dependence were chosen as controls, in five studies patients with alcohol dependence but without DT, in five studies both these groups, and finally in five studies alcohol dependent patients with only mild withdrawal symptoms. A correction for multiple testing was done in six studies by the Bonferroni method. The 30 studied polymorphisms are described in detail in Table 2. There were 23 different polymorphisms. The authors mentioned that of the 30 described polymorphisms, 17 polymorphisms had proven functionality, five polymorphisms were silent and of eight polymorphisms, the functionality was unknown or not described. The authors reported polymorphisms located in 19 different genes (Table 3). Positive associations were found in three different candidate genes involved in the dopamine transmission, besides one association involved in the glutamate pathway, one in a neuropeptide gene, and one in a cannabinoid gene. There were 2 candidate genes each associated with DT in 2 different study populations. The first was the functional polymorphism in the DRD3 gene. One study did compare allele frequencies between the various samples of alcoholic subjects and controls of both men and women 20. The A allele frequency, coding for glycine (instead of the G allele coding for serine), was significantly higher in alcohol dependent individuals with DT compared with the nonalcoholic controls (79 vs. 68%, p=0.042). The second study compared frequencies of Gly/Gly genotypes in alcoholic subjects with DT to alcoholic subjects without DT, and did only found a significant difference in the male alcoholic 63

7 Chapter 6 subjects (14.9 vs. 7.1%, p<0.05) 207. The second candidate gene was the SLC6A3 gene. Sander et al., found an increased prevalence of the nine repeat allele of 54% in 59 alcoholic subjects with DT, compared with 32% in nonalcoholic controls (p=0.007) 167. In a more extended population of this same study group, a second polymorphism in the SLC6A3 gene, the G2319A polymorphism, was studied. They did not show a significant difference between alcoholic subjects with DT and controls without alcoholism 204. However, in a subgroup of individuals with VNTR homozygosity for the ten repeat allele, a higher prevalence of homozygosity A/A was found in patients with DT (9.4%) compared with nonalcoholic controls (2.2%; p=0.032). Genetic variants in brain derived neurotrophic factor (BDNF) 121, cannabinoid receptor 1 (CNR1) 171, GR ionotropic, kainate 3 (GRIK3) 148 and tyrosine hydroxylase (TH) 168 could possibly influence vulnerability to DT, as there was an association found in one study. Alcoholic subjects with a history of DT had significantly higher frequency of the AA genotype in the BDNF gene (21.6%), compared with alcoholic subjects without DT (13.1%) (p=0.008) 121. In the CNR1 gene, there was a significant excess of the A/A genotype found in patients with a history of DT (24%) compared with controls without alcohol dependence (11%) (p=0.031) 171. Alcohol dependent subjects who reported a history of DT were significantly more often carriers of the Ser310 allele in the GRIK3 gene compared with alcoholic subjects without DT 148. In the TH gene the frequency of the A10 allele was significantly increased in the alcoholic subjects with DT (3.2%) compared with nonalcoholic controls (0.5%) 168. No association with DT was found with polymorphisms in the GR subunit genes, ionotropic NMDA 1 (GRIN1) 162, ionotropic NMDA 2B (GRIN 2B) 182, GR metabotropic 7 (GRM7), and GR metabotropic 8 (GRM8) 146. The remaining polymorphisms showed no significant association with DT as well. Clarification Table 1 (next page): Study population: female (F); male (M) Diagnosis DT: DSM III R, Diagnostic and Statistical Manual of Mental Disorders III revised; DSM IV, Diagnostic and Statistical Manual of Mental Disorders IV; ICD 10, International Classification of Diseases Controls: I, nonalcoholic subjects; II, alcoholic subjects without DT; III, alcoholic subjects with mild alcohol withdrawal without DT. a % DT in alcoholic subjects with withdrawal syndrome b % DT in all participating alcoholic subjects c subanalysis was performed, apart from the total DT group d two different polymorphisms DT, delirium tremens; DRD1, dopamine receptor D1; SLC6A3, solute carrier family 6, member 3; COMT, catechol Omethyltransferase; NPY, neuropeptide Y; DBH, dopamine b hydroxylase; BDNF, brain derived neurotrophic factor; CCK, cholecystokinin; GSTM1, glutathion S transferase M1; NQO1, NAD(P)H dehydrogenase quinone 1; GRM7, glutamate receptor metabotropic 7; GRIK3, glutamate receptor ionotropic, kainate 3; CNR1, cannabinoid receptor 1; GRIN1, glutamate receptor, ionotropic, N methyl D aspartate 1; TH, tyrosine hydroxylase. 64

