Kidney, thyroid and other organ functions after 40 years or more of lithium therapy: a case series of five patients

Abstract

We present the cases of five patients (two men aged 64 years and 79 years) and three women (aged 64 years, 65 years and 75 years) who have received lithium treatment for 40–45 years, with particular regard to kidney and thyroid functions, hypercalcaemia and cognition, in the context of disease course and overall functioning. Lithium was initiated in the early phase of the illness (in three patients within the first 2 years). In four patients, lithium concentration was between 0.60 and 0.65 mmol/l and in one patient, between 0.7 and 0.8 mmol/l. Four were very good lithium responders. One man had stage 3 chronic kidney disease, and the other stage 2/3 chronic kidney disease. All three women had asymptomatic stage 2 chronic kidney disease. One woman had severe thyroid dysfunction (Hashimoto’s disease) with extremely high levels of antithyroid peroxidase antibodies and antithyroglobulin antibodies and was receiving thyroxine. Serum calcium levels were normal or borderline in all five patients, and most cognitive functions were comparable to healthy persons of similar gender, age and years of education. All the patients were professionally active until 55–65 years and their family and social functioning were satisfactory. It was concluded that, in good lithium responders, ultra-long-term treatment with lithium enables good professional and psychosocial functioning, and the possible somatic side effects are manageable.

Introduction

More than 50 years ago, the British psychiatrist, Geoffrey Hartigan presented his observation concerning the prolonged use of lithium for 3 years in 15 patients with affective disorder: 7 with bipolar affective disorder and 8 with recurrent depression [Hartigan, 1963]. He reported that recurrences of illness were not observed among six patients from the first group and six patients from the other group during that time. He suggested that the long-term use of lithium has a ‘prophylactic’ effect on recurrences of affective illness.

Nowadays, lithium is still the cornerstone of long-term treatment of bipolar disorder. In the most recent meta-analysis it was found that lithium was significantly more effective than a placebo for the prevention of both manic and depressive episodes and, for the former, lithium showed superiority over anticonvulsants [Severus et al. 2014].

Those patients with bipolar mood disorder who benefit most from long-term lithium treatment are described as ‘excellent lithium responders’. This term was introduced by Grof to describe patients in whom monotherapy with lithium prevents the further recurrence of episodes for 10 years or more [Grof, 1999]. The clinical profile of such patients may include complete remissions and other characteristics of an episodic clinical course, bipolar family history, low psychiatric comorbidity and a characteristic presenting psychopathology, approximating to the classical Kraepelin’s description of a manic-depressive patient [Grof, 2010]. Excellent lithium responders make up about one out of three lithium-treated patients. In our previous study [Rybakowski et al. 2001], the percentage of excellent lithium responders, defined as those having an absence of recurrence in 10 years of lithium monotherapy, remained similar in patients entering lithium treatment in the 1970s and 1980s (34% versus 27%, respectively).

The most frequent adverse side effects of long-term lithium therapy include kidney and thyroid dysfunctions. A reduction in urinary concentrating capacity can occur after only a few weeks of lithium administration. In some patients receiving lithium for more than 10 years, chronic interstitial nephropathy may develop resulting in an increase in serum creatinine and a reduction in the glomerular filtration rate (GFR) which, in a small proportion of patients, can result in end-stage renal disease. In recent reviews on lithium neurotoxicity in long-term, lithium-treated patients McKnight and colleagues found a reduction in maximum urinary concentrating ability of about 15%, and a reduction in GFR of 0–5 ml/min/year [McKnight et al. 2012], while Shine and colleagues observed an increased risk, with hazard ratio (HR) 1.93, of stage 3 chronic kidney disease [Shine et al. 2015]. The duration of lithium treatment may be the most important causative factor for chronic kidney disease. Recently, Clos and colleagues, in their population-based cohort study, found that neither the duration of exposure to lithium nor the mean serum lithium level were significant predictors for GFR decline [Clos et al. 2015]. However, the mean duration of lithium treatment in their study was 4.6 years, therefore such a finding is not relevant to patients taking lithium for 10 years or more. On the other hand, Azab and colleagues, in their analysis of 180 patients receiving long-term lithium therapy in southern Israel, identified 2 patients on dialysis and 13 patients with stage 3 chronic kidney disease [Azab et al. 2014].

