Many clinicians perceive prescribing lithium as problematic because of the need for monitoring of plasma levels and the risk of opening the door to a myriad of additional health concerns. However, when lithium is prescribed to a suitable candidate and managed judiciously, the therapeutic return is worthwhile and far outweighs the effort expended in initial assessment and the refinement of dosage and administration. Prior to the initiation of lithium, a number of factors must be considered to assess a patient’s suitability for treatment. These concern the psychiatric profile, episodic course and presentation of bipolar disorder, all of which must be weighed up in a probabilistic approach to determine the likelihood of lithium responsiveness (see Fig. 11.3). Once lithium has been selected as the therapy of choice, appropriate medical screening must be conducted to detect potential contraindications, inform the clinician of likely side effects and schedule subsequent monitoring (Malhi et al. 2009). In particular, this screening should address renal, thyroid, cardiac and metabolic functioning, all of which might affect, or be affected by, lithium treatment (American Psychiatric Association 2002; NICE 2014). At the commencement of treatment, baseline measures should include an assessment of cardiovascular functioning by ECG, as well as complete blood count, to inform the interpretation of subsequent monitoring. In addition, thyroid function evaluation, pregnancy testing, physical examination and tests for urea and electrolytes are required to determine appropriate dosage and possibility of contraindications (American Psychiatric Association 2002; Malhi et al. 2009). Precautions must be communicated to the patient with regard to maintaining a normal diet with adequate salt and fluid intake and avoiding excessive sweating that may result for example from heavy exercise (Malhi et al. 2009). Patient education is essential in emphasising the importance of treatment adherence and administration of lithium, and the patient must be made aware of managing the potential side effects and risk factors of long-term lithium therapy.

Both the narrow therapeutic index of lithium and its associated risk of toxicity indicate the imperative for appropriate dosing and careful monitoring of serum levels (Malhi et al. 2012b). The appropriate dose should be informed by illness course and severity, and the aim should be to achieve the lowest level whilst balancing effectiveness and adverse effects. At the commencement of treatment, higher lithium plasma levels can be used to treat acute symptoms, and lithium plasma levels should be measured weekly and eventually stabilised at 0.6–0.8 mmol/L (NICE 2014; Malhi et al. 2011, 2012b).

Lithium is best administered early in the course of illness (Kessing et al. 2014), but this also means that it may be administered to patients who do not present with a classic bipolar I disorder pattern. It is no surprise that lithium is much less effective when used in patients who have significant personality problems or additional psychiatric pathology, but, unfortunately, this is often misinterpreted as lithium ‘lacking efficacy’, or occurring because it is difficult to prescribe. Therefore it is essential to identify those patients who are likely to be lithium responsive (see Fig. 11.3; Malhi and Geddes 2014). Small, divided daily doses are recommended in APA guidelines to minimise side effects in the early course of the treatment, gradually titrating serum levels to achieve the highest dose tolerated without side effects, ranging between 0.5 and 1.2 mmol/L (American Psychiatric Association 2002). However, if tolerated, a single daily dose may be preferred. Compared to a multiple dose, a single-dose regimen induces less polyuria and thus reduces the long-term risk of renal damage (Malhi et al. 2012b). If tolerated, a single dose taken at night can be trialled (American Psychiatric Association 2002; Goodwin et al. 2016), and in some cases this dose can be reduced by up to 25 % in order to limit side effects, such as fatigue, and increase compliance (Letica-Crepulja et al. 2008; Malhi et al. 2009, 2012b).

To optimise the effectiveness of lithium treatment, it is important to appropriately manage its potential for adverse side effects and toxicity. Therefore, it is necessary to ensure that the correct blood level concentrations are maintained, and to achieve this lithium plasma concentration, testing must be conducted upon initiation of therapy and at regular intervals subsequently (Malhi and Tanious 2011). This is particularly important for those who respond to lithium, and it is maintained long term for prophylaxis. By instituting adequate monitoring and ensuring that optimal levels are maintained (tailored to the needs of the individual), the likelihood of recurrences will be greatly diminished, and side effects can be promptly managed if they arise. The ‘lithiumeter’ (Fig. 11.4) depicts indications and risks associated with lithium therapy according to its blood plasma levels. Optimal levels for treatment initiation and maintenance are between 0.6 and 0.8 mmol/L and should be individually tailored on the basis of recurrences and polarity of symptoms (Malhi et al. 2015). In prophylaxis, serum lithium levels can be increased up to 1.0 mmol/L for mania and decreased to 0.4 mmol/L for bipolar depression (Malhi et al. 2011, 2012b). Maintaining optimal plasma levels in lithium prophylaxis is central to the effective maintenance of remission and the prevention of future episodes, as well as minimising side effects and ensuring treatment adherence.

Lithium is associated with both acute and chronic side effects that can limit its tolerability. Nonetheless, correct monitoring practice can enable patients to achieve remission with the fewest side effects possible. At the initiation of lithium treatment, it is common for a rapid increase of plasma lithium concentration to cause acute side effects that include tremor, general fatigue, diarrhoea, thirst, polyuria, nausea, headache and vomiting (Fyrö et al. 1970; Lydiard and Gelenberg 1982). In practice, these normally subside after a day or so and are rarely troublesome especially if patients are forewarned.

Over the long term, clinical consideration must be given to renal and endocrine function, obesity and metabolic syndrome, cardiovascular disease, neuroprogression and cognitive impairment. Although modest in most cases, renal function, and in particular renal concentrating ability, can be impaired by long-term lithium use, leading to polyuria and secondary thirst (Grandjean and Aubry 2009). The risk of renal failure with lithium therapy is often at the forefront of clinicians’ minds and no doubt in many cases prevents or limits its use. In actuality, end-stage renal failure only occurs in 0.53 % of lithium-treated individuals, compared to 0.2 % in the general population (Coresh et al. 2003). The vast majority of patients on long-term lithium therapy experience a gradual decrease in glomerular function that is comparable to that of normal ageing (Gitlin 1999; Silverstone 2000). Nonetheless, routine examinations of urea, creatinine and electrolytes are recommended both upon initiation of lithium therapy and during ongoing maintenance treatment (see Table 11.2). Similarly, because clinically significant hypothyroidism is observed in 8–19 % of patients receiving lithium therapy, endocrine monitoring is essential to detect and manage thyroid abnormalities, especially since hypothyroidism is associated with depression and rapid cycling (Yatham et al. 2005). Furthermore, patients with hypothyroidism whilst on lithium therapy should be prescribed thyroxine (Freeman and Freeman 2006). Another endocrine problem that patients on long-term lithium therapy are prone to develop is that of hypercalcaemia secondary to elevated parathyroid hormone levels, which is also associated with a decrease in bone resorption (Grandjean and Aubry 2009). Therefore, careful monitoring of the approximate endocrine measures is necessary to inform long-term management—involving adjustments in lithium dose to mitigate potential adverse effects. Regular monitoring (see Table 11.2) should also include assessments of obesity, metabolic dysfunction and cardiovascular illness, given that these illnesses are associated with bipolar disorder and have shared pathogenetic pathways (Outhred et al. 2016; Vancampfort et al. 2013).