- Email: [email protected]
Low serum cholesterol in violent but not in non-violent suicide attempters
Psychiatry Research 95 Ž2000. 103᎐108
Low serum cholesterol in violent but not in non-violent suicide attempters Jean-Claude Alvarez a,b,U , Didier Cremniter c , Nathalie Gluck d, Phillipe Quintin e, Marion Leboyer e, Ivan Berlin f , Patrice Therond b, Odile Spreux-Varoquaux b a
Faculte´ de Medecine Paris-Ouest et Laboratoire de Biochimie, Hopital R. Poincare, ´ ˆ ´ AP-HP, 104 B¨ d R. Poincare, ´ 92380 Garches, France b Faculte´ de Medecine Paris-Ouest et Departement de Biochimie-Pharmacologie-Toxicologie, Centre Hospitalier de Versailles, ´ ´ 177 rue de Versailles, 78157 Le Chesnay, France c Departement de Psychiatrie, Hopital H. Mondor, 51 A¨ enue du Marechal de Lattre de Tassigny, 94010 Creteil, ´ ˆ ´ ´ France d Faculte´ de Medecine Paris-Ouest et Departement de Psychiatrie, Centre Hospitalier de Versailles, 177 rue de Versailles, ´ ´ 78157 Le Chesnay, France e Psychiatrie, Groupe Hospitalier Pitie-Salpetriere, 75651 Paris Cedex 13, France ´ ˆ ` 47-83 B¨ d de l’Hopital, ˆ f Pharmacologie, Groupe Hospitalier Pitie-Salpetriere, 75651 Paris Cedex 13, France ´ ˆ ` 47-83 B¨ d de l’Hopital, ˆ Received 13 December 1999; received in revised form 10 April 2000; accepted 8 May 2000
Abstract Many previous studies have suggested that low or lowered serum cholesterol levels may increase the risk of mortality not due to somatic disease: principally, suicide and violent death. Because violent death is rare, some studies have investigated afterwards the relation between cholesterol levels and either suicide attempts in psychiatric populations or violence in criminally violent populations. However, none of these studies have compared cholesterol levels in violent and non-violent suicide attempters. The blood of 25 consecutive drug-free patients following a violent suicide attempt and of 27 patients following a non-violent suicide attempt by drug overdose was drawn in the 24 h following admission. Patients with a diagnosis of alcohol abuse and with cholesterol-lowering therapy were excluded. Age, sex, body mass index, psychiatric diagnosis and the physical conditions of the suicide attempt were investigated. Thirty-two healthy subjects were used as a control group. There were no differences between the groups in age, frequency of psychiatric diagnoses or body mass index. There was more women in the group of non-violent suicide attempters than in that of violent suicide attempters Ž P- 0.001.. In analyses controlling for sex and age, the serum cholesterol concentration was 30% lower Ž F2,82 s 15.8; P- 0.0001. in the group of violent suicide attempters Ž147 " 54 mgrdl. than in the group of non-violent suicide attempters Ž209 " 38 mgrdl. or control subjects Ž213 " 46 U
Corresponding author. Tel.: q33-1-47-10-79-38; fax: q33-1-47-10-79-23. E-mail address: [email protected] ŽJ.C. Alvarez..
0165-1781r00r$ - see front matter 䊚 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 7 1 - 2
104
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
mgrdl.. Our results showed that low serum cholesterol level is associated with the violence of the suicide attempt and not with the suicide attempt itself. Further investigations are necessary to determine the usefulness of this easily accessible parameter as a potential risk indicator for violent acts such as violent suicidal behavior in susceptible individuals. 䊚 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Violence; Suicide attempt; Cholesterol
1. Introduction There has been considerable controversy about a possible association between low serum cholesterol levels and increased mortality from suicide or violence. The background to this issue arises from randomized clinical trials of cholesterollowering interventions ŽMuldoon et al., 1990; Davey Smith and Pekkanen, 1992. and from observational cohort studies in which a follow-up was done on patients whose cholesterol levels had been measured some years earlier ŽJacobs et al., 1992; Lindberg et al., 1992; Neaton et al., 1992.. Because violent death is rare, many studies focused on patients with high rates of violence and have investigated the relationship between cholesterol levels and suicide attempts in psychiatric populations, or between cholesterol levels and violence in criminally violent subjects. Low cholesterol levels were found in many, but not all, studies among patients who had attempted suicide Žreview, Golomb, 1998., among patients with violent or aggressive conduct disorders ŽVirkkunen, 1979; Virkkunen and Penttinen, 1984., or among violent criminals with histories of aggression ŽSpitz et al., 1994., but not among patients with schizophrenia ŽSteinert et al., 1999.. In all the studies, violent and non-violent suicide attempters were considered either together or separately, and none of these studies compared cholesterol levels in violent and non-violent suicide attempters. Since low cholesterol levels have been associated with violence as well as with suicide attempts, we hypothesized that low cholesterol levels could be associated with violent suicide attempts and not with non-violent suicide attempts.
