Original Author: Paul Perry, Ph.D, BCPP
Latest Revisers: Paul Perry, Ph.D, BCPP, Brian C. Lund,
Pharm.D.
Creation Date: 1996
Peer Review Status: Internally Peer Reviewed
Lithium adverse effects are divided by therapeutic- (i.e., < or = 1.5 mEq/L) and toxic-related (i.e., > 1.5 mEq/L) levels. This handout reviews non- renal lithium adverse effects that occur at therapeutic levels. Renal adverse effects and toxic signs and symptoms are reviewed later.
Incidence and Prevalence
The following studies are representative of several published lithium adverse effect assessment studies.
Schou et al (1970) retrospectively surveyed 30 charts and 100 charts for adverse effects in patients who had been receiving lithium for one week and one to two years, respectively. Adverse effect complaints after one week of lithium treatment was GI irritation (33%), tremor (53%), muscle weakness (40%), thirst and polyuria (60%), and weight gain > 5 kg (0%). Thirteen percent of patients had no complaints (13%). In the long-term group there were no gastrointestinal complaints or complaints of muscular weakness; 4% of the patients complained of tremor; weight gains of > 5 kg were listed in 11%; and polyuria and polydipsia was a complaint in 24% of the patients. Sixty-five percent of patients had no complaints.
Vestergaard et al (1980) questioned 237 patients on long-term lithium. The following adverse effects were noted: polyuria and polydipsia, 70%; tremor, 45%; loose stools, 20%; weight gain > 10 kg, 20%; edema, 10%; dermatitis, 3%; and muscle weakness, 1.6%. He noted 25% of patients had > or = 3 complaints, 66% had > or = 1 complaint, and only 10% had no complaints.
Gastrointestinal
Milder GI complaints include epigastric bloating and slight abdominal pain. More severe symptoms such as nausea, vomiting, and anorexia may occur. The most serious problems are loose stools, diarrhea, and occasional bloody stools (Jefferson and Greist 1977, Vacaflor 1975). These side effects can produce water and electrolyte disturbances, most significantly loss of sodium, that may produce lithium retention and toxicity.
GI disturbances are usually transient in duration and often subside after the initial treatment period (Prien et al 1971). Early literature suggests that GI symptoms are related to the transient, rapid rise in serum lithium levels (Gattozi 1970, Trautner et al 1955). However, more recent work indicates nausea is the only GI side effect related to the rate of rise of the serum lithium level (Persson 1977).
Lithium should be administered with food to minimize GI adverse effects. Readjustment of the dosage schedule by dividing the total daily dose into small divided doses or utilizing a slow-release dosage form of lithium may reduce these disturbances (Vacaflor 1975). Lyskowski and Nasrallah (1981) reported a double-blind 4-week crossover study comparing side effects of Lithobid (a sustained-release product) and regular-release lithium in 29 patients. Overall, no significant differences in the incidence of nausea, diarrhea, or tremor were found between preparations. However, individually, some patients had less GI complaints on the sustained-release preparation. Diarrhea may be the result of unabsorbed lithium in the large intestine pulling water into the lumen by osmotic pressure (Jeppsson and Sjogren 1975). Therefore, diarrhea may appear or worsen when a patient changes from a standard to a slow-release preparation.
Bone et al noted when patients are manic or depressed their GI complaints related to lithium increase 2 - 20 times compared to their euthymic state (Bone et al 1980).
NEUROMUSCULAR
Tremor
Jarrett et al. (1975) reported 39 of 130 patients (30%) developed tremor, but this side effect did not often handicap their daily activities. The reported range is 10 to 65 percent (Carroll et al 1987).
Lithium tremor may occur both at rest and while moving, although the tremor may be aggravated by the performance of delicate hand movements (Carroll et al 1987). An exacerbation of the tremor or extension of the tremor to other parts of the body may signify impending lithium toxicity. Emotional stress and caffeine intake may also worsen the tremor. However, reducing caffeine intake may increase lithium levels and worsen tremor (Jefferson 1988).
Lithium tremor is not benefited by L-dopa or anticholinergic agents (Carroll et al 1987, Vacaflor 1975). A lowering of the lithium dose may be effective and it is reversible with drug discontinuation.
