Prediction of lithium response using clinical data.

OBJECTIVE: Promptly establishing maintenance therapy could reduce morbidity and mortality in patients with bipolar disorder. Using a machine learning approach, we sought to evaluate whether lithium responsiveness (LR) is predictable using clinical markers. METHOD: Our data are the largest existing sample of direct interview-based clinical data from lithium-treated patients (n = 1266, 34.7% responders), collected across seven sites, internationally. We trained a random forest model to classify LR-as defined by the previously validated Alda scale-against 180 clinical predictors. RESULTS: Under appropriate cross-validation procedures, LR was predictable in the pooled sample with an area under the receiver operating characteristic curve of 0.80 (95% CI 0.78-0.82) and a Cohen kappa of 0.46 (0.4-0.51). The model demonstrated a particularly low false-positive rate (specificity 0.91 [0.88-0.92]). Features related to clinical course and the absence of rapid cycling appeared consistently informative. CONCLUSION: Clinical data can inform out-of-sample LR prediction to a potentially clinically relevant degree. Despite the relevance of clinical course and the absence of rapid cycling, there was substantial between-site heterogeneity with respect to feature importance. Future work must focus on improving classification of true positives, better characterizing between- and within-site heterogeneity, and further testing such models on new external datasets.

Fifteen-year follow-up of thyroid function in lithium patients.

OBJECTIVE: To study prospectively the course of clinically relevant thyroid dysfunction in a cohort of patients on long-term lithium treatment. METHOD: Patients (no.=150) who had undergone a cross-sectional evaluation of their thyroid function in 1989, when they were at different stages of lithium treatment, were followed up for thyroid circulating thyroid antibodies, hypothyroidism, hyperthyroidism, and thyroidectomy, during a further period of lithium exposure of up to 15 yr. RESULTS: Annual rates of newly developed circulating thyroid antibodies and hypothyroidism were 1.7 and 1.5%, respectively. Subjects with thyroid antibodies had a higher chance of requiring substitution treatment with levothyroxine for hypothyroidism compared with subjects with no evidence of thyroid antibodies (6.4% annual rate compared to 0.8%; relative risk: 8.4; 95% confidence interval: 2.9-24.0). One case of hyperthyroidism was observed over 976 patient-yr. Three patients underwent thyroidectomy during followup (two for multinodular goiter and one for multicentric papillary carcinoma). CONCLUSIONS: Lithium may be associated with hypothyroidism in particular in the presence of circulating thyroid antibodies. Incidence of thyroid antibodies is comparable with that reported for the general population. Hyperthyroidism and thyroid cancer are rare.

Association of personal and familial suicide risk with low serum cholesterol concentration in male lithium patients.

OBJECTIVE: We sought to establish whether low cholesterol concentration may be associated with a personal history of attempted suicide or a family history of completed suicide in psychiatric out-patients on maintenance lithium treatment, who represent a population at risk for suicide. METHOD: We retrospectively reviewed charts regarding 783 out-patients consecutively admitted to a lithium clinic from 1976 to 1999. Individual age- and gender-specific quartile of serum cholesterol concentration were correlated against personal lifetime suicide attempts and completed suicide in first-degree relatives. RESULTS: The proportion of men with a personal lifetime history of attempted suicide, especially if violent, and that of men with history of completed suicide in a first-degree relative were significantly higher among the group with cholesterol concentration in the lowest quartile compared to the group with cholesterol levels above the 25th percentile. CONCLUSION: Low cholesterol concentration should be studied further as a potential biological/genetic marker of suicide risk.

Ten-year follow-up of thyroid function in lithium patients.

The objective of this paper was to study prospectively the course of clinically relevant thyroid dysfunction in a cohort of patients on long-term lithium treatment. Patients (N = 150) who had undergone a cross-sectional evaluation of their thyroid function in 1989, when they were at different stages of lithium treatment, were followed up for the presence of thyroid autoimmunity, hypothyroidism, and goiter during a further period of lithium exposure of up to ten years. The following annual rates of newly developed thyroid dysfunction were observed: autoimmunnity (1.4%), subclinical hypothyroidism (1.7%), and goiter (2.1%). Subjects with thyroid autoimmunity had a higher chance of requiring substitution treatment with levothyroxine for subclinical hypothyroidism compared with subjects with no evidence of thyroid autoimmunity (13/32 = 41% versus 7/118 = 6%). Subjects (N = 15) who were prescribed carbamazepine in addition to lithium showed a significant decrease of TSH concentrations. In patients already being treated with lithium for several years, the overall incidence of hypothyroidism, goiter, and thyroid autoimmunity were comparable with those reported for the general population. However, lithium exposure may represent an additional risk factor for hypothyroidism in women and/or in the presence of thyroid autoimmunity.