2. Subjects and methods 2.1. Subjects and clinical assessment None of the patients and control subjects had ever undergone a cholesterol-lowering pharmacological treatment. All of the patients were free of antidepressant medication at the time of the suicide attempt. The inpatients and control subjects were included if they provided written informed consent. The study was approved by the ethics committee of the Henri Mondor Hospital, Creteil, ´ France. 2.1.1. Patients Patients admitted to the intensive care unit following a violent suicide attempt, according to the criteria of Traskman et al. Ž1981. Ži.e. exclu¨ sion of drug overdose or superficial phlebotomy., and to the emergency ward following a suicide attempt by an overdose of benzodiazepines were eligible for the study. For all the patients, a clinical evaluation was performed by one trained psychiatrist as soon as possible Žbetween 1 and 3 days. after admission. Previous psychiatric history, medication, alcohol intake and history of suicide attempts were assessed, and a semi-structured interview was carried out in order to establish a DSM IV diagnosis. 2.1.2. Control subjects A group of 32 persons from the medical staff without any history of psychiatric illness and free of any previous suicide attempt were recruited as a control group. They were free of any medication and had never undergone cholesterol-lowering therapy.
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
105
Table 1 General characteristics and serum cholesterol levels of the violent, non-violent suicide attempters and control subjects a
Sex ratio ŽMrF. Age Žyears " S.D.. BMI Žkgrm2 . Psychiatric diagnosis Personality disorder Schizophrenia Mood disorder Physical condition Firearm Knife wound Jumping from a high place Drowning Caustic ingestion Getting run over by a train Benzodiazepine ingestion Serum cholesterol Žmgrdl.
Violent suicide attempters Ž n s 25.
Non-violent suicide attempters Ž n s 27.
Control subjects Ž n s 32.
Statistic; P
20r5 41.0" 13.9 24.9" 3.9
6r21 39.3" 12.3 23.5" 3.4
14r18 44.5" 14.5 23.1" 4.1
17.6; 0.001 1.15; 0.3 1.63; 0.2
4 2 19 11 2 7 3 1 1 0 147 " 54
2 0 25
0 0 0
0 0 0 0 0 0 27 209 " 38
᎐ ᎐ ᎐ ᎐ ᎐ ᎐ ᎐ 213 " 46
0.94; 0.33 2.25; 0.13 2.75; 0.10
15.8; 0.0001
a
For age, body mass index ŽBMI. and serum cholesterol, the values shown are mean " S.D., and the statistic is the F test with d.f.s 2.82. For sex ratio and psychiatric diagnosis, the values shown are the numbers and the statistic is the chi-square test with d.f.s 2.
2.2. Collection of samples and biochemical determination All of the patients and the control subjects fasted overnight. Samples of blood were drawn from the anterocubital vein at 09.00 h. For the suicidal patients, blood was collected within 24 h after the suicide attempt. The blood Ž5 ml. was collected into Vacutainer tubes with no additive for cholesterol determination. The serum was separated by centrifugation at 3000 = g at room temperature for 5 min, and stored at y 80⬚C until analysis. The cholesterol levels were determined by the Trinder Ž1969. method, using a Hitachi 917 analyzer. 2.3. Statistical analysis The frequencies of psychiatric diagnoses and gender were compared with chi-square tests and serum cholesterol levels, age and body mass index with analysis of variance ŽANOVA.. Because of a difference in the sex ratio between the groups, and because age could have an influence on cholesterol levels, these variables were included as covariates in the comparison of serum choles-
terol levels. Statistical analyses were performed with the statistical package SPSS.
3. Results Twenty-five violent suicide attempters, 27 nonviolent suicide attempters and 32 control subjects were included in the study. As shown in Table 1, there was a significant difference between the groups in the sex ratio, with more women in the group of non-violent suicide attempters than in the group of violent suicide attempters. There were no significant differences across the groups for age, frequency of psychiatric diagnoses, or body mass index. In analyses controlling for sex and age, the serum cholesterol concentration was found to be significantly lower Ž F2,82 s 15.8; P0.0001. in the group of violent suicide attempters Ž147 " 54 mgrdl, n s 25. than in the groups of non-violent suicide attempters Ž209 " 38 mgrdl; n s 27. or control subjects Ž213 " 46 mgrdl; n s 32.. To explore the effects of potential confounders, we fitted a linear regression model with serum
106
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
cholesterol as the dependent variable. The independent variables we controlled for were the groups Žviolent, non-violent suicide attempters and control subjects., sex, age and body mass index. The only significant association was between low serum cholesterol and violent suicide attempters Ž Ps 0.0001., with the probability values for all other variables ) 0.15. There was no significant difference between males and females for serum cholesterol Ž F1,83 s 2.07, Ps 0.15..