In a review of 77 studies on the effect of long-term lithium treatment on thyroid function, McKnight and colleagues found a six-fold increased risk of clinical hypothyroidism in patients given lithium for a mean of 5.8 years and a four-fold increased risk of increased thyroid-stimulating hormone (TSH) with lithium therapy for a mean of 1.5 years [McKnight et al. 2012]. Similarly, Shine and colleagues, analysing laboratory data from patients treated with lithium for 20 years (mean 5.3 years), showed an increased risk (HR 2.31) of hypothyroidism in these patients [Shine et al. 2015]. However, it seems that hypothyroidism occurs most frequently during the initial years of lithium therapy. In our study of 66 patients (21 men, 45 women) with bipolar mood disorder, who had received lithium for 10–44 years, there were no differences in thyroid function between those patients receiving lithium for 10–20 years and those taking the drug for more than 20 years. Some features of hypothyroidism were found in 10 (22%) female patients, the majority of whom were receiving levothyroxine [Kraszewska et al. 2015].

Another frequent lithium side effect is connected with parathyroid dysfunction resulting in hypercalcaemia. A meta-analysis of 14 observational studies by McKnight and colleagues showed a 10% increase in both serum calcium and parathyroid hormone levels in patients taking lithium long term [McKnight et al. 2012]. In our study involving 90 bipolar patients receiving lithium for an average period of 16 years, hypercalcaemia was found in 10% [Abramowicz et al. 2014]. In Shine and colleagues’ study analysing laboratory data from lithium-treated patients, an increased risk (HR 1.43) of raised total serum calcium concentration was estimated [Shine et al. 2015].

Meta-analyses, comparing bipolar patients treated with lithium with those not receiving the drug, show a moderate negative effect of lithium on cognition which, among clinicians, is considered one of the side effects of the drug [Wingo et al. 2009]. However, in our own studies, the effect of lithium on cognitive function was shown to be connected with the efficacy of the drug in preventing affective episodes. Excellent lithium responders, with no affective recurrences throughout the period of lithium therapy, perform in cognitive function tests similarly to age-matched, healthy control subjects [Rybakowski and Suwalska, 2010].

In this paper, we present a series of five patients who have been receiving lithium for 40 years or more in which an assessment of the status of kidney, thyroid and other functions was made, in the context of course of the disease and psychosocial functioning. Since such a subpopulation of lithium-treated patients is rare we want to share our experience in this respect with those interested in the ultra-long-term administration of lithium.

Case series

Patient 1, MM (a man, 64 years old, 12 years of education). The patient started lithium treatment in November 1971 aged 22 years. Previously, he had had a depressive episode in February and a manic episode 6 months after, with hospitalization. In February 1972, he had a 1-month period of mixed state and since then he has been stable. His lithium carbonate dose was 1000 mg/day and the mean serum level was 0.65 mmol/l. He was professionally active until the age of 60 years, initially as a carpenter and later as the manager of a storehouse. His family and social functioning were both very satisfactory. Cognitive assessment performed in 2007 showed a Wechsler Adult Intelligence Scale revised (WAIS-R) total of 108 points, 112 points on the verbal scale and 102 points on the nonverbal scale. The CANTAB tests for spatial memory produced results in the normal range. As to thyroid function assessments, in 2013 both hormone and antibodies were in the normal range. Examination of kidney function in 2012 revealed stage 3 chronic kidney disease with a serum creatinine level of 129.0 µmol/l and an estimated GFR (eGFR) of 49.5 ml/min/1.73 m². The calcium level was 0.6 mmol/l. He was referred to a nephrologist who did not find a contraindication for continuing lithium therapy, providing there were periodic nephrological check ups. In 2015 his serum creatinine was 137.0 µmol/l and the eGFR was 45 ml/min.

Patient 2, MWA (a woman, 64 years old, 19 years of education). In February 1973 the patient had a severe postpartum manic episode with hospitalization, after which she was started on lithium in May 1973. She discontinued lithium at the turn of 1974/75 for several months because of a second pregnancy, and after giving birth, she had another manic episode in May 1975. Lithium was resumed in June 1975 and the patient has been stable until now. Her lithium carbonate dose was 750 mg/day and the mean serum level was 0.65 mmol/l. She was professionally active as an administrative worker until the age of 55 years. Her family and social functioning were satisfactory. A cognitive assessment performed in 2007 showed a WAIS-R total of 125 points, 129 points in the verbal scale and 120 points in the nonverbal scales. The CANTAB tests for spatial memory were within the normal range. Examination of kidney function in 2012 revealed asymptomatic stage 2 chronic kidney disease with serum creatinine 83.0 µmol/l, an eGFR of 60.5 ml/min/1.73 m² and a calcium level of 0.6 mmol/l. Thyroid hormones and antibodies assessed in 2013 were in the normal range.