Case reports and controlled studies demonstrate propranolol to be effective in treating lithium-induced tremor (Vacaflor 1975). Kirk et al. (1973) conducted a single- blind, placebo-controlled study with ten patients, each treated with propranolol 30-80 mg/d for two weeks and with placebo for another two weeks. The propranolol was then restarted for two weeks followed by a further two weeks of placebo. The patients were questioned at the end of each treatment period about their preference to the two identical tablets. Five patients preferred propranolol in both propranolol treatment periods and an additional three patients showed preference to propranolol in one of the two propranolol treatment periods. In contrast, none of the patients preferred placebo at any time. The difference was statistically significant (p < 0.01).
Metoprolol, a cardioselective beta-blocker, can improve lithium-induced tremor (Gaby et al 1983). However, in one case bronchospasm occurred at doses required to improve the tremor (Zubenki et al 1984).
Lithium-induced tremor can be improved by lowering the lithium level. If this is not therapeutically possible or if it fails to produce improvement, propranolol can be instituted. Though propranolol's dose is low, exacerbation of depression should be considered before institution of this treatment.
Muscle Weakness
Muscle weakness has been reported as a transient side effect (Vestergaard et al 1980). Schou et al. (1970) noted 12 of 30 patients (40%) complained of muscle weakness during the first two weeks of lithium therapy. But this side effect was not observed in the 100 patients who had been on lithium for one to two years and in only 2 of the 100 patients in the follow-up study of Vestergaard (1980). Muscle weakness appears to be dose-related and disappears with reduction or discontinuation of lithium. Case reports of a severe neuromuscular disorder resembling myasthenia gravis during the treatment with lithium have been reported (Neil et al 1976). The symptom improved with lithium discontinuation.
ENDOCRINE
Weight Gain
Six studies have reported the prevalence of lithium-induced weight gain (Garland et al 1988). The duration of lithium treatment ranged from 6 months to 17 years. Four studies reported weight gain of >4.5 kg ranged from 11 to 64%. Two studies reported 20% of patients gained more than 10 kg. This compares to a weight gain of >4.5 kg in 8% of affective disorder patients treated with placebo. Overall, the range of weight gained was 3 to 28 kg with an average of 8.5 kg.
Factors associated with weight gain were increased thirst, a previous history of weight problems, and edema, though the latter is not a consistent finding (Garland et al 1988). It is of interest that lithium responders may gain more weight than non-responders.
The etiology of weight gain with lithium is not clear. Lithium may have a insulin-like effect to decrease blood glucose and inhibit adenyl cyclase to decrease lipolysis. Lithium is not contraindicated in diabetes, but adjustment of antidiabetic treatment maybe required because of improved glucose tolerance.
Weight gain can be minimized by caloric restriction and restriction of excessive fluid intake (Dempsey et al 1976). Once developed caloric restriction will promote weight loss. It is important to inform patients of the potential for weight gain and to avoid high calorie soft drinks in replacement of fluid loss secondary to polyuria.
Thyroid
Lithium has rarely been associated with thyrotoxicosis (Lazarus 1986, Salata and Klein 1987). One report found that the risk of thryotoxicosis was slightly, but significantly, elevated among lithium treated patients followed over an 18 year period compared to controls (Barcalay et al, 1994).
More commonly, the thyroid abnormality associated with lithium is hypofunction. Presentation includes abnormal laboratory tests (i.e., "chemical hypothyroidism"), goiter without hypothyroidism, and hypothyroidism.
The incidence of chemical hypothyroidism, if response to TRH stimulation tests is included, is approximately 50% (Lazarus 1986, Salata and Klein 1987). Therefore, abnormal thyroid function tests (TFT) without clinical consequences are common with lithium.
Emerson et al. evaluated 27 patients in whom lithium therapy was initiated after baseline data collection (Emerson et al 1973). The mean TSH increased for the first three months of therapy (p<0.001) and remained constant over the next 18 months, except in one patient who developed hypothyroidism after 13 months of lithium. Overall, approximately 15% of the patients had elevated TSH values in the range of 8-20 mU/ml. The TSH values remained stable over the next four years.
Two additional reports indicated early in treatment laboratory values may only be transiently changed. Smigan et al. studied 51 patients at baseline, 4 months, and 12 months after initiating lithium (Smigan et al 1984). After 4 months, the means T4 and T3 values decreased and the mean TSH increased. One patient developed clinical signs of hypothyroidism and received thyroid supplementation. After 12 months, the mean T4 and T3 values had returned to baseline and mean TSH values had decreased, but not to baseline.