4. Discussion The main result of the present study is that violent suicide attempters had a marked reduction of serum cholesterol levels compared to control subjects, whereas non-violent suicide attempters have serum cholesterol levels similar to those of the control subjects as hypothesized a priori. This association remained statistically significant in analyses controlling for age and sex ratio differences across the groups. Different confounding factors may have biased our data. Confounding occurs when the observed association between two variables is due to the causal influence of a third variable. For example, it has been pointed out that a possible explanation for the relationship between low serum cholesterol and suicidal behavior is that depressed patients, suffering from a decrease in appetite which lowers blood cholesterol levels, are more liable to commit suicide because of mood, which is attributable to their illness but not to the low cholesterol ŽGoble and Worcester, 1992; Davey Smith and Shipley, 1993.. In our results, body mass index, which is a commonly used clinical parameter for nutritional status, was similar between the two groups of suicide attempters and control subjects, and was included as a covariate in the multivariate analysis. Therefore, the low serum cholesterol levels observed only in violent suicide attempters cannot be accounted for by loss of appetite. We did not include the affective status of the subjects in our model of multivariate analysis. The use of a depression rating as a covariate could be potentially
confounding, since no result allowed us to conclude that the presence of depressive disorder may lead to a decrease in total cholesterol levels. In our study, the same proportion of patients in the two groups of suicide attempters suffered from mood disorder Ž76% in violent suicide attempters and 88% in non-violent suicide attempters, Ps 0.2.. Moreover, in a recent study, Papassotiropoulos et al. Ž1999. reported that suicidal depressed patients have significantly lower cholesterol levels than non-suicidal depressed patients, showing that low cholesterol levels were not related to depressive status. Another confounding factor is that low cholesterol could be due to physical damage subsequent to a violent suicide attempt. However, Kunugi et al. Ž1997. demonstrated that the low total cholesterol observed in suicide attempters, more particularly after a violent suicide attempt, cannot be ascribed to any bias related to physical damage, since the authors reported low serum cholesterol levels in violent suicide attempters but not in patients who had accidental physical injury. A third potential confounding factor could be the sex ratio difference between the groups of violent and non-violent suicide attempters, as one study reported that the risk for violent death associated with low cholesterol was lower in women than that in men ŽLindberg et al., 1992., and one study showed a non-significant relationship in women ŽGolier et al., 1995.. However, some other studies showed an association between low cholesterol levels and suicide attempts in both men or women ŽModai et al., 1994; Sullivan et al., 1994; Gallerani et al., 1995.. There was no statistical difference between males and females in our results for serum cholesterol, and the relationship between low cholesterol and violent suicide attempters remained statistically significant Ž Ps 0.0001. when the effect of sex ratio was taken into account. A final confounding factor could be cancer, but none of our subjects Žpatients and control subjects. suffered from such a disease. The cholesterol levels found in our group of violent suicide attempters Ž147 " 54 mgrdl. were consistently lower than those in the two other groups and than the normal values of the laboratory Ž174᎐255 mgrdl.. This proportion of choles-
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
terol reduction Ž30%. is even higher than in other studies Žreview, Golomb, 1998.. In a previous study ŽAlvarez et al., 1999., we evaluated only 17 violent suicide attempters, and we concluded that decreased serum cholesterol levels may reflect more the dimension of suicidality andror violence than the dimension of impulsivity, which was assessed using the Impulsivity Rating Scale ŽLecrubier et al., 1995.. In the present study, our results seem to show that low serum cholesterol level is associated with the violence of the suicide attempt and not with suicidality. They could explain why different studies have not shown a relation between low serum cholesterol and suicide attempts, since many studies have investigated either violent and non-violent suicide attempters together or non-violent suicide attempters alone ŽSeefried and Gumpel, 1997.. Our results are in agreement with the previous studies of Virkkunen, in which a low cholesterol concentration was found among patients with violent or aggressive conduct disorder ŽVirkkunen, 1979; Virkkunen and Penttinen, 1984. and in homicidal offenders with histories of violence or habitually violent tendencies under the influence of alcohol ŽVirkkunen, 1983.. Our study gives support to the theory that low cholesterol levels are associated with the violence of the suicide attempt, but does not allow us to understand the biological mechanisms responsible for this association. Several studies in humans ŽRingo et al., 1994; Steegmans et al., 1996. and non-human primates ŽKaplan et al., 1991, 1994. suggest a specific relationship between low or lowered cholesterol levels and low or lowered serotonergic activity through modification of neuron membrane composition. Moreover, some data support a causal link between low or lowered brain serotonergic activity and violence ŽCoccaro, 1989; Brown and Linnoila, 1990.. Thus, a relationship between low serum cholesterol levels and increased violence, mediated by low serotonergic activity, is biologically possible. If this plausible causality is present in that relationship, low or lowered cholesterol could lead to violent behavior such as violent suicide attempts in susceptible individuals. Many questions remain about the relationship between low or lowered cholesterol
107
levels and violence, in particular why all the patients did not present this low or lowered cholesterol-associated violence. Future research should focus on the determination of biological or clinical markers which allow the identification of patients who are at risk for violent acts when their serum cholesterol levels are low or lowered by cholesterol-lowering therapy. It will also be important to investigate the usefulness of the serum cholesterol level as a potential risk indicator for violent suicidal behavior. With this object in mind, a follow-up of all violent suicide attempters is in progress, in order to establish whether subsequent suicide attempts are more frequent in those patients who presented lower blood cholesterol levels. Moreover, since recent data seem to show a decrease particularly in high-density lipoprotein cholesterol in suicide attempters ŽMaes et al., 1997., it would be important to determine the changes in serum lipid composition which could be related more precisely to violent suicide attempt. References Alvarez, J.C., Cremniter, D., Lesieur, P., Gregoire, A., Gilton, A., Macquin-Mavier, I., Jarreau, C., Spreux-Varoquaux, O., 1999. Low blood cholesterol and low platelet serotonin levels in violent suicide attempters. Biological Psychiatry 45, 1066᎐1069. Brown, G.L., Linnoila, M.I., 1990. CSF serotonin metabolite Ž5-HIAA. studies in depression, impulsivity, and violence. Journal of Clinical Psychiatry 51, 31᎐41. Coccaro, E.F., 1989. Central serotonin and impulsive aggression. British Journal of Psychiatry 155, 52᎐62. Davey Smith, G., Pekkanen, J., 1992. Should there be a moratorium on the use of cholesterol-lowering drugs? British Medical Journal 304, 431᎐434. Davey Smith, G., Shipley, M.J., 1993. Serum lipids and depression. Lancet 341, 434. Gallerani, M., Manfredini, R., Caracciolo, S., Scapoli, C., Molinari, S., Fersini, C., 1995. Serum cholesterol concentrations in parasuicide. British Medical Journal 310, 1632᎐1636. Goble, A.J., Worcester, M.C., 1992. Low serum cholesterol and violent death. British Medical Journal 305, 773. Golier, J.A., Marzuk, P.M., Leon, A.C., Weiner, C., Tardiff, K., 1995. Low serum cholesterol level and attempted suicide. American Journal of Psychiatry 152, 419᎐423. Golomb, B.A., 1998. Cholesterol and violence: is there a connection? Annals of Internal Medicine 128, 478᎐487.