Patient 3, FI (a woman, 65 years old, 12 years of education). The patient was first hospitalized for depression in 1964 when she was 16 years old. In the following years she was hospitalized for mania at the turn of 1969/1970 and again in 1970/1971. In June 1971 lithium was started and the patient was stable until 1974 when she discontinued lithium because of pregnancy. After postpartum mania, lithium was resumed and continued until 1989. During that time, she was hospitalized for mania in 1976 and 1977. In 1989 she discontinued lithium but in 1991 she resumed it and was stable until 2010. She was professionally active as a teacher until 55 years of age. Her lithium carbonate dose was 750 mg/day and the mean serum level was 0.6 mmol/l. A cognitive assessment performed in 2007 showed a WAIS-R total of 125 points, 129 points in the verbal scale and 120 points in the nonverbal scale. The CANTAB tests for spatial memory were within the normal range. Another hospitalization in 2010 was due to a sudden worsening of psychiatric status in the form of a mixed state with periodic disturbances of consciousness. She improved within 6 weeks and was released with lithium carbonate, 750 mg/day, and divalproex, 600 mg/day, and has been stable until now. During hospitalization Hashimoto’s disease was diagnosed, with increased TSH and extremely high levels of antithyroid antibodies, that is, antithyroid peroxidase (anti-TPO) antibodies, 3500 IU/ml, and antithyroglobulin (anti-TG) antibodies, 2500 IU/ml, and she was therefore started on thyroxine at 75 µg/day. Examination of kidney function in 2012 revealed asymptomatic stage 2 chronic kidney disease with serum creatinine 70.7 µmol/l and an eGFR of 75 ml/min/1.73 m². Furthermore, in ultrasound examination of the kidneys, microcysts less than 2 mm in diameter, reflecting hyperechogenic foci in the renal parenchyma, characteristic of lithium-induced nephropathy, were found. The calcium level was 0.58 mmol/l.

Patient 4, DB (a woman, 75 years old, 19 years of education, physician, ophthalmologist). Her personality features were assessed as hyperthymic. From the beginning of the 1960s she suffered from labial herpes on average once a year. In 1966 the patient suffered from shingles and, in 1968, from Wilson’s lichen planus. Her first depressive episode appeared in 1965 without apparent cause and receded after 2 months of treatment with imipramine. The next depressive episodes occurred in 1968 and at the end of 1970. She was again treated with imipramine with beneficial results. After the third depressive episode, in April 1970, lithium carbonate medication was started and continued, without a break, until today. For the first 20 years the dose was 1000 mg/day, taken once during the evening, and the average concentration of lithium in the serum was 0.6 mmol/l. In 1990 this dose was reduced to 750 mg/day, taken until today. The mental state of the patient is stable and has not shown any significant fluctuations throughout the entire period of using lithium. In 1985, after the death of the patient’s mother, a 2-week period of increased emotional lability, likely hypomania, was observed in the patient but this did not require additional pharmacological intervention. The patient successfully worked professionally until 2007. During the period of taking lithium labial herpes occurred only twice and she did not report the occurrence of colds and other viral infections. Examination of kidney function in 2012 revealed asymptomatic stage 2 chronic kidney disease with serum creatinine 85.7 µmol/l, an eGFR of 60 ml/min/1.73 m² and a calcium level of 0.61 mmol/l. As regards thyroid function, both hormones and antibodies assessed in 2013 were in the normal range.

Patient 5, BR (a man, 79 years old, 15 years of education). The patient was started on lithium in 1973 at the age of 36 years, during a first psychiatric hospitalization for mania. Previously, he had had several mild episodes of hypomania or depression, which were not treated pharmacologically. While on lithium he had one hypomanic episode in 1975, and one manic episode, in 1980, for which he was hospitalized. Since 1980, he has been stable until today. His lithium carbonate dose was 750 mg/day and the mean serum level was 0.8 mmol/l. He was professionally active as a teacher until 60 years old. His family and social functioning were very satisfactory. A cognitive assessment, performed in 2007, showed a WAIS-R total of 95 points, 99 points in the verbal scale and 93 points in the nonverbal scale. The CANTAB test scores for spatial memory were in the normal range. Examination of kidney function in 2012 revealed stage 2/3 chronic kidney disease, with serum creatinine 110.5 µmol/l and an eGFR of 59 ml/min/1.73 m². The calcium level was 0.66 mmol/l (the upper limit of normal). As to thyroid function, both hormones and antibodies assessed in 2013 were in the normal range.