Maarbjerg et al. (1987) evaluated T4 and TSH at baseline, 6 months, and 12 months, then yearly after start of lithium in 430 patients. Mean T4 levels decreased and TSH increased at 6 months (p>0.05) and returned to baseline at 12 months. Thereafter, mean T4 increased and mean TSH decreased yearly (p<0.001). Clinical hypothyroidism developed in 8 patients after 6 months (N=3), 12 months (N=2), 2 years (N=1), and 3 years (N=2). All patients developing hypothyroidism were women and most were over 40 years of age.
In summary, many patients will develop abnormal TFT. These often present within 4 months of lithium initiation and normalize around 12 months. However, some patient's TFT remain abnormal throughout lithium treatment, without clinical consequences.
The prevalence of goiter without hypothyroidism ranges from 0 - 61% (mean 6%) in 4 studies (Lazarus 1986). The higher reported percentage is based on ultrasound findings, where the lower figures were made by clinical assessment (i.e., palpation).
Usually the goiter is diffuse, non-tender, and may spontaneously resolved with continued treatment. The goiter often is unnoticed unless the thyroid is examined prior to and during the course of lithium therapy. However, in 4 of 45 cases the enlarged thyroid gland produced difficult swallowing (Berens and Wolff 1975). No relationship has been demonstrated between lithium treatment and thyroid cancer.
Sixteen studies surveyed report a prevalence of hypothyroidism to range from 0 to 23% (mean 3.4%). These studies which included patient numbers >50 involved 4681 lithium-treated patients. Of hypothyroid patients the female/male ratio was 5:1. Age > 40 years been associated with a higher incidence of hypothyroidism. Lithium-induced hypothyroidism can occur anytime, but is more likely to occur early in treatment. Of 41 cases reporting time of onset, 36/41 (88%) occurred within the first four years of lithium therapy. Attempts to correlate the incidence and severity of hypothyroidism with serum lithium levels have not been successful (Berens and Wolff 1975).
The mechanism of lithium's effect on the thyroid gland is not fully known (Salata and Klein 1987). Lithium has been shown to reduce iodine uptake into the gland, inhibit iodine addition to tyrosine, reduce T3 and T4 release, reduce peripheral metabolism of thyroid hormones, and decrease the thyroid gland's sensitivity to TSH. It is unlikely that lithium has a direct effect on the hypothalamic-pituitary axis to reduce TSH release.
Seven studies report a higher incidence (10-33%, average 21%) of thyroid autoantibodies in patients receiving lithium compared to controls (10%) (Lazarus 1986). Lithium may accelerate the production of autoantibodies that are present prior to lithium initiation. Though pre-existing thyroid dysfunction is not a prerequisite for the development of lithium-induced hypothyroidism, patients with impaired function maybe at a higher risk. Autoimmune thyroiditis is not the sole cause of lithium-induced hypothyroidism as not all patients have thyroid antibodies. Also, increased levels of antibodies are observed in patients without hypothyroidism. It appears that thyroid failure due to lithium is usually but not necessarily dependent on antibody mediated damage.
Monitoring should include baseline free T4 and TSH. Repeat values should be obtained at six months, one year, and then yearly. Monitoring has been recommended in the literature for the duration of lithium treatment. However, since the majority of patients on lithium develop hypothyroidism within the first 3 years, routine laboratory monitoring after this period yields a low rate of return for considering the expense. Patients may develop symptoms between routine laboratory assessment. Therefore, patients should be questioned concerning signs and symptoms of hypothyroidism. This is especially true for women over 40 years of age on maintenance treatment.
Treatment may not be necessary as the majority of cases of lithium- induced thyroid abnormalities are transient and often without clinical symptoms (Lazarus 1986, Salata and Klein 1987). Despite the lack of a dose-response relationship, some clinicians report lowering the dose may reverse the hypothyroid symptoms. Goiter and the hypothyroid state can be reversed with thyroid supplementation (i.e., levothyroxine). Pre-existing hypothyroidism is not an absolute contraindication to lithium treatment. Lithium discontinuation will reverse the thyroid abnormalities.