108
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
Jacobs, D., Blackburn, H., Higgins, M., Reed, D., Iso, H., McMillan, G., Neaton, J., Nelson, J., Potter, J., Rifkind, B., 1992. Report of the conference on low blood cholesterol: mortality associations. Circulation 86, 1046᎐1060. Kaplan, J.R., Manuck, S.B., Shively, C., 1991. The effects of fat and cholesterol on social behavior in monkeys. Psychosomatic Medicine 53, 634᎐642. Kaplan, J.R., Shively, C.A., Fontenot, M.B., Morgan, T.M., Howell, S.M., Manuck, S.B., Muldoon, M.F., Mann, J.J., 1994. Demonstration of an association among dietary cholesterol, central serotonergic activity, and social behavior in monkeys. Psychosomatic Medicine 56, 479᎐484. Kunugi, H., Takei, N., Aoki, H., Nanko, S., 1997. Low serum cholesterol in suicide attempters. Biological Psychiatry 41, 196᎐200. Lecrubier, Y., Braconnier, A., Said, S., Payan, C., 1995. The impulsivity rating scale ŽIRS.: preliminary results. European Psychiatry 10, 331᎐338. Lindberg, G., Rastam, L., Gullberg, B., Eklund, G.A., 1992. Low serum cholesterol concentration and short-term mortality from injuries in men and women. British Medical Journal 305, 277᎐279. Maes, M., Smith, R., Christophe, A., Vandoolaeghe, E., Van Gastel, A., Neels, H., Demedts, P., Wauters, A., Meltzer, H.Y., 1997. Lower serum high-density lipoprotein cholesterol ŽHDL-C. in major depression and in depressed men with serious suicidal attempts: relationship with immuneinflammatory markers. Acta Psychiatrica Scandinavica 95, 212᎐221. Modai, I., Valevski, A., Dror, S., Weizman, A., 1994. Serum cholesterol levels and suicidal tendencies in psychiatric inpatients. Journal of Clinical Psychiatry 55, 252᎐254. Muldoon, M.F., Manuck, S.B., Matthews, K.A., 1990. Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials. British Medical Journal 301, 309᎐314. Neaton, J.D., Blackburn, H., Jacobs, D., Kuller, L., Lee, D.J., Sherwin, R., Shih, J., Stamler, J., Wentworth, D., 1992. Serum cholesterol level and mortality findings for men screened in the multiple risk factor intervention trial. Multiple risk factor intervention trial research group. Archives of Internal Medicine 152, 1490᎐1500.
Papassotiropoulos, A., Hawelleck, B., Frahnert, C., Rao, G.S., Rao, M.L., 1999. The risk of acute suicidality in psychiatric inpatients increases with low plasma cholesterol. Pharmacopsychiatry 32, 1᎐4. Ringo, D.L., Lindley, S.E., Faull, K.F., Faustman, W.O., 1994. Cholesterol and serotonin: seeking a possible link between blood cholesterol and CSF 5-HIAA. Biological Psychiatry 35, 957᎐959. Seefried, G., Gumpel, K., 1997. Low serum cholesterol and triglycerides and risk of death from suicide. Archives of Gerontology and Geriatrics 25, 111᎐117. Spitz, R.T., Hillbrand, M., Foster Jr., H.J., 1994. Serum cholesterol levels and frequency of aggression. Psychological Reports 74, 622. Steegmans, P.H., Fekkes, D., Hoes, A.W., Bak, A.A., Van der Does, E., Grobbee, D.E., 1996. Low serum cholesterol concentration and serotonin metabolism in men. British Medical Journal 312, 221. Steinert, T., Woelfle, M., Gebhardt, R.P., 1999. No correlation of serum cholesterol levels with measures of violence in patients with schizophrenia and non-psychotic disorders. European Psychiatry 14, 346᎐348. Sullivan, P.F., Joyce, P.R., Bulik, C.M., Mulder, R.T., OakleyBrowne, M., 1994. Total cholesterol and suicidality in depression. Biological Psychiatry 36, 472᎐477. ˚ Traskman, L., Asberg, M., Bertilsson, L., Sjostrand, L., 1981. ¨ ¨ Monoamine metabolites in CSF and suicidal behaviour. Archives of General Psychiatry 38, 631᎐636. Trinder, P., 1969. Determination of glucose in blood using glucose oxidase with an alternative oxygen acceptor. Annals of Clinical Biochemistry 6, 24. Virkkunen, M., 1979. Serum cholesterol in antisocial personality. Neuropsychobiology 5, 27᎐30. Virkkunen, M., 1983. Serum cholesterol levels in homicidal offenders. A low cholesterol is connected with habitually violent tendency under the influence of alcohol. Neuropsychobiology 10, 65᎐69. Virkkunen, M., Penttinen, H., 1984. Serum cholesterol in aggressive conduct disorder: a preliminary study. Biological Psychiatry 19, 435᎐439.
