Our results, even if limited by the cross-sectional design and relatively small sample size, underscore that the duration of lithium exposure contributes to the effect of advancing age in the decline of renal function. The trend of our results is similar to that reported in a recent paper with a very similar design, that is, a cross-sectional study of 61 patients treated with lithium for a mean of 11.5 years whose eGFR was compared to 62 control patients .
With regard to end-stage renal disease, it had long been considered an unlikely event in lithium patients  until recently, when Bendz et al. suggested that renal failure does occur in chronic lithium treatment, even if it is uncommon .
In a systematic review and meta-analysis , McKnight and colleagues concluded that the reduction in eGFR in lithium patients is small (a maximum of 5 ml/min/year), representing only 5% of the minimum normal GFR. However, assuming that the decline is linear, 20 years of lithium therapy (a duration which is often attained in prophylaxis) would result in 100% loss of glomerular filtration rate; their estimate, however, was based on prospective studies with a mean observation time of one year on lithium. We maintain that reduced glomerular filtration rate is not clinically negligible even if not falling in CKD stage 5, namely end-stage renal failure, the condition that raised the greatest concern after recent reports . Patients at CKD stage 3 (GFR of 30 to 60 ml/min) raise clinical, ethical and legal questions regarding the decision to continue or discontinue lithium prophylaxis. Moreover, evidence of the decline in renal function with duration of lithium treatment and advancing age may argue for the revision of current guidelines recommending early maintenance in recurrent affective disorder. In our study, we can estimate according to the multivariate analysis, that, for example, all patients 60 years old or older may undergo CKD stage 3 or more severe stages if treated with lithium for 30 years. On the other hand, the incidence of ESRD is considered uncommon based on data from renal dialysis and transplant registers . In any case, the risk of renal failure needs to be weighed against the benefits obtained, as recently reported by Werneke et al. , who conducted a decision analysis simulating the decision process between physicians and patients, comparing the relative risks and utility of maintenance treatment. The analysis addressed two questions: 'Should lithium be recommended at the beginning of treatment in view of a small but significant risk of ESRD later in life?' and 'Should lithium continuation be recommended even in the presence of long-term adverse renal effects?' This involved weighing the need for effective relapse and suicide prevention from the very beginning of treatment against the risk of lithium-associated ESRD occurring many years later. Indeed, besides its potentially detrimental effects on renal function, lithium has been found to protect not only from recurrence of affective disorders but also from their otherwise high mortality, principally due to suicide [12, 18]. In a follow-up of cause-specific mortality in 1,411 patients registered at our lithium clinic between 1980 and 2000, renal failure was recorded as the main cause in two of 201 death certificates, compared with 43 definite suicides (42 of which committed after abandoning lithium prophylaxis) . The progression of renal dysfunction may be very slow, even in patients continuing lithium. In patients at risk of ESRD from other causes, the eGFR point of no return may be set at 30 ml/min. In this cohort, five women fall into this category and are being closely monitored with regard to lithium treatment and renal function. Perhaps it is safe to start early lithium maintenance in patients with normal renal function, provided the function is followed up regularly. Discontinuation of lithium should be considered if the decline in glomerular filtration rate exceeds the expected age-related decline, even though there may be exceptions. The following emblematic case shows the importance of interaction between treating psychiatrists and consultant nephrologists: Belgamwar et al.  reported a 73-year-old woman with bipolar disorder who was stable on lithium. After long-term use of lithium, she developed chronic renal failure. Doctors decided to stop the lithium and try alternatives, but this proved unsuccessful and resulted in a very poor course, including long hospital admissions. They, therefore, respected the patient's wishes and ability to make a decision and retried lithium. The patient herself and her family accepted this decision despite a high risk of going into dialysis in the future.
We now plan to study the entire cohort of lithium patients registered at our facility over the last three decades (approximately 1,500) to estimate the incidence and time-to-event of the various CKD stages, including patients who had to stop lithium due to very low eGFR or required dialysis or renal transplant. This larger, more comprehensive study should, in part, amend the limitations of this cross-sectional study of 139 lithium patients.