DERMATOLOGIC EFFECTS
Lithium has been associated with a wide range of dermatological reactions of varying clinical significance (Deandrea et al 1982; Sarantidis and Waters, 1983). In one controlled study comparing the occurrence of cutaneous conditions between chronic lithium users and a control group of patients on minor tranquilizers or tricyclic antidepressants, the lithium patients experienced dermatologic adverse effects attributed to lithium in 34% (31/91) cases compared to 14% (6/44) in the control patients (Sarantidis and Waters, 1983). Maculopapular eruptions, which usually occur within one to three weeks of lithium initiation, may clear without discontinuation of lithium or any specific treatment. Acneiform eruptions may emerge or grow worse with lithium therapy, may be treatment resistant, and may require a decrease or discontinuation of the drug depending upon the patient. Sarantidis and Waters (1983) attributed acneiform eruptions to lithium in 11% (10/91) of lithium treated patients. Follicular eruptions may occur in one-third of patients, but go unnoticed due to lack of symptoms and spontaneous termination despite continued lithium treatment. Lithium may induce or, more commonly, exacerbate psoriasis, which may prove to be resistant to usual treatments and therefore require lithium discontinuation. Sarantidis and Waters (1983) attributed psoriatic exacerbations to lithium in 2% (2/91) of lithium treated patients. Exfoliative dermatitis exists as a rare, but potentially serious reaction to lithium. The mechanism of lithium-induced dermatological reactions is unknown.
HEMATOLOGIC EFFECTS
Hematologic effects are not related to age, sex, or psychiatric diagnosis (Prakash 1985). RBCs are not affected by lithium.
White Blood Cells
It is reported 75-100% of patients on lithium will demonstrate some leukocytosis (Joffe et al 1984). All WBCs are affected by lithium, with the exception of basophils (Prakash 1985). The increase in WBC count is usually 30-45% and is primarily due to neutrophilia. The maximum reported WBC count is 24,000. There is no shift to the left. The peak elevation typically occurs within one week and may be maintained long-term (Prakash 1985). One reports suggests that WBC count will gradually normalize over one year of treatment (Özdemir et al, 1994). Upon discontinuation the effect is reversible within 1 to 2 weeks. It does not appear to be dose-related.
The mechanism of action is unknown. However, the effect is not due simply to a redistribution of the granulocyte pool as increased production has been demonstrated.
Lithium has not been demonstrated to produce leukemia with long-term treatment.
Platelets
Eighty-one percent of patients have thrombocytosis with lithium (Joffe et al 1984). In one study after 4 weeks of treatment, platelet counts increased an average of 13%. Twenty-four percent of patients had platelet counts above the upper range of normal.
Two patients maintained an elevated platelet count 2 to 4 months after lithium discontinuation.
CNS EFFECTS
Organic changes may occur with therapeutic levels of lithium (Sansone and Ziegler 1985). These include distractibility, poor memory, disorientation, incoherence, poor concentration, and impaired judgment. These maybe accompanied by involuntary movements, ataxia, and dysarthria. The symptoms often appear insidiously and maybe unrecognized as lithium-related. Often the EEG is abnormal with minor asymmetries of alpha frequency or increase in 4-6 theta activities. With higher levels increasing episodes of intermittent high amplitude diffuse delta (below 4 Hz) waves with accentuation of previous focal abnormalities are common findings. Patients with organic or psychotic diagnoses maybe predisposed to this adverse effect (Shopsin et al 1970, Tucker et al 1965). Lithium-induced encephalopathy maybe reversed with dose-reduction or require discontinuation of the drug.
Neurotoxicity has been reported when lithium is combined with haloperidol, thioridazine, chlorpromazine, and fluphenazine (Sansome and Ziegler 1985). Three studies have not supported these case reports and many authors believe these represent cases of neuroleptic malignant syndrome related to antipsychotics.
CARDIAC EFFECTS
Lithium-induced T-wave flattening and inversion and widening of the QRS complex have been reported (Mitchell and Mackenzie 1982). T-wave changes are noted in 13-100% of patients. In healthy individuals, no clinical symptoms are related to these changes, they frequently disappear on continued treatment, and are readily reversible with lithium discontinuation. These appear to be related to replacement of intracellular potassium rather than a direct pharmacologic effect. Though little risk of an adverse event exits, patients with pre-existing conduction abnormalities should be monitored with an EKG after steady-state lithium levels are achieved. Sinus node dysfunction, in particular, may be frequently present in lithium treated patients. In one study of 45 chronic lithium treated patients, sinus node dysfunction, measured as sinus arrest > 1.5 seconds or heart rate < 50 beats/min, was more common among lithium patients versus controls (Rosenqvist et al, 1993). Among lithium patients (mean age 50.3 years old, range 34-72), sinus arrest and bradycardia occurred in 56% and 78%, respectively, compared to 30% and 30%, respectively, among age stratified controls, a significant difference. Clinically significant sinus node dysfunction, however, was very rare among lithium treated patients in this study.
