[Predictors of prophylactic response to lithium].

INTRODUCTION: Since their first utilization in psychiatry as mood stabilizers in the 1940s, lithium salts have been widely studied in the medical literature. The considerable amount of data available to date, supports the use of lithium salts as first-line mood stabilizing agents, with acute antimanic and antidepressant properties and proven efficacy in the long term prevention of manic and depressive relapses.

LITERATURE FINDINGS: Several predictors were reported by different authors in early articles and were confirmed later on by the medical literature. All the psychopathological, environmental, biological, neurophysiologic and genetic predictors known to date are reviewed here. PSYCHOLOGICAL PREDICTORS: Psychopathological predictors of a good response to lithium prophylaxis include: the initial good response to lithium during the first 6-12 months of treatment, considered to date to be the most reliable predictor of a favourable response to lithium; the classical pattern of elated manic episodes; a positive familial history of bipolar disorders, especially those known to be responsive to lithium; the absence of comorbid personality disorders; bipolar type I disorders; melancholic features during depressive episodes; MDI pattern in the illness course and early onset of lithium treatment. In contrast, the following have been confirmed as psychopathological predictors of poor prophylactic lithium response: mixed episodes, considered to be one of the most reliable predictors of poor response to lithium since Kraeplin's description; rapid cycling bipolar disorders; comorbid alcohol and/or drug abuse; mood disorders with incongruent psychotic features; early onset bipolar disorder before the age of 18; discontinuation of lithium treatment; high number of previous affective episodes in the illness course before lithium initiation and DMI pattern. ENVIRONMENTAL PREDICTORS: Among environmental factors, being single was found to be the only predictor of a poor response to lithium treatment in prophylaxis. BIOLOGICAL PREDICTORS: Biological predictors of a good prophylactic response to lithium include a high RBC/plasma-lithium ratio, one of the most controversial predictors of a favourable response to lithium in the literature, a higher platelet serotonin-induced calcium mobilization, and a high rate of red blood cell membrane phospholipids, especially of phosphatidylcholine, and a phospholipid implicated in lithium intracellular transport. Among neurophysiologic predictors of a favourable response to lithium, the following have been reported: brain lithium concentrations above 0.2 mEq/L when measured by 7Li-MRS; decreased cerebral intracellular pH and white matter hyper intensity at (31)P-MRS and a high intensity of loudness dependence auditory-evoked potentials (LDAEP), the latter being one of the best indicators of human cerebral serotoninergic functioning. In contrast, the following have been reported as neurophysiological predictors of a poor lithium response in prophylaxis: epileptiform anomalies with diffuse theta waves on electroencephalography, a predictor of poor response to lithium known since the descriptions of Dalen in 1965 and decreased cerebral phosphocreatine levels at (31)P-MRS, the latter being an indicator of cerebral mitochondrial dysfunction. GENETIC PREDICTORS: Genetic predictors of good response to lithium in prophylaxis include a lower-inositol-monophosphatase (IMPase-2) mRNA expression, IMPase-2 being a key enzyme of the calcium-intracellular-signalling pathway and IMPase-2 gene being studied recently as a candidate gene in bipolar disorder. A higher frequency of phospholipase C isoenzyme gamma1 (PLCG1)-5 repeat allele genes has also been associated with a good response to lithium, PLCG1 being a major enzyme of the phosphatidylinositol second messenger system. Genetic predictors of negative prophylactic lithium response include the homozygotic forms of the short allele of the serotonin transporter gene (5-HTT), the presence of the A/A subtype of tryptophan hydroxylase (TPH) gene and a high frequency of human leukocyte antigens type A3 (HLA-A3), this genotype being associated with cellular membrane anomalies implicated in alteration of lithium intracellular transport.

DISCUSSION: The search for new predictors of lithium prophylactic response is currently facing several methodological problems: lack of representativity of the samples of bipolar patients enrolled in research studies, poor reliability of retrospective reconstructions of the course of the bipolar disorder before initiation of lithium treatment, absence of consensus on tools used to assess response to lithium prophylaxis in study designs, difficult access and high costs of most of the laboratory and neuroimaging techniques used in recent studies such as magnetic resonance spectroscopy and LDAEP measures, and problematic evaluation of the impact of treatment on a disorder whose natural intrinsic course is often irregular.