Abstract
Objective
To review the relationship between lithium-related renal dysfunction and microcysts.
Method
Electronic databases (PubMed and Google Scholar) were queried.
Results
From a total of 12,425 publications, 76 were reviewed.
Discussion
Glomerular renal dysfunction occurs after an average of 20 years of continuous lithium treatment, and the severity is related to the total lithium load as measured by dose and duration. Recently, several reports have highlighted the relationship between renal microcyst formation and significant reductions in glomerular filtration rate. Radiologically visible lithium-related microcysts are usually 1–2 mm and occasionally 3 mm. Smaller cysts, which are impossible to resolve, are probably more common than the visible cysts, based on observations of renal needle biopsies. Increases in the number of microcysts and the space they occupy within kidney volume appear to be related to both the duration of lithium treatment and the reduction in kidney function. The proposed mechanism of microcyst formation is related to the antiapoptotic effect of lithium. Specifically, by preventing renal tubular epithelial cells from undergoing apoptosis as part of the process of normal renal maintenance, lithium may be allowing the inappropriate growth of the surface area of tubules to form invaginations and ultimately cysts. It is proposed that the physical space occupied by these cysts in the limited volume within the renal capsule compromises the function of otherwise healthy renal tissue. Monitoring of kidneys utilizing radiographic imaging may be more sensitive than monitoring laboratory values. Additional research is required to optimize this new monitoring tool.
Keywords
Lithium was the first specific treatment for bipolar disorder. Its efficacy in this illness was originally discovered by John Cade in 1949. 1 By the 1950s, lithium was in widespread use in Europe. 2 In the late 1960s, lithium was being administered experimentally in the United States under a research program sponsored by the National Institute of Health (since the lithium ion was not patentable, the pharmaceutical industry would not fund the U.S. Food and Drug Administration-mandated studies 3 ). In 1970, lithium was approved by the Food and Drug Administration in the United States. Today, nearly 65 years after its initial modern introduction, lithium is still widely used to treat bipolar disorder and is considered to be the gold standard by many.4,5
Despite its clear utility, lithium is prescribed for only about one third of bipolar patients in the United States (29.5% among older bipolar patients and 37.8% among younger ones 6 ), and its use may be decreasing (from 27.6% in the 2000–2005 epoch to 22.7% in the 2006–2011 epoch7,8). However, in developing countries and some European countries (e.g. Spain and Britain9,10), lithium use is more common at over half of bipolar patients (e.g. 57% in northern India 11 ). One of the major obstacles to the use of lithium is its narrow therapeutic index with the potential for acute toxicity and problematic adverse effects. 12 While lithium may induce many reversible adverse effects—such as tremor, dyscoordination, gastrointestinal disturbance, and thyroid resistance—it is the irreversible renal dysfunction that is most troubling. 12
Renal consequences of lithium treatment became apparent soon after its introduction into psychiatry. That lithium causes tubular dysfunction is widely known, and patients taking lithium are frequently unable to concentrate their urine and may experience diabetes insipidus. 13 However, it is the more disconcerting renal insufficiency that worries clinicians and patients.
Throughout the 1950s, before lithium’s introduction into the United States, there was a spate of cases in which patients presented with lithium toxicity and renal shutdown. 14 A detailed review of these cases revealed that most of these patients had become dehydrated, frequently secondary to polyuria, and then developed prerenal, renal shutdown. 14 The combination of dehydration and acute renal shutdown resulted in elevated lithium levels and lithium toxicity. In all cases in which follow-up was reported in these initial cases, renal recovery occurred with rehydration, 14 leading to the conclusion that renal complications with lithium are reversible. 2 The idea that chronic lithium use can induce chronic renal injury was introduced in 1977.15,16 And while indeed there were associations with lithium and renal disease, 17 the issue was confirmed with studies that compared patients with mood disorder taking lithium with patients with a mood disorder but not taking lithium. These studies revealed a lithium-related decline in renal function. 18 . The most severe consequence, end-stage renal disease (ESRD), is clearly uncommon. It appears to be related to the duration of lithium treatment (generally over 10 years of continuous lithium use18,19) and perhaps the levels at which lithium is maintained. 19 There is evidence that monitoring greatly reduces the risk of ESRD; 19 nonetheless, the fear of eventual ESRD may prevent the recommendation of initiating lithium treatment when clinically indicated.