PARATHYROID
Lithium can increase serum calcium, reduce serum phosphorous, and increase PTH (Salata and Klein 1987). Ten to fifteen percent of patients will develop slightly elevated Ca++ and PTH within 4 weeks of lithium initiation. These values may reverse within one week of lithium discontinuation and should be normal within 2 to 4 weeks.
It is unknown if lithium unmasks parathyroid pathology. Complications of primary hyperparathyroidism do not occur, though osteopenia has been reported.
The mechanism of lithium effect is not known. However, lithium may increase the threshold calcium level necessary to suppress PTH by preventing calcium binding to the parathyroid gland. A recent in vitro study indicated lithium increases the release of PTH at therapeutic concentrations (Birnbaum et al 1988).
PREGNANCY AND LACTATION
Teratogenicity of Lithium
Lithium has traditionally been considered a significant human teratogen (Schou et al 1973b). Most of the data that support this assertion, however, come from retrospective studies, a conservative approach that overestimates the true incidence of malformations. More recent cohort and case-control studies suggest that the risk of teratogenicity due to lithium has been substantially less than initially thought (Cohen et al, 1994).
In animal studies, lithium has been shown to be teratogenic when administered in high doses (Weinstein and Goldfield 1975a). Because of difficulties in experimental design, a direct link between lithium at therapeutic levels and teratogenicity has not been definitely established. Chromosome studies suggest that lithium carbonate has no significant adverse effect in therapeutic concentrations (Weinstein and Goldfield 1975a).
In humans, indirect evidence of the effect of lithium on the fetus initially came from the International Register of Lithium Babies, a collaborative program developed by Schou et al (Weinstein and Goldfield 1975a, Schou et al 1973a, Schou et al 1973b, Weinstein and Goldfield 1975b). The Register requested physicians report any pregnant patients treated with lithium at least during the first trimester. The authors acknowledged that deviation from normal are more likely to be reported to the register than are reports of normal babies. This may result in an over-estimation of the incidence figure.
Overall, malformation rate estimates in humans range between 1 and 7%, with 3% being average. As of 1980, 225 lithium babies had been reported to the Register. Twenty-five newborns (11.1%) had malformations, of which 18 (8%) involved the cardiovascular system (Weinstein 1980). The cardiovascular abnormalities included Ebstein's anomaly--a rare malformation consisting of a distorted tricuspid valve with secondary abnormalities of the right ventricle and atrium as well as other major cardiovascular malformations. Atrial septal defect of the patent foramen ovale is also reported.
The incidence of all cardiovascular abnormalities in the general population is 0.004%, compared to 8% in the Lithium Registry. In 1975 the reported incidence of Ebstein's anomaly in lithium babies was 2.8%. Through 1985, with more exposures reported, the figure dropped to 1.0%, (Elia et al 1987). This contrasts with a normal incidence of 0.0005% (Weinstein and Goldfield 1975). An incidence of Ebsein's anomaly of 1% among lithium exposures is approximately 200-400 times greater than normal (Cohen et al, 1994, Zalzstein et al, 1990). Therefore, the Registry's data suggest that a fetus exposed to lithium has an increased risk of a congenital abnormality, especially of the cardiovascular system
In addition to the reports of in utero exposure, 60 "lithium children" who did not have malformations at birth were identified (Schou and Amdisen 1975). At a mean age of 7.3 years, there was no difference in the incidence of physical or mental abnormalities in the lithium children as compared to siblings who had not been exposed to lithium. It is important to note this information was obtained by survey and not direct examination of the children.