Discussion

In this paper, the effects on the kidneys, thyroid gland and other organs of five individuals who have taken lithium for more than 40 years are assessed. These findings allow us to make a few conclusions.

As to bipolar diagnosis, four of the patients (1, 2, 3, 5) can be characterized as bipolar I, having severe manias requiring hospitalization. Patient 4 can be diagnosed as bipolar II, or in the bipolar spectrum, due to a hyperthymic personality and probable hypomania after her mother’s death, and frequent depressive episodes which, according to some authors, may also belong to the bipolar spectrum [Saggese et al. 2006]. The course of illness was clearly periodic in all of the patients with distinct periods of remission between episodes. No comorbid conditions, such as an anxiety disorder or substance abuse, occurred in any of them. A periodic course with distinct periods of remissions and lack of comorbid conditions are regarded as favourable factors for a good long-term response to lithium [Grof, 2010]. In three patients, lithium treatment was started within the first 1–2 years of illness, in one patient after 5 years and, in another patient, after 7 years. Franchini and colleagues suggest that an early start of lithium prophylaxis (within the first 10 years from the onset of the illness) is a predictor of a good long-term outcome [Franchini et al. 1999]. In patient 5, the late onset of illness (35 years of age) can be a factor for a good response to lithium [Kleindienst et al. 2005]. For all five patients the serum lithium concentration has been maintained at a moderate level. In four patients, this concentration was 0.60–0.65 mmol/l throughout the study, a level which was recommended several decades ago for minimizing possible side effects of lithium prophylaxis [Schou, 1981]. In the other patient, despite a low lithium carbonate dose (750 mg/day) the serum lithium level was between 0.7 and 0.8 mmol/l, without significant somatic side effects. All these patients, except patient 3, were good or excellent lithium responders. However, in patient 3, despite recurrences and hospitalizations, the overall effect of long-term lithium treatment can also be assessed as relatively good.

The results of kidney function showed in one man (patient 1), stage 3 chronic kidney disease, and in another (patient 5), stage 2/3 chronic kidney disease. On the other hand, all three women had asymptomatic stage 2 chronic kidney disease. Analysing a large population of 80 long-term, lithium-treated patients, we demonstrated worse parameters of kidney function in men rather than in women [Rybakowski et al. 2012]. However, in the study of Shine and colleagues, it was found that women were at higher risk of developing kidney damage than men [Shine et al. 2015].

Many studies have suggested that female patients are more prone to developing thyroid dysfunction [Bauer et al. 2014]. In our sample, only one female patient (3) had severe thyroid dysfunction (Hashimoto’s disease) with extremely high levels of anti-TPO and anti-TG antibodies, for which she has been receiving thyroxine. The levels of thyroid hormones and antibodies in all the four remaining patients were in the normal range.

In all the patients, the calcium levels were in the normal range (at the upper limit of normal in one man). Also, in all patients, the preservation of cognitive functions was satisfactory enabling them to continue their professional activity until 55–65 years of age. This is in line with our studies showing the preservation of normal cognitive functions in good lithium responders [Rybakowski et al. 2009; Rybakowski and Suwalska, 2010]. Also worth noticing is a favourable effect of long-term lithium treatment on general health, especially in relation to the viral and other respiratory infections in patient 4. In our retrospective study of Polish and US patients, we demonstrated that, in patients with mood disorders and recurrent labial herpes infections, lithium exerts an antiviral action, preventing or reducing recurrences of labial herpes [Rybakowski and Amsterdam, 1991]. We also observed a reduction in the mean rate of reported flu-like illness during lithium therapy [Amsterdam et al. 1998].

The results obtained may confirm a powerful beneficial effect of ultra-long-term treatment with lithium in selected patients with bipolar disorder. It is concluded that, in good responders, such long-term treatment with lithium enables them to live without the disease, and that possible somatic side effects are manageable.