Low serum cholesterol in violent but not in non-violent suicide attempters Jean-Claude Alvarez a,b,U , Didier Cremniter c , Nathalie Gluck d, Phillipe Quintin e, Marion Leboyer e, Ivan Berlin f , Patrice Therond b, Odile Spreux-Varoquaux b a
Faculte´ de Medecine Paris-Ouest et Laboratoire de Biochimie, Hopital R. Poincare, ´ ˆ ´ AP-HP, 104 B¨ d R. Poincare, ´ 92380 Garches, France b Faculte´ de Medecine Paris-Ouest et Departement de Biochimie-Pharmacologie-Toxicologie, Centre Hospitalier de Versailles, ´ ´ 177 rue de Versailles, 78157 Le Chesnay, France c Departement de Psychiatrie, Hopital H. Mondor, 51 A¨ enue du Marechal de Lattre de Tassigny, 94010 Creteil, ´ ˆ ´ ´ France d Faculte´ de Medecine Paris-Ouest et Departement de Psychiatrie, Centre Hospitalier de Versailles, 177 rue de Versailles, ´ ´ 78157 Le Chesnay, France e Psychiatrie, Groupe Hospitalier Pitie-Salpetriere, 75651 Paris Cedex 13, France ´ ˆ ` 47-83 B¨ d de l’Hopital, ˆ f Pharmacologie, Groupe Hospitalier Pitie-Salpetriere, 75651 Paris Cedex 13, France ´ ˆ ` 47-83 B¨ d de l’Hopital, ˆ Received 13 December 1999; received in revised form 10 April 2000; accepted 8 May 2000
Abstract Many previous studies have suggested that low or lowered serum cholesterol levels may increase the risk of mortality not due to somatic disease: principally, suicide and violent death. Because violent death is rare, some studies have investigated afterwards the relation between cholesterol levels and either suicide attempts in psychiatric populations or violence in criminally violent populations. However, none of these studies have compared cholesterol levels in violent and non-violent suicide attempters. The blood of 25 consecutive drug-free patients following a violent suicide attempt and of 27 patients following a non-violent suicide attempt by drug overdose was drawn in the 24 h following admission. Patients with a diagnosis of alcohol abuse and with cholesterol-lowering therapy were excluded. Age, sex, body mass index, psychiatric diagnosis and the physical conditions of the suicide attempt were investigated. Thirty-two healthy subjects were used as a control group. There were no differences between the groups in age, frequency of psychiatric diagnoses or body mass index. There was more women in the group of non-violent suicide attempters than in that of violent suicide attempters Ž P- 0.001.. In analyses controlling for sex and age, the serum cholesterol concentration was 30% lower Ž F2,82 s 15.8; P- 0.0001. in the group of violent suicide attempters Ž147 " 54 mgrdl. than in the group of non-violent suicide attempters Ž209 " 38 mgrdl. or control subjects Ž213 " 46 U
Corresponding author. Tel.: q33-1-47-10-79-38; fax: q33-1-47-10-79-23. E-mail address: [email protected] ŽJ.C. Alvarez..
0165-1781r00r$ - see front matter 䊚 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 5 - 1 7 8 1 Ž 0 0 . 0 0 1 7 1 - 2
104
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
mgrdl.. Our results showed that low serum cholesterol level is associated with the violence of the suicide attempt and not with the suicide attempt itself. Further investigations are necessary to determine the usefulness of this easily accessible parameter as a potential risk indicator for violent acts such as violent suicidal behavior in susceptible individuals. 䊚 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Violence; Suicide attempt; Cholesterol
1. Introduction There has been considerable controversy about a possible association between low serum cholesterol levels and increased mortality from suicide or violence. The background to this issue arises from randomized clinical trials of cholesterollowering interventions ŽMuldoon et al., 1990; Davey Smith and Pekkanen, 1992. and from observational cohort studies in which a follow-up was done on patients whose cholesterol levels had been measured some years earlier ŽJacobs et al., 1992; Lindberg et al., 1992; Neaton et al., 1992.. Because violent death is rare, many studies focused on patients with high rates of violence and have investigated the relationship between cholesterol levels and suicide attempts in psychiatric populations, or between cholesterol levels and violence in criminally violent subjects. Low cholesterol levels were found in many, but not all, studies among patients who had attempted suicide Žreview, Golomb, 1998., among patients with violent or aggressive conduct disorders ŽVirkkunen, 1979; Virkkunen and Penttinen, 1984., or among violent criminals with histories of aggression ŽSpitz et al., 1994., but not among patients with schizophrenia ŽSteinert et al., 1999.. In all the studies, violent and non-violent suicide attempters were considered either together or separately, and none of these studies compared cholesterol levels in violent and non-violent suicide attempters. Since low cholesterol levels have been associated with violence as well as with suicide attempts, we hypothesized that low cholesterol levels could be associated with violent suicide attempts and not with non-violent suicide attempts.
2. Subjects and methods 2.1. Subjects and clinical assessment None of the patients and control subjects had ever undergone a cholesterol-lowering pharmacological treatment. All of the patients were free of antidepressant medication at the time of the suicide attempt. The inpatients and control subjects were included if they provided written informed consent. The study was approved by the ethics committee of the Henri Mondor Hospital, Creteil, ´ France. 2.1.1. Patients Patients admitted to the intensive care unit following a violent suicide attempt, according to the criteria of Traskman et al. Ž1981. Ži.e. exclu¨ sion of drug overdose or superficial phlebotomy., and to the emergency ward following a suicide attempt by an overdose of benzodiazepines were eligible for the study. For all the patients, a clinical evaluation was performed by one trained psychiatrist as soon as possible Žbetween 1 and 3 days. after admission. Previous psychiatric history, medication, alcohol intake and history of suicide attempts were assessed, and a semi-structured interview was carried out in order to establish a DSM IV diagnosis. 2.1.2. Control subjects A group of 32 persons from the medical staff without any history of psychiatric illness and free of any previous suicide attempt were recruited as a control group. They were free of any medication and had never undergone cholesterol-lowering therapy.