The Cohen et al review (1994), however, suggests that the Lithium Registry's retrospective data dramatically overestimates the true incidence of lithium teratogenicity. Among the four case control studies, no cases of lithium exposure were identified among 207 cases of Ebstein's anomaly. Each of the studies had a power of at least 97% to identify a risk of Ebstein's anomaly 400 times greater than normal among lithium treated mothers. Therefore, it would be very unlikely for each of these studies to fail to find a link between Ebstein's anomaly and lithium exposure during pregnancy. In one report, the risk of Ebstein's anomaly was no more than 28 times normal at an a = 0.05 and b = 0.8 (Zalzstein et al, 1990) Between two additional studies cited, there was less than a 2.5% chance of an 8.4-fold increase of congenital heart defects (Kallen, 1991) and an eight-fold increase of congenital anomalies, overall (Czeizel and Racz, 1990), in mothers treated with lithium during pregnancy.
In the two cohort studies cited by Cohen et al (1994), a significant risk of congenital anomalies (all types) and cardiac anomalies was found in one study (Kallen and Tandberg, 1983). In this study, the relative risk of all types of congenital anomalies and cardiac anomalies was three times and 7.7 times control, respectively. The other study failed to find a similar relationship, but, of interest, one case of Ebstein's anomaly occurred in the lithium-exposed group (Jacobson et al, 1992).
Lithium Administration During Pregnancy
Cohen et al have published guidelines for lithium use in women with bipolar affective disorder (1994):
If administered during the second or third trimesters, the lithium dosage should be adjusted to avoid high serum levels. It is important to note salt restriction and diuretics potentially may increase lithium levels. Serum concentrations of lithium in neonates have been found to be equal to that in the mother (Schou and Amdisen 1975). Woody et al (1971) reported a case of lithium intoxication in a neonate that showed signs of sedation and listlessness. Thyroid function disturbances have also been found in infants of mothers treated with lithium (Karlsson et al 1975, Schou et al 1968).
Schou et al. (1973) reported renal lithium clearance increase 50-100% during gestation, which may necessitate dosage increases to maintain a desired level. Lithium clearance returns to baseline at delivery (Schou et al 1973). Maternal lithium intoxication has been reported a few hours before delivery (Vacaflor et al 1970, Wilbanks et al 1970). Weinstein and Goldfield (1975) recommend reducing the daily lithium dose by 50% in the last week of gestation, discontinuing the drug entirely at the onset of labor, and reinstituting it at the pre-pregnancy dose immediately after delivery. Lithium levels should be monitored more frequently (weekly or bi-weekly) as the women approaches delivery.
Lithium and Breast-Feeding
Lithium is present in breast milk at a concentration about 30-100% of the mother's serum (Schou and Amdisen 1973, Weinstein and Goldfield 1975). A similar concentration of lithium has been found in the serum of the infant breast-fed by a lithium-treated mother (Schou and Amdisen 1973). It can be argued that breast-feeding need not be restricted since the infant has been exposed to the same concentration of lithium before birth. It is the opinion of most authors, however, this practice is hazardous to the infant because the infant's regulatory and excretory mechanisms are not well developed (Ananth 1978, Weinstein and Goldfield 1975). The infant may be especially susceptible to alterations in fluid-electrolyte imbalance and lithium toxicity. Tunnessen and Hertz (1972) reported that a breast- fed baby with a serum level of 0.6 mEq/L developed cyanosis and poor muscle tone. The infant also manifested T wave changes on EKG.
CONCLUSION
The evidence linking lithium and birth defects is equivocal based on available data. Conservative clinical judgment suggests the drug be discontinued at least during the first trimester of pregnancy. Lithium needs close monitoring during the second and third trimesters and, especially, at the time of delivery. Breast feeding may produce significant lithium concentrations in the infant.
REFERENCES
Ananth J (1978). Side effect in the neonate from psychotropic agents excreted through breast-feeding. Am J Psychiatry 135:801-4.
Barclay ML, Brownlie BE, Turner JG, et al (1994). Lithium associated thyrotoxicosis: a report of 14 cases, with statistical analysis of incidence. Clinical Endocrinology 40:759-64.
Berens SC, Wolff J (1975). The endocrine effects of lithium. In: Johnson FN, ed. Lithium Research and Therapy, London: Academic Press 443-72.
Birnbaum J, Klandorf H, Builian A, et al (1988). Lithium stimulates the release of human parathyroid hormone in vitro. J Clin Endocrinol Metab 66: 1187-91.
Bone S, Roose SP, Dunner DL, et al (1980). Incidence of side effects in patients on long-term lithium. Am J Psychiatry 137:103-4.