8 Genetic polymorphisms related to delirium tremens Table 1: Characteristics of studies included. Number of controls % DT in alcoholic subjects Number DT Diagnosis DT Study population Candidate gene Study (Cruz et al., 1995) 29 DRD2 Mexican, M (control DSM III R (I) F/M) (Ishiguro et al., 1999 a ) 81 COMT Japanese, F/M (I) (Ishiguro et al., 1999 b ) 82 SLC6A4 Japanese, F/M DSM IV a 290 (I) (Johann et al., 2004) 83 DRD2 German Caucasian, F/M (II) (Koehnke et al., 2002) 91 NPY German, F/M DSM IV (III) (Kohnke et al., 2002) 95 DBH German, F/M DSM IV (III) (Kohnke et al., 2003) 94 COMT German, F/M DSM IV (III) (Kohnke et al., 2005) 92 SLC6A3 German, F/M DSM IV (III) (Kohnke et al., 2006) 93 DBH German, F/M DSM IV (III) (Limosin et al., 2004) 111 SLC6A3 French Caucasian, F DSM III R (II) (Matsushita et al.,2001) 122 SLC6A4 Japanese, M (II) (Matsushita et al.,2004) 121 BDNF Japanese, M (II) (Okubo et al., 2000) 136 CCK Japanese, M (I) (Okubo et al., 2003) 138 GSTM1, NQO1 Japanese, M (I) (Preuss et al., 2002) 146 GRM7, GRM8 German Caucasian, F/M ICD (I), 145 (II) (Preuss et al., 2006) 148 GRIK3 German, F/M (I), 188 (II) (Preuss et al., 2003) 147 CNR1 German Caucasian, F/M ICD (I), 165 (II) (Rujescu et al., 2005) 162 GRIN1 German, F/M (I), 184 (II) (Sander et al., 1995) 166 DRD1 German Caucasian, F/M b 101 (I) DRD2 German Caucasian, F/M b 113 (I) DRD3 German Caucasian, F/M b 111 (I) c (Sander et al., 1997) 167 SLC6A3 German Caucasian, F/M (I) (Sander et al., 1998) 168 TH German, F/M (I) (Schmidt et al., 2002) 171 CNR1 German Caucasian, F/M (I) (Tadic et al., 2005) 182 GRIN2B d Caucasian, F/M (I), 239 (II) (Wernicke et al., 2002) 204 SLC6A3 German Caucasian, F/M (I) (Wodarz et al., 2003) 207 DRD3 German Caucasian, F/M (II) c 65

9 Chapter 6 Table 2: Characteristics of the included polymorphisms. Reference Asscociation DT Functionality Type Chromosome HGNC symbol Polymorphism 5 HTT prom 44bp ins/del SLC6A4 17q11.1 q12 Deletion y 82 5 HTT prom 44bp ins/del SLC6A4 17q11.1 q12 Deletion y 122 BDNF G196A BDNF 11p14.1 Nonsyn coding y CCK 45 C/T CCK 3pter p21 Non coding u 136 CNR1 G1359A CNR1 6q14 q15 Syn coding u 147 CNR1 G1359A CNR1 6q14 q15 Syn coding n COMT Val158Met COMT 22q11.21 q11.23 Nonsyn coding y 94 COMT Val158Met COMT 22q11.21 q11.23 Nonsyn coding y 81 DAT A9 SLC6A3 5p15.3 VNTR y DAT A9 SLC6A3 5p15.3 VNTR y 111 DAT A9 SLC6A3 5p15.3 VNTR y 92 DAT G2319A SLC6A3 5p15.4 Non coding u + a 204 DBH 1021C/T DBH 9q34 Non coding y 95 DBH 444 G/A DBH 9q34 Coding y 93 DRD1 Bsp 1286I RFLP DRD1 5q34 q35 Non coding u 166 DRD2 TaqA RFLP DRD2 11q22 q23 u 29 DRD2 TaqA RFLP DRD2 11q22 q23 Non coding u 166 DRD2 141CDel DRD2 11q22 q24 Deletion y 83 DRD3 BalI DRD3 3q13.3 Nonsyn coding y DRD3 BalI DRD3 3q13.3 Nonsyn coding y + a 207 GRIK3Ser310Ala GRIK3 1p34 p33 Nonsyn coding y GRIN1 G2108A GRIN1 9q34.3 Syn coding n 162 GRIN2B A3743G GRIN2B 12p12 Syn coding n 182 GRIN2B C2873T GRIN2B 12p12 Syn coding n 182 GSTM1 GSTM1 1p13.3 Deletion u 138 mglurr7 Tyr433Phe GRM7 3p26 p25 Nonsyn coding y 146 mglurr8 C2756T GRM8 7q31.3 q32.1 u 146 NPY T1128C NPY 7p15.3 n 91 NQO1 609 C>T NQO1 16q12 q22 Nonsyn coding y 138 TH A10 TH 11p15.5 VNTR y Type: type of polymorphism as noted in the article; VNTR, variable number of tandem repeats; Syn coding, synonymous in coding sequence; Nonsyn coding, nonsynonymous in coding sequence Functionality: y, functional polymorphism; u, functionality not mentioned; n, no proved functionality yet. a Positive association found in subgroup analysis BDNF, brain derived neurotrophic factor; CCK, cholecystokinin; CNR1, cannabinoid receptor 1; COMT, catechol O methyltransferase; DBH, dopamine b hydroxylase; DRD1, dopamine receptor D1; GRIK3, glutamate receptor ionotropic, kainate 3; GRIN1, glutamate receptor, ionotropic N methyl D aspartate 1; GSTM1, glutathion S transferase M1; NPY, neuropeptide Y; NQO1, NAD(P)H dehydrogenase quinone 1; TH, tyrosine hydroxylase 66