Family-based association study between bipolar disorder and DRD2, DRD4, DAT, and SERT in Sardinia.

Association analysis of candidate genes may represent a strategy for clarifying the genetic components involved in bipolar disorder. Polymorphism at dopamine receptor genes DRD2, DRD4, and dopamine and serotonin transporter genes (DAT, SERT) has been used in previous association studies. Some authors have reported positive association between certain alleles and bipolar disorder, using the case-control design. In this family-based association study of DRD2, DRD4, DAT, and SERT, the distribution of parental nontransmitted alleles was compared with that of alleles transmitted to 53 Sardinian probands suffering from bipolar disorder. The transmission disequilibrium test (TDT) was used to detect any disproportionate transmission of alleles by heterozygous parents to affected children. No differences were found between the allele distribution of polymorphisms at DRD2, DRD4, DAT, and SERT in probands and parental nontransmitted chromosomes. TDT did not reveal any difference between transmitted and nontransmitted alleles. Our results do not support the hypothesis of a role for DRD2, DRD4, DAT, or SERT in bipolar disorder. Previously reported positive associations between DRD2 or SERT and bipolar disorder were conceivably due to stratification dependent on the case-control design, even though our sample might have failed to detect small associations due to limited power.

Maternal inheritance of manic depression in hemizygotes for the G6PD-Mediterranean mutation. Indirect evidence for Xq28 transmission in Sardinia.

Both X-linkage and a parent-of-origin effect have been hypothesized in manic-depressive disorder. We have previously shown an allelic association between X-linked G6PD deficiency and manic depression in Mediterranean populations. To test both X-linkage and a parent-of-origin effect in manic depression further, we have studied 274 Sardinian manic-depressive probands and their parents. Excess of maternal transmission (P = 0.005) of major affective disorder was found in male probands carrying the G6PD-Mediterranean mutation. Our results provide indirect molecular support for an association between manic depression and the Xq28 chromosome region in Sardinia. Further studies on Xq28 using tests of allelic association and transmission disequilibrium with multiple DNA markers are required, to clarify the nature of the association we have found. Our study cannot implicate or exclude a direct role for G6PD deficiency in the aetiology of manic depression.

Suicidal behavior on and off lithium prophylaxis in a group of patients with prior suicide attempts.

One hundred patients who had attempted suicide before commencing lithium prophylaxis were followed up. Outcome was analyzed in terms of attempted or completed suicide after a mean of 10 years since admission to the lithium clinics. Of 10 patients who committed suicide, 9 had discontinued adequate lithium prophylaxis for a period ranging from 2 weeks to 7 years before death. Having discontinued lithium therapy was associated with suicide also in the subgroup of patients for whom lithium had not completely prevented episodes during lithium treatment. Suicide risk was 24 times as high during periods off compared with periods on adequate lithium prophylaxis. Incidence of attempting suicide was similar during the periods before receiving or after discontinuing lithium treatment, whereas it was 5 to 6 times lower during prophylaxis. Continuous and adequate lithium prophylaxis should be considered in the presence of high suicide risk, even if the prophylactic effect on the underlying mood disorder may be incomplete.

Association between dopamine receptor genes and migraine without aura in a Sardinian sample.