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
105
Table 1 General characteristics and serum cholesterol levels of the violent, non-violent suicide attempters and control subjects a
Sex ratio ŽMrF. Age Žyears " S.D.. BMI Žkgrm2 . Psychiatric diagnosis Personality disorder Schizophrenia Mood disorder Physical condition Firearm Knife wound Jumping from a high place Drowning Caustic ingestion Getting run over by a train Benzodiazepine ingestion Serum cholesterol Žmgrdl.
Violent suicide attempters Ž n s 25.
Non-violent suicide attempters Ž n s 27.
Control subjects Ž n s 32.
Statistic; P
20r5 41.0" 13.9 24.9" 3.9
6r21 39.3" 12.3 23.5" 3.4
14r18 44.5" 14.5 23.1" 4.1
17.6; 0.001 1.15; 0.3 1.63; 0.2
4 2 19 11 2 7 3 1 1 0 147 " 54
2 0 25
0 0 0
0 0 0 0 0 0 27 209 " 38
᎐ ᎐ ᎐ ᎐ ᎐ ᎐ ᎐ 213 " 46
0.94; 0.33 2.25; 0.13 2.75; 0.10
15.8; 0.0001
a
For age, body mass index ŽBMI. and serum cholesterol, the values shown are mean " S.D., and the statistic is the F test with d.f.s 2.82. For sex ratio and psychiatric diagnosis, the values shown are the numbers and the statistic is the chi-square test with d.f.s 2.
2.2. Collection of samples and biochemical determination All of the patients and the control subjects fasted overnight. Samples of blood were drawn from the anterocubital vein at 09.00 h. For the suicidal patients, blood was collected within 24 h after the suicide attempt. The blood Ž5 ml. was collected into Vacutainer tubes with no additive for cholesterol determination. The serum was separated by centrifugation at 3000 = g at room temperature for 5 min, and stored at y 80⬚C until analysis. The cholesterol levels were determined by the Trinder Ž1969. method, using a Hitachi 917 analyzer. 2.3. Statistical analysis The frequencies of psychiatric diagnoses and gender were compared with chi-square tests and serum cholesterol levels, age and body mass index with analysis of variance ŽANOVA.. Because of a difference in the sex ratio between the groups, and because age could have an influence on cholesterol levels, these variables were included as covariates in the comparison of serum choles-
terol levels. Statistical analyses were performed with the statistical package SPSS.
3. Results Twenty-five violent suicide attempters, 27 nonviolent suicide attempters and 32 control subjects were included in the study. As shown in Table 1, there was a significant difference between the groups in the sex ratio, with more women in the group of non-violent suicide attempters than in the group of violent suicide attempters. There were no significant differences across the groups for age, frequency of psychiatric diagnoses, or body mass index. In analyses controlling for sex and age, the serum cholesterol concentration was found to be significantly lower Ž F2,82 s 15.8; P0.0001. in the group of violent suicide attempters Ž147 " 54 mgrdl, n s 25. than in the groups of non-violent suicide attempters Ž209 " 38 mgrdl; n s 27. or control subjects Ž213 " 46 mgrdl; n s 32.. To explore the effects of potential confounders, we fitted a linear regression model with serum
106
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
cholesterol as the dependent variable. The independent variables we controlled for were the groups Žviolent, non-violent suicide attempters and control subjects., sex, age and body mass index. The only significant association was between low serum cholesterol and violent suicide attempters Ž Ps 0.0001., with the probability values for all other variables ) 0.15. There was no significant difference between males and females for serum cholesterol Ž F1,83 s 2.07, Ps 0.15..
