Understanding and Preventing Violence, Volume 2: Biobehavioral Influences (1994)
Chapter: Lithium
given chronically to rats, may actually increase aggressive-defensive responses in reaction to noxious stimuli such as pain or sleep deprivation (see Table 8A; e.g., Eichelman and Barchas, 1975; Carlini et al., 1976; Prasad and Sheard, 1982, 1983a,b; Willner et al., 1981; Mogilnicka et al., 1983). The increased sensitivity to aggression-provoking aversive events during chronic treatment with antidepressants in mice and rats may be based on regulatory changes at noradrenergic or serotonergic receptors (e.g., Eichelman, 1979; Sheard, 1981). It is conceivable that noradrenergic receptor hypersensitivity may be responsible for the occasional instances of so-called paradoxical rage in human patients (e.g., Pallmeyer and Petti, 1979; Rampling, 1978; Tupin, 1985).
Similar to the overall pattern of results with animal preparations, in clinical studies of inpatients and outpatients, imipramine and similar antidepressants are of poor and inconsistent efficacy in reducing aggressive behavior (e.g., Itil and Seaman, 1978; see Table 8B). In a small number of obsessive-compulsive patients, clomipramine reduces aggressive thoughts (Rapoport, 1989). The most clear effects on aggressive behavior by imipramine and amitriptyline are observed in hyperactive aggressive children (see Table 8A; e.g., Winsberg et al., 1972; Yepes et al., 1977; Puig-Antich, 1982). In this group of patients, tricyclic antidepressants may be a suitable alternative to the more commonly used methylphenidate-type medications.
Lithium
More than 40 years ago, Cade (1949) recommended lithium as a beneficial treatment to ''control … restless impulses and ungovernable tempers" in preference over prefrontal leukotomy! There is remarkably consistent evidence for the selective antiaggressive effects of lithium from experimental studies in several animal preparations and from psychiatric inpatient studies with various diagnoses (see Tables 9A and 9B). The most relevant evidence comes from experimental studies in violent prisoners that show long-lasting decrements in assaultive episodes while being maintained on lithium (e.g., Sheard, 1983, 1988).
The experimental evidence from animal studies delineates several important characteristics of lithium's effects on aggressive behavior (see Table 9A). (1) The suppression of aggressive behavior is most reliable when lithium is given chronically such as in drinking water or food. (2) It is essential to monitor lithium in the blood to define the therapeutic window for a given individual.
Very low doses of lithium either are ineffective or can actually increase certain aggressive responses, and high doses induce renal and thyroid toxicity (e.g., Ozawa et al., 1975; Broderick and Lynch, 1982; Sovner and Hurley, 1981; Glenn et al., 1989; Craft et al., 1987). (3) Lithium is particularly effective in reducing attack behavior by isolated mice; threats and attacks by resident mice, hamsters, or fish toward an intruder; defensive responses to noxious stimuli; and irritability due to brain injury (see Table 9A; e.g., Weischer, 1969; Sheard, 1973; Eichelman et al., 1977; Oehler et al., 1985a). However, other types of aggressive behavior, such as maternal aggression and predatory killing, remain unaltered by lithium, even when given chronically (e.g., Brain and Al-Maliki, 1979).
The animal studies also highlight important limitations of the use of lithium in the management of violent and aggressive individuals. The potential for renal and thyroid toxicity at too high lithium doses requires close monitoring of adequate dosing and blood levels. Lithium may induce nausea that is readily conditioned to features of the environment surrounding the drug administration. For example, conditioned taste aversion, initially demonstrated in laboratory rats, inhibits coyotes from attacking sheep after they have consumed as little as a single meal of lithium-laced sheep meat (e.g., Gustavson et al., 1974; Krames et al., 1973; O'Boyle et al., 1973). Lithium-induced nausea may be a further reason for the poor compliance with this type of medication in outpatients.
There are numerous demonstrations of antiaggressive effects in lithium responders among institutionalized individuals with diagnoses ranging from mental retardation to epilepsy, psychosis, and antisocial personality (see Table 9B; e.g., Dostal and Zvolsky, 1970; Tupin, 1972; Goetzl et al., 1977; Dale, 1980; Craft et al., 1987; Glenn et al., 1989; Luchins and Dojka, 1989). The most convincing evidence for lithium's effectiveness as an antiaggressive medication was gathered in violent prisoners (see Table 9B; Sheard, 1971, 1977b; Sheard and Marini, 1978; Marini and Sheard, 1976, 1977; Tupin et al., 1973). Sheard's initial placebo-controlled study on 12 male volunteer prisoners in whom lithium given three times a day decreased assaultive behavior and verbal hostility as assessed by the prison staff was confirmed in a larger double-blind study with 66 highly aggressive prisoners (Sheard et al., 1976). In both a Connecticut and a California sample of violent convicts, lithium achieved a near elimination of aggressive feelings and violent behavior over 3-18 months of treatment (Tupin et al., 1973;
