• long-term treatment
• relapses
• Bipolar Disorder

Bipolar disorder (BD) is a chronic mood disorder, despite modern pharmacological therapeutic strategies more than half of all bipolar patients relapse within 2 years, with over 90% of bipolar patients having experienced at least 1 additional affective episode during their lifetime (Perlis et al., 2006). Many studies corroborate the importance to investigate and treat predictors of recurrences in bipolar-II and bipolar-I patients such as substance abuse or dependence, the presence of anxiety symptoms or of a comorbid eating disorder and residual mood symptoms early in recovery whose presence appears to increase the risk for depressive recurrence (Perlis et al., 2006; Judd et al., 2002).
Significant predictors of relapses in BD patients, are also previous mixed episodes and studies showed for example that early BD mixed-states could be followed by particularly severe later morbidity (Berk, Dodd et al. 2005, Vieta 2005, Gonzalez-Pinto, Aldama et al. 2007).
Management of bipolar patients is often complicated because of the risk of serious recurrences and very low percentage of patients who remain in remission (bipolar I patients: 28% for 4 years and about 10% for 5 years) (Goodwin, Jamison, 2007; Keller et al., 1993, Tohen et al., 1990).
The main goal of maintenance treatment of bipolar disorder is stabilizing mood and reducing inter-episodic symptoms. Lithium salts are the gold standard for acute and long-term management of bipolar disorder because of its prophylactic and antisuicidal properties. The aim of our prospective naturalistic study was to investigate the presence of a relation among age of onset, polarity of onset, previous clinical course, comorbidity and high frequency of recurrences in a sample of 266 patients and examine the clinical characteristics and the course of illness of a subgroup of 234 BD-I or BD-II outpatients treated with lithium salts or anti-epileptics drugs or with a combination of both. We studied a possible relationship between the kind of stabilization utilized and the characteristics of illness. Patients have been followed for a period of at least 8 months at Section of Psychiatry, Department of Clinical and Experimental Medicine, University of Pisa and at the Istituto di Psicopatologia in Rome. Patient enrollment criteria were age 18-65 years and meeting DSM IV criteria for BD-I or BD-II or Koukopoulos’s criteria for mixed attenuated episodes. We used the SCID-I / P, SIMD and the LIFE scale to evaluate the diagnostic, clinical and therapeutic aspects. We divided the 234 patients into three groups according to the different treatments utilized for long term stabilization of Bipolar disorder: 51 patients with Lithium only, 56 patients stabilized with anticonvulsant drugs and 127 patients stabilized with a combination of Lithium and anticonvulsant drugs. Our study highlights that Lithium and antiepileptic drugs combination was the most common treatment strategy used to stabilize patients in the sample.
From our findings, gender female, age, duration of follow-up, total mean number of previous episodes, frequency of past mixed episodes, resulted related to risk of recurrences..
Furthermore, patients taking lithium and anti-epileptics combination were more frequently affected by BD-I, had higher rates of mixed episodes in retrospective and prospective course, higher rates of lifetime psychotic symptoms, and showed higher rates of concomitant atypical antipsychotic prescription. Patients stabilized with anti-epileptics or lithium alone were more frequently BD-II, had higher rate of comorbidity and higher antidepressants concomitant prescription.
No significant difference emerged according to the duration of the follow up period among the three groups of treatment, so that we could suppose the long-term tolerance of lithium associated with anti-epileptic drugs was similar to the tolerability expected by using a single mood stabilizer. Overall, in our study, patients on a combination therapy, had evidenced the major improvement measured as the difference between the medium frequency of relapses in retrospective and prospective course, rather than the other patients treated on a single mood stabilizer.
Several limitations in the present study should be considered.
The most important are its small sample, the lack of randomization and the absence of a placebo group which not permitted to evaluate and draw conclusions about the efficacy of the treatments. Further investigations are required to confirm the present study's results.