4. Discussion The main result of the present study is that violent suicide attempters had a marked reduction of serum cholesterol levels compared to control subjects, whereas non-violent suicide attempters have serum cholesterol levels similar to those of the control subjects as hypothesized a priori. This association remained statistically significant in analyses controlling for age and sex ratio differences across the groups. Different confounding factors may have biased our data. Confounding occurs when the observed association between two variables is due to the causal influence of a third variable. For example, it has been pointed out that a possible explanation for the relationship between low serum cholesterol and suicidal behavior is that depressed patients, suffering from a decrease in appetite which lowers blood cholesterol levels, are more liable to commit suicide because of mood, which is attributable to their illness but not to the low cholesterol ŽGoble and Worcester, 1992; Davey Smith and Shipley, 1993.. In our results, body mass index, which is a commonly used clinical parameter for nutritional status, was similar between the two groups of suicide attempters and control subjects, and was included as a covariate in the multivariate analysis. Therefore, the low serum cholesterol levels observed only in violent suicide attempters cannot be accounted for by loss of appetite. We did not include the affective status of the subjects in our model of multivariate analysis. The use of a depression rating as a covariate could be potentially
confounding, since no result allowed us to conclude that the presence of depressive disorder may lead to a decrease in total cholesterol levels. In our study, the same proportion of patients in the two groups of suicide attempters suffered from mood disorder Ž76% in violent suicide attempters and 88% in non-violent suicide attempters, Ps 0.2.. Moreover, in a recent study, Papassotiropoulos et al. Ž1999. reported that suicidal depressed patients have significantly lower cholesterol levels than non-suicidal depressed patients, showing that low cholesterol levels were not related to depressive status. Another confounding factor is that low cholesterol could be due to physical damage subsequent to a violent suicide attempt. However, Kunugi et al. Ž1997. demonstrated that the low total cholesterol observed in suicide attempters, more particularly after a violent suicide attempt, cannot be ascribed to any bias related to physical damage, since the authors reported low serum cholesterol levels in violent suicide attempters but not in patients who had accidental physical injury. A third potential confounding factor could be the sex ratio difference between the groups of violent and non-violent suicide attempters, as one study reported that the risk for violent death associated with low cholesterol was lower in women than that in men ŽLindberg et al., 1992., and one study showed a non-significant relationship in women ŽGolier et al., 1995.. However, some other studies showed an association between low cholesterol levels and suicide attempts in both men or women ŽModai et al., 1994; Sullivan et al., 1994; Gallerani et al., 1995.. There was no statistical difference between males and females in our results for serum cholesterol, and the relationship between low cholesterol and violent suicide attempters remained statistically significant Ž Ps 0.0001. when the effect of sex ratio was taken into account. A final confounding factor could be cancer, but none of our subjects Žpatients and control subjects. suffered from such a disease. The cholesterol levels found in our group of violent suicide attempters Ž147 " 54 mgrdl. were consistently lower than those in the two other groups and than the normal values of the laboratory Ž174᎐255 mgrdl.. This proportion of choles-
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
terol reduction Ž30%. is even higher than in other studies Žreview, Golomb, 1998.. In a previous study ŽAlvarez et al., 1999., we evaluated only 17 violent suicide attempters, and we concluded that decreased serum cholesterol levels may reflect more the dimension of suicidality andror violence than the dimension of impulsivity, which was assessed using the Impulsivity Rating Scale ŽLecrubier et al., 1995.. In the present study, our results seem to show that low serum cholesterol level is associated with the violence of the suicide attempt and not with suicidality. They could explain why different studies have not shown a relation between low serum cholesterol and suicide attempts, since many studies have investigated either violent and non-violent suicide attempters together or non-violent suicide attempters alone ŽSeefried and Gumpel, 1997.. Our results are in agreement with the previous studies of Virkkunen, in which a low cholesterol concentration was found among patients with violent or aggressive conduct disorder ŽVirkkunen, 1979; Virkkunen and Penttinen, 1984. and in homicidal offenders with histories of violence or habitually violent tendencies under the influence of alcohol ŽVirkkunen, 1983.. Our study gives support to the theory that low cholesterol levels are associated with the violence of the suicide attempt, but does not allow us to understand the biological mechanisms responsible for this association. Several studies in humans ŽRingo et al., 1994; Steegmans et al., 1996. and non-human primates ŽKaplan et al., 1991, 1994. suggest a specific relationship between low or lowered cholesterol levels and low or lowered serotonergic activity through modification of neuron membrane composition. Moreover, some data support a causal link between low or lowered brain serotonergic activity and violence ŽCoccaro, 1989; Brown and Linnoila, 1990.. Thus, a relationship between low serum cholesterol levels and increased violence, mediated by low serotonergic activity, is biologically possible. If this plausible causality is present in that relationship, low or lowered cholesterol could lead to violent behavior such as violent suicide attempts in susceptible individuals. Many questions remain about the relationship between low or lowered cholesterol
107
levels and violence, in particular why all the patients did not present this low or lowered cholesterol-associated violence. Future research should focus on the determination of biological or clinical markers which allow the identification of patients who are at risk for violent acts when their serum cholesterol levels are low or lowered by cholesterol-lowering therapy. It will also be important to investigate the usefulness of the serum cholesterol level as a potential risk indicator for violent suicidal behavior. With this object in mind, a follow-up of all violent suicide attempters is in progress, in order to establish whether subsequent suicide attempts are more frequent in those patients who presented lower blood cholesterol levels. Moreover, since recent data seem to show a decrease particularly in high-density lipoprotein cholesterol in suicide attempters ŽMaes et al., 1997., it would be important to determine the changes in serum lipid composition which could be related more precisely to violent suicide attempt. References Alvarez, J.C., Cremniter, D., Lesieur, P., Gregoire, A., Gilton, A., Macquin-Mavier, I., Jarreau, C., Spreux-Varoquaux, O., 1999. Low blood cholesterol and low platelet serotonin levels in violent suicide attempters. Biological Psychiatry 45, 1066᎐1069. Brown, G.L., Linnoila, M.I., 1990. CSF serotonin metabolite Ž5-HIAA. studies in depression, impulsivity, and violence. Journal of Clinical Psychiatry 51, 31᎐41. Coccaro, E.F., 1989. Central serotonin and impulsive aggression. British Journal of Psychiatry 155, 52᎐62. Davey Smith, G., Pekkanen, J., 1992. Should there be a moratorium on the use of cholesterol-lowering drugs? British Medical Journal 304, 431᎐434. Davey Smith, G., Shipley, M.J., 1993. Serum lipids and depression. Lancet 341, 434. Gallerani, M., Manfredini, R., Caracciolo, S., Scapoli, C., Molinari, S., Fersini, C., 1995. Serum cholesterol concentrations in parasuicide. British Medical Journal 310, 1632᎐1636. Goble, A.J., Worcester, M.C., 1992. Low serum cholesterol and violent death. British Medical Journal 305, 773. Golier, J.A., Marzuk, P.M., Leon, A.C., Weiner, C., Tardiff, K., 1995. Low serum cholesterol level and attempted suicide. American Journal of Psychiatry 152, 419᎐423. Golomb, B.A., 1998. Cholesterol and violence: is there a connection? Annals of Internal Medicine 128, 478᎐487.