BACKGROUND: Migraine seems to be caused by a combination of environmental and genetic factors. Clinical and pharmacologic evidence supports the hypothesis that dopaminergic transmission is involved in the pathogenesis of migraine. OBJECTIVE: The current report concerns a genetic study to test the involvement of genes for dopamine (DA) receptors D2 (DRD2), D3 (DRD3), and D4 (DRD4) in migraine without aura, particularly in a subgroup with enhanced DA sensitivity. METHODS: For the first time, a family-based association method--the Transmission Disequilibrium Test (TDT)--was used to examine an isolated population, such as Sardinians. We studied 50 nuclear families of patients affected by migraine without aura. The subgroup of dopaminergic migraineurs was selected based on the presence of both nausea and yawning immediately before or during the pain phase of migraine. RESULTS: No association was detected using the TDT between DRD3, DRD4, and migraine without aura either in the overall sample or in the subgroup. No difference was observed in DRD2 allelic distribution in the overall sample, although the allelic distribution at the DRD2 locus differed significantly in the subgroup of dopaminergic migraineurs (p = 0.004). Allele 1 of the TG dinucleotide intronic noncoding polymorphism of the DRD2 locus was the individual allele that appeared to be in disequilibrium with migraine without aura (p = 0.02). CONCLUSIONS: Our data suggest that a genetic approach could be useful in providing molecular support to the hypothesis that hypersensitivity of the dopaminergic system may represent the pathophysiologic basis of migraine, at least in a subgroup of patients.

Calcitonin decreases lithium plasma levels in man. Preliminary report.

Lithium kinetics were studied in four manic-depressive women treated with calcitonin for post-menopausal osteoporosis. During calcitonin therapy all patients showed a significant decrease of lithium blood levels, and an increase in lithium renal clearance was observed in two out of two patients. According to these data, lithium therapy needs to be adjusted in patients undergoing concomitant treatment with calcitonin.

No evidence of association between dopamine D3 receptor gene and bipolar affective disorder.

A recent study reported a possible association between allele 1 of the dopamine D3 receptor gene and bipolar affective disorder using the haplotype relative risk approach. In attempt to replicate these findings, we used similar family-based methods, such as the Haplotype-Based Haplotype Relative Risk method and the Transmission Disequilibrium Test, in a sample of 44 bipolar probands from Sardinia with both parents available. Using the Bal I restriction enzyme site polymorphism of Lannfelt et al. (1992), no differences were found between transmitted and non-transmitted alleles and no evidence of linkage disequilibrium was observed.

Carbamazepine augmentation in lithium-refractory bipolar patients: a prospective study on long-term prophlyactic effectiveness.

Twenty-two patients affected by bipolar or schizoaffective disorder, in whom carbamazepine was added to lithium after recurrence when on maintenance with lithium alone, were followed up prospectively for 2 to 13 years. The number of episodes, hospitalizations, and cumulative affective morbidity was markedly reduced after carbamazepine augmentation. Seventeen patients presented a better course during combined treatment than during lithium alone, and of these 15 had no further recurrences. Four patients did not appear to improve after carbamazepine augmentation, whereas one featured reemergence of affective episodes after having derived satisfactory benefit from combination for 7 years (delayed tolerance). Carbamazepine augmentation was associated with a reduction of lithium doses in some patients, including a subgroup who had not tolerated lithium at usual therapeutic levels. Carbamazepine significantly reduced serum thyrotropin concentrations, which were abnormally high in approximately one half of patients when on lithium alone. Total serum thyroxine concentrations were also decreased after carbamazepine augmentation, but free thyroid hormone concentrations did not change. Other significant carbamazepine-induced changes in laboratory tests included increases in total cholesterol concentrations and decreases in white blood cell counts.

Value of thyroid echography in the long-term follow-up of lithium-treated patients.