10 Genetic polymorphisms related to delirium tremens Table 3: Associations candidate genes with DT. Reference Neuro transmitter Association DT Number of polymorphisms Number of studies HGNC name Candidate gene CNR1 cannabinoid receptor Cannabionid 147, 171 GSTM1 glutathion S transferase M Detoxification enzyme 138 NQO1 NAD(P)Hdehydrogenase quinone Detoxification enzyme 138 DRD1 dopamine receptor D Dopamine 166 DRD2 dopamine receptor D Dopamine 29, 83, 166 DRD3 dopamine receptor D a Dopamine 166, 207 SLC6A3 solute carrier family 6, member a Dopamine 92, 111, 167, 204 COMT catechol O methyltransferase Dopamine/norepinephrine 81, 94 DBH dopamine beta hydroxylase Dopamine/norepinephrine 93, 95 TH tyrosine hydroxylase Dopamine/norepinephrine 168 GRIK3 glutamate receptor ionotropic, Glutamate 148 kainate 3 GRIN1 glutamate receptor, ionotropic, N Glutamate 162 methyl D aspartate 1 GRIN2B glutamate receptor, ionotropic, N Glutamate 182 methyl D aspartate 2B GRM7 glutamate receptor metabotropic Glutamate 146 GRM8 glutamate receptor metabotropic Glutamate 146 NPY neuropeptide Y Neuropeptide 91 CCK cholecystokinin Neuropeptide 136 BDNF brain derived neurotrophic factor Neuropeptide 121 SLC6A4 solute carrier family 6, member Serotonin 82, 122 a One of the found associations is in subgroup analysis Discussion Positive associations were found in three different candidate genes involved in the dopamine transmission, one gene involved in the glutamate pathway, one neuropeptide gene and one cannabinoid gene. The postulated pathophysiologic mechanism in AW, however, involves mainly the neurotransmitter systems GABA and glutamate. Preuss et al., found a positive association in the glutamate pathway and reported that alcohol 67

11 Chapter 6 dependent subjects with a history of DT were significantly more often carriers of the Ser310 allele in the GRIK3 gene 148. However, this finding was not confirmed in a study with a Polish population that studied DT and/or seizures 163. This discrepancy might be caused by differences in case mix, as the latter study also included patients without DT but with seizures. Despite the existence of polymorphisms in the GABA and glutamate systems, the diverse published association studies on DT mostly focused on polymorphisms in the dopamine transmission system, namely 10 of the 23 studied polymorphisms. An explanation for the high interest in the dopamine system might be that most of the studies were aimed at finding an association between alcohol dependence and the studied polymorphism. The dopaminergic system is critically involved in rewarding mechanisms that might mediate the reinforcing properties of alcohol, leading to alcohol dependence 97. However, it is to be expected that the individual risk for the occurrence of clinical entities, such as DT, might be quite different from the risk for the development of dependence. The associations between DT and two of the candidate genes involved in dopaminergic transmission, DRD3 and SLC6A3, were replicated. Moreover the studied polymorphisms in these candidate genes were both shown to be functional and may therefore be of clinical significance. The allelic variant of the DRD3 may effect protein insertion into the membrane 166. The dopamine transporters perform presynaptic dopamine reuptake and thus regulate extracellulair dopamine concentrations. The genetic constitution of the SLC6A3 gene affects availability of dopamine transporters in the striatum 69. This functionality makes a pathogenetic role in the development of DT more probable, as there is a proven change in the final protein product. The other positive results reported in the included studies need replication in larger samples to exclude spurious associations. The absence of an association between a candidate gene and DT does not exclude a role of the final gene product in the pathophysiology. The hypothesis that dopaminergic transmission influences DT is supported by the observation of Sano et al., who found significantly higher plasma levels of the dopamine metabolite homovanillic acid in alcoholic subjects with DT after withdrawal compared with patients whose consciousness was not clouded 169. During AW, a decreased GABAergic neurotransmission may disinhibit dopaminergic neurons and increase the severity of AW 70,164. The influence of dopamine on the severity of AW (i.e. DT and/or alcohol seizures) was also suggested by a tendency towards an association of the exon 8 A/A DRD2 genotype with more severe withdrawal symptoms 50. Glue et al., already suggested three different areas of AW, with hallucinations and DT as one of the distinct areas 58. They speculated over the role of dopamine in the pathophysiology of this specific area of the AW, based on the difference in treatment regimes. Haloperidol, a dopamine receptor (DR) antagonist, is an effective treatment for DT, in contrast with the little effect this drug has on other withdrawal syndromes. The total of 8 positive associations out of 30 polymorphisms makes a genetic base for DT plausible, although there may be an overestimation of positive associations because of publication bias of positive results. Heterogeneity in authors definitions of DT is one of 68