108
J.C. Al¨ arez et al. r Psychiatry Research 95 (2000) 103᎐108
Jacobs, D., Blackburn, H., Higgins, M., Reed, D., Iso, H., McMillan, G., Neaton, J., Nelson, J., Potter, J., Rifkind, B., 1992. Report of the conference on low blood cholesterol: mortality associations. Circulation 86, 1046᎐1060. Kaplan, J.R., Manuck, S.B., Shively, C., 1991. The effects of fat and cholesterol on social behavior in monkeys. Psychosomatic Medicine 53, 634᎐642. Kaplan, J.R., Shively, C.A., Fontenot, M.B., Morgan, T.M., Howell, S.M., Manuck, S.B., Muldoon, M.F., Mann, J.J., 1994. Demonstration of an association among dietary cholesterol, central serotonergic activity, and social behavior in monkeys. Psychosomatic Medicine 56, 479᎐484. Kunugi, H., Takei, N., Aoki, H., Nanko, S., 1997. Low serum cholesterol in suicide attempters. Biological Psychiatry 41, 196᎐200. Lecrubier, Y., Braconnier, A., Said, S., Payan, C., 1995. The impulsivity rating scale ŽIRS.: preliminary results. European Psychiatry 10, 331᎐338. Lindberg, G., Rastam, L., Gullberg, B., Eklund, G.A., 1992. Low serum cholesterol concentration and short-term mortality from injuries in men and women. British Medical Journal 305, 277᎐279. Maes, M., Smith, R., Christophe, A., Vandoolaeghe, E., Van Gastel, A., Neels, H., Demedts, P., Wauters, A., Meltzer, H.Y., 1997. Lower serum high-density lipoprotein cholesterol ŽHDL-C. in major depression and in depressed men with serious suicidal attempts: relationship with immuneinflammatory markers. Acta Psychiatrica Scandinavica 95, 212᎐221. Modai, I., Valevski, A., Dror, S., Weizman, A., 1994. Serum cholesterol levels and suicidal tendencies in psychiatric inpatients. Journal of Clinical Psychiatry 55, 252᎐254. Muldoon, M.F., Manuck, S.B., Matthews, K.A., 1990. Lowering cholesterol concentrations and mortality: a quantitative review of primary prevention trials. British Medical Journal 301, 309᎐314. Neaton, J.D., Blackburn, H., Jacobs, D., Kuller, L., Lee, D.J., Sherwin, R., Shih, J., Stamler, J., Wentworth, D., 1992. Serum cholesterol level and mortality findings for men screened in the multiple risk factor intervention trial. Multiple risk factor intervention trial research group. Archives of Internal Medicine 152, 1490᎐1500.
Papassotiropoulos, A., Hawelleck, B., Frahnert, C., Rao, G.S., Rao, M.L., 1999. The risk of acute suicidality in psychiatric inpatients increases with low plasma cholesterol. Pharmacopsychiatry 32, 1᎐4. Ringo, D.L., Lindley, S.E., Faull, K.F., Faustman, W.O., 1994. Cholesterol and serotonin: seeking a possible link between blood cholesterol and CSF 5-HIAA. Biological Psychiatry 35, 957᎐959. Seefried, G., Gumpel, K., 1997. Low serum cholesterol and triglycerides and risk of death from suicide. Archives of Gerontology and Geriatrics 25, 111᎐117. Spitz, R.T., Hillbrand, M., Foster Jr., H.J., 1994. Serum cholesterol levels and frequency of aggression. Psychological Reports 74, 622. Steegmans, P.H., Fekkes, D., Hoes, A.W., Bak, A.A., Van der Does, E., Grobbee, D.E., 1996. Low serum cholesterol concentration and serotonin metabolism in men. British Medical Journal 312, 221. Steinert, T., Woelfle, M., Gebhardt, R.P., 1999. No correlation of serum cholesterol levels with measures of violence in patients with schizophrenia and non-psychotic disorders. European Psychiatry 14, 346᎐348. Sullivan, P.F., Joyce, P.R., Bulik, C.M., Mulder, R.T., OakleyBrowne, M., 1994. Total cholesterol and suicidality in depression. Biological Psychiatry 36, 472᎐477. ˚ Traskman, L., Asberg, M., Bertilsson, L., Sjostrand, L., 1981. ¨ ¨ Monoamine metabolites in CSF and suicidal behaviour. Archives of General Psychiatry 38, 631᎐636. Trinder, P., 1969. Determination of glucose in blood using glucose oxidase with an alternative oxygen acceptor. Annals of Clinical Biochemistry 6, 24. Virkkunen, M., 1979. Serum cholesterol in antisocial personality. Neuropsychobiology 5, 27᎐30. Virkkunen, M., 1983. Serum cholesterol levels in homicidal offenders. A low cholesterol is connected with habitually violent tendency under the influence of alcohol. Neuropsychobiology 10, 65᎐69. Virkkunen, M., Penttinen, H., 1984. Serum cholesterol in aggressive conduct disorder: a preliminary study. Biological Psychiatry 19, 435᎐439.