Psychiatric patients on long-term lithium (Li) therapy frequently develop goiter and/or hypothyroidism. It has also been suggested that Li may trigger/exacerbate thyroid autoimmunity. Previous studies provided evidence that underlying thyroid diseases represent important predisposing factors for the development of Li-induced thyroid dysfunction. The aim of the present paper was to assess the value of thyroid ultrasound-a simple and reliable tool to detect subtle thyroid abnormalities-in the longitudinal evaluation of 23 Li-treated psychiatric patients without evidence of biochemical thyroid abnormalities before therapy. For this purpose, thyroid ultrasound was associated with a clinical and laboratory (serum thyroxine, serum triiodothyronine, serum TSH, antithyroglobulin (AbTg), antithyroid microsomal (AbM) and antithyroid peroxidase autoantibodies) evaluation prior to and at 6- to 12-month intervals during Li treatment. On the basis of thyroid ultrasound before Li, patients were subdivided into two groups: group A (n = 15, 7 males, 8 females) with a normal echography and group B (n = 8, 5 males, 3 females) with mild ultrasound abnormalities. In group A the development of a small diffuse goiter was confirmed by physical examination during Li therapy; 2 patients displayed a transient increase of serum TSH concentration and none developed detectable serum antithyroid autoantibodies. Beside the small volumetric increase, no other ultrasound abnormalities were observed during the entire follow-up. In all group B patients a mild diffuse goiter was clinically detected before and on Li administration and no significant volumetric changes were observed during follow-up. Two patients developed high titers of AbM and AbTg 12 and 18 months after the beginning of Li, respectively; in 1 a persistent increase of serum TSH concentration was also observed. Thyroid echography before Li displayed different degrees of scattered or diffuse hypoechogenicity and a further decrease in echogenicity was detected during Li therapy in 2 patients. In conclusion, we provided further evidence that long-term Li administration is not associated with de novo appearance of thyroid autoimmune phenomena in humans, but rather with an exacerbation of underlying thyroid autoimmunity. In addition to thyroid autoantibody and TSH measurements, thyroid echography appears to be a sensitive tool in the identification of patients at risk of developing autoimmune hypothyroidism during long-term Li therapy.

Six-year follow-up of thyroid function during lithium treatment.

A cohort of patients at various stages of lithium treatment was followed up for 6 years in order to evaluate the course of thyroid abnormalities. Ultrasonography confirmed that lithium can increase thyroid size, especially in cigarette smokers, and that it can affect the texture of the gland. However, the incidence of clinical hypothyroidism or specific thyroid autoimmunity does not exceed that found in the general population. Repeated determinations of thyrotrophin (TSH) concentrations can prevent clinically relevant consequences. Addition of carbamazepine to lithium can counteract lithium-induced subclinical hypothyroidism, possibly improving prophylactic efficacy in recurrent affective disorders.

Heterogeneity of monoaminergic vesicular carriers: pharmacological evidence using MPP+ as a marker.

MPP-production and uptake by dopaminergic terminals are critical steps in MPTP-induced Parkinson-like disorder. We reported evidence for a specific uptake of MPP by synaptic vesicles from mouse striatum. Its regional distribution suggests it as a marker of the dopamine vesicular carrier. We decided to further characterize such an MPP uptake. Tetrabenazine inhibits the dopamine uptake both in the striatum and in the cerebellum with similar Km values suggesting an identify of the vesicular carrier in these areas. On the contrary, 3H-MPP vesicular uptake had in the striatum a t1/2 of 60 sec, but was not detectable at any time in the cerebellum. Moreover, MPP inhibited the uptake of 3H-DA (Ki: 1.6 +/- 0.03 microM) and 3H-NE (Ki 2.6 +/- 0.01 microM) in the striatum but not in the cerebellum, even at molar concentration. These pharmacological data indicate that in nondopaminergic areas the monoamine carrier may be similar but not identical from that located in dopaminergic areas.

A double-blind study of L-sulpiride versus amitriptyline in lithium-maintained bipolar depressives.

A double-blind group comparison study of L-sulpiride and amitriptyline was carried out in 30 bipolar outpatients on maintenance treatment with lithium suffering from a major depressive recurrence. L-sulpiride showed equivalent antidepressant activity to amitriptyline at 4 weeks using the Hamilton Rating Scale for Depression. However, the onset of action was faster in the L-sulpiride group, showing a significant improvement at 1 week in both anxiety-somatization and specific depression items, including depressed mood, feelings of guilt, work & activities and retardation. The incidence of anticholinergic side effects was significantly higher in the amitriptyline treatment group.

The course of thyroid abnormalities during lithium treatment: a two-year follow-up study.

A total of 116 patients on lithium treatment were followed up for 2 years to determine the course and the clinical relevance of thyroid abnormalities. Elevated thyroid-stimulating hormone (TSH) concentrations were transitory in most patients, except those with serum antithyroid antibodies. The patients who initially had microsomal antibodies remained positive, with an increase in titre in two-thirds of cases. Three young patients of both sexes developed thyroid autoimmunity early in the treatment. The risk of developing hypothyroidism was higher in women, especially in the presence of antibodies. TSH concentrations were significantly lower when carbamazepine was combined with lithium.