Carroll JA, Jefferson JW, Greist JH (1987). Treating tremor induced by lithium. Hosp Comm Psychiatry 38:1280, 1288.
Cohen LS, Friedman JM, Jefferson JW, et al (1994). A reevaluation of risk of in utero exposure to lithium. JAMA 271:146-150.
Czeizel A, Racz J (1990). Evaluation of drug intake during pregnancy in the Hungarian case-control surveillance of congenital anomalies. Teratology 42:505-12.
Deandrea D, Walker N, Mahlmaker M, et al (1982). Dermatological reactions to lithium: A critical review of the literature. J Clin Psychopharmacol 2:199-204.
Dempsey GM, Dunner DL, Fieve RR, et al (1976). Treatment of excessive weight gain in patients taking lithium. Am J Psychiatry 133:1082-4.
Elia J, Katz IR, Simpson GM (1987). Teratogenicity of Psychotherapeutic medication. Psychopharmacol Bull 23:531-86.
Emerson GH, Dyson WL, Utiger RD (1973). Serum thyrotropin and thyroxin concentrations in patients receiving lithium carbonate. J Clin Endocrinol Metab 36:338-46.
Gaby NS, Lefkowitz DS, Israel JR (1983). Treatment of lithium tremor with metoprolol. Am J Psychiatry 140:593-5.
Garland EJ, Remick RA, Zis AP (1988). Weight gain with antidepressants and lithium. J Clin Psychopharmacol 8:323-30.
Gattozi AA (1970). Lithium in the treatment of mood disorders. National Clearinghouse for Mental Health Information, NIMH Publication Number 5033.
Jacobson SJ, Jones K, Johnson X, et al (1992). Prospective multicenter study of pregnancy outcome after lithium exposure during first trimester. Lancet 339:530-3.
Jarrett DB, Serry J, Graham DB (1975). Lithium-induced tremor (letter). Med J Aust 62:21.
Jefferson JW Greist JH (1977). Other adverse reactions. Primer of Lithium Therapy, Baltimore: The Williams and Wilkins Co. 189-200.
Jefferson JW (1988). Lithium tremor and caffeine intake: two cases of drinking less and shaking more. J Clin Psychiatry 49:72-3.
Jeppsson J, Sjogren J (1975). The influence of food on side effects and absorption of lithium. Acta Psychiatr Scand 51:285-8.
Joffe RT, Kellner CG, Post RM, et al (1984). Lithium increases in platelet count (letter). N Engl J Med 311:674-5.
Kallen B, Tandberg A (1983). Lithium and pregnancy: a cohort study on manic-depressive women. Acta Psychiatr Scand 68:134-9.
Kallen B (1991). Lithium therapy and congenital malformations. In Schrauzer GM, Klippel KF, eds. Lithium in biology and medicine: new applications and developments. Weinheim, Germany: VCH Verlagsgesellschaft:121-30.
Karlsson K, Lindstedt, Lundberg PA, et al (1975). Transplacental lithium poisoning: reversible inhibition of fetal thyroid (letter). Lancet 1:1295.
Kirk L, Baastrup PC, Schou M (1973). Propranolol treatment of lithium-induced tremor. Lancet 2:1086-7.
Lazarus JH (1986). Endocrine and metabolic effects of lithium. New York: Plenum 99-124.
Lyskowski J, Nasrallah HA (1981). Slowed release lithium: A review and a comparative study. J Clin Psychopharmacol 1:406-8.
Maarbjerg K, Vestergaard P, Schou M (1987). Changes in serum thyroxin (T4) and serum thyroid stimulating hormone (TSH) during prolonged lithium treatment. Acta Psychiatr Scand 75:217-21.
Mitchell JE, Mackenzie TB (1982). Cardiac effects of lithium in man: a review. J Clin Psychiatry 43:47-51.
Neil JF, Himmelhoch JM, Licata SM (1976). Emergence of myasthenia gravis during treatment with lithium carbonate. Arch Gen Psychiatry 33:1090-2.
özdemir MA, Sofuoglu S, Tanridulu G, et al (1994). Lithium-induced hematologic changes in patients with bipolar affective disorder. Biol Psychiatry 35:210-13.
Persson G (1977). Lithium side effects in relation to dose and to levels and gradients of lithium in plasma. Acta Psychiatr Scand 55:208-13.