12 Genetic polymorphisms related to delirium tremens the main obstacles in identifying genes that produce a minor effect. The frequency of observed DT in the included studies varied substantially, depending on the different population studied, the administration and timing of treatment, but more importantly on the definition used. There is no golden standard for the diagnosis of DT besides definitions such as the Diagnostic and Statistical Manual of Mental Disorders criteria (DSM IV; American Psychiatric Association, 1994) or the International Classification of Diseases (ICD 10; World Health Organization, 1992). Unfortunately, most of the studies did not report the diagnostic criteria for DT. Moreover, frequently DT was diagnosed retrospectively using a questionnaire or based solely on symptoms in the medical chart. However, diagnosing DT without performing a thorough physical and psychiatric examination of the patient may be prone to misdiagnosis. For example hallucinations that are not connected with DT occur in 3 to 10 percent of patients during severe AW 143. Fulfilling the DSM IV criteria, the syndrome is still heterogeneous and a subdivision in subtypes is needed for a precise definition and description of phenotypic characteristics. Studies conducted with nonalcoholic subjects as a control group could possibly look at the association between the studied polymorphism and alcohol dependence. However, in the studies taking nonalcoholic subjects as control group, there was also no association found between the studied polymorphism and alcohol dependence. The most suitable control group seems to be patients with alcohol dependence without DT. Moreover, the power of the included studies was low, with a median number of cases of 66. These avoidable problems in power and choice of controls are frequently encountered in genetic association studies of alcoholism 23. Most risk factors for developing DT 1, among others severity and duration of alcohol abuse, were not described in the studies and could possibly have influenced the association between the studied polymorphism and DT. The only described risk factor was gender, as male alcohol dependent subjects tend to suffer from more AW symptoms than their female counterparts 37. Most study populations included more males than females, because the majority of alcohol dependent individuals are men. The study of Wodarz et al. 207, showed that gender is an important confounder that should be taken into account. The replication of their findings in the study of Sander et al. 166, could in this perspective, be explained by the high percentage of male patients (92%). Findings from genetic studies should be interpreted with caution, as the probability of finding an association is relatively high due to chance only. Lander et al., propose to use a more stringent p value than the usual At the same time, they warn not to be too cautious, as it runs the risk of missing true hints of associations. In this review most studies do not conform to this higher value for significance. Therefore, we need to be careful with a definite conclusion regarding the role of genetic polymorphisms in the pathophysiology of delirium. Yet, the findings regarding two candidate genes involved in the dopamine transmission (DRD3 and SLC6A3) were confirmed in a second sample, supporting a possible role of this neurotransmitter system 23. To draw a final conclusion, other studies, including an appropriate control for potential population differences, or including affected 69

13 Chapter 6 and non affected offspring are needed 172. Moreover, the candidate genes not involved in dopamine transmission should not be totally dismissed as potentially involved in the pathogenetic process of DT 103. Nevertheless, based on the results of our review with replication of findings in two candidate genes, we believe that the insights into the pathogenesis of DT may indeed be augmented by the performed genetic association studies, indicating a role for dopaminergic neurotransmission in the development of DT. Acknowledgements The authors wish to thank Onno C. Meijer of the Division of Medical Pharmacology from the Leiden University Medical Center for his helpful and critical comments. 70