Lithium-related ESRD (Li-ESRD) appears to be associated with renal microcyst formation. There is accumulating evidence that microcyst formation predates the increase in serum creatinine. Recent advances in the imaging technologies that allow noninvasive visualization of the kidneys raise the potential new method of monitoring patients on lithium treatment for future development of Li-ESRD. This article reviews the literature regarding microcyst formation and Li-ESRD and puts forward the idea of periodic imaging with computed tomography or magnetic resonance imaging (MRI) of the kidneys of patients on chronic lithium treatment to augment routine laboratory measures, to further reduce the likelihood of Li-ESRD.
Methods
A selected review of the literature was made to investigate microcyst formation and lithium treatment. Computerized databases of PubMed and Google Scholar were queried with the key words of “micro cysts,” “microcysts,” “lithium,” “kidney,” and “kidneys.” Relevant articles were examined by the two coauthors and included in the review where indicated. Additional appropriate references were investigated as needed.
Results
PubMed search produced a total of 25 citations in six separate searches. Google Scholar resulted in 12,400 references, of which only 70 were relevant.
Discussion
Renal dysfunction
Chronic lithium use is associated with renal dysfunction, which can be divided into two broad clinical categories. Lithium may reduce the kidney’s ability to concentrate urine, usually with minimal changes in creatinine clearance. This is generally described as nephrogenic diabetes insipidus and is experienced as polyuria and polydipsia. 20 This is the most common renal complaint in patients receiving lithium and occurs in nearly a quarter of patients receiving lithium. 20 Alternatively, lithium may be associated with reduction in the glomerular filtration rate (GFR) with progression to ESRD. ESRD is probably uncommon, but some level of significantly reduced glomerular filtration may occur in 15% of patients receiving lithium 20 and may represent 0.22% of all patients on dialysis. 21 Some decline in GFR may occur in many patients within one year of initiating lithium treatment. 22 Tubular and glomerular dysfunction are not mutually exclusive and may frequently co-occur. But they are also not mutually predictive; that is, development of reduced concentrating ability is not predictive of ultimate reduction in GFR. These complications may be associated with other renal manifestations, which are probably related, such as renal tubular acidosis, tubulointerstitial nephritis, and nephrotic syndrome. On renal biopsy in patients with lithium-associated reduction in GFR, there are some individuals with evidence of glomerulonephritis, suggesting the possibility of more than one mechanism of lithium-associated reduction in GFR. 23
ESRD is clearly the most problematic consequence of chronic lithium treatment. Among susceptible individuals, decline in GFR appears to be a slowly progressive condition with an estimated latency of 20 years between onset of lithium treatment and ESRD. 21 The average age of patients who develop ESRD is 65 years. 21 In these individuals, creatinine clearance may decline at a rate of 2.29 mL/min/year, and this decline is related to the total load of lithium, as expressed in both cumulative dose and duration of treatment. 21 It is unclear what may predispose patients to this type of kidney dysfunction, but the best predictor of who will develop ESRD is a creatinine of >2.5 mg/dL. Among patients with some level of kidney dysfunction, 77.8% of patients with serum creatinine >2.5 mg/dL will develop ESRD, compared with 10% of people with creatinine <2.5 mg/dL. 23
Microcysts
Early attempts to understand lithium-related nephropathy led to studies of renal tissue by needle biopsy. While these studies have revealed a wide range of abnormalities, the appearance of microcysts was notable and occurred in the majority of patients (62.5% 23 ). This also occurs in rats treated with lithium for at least eight weeks postnatally.24,25 More recently, as noninvasive imaging technologies have become more powerful, the presence of microcysts in patients developing lithium-related ESRD has been documented with ultrasound,26–28 computed tomography,28,29 and MRI.27,29–31 Ultrasonography is sometimes difficult to utilize since classic cysts appear as anechoic, but lithium-related microcysts may have the appearance of small echogenic foci. 28 T2-weighted MRI may be particularly effective at detecting microcysts because the cysts appear hyperintense.27,29 Most microcysts were measured at 1–2 mm and occasionally reaching 3 mm. There may be smaller cysts, but the lower limit of visualization with the most sensitive method, MRI, is 0.9 mm. 32 Kidney size is normal, which frequently rules out other degenerative conditions.29,31 Furthermore, lithium-associated cysts also tend to be equally distributed throughout the kidney in both the cortex and medulla and distributed equally in both kidneys.29,31 Furthermore, the number of cysts is generally related to the longitudinal exposure to lithium, so that microcysts are present long before the actual elevation in creatinine but increase in number over time. 33
In autosomal-dominant polycystic kidney disease34–36 and cyclosporine-related cystic disease, 37 there appears to be an inverse relationship between the size of cysts and the GFR of