Prakash R (1971). A review of the hematologic side effects of lithium. Hosp Commun Psychiatry 1985;36:127-8.
Prien RF, Caffey EM, Klett CJ. Lithium carbonate: A survey of the history and current status of lithium in treating mood disorders. Dis Nerv Syst 32:521-31.
Rosenqvist M, Bergfeldt L, Aili H, et al (1993). Sinus node dysfunction during long-term lithium treatment. Br Heart J 70:371-5.
Salata R, Klein I (1987). Effects of lithium on the endocrine system: a review. J Lab Clin Med 110:130-6.
Sansone MEG, Ziegler DK (1985). Lithium toxicity: a review of neurologic complications. Clin Neuropharmacol 8:242-8.
Sarantidis D, Waters B (1983). A review and controlled study of cutanious conditions associated with lithium carbonate. Br J Psychiatry 143:42-50.
Schou M, Amdisen A, Jensen SE, et al (1968). Occurrence of goiter during lithium treatment. Brit Med J 3:710-3.
Schou M, Baastrup PC, Grof P, et al (1970). Pharmacological and clinical problems of lithium prophylaxis. Br J Psychiatry 116:615-9.
Schou M, Amdisen A, Steenstrup OR (1973a). Lithium and pregnancy-II, hazards to women given lithium during pregnancy and delivery. Br Med J 2:137-8.
Schou M, Amdisen A (1973). Lithium and pregnancy-III, lithium ingestion by children breast-fed by women on lithium treatment (letter). Br Med J 2:138.
Schou M, Goldfield MD, Weinstein MR et al (1973b). Lithium and pregnancy-I, report from the register of lithium babies. Br Med J 2:135-6.
Schou M, Amdisen A (1975). Lithium and the placenta (letter). Am J Obstet Gynecol 122:541.
Schou M (1976). What happened later to the lithium babies? A follow-up study of children born with malformations. Acta Psychiatr Scand 54:193-7.
Shopsin B, Johnson G, Gershon S (1970). Neurotoxicity with lithium: differential drug responsiveness. Int Pharmacopsychiatry 5:170-82.
Smigan L, Wahlin A, Jacobsson L, et al (1984). Lithium therapy and thyroid function tests: a prospective study. Neuropsychobiology 11:39-43.
Trautner EM, Morris R, Noack CH, et al (1955). The excretion and retention of ingested lithium and its effect on the ionic balance of man. Med J Aust 2:280-91.
Tucker GJ, Detre T, Harrow M, et al (1965). Behavior and symptoms of psychiatric patients and the EEG. Arch Gen Psychiatry 12:278-86.
Tunnessen WW, Hertz CG (1972). Toxic effects of lithium in newborn infants: A commentary. J Pediatr 81:804-7.
Vacaflor L, Lehmann HE, Ban TA (1970). Side effects of teratogenicity of lithium carbonate treatment. J Clin Pharmacol Ther 10:387-9.
Vacaflor L (1975). Lithium side effects and toxicity: The clinical picture. In: Johnson FN, ed. Lithium Research and Therapy, London: Academic Press 211-25.
Vestergaard P, Amdisen A, Schou M (1980). Clinically significant side effects of lithium treatment. A survey of 237 patients in long-term treatment. Acta Psychiatric Scand 62:193-200.
Weinstein MR, Goldfield MD (1975a). Administration of lithium during pregnancy. In: Johnson FN, ed. Lithium Research and Therapy, London: Academic Press 237-64.
Weinstein MR, Goldfield MD (1975b). Cardiovascular malformations with lithium use during pregnancy. Am J Psychiatry 132:529-31.
Weinstein MR (1980). Lithium treatment of women during pregnancy and in the post- delivery period. in Johnson FN, ed. Handbook of Lithium Therapy, London: MTP Press 421-9.
Wilbanks GD, Bressler B, Peete CH, et al (1970). Toxic effects of lithium carbonate in a mother and newborn infant. JAMA 213:865-7.
Woody JH, London WI, Wilbanks GD (1971). Lithium toxicity in a newborn. Pediatrics 47:94-6.
Zubenki GS, Cohen BM, Lipinski JF (1984). Comparison of metoprolol and propranolol in the treatment of lithium tremor (letter). Psychiatry Research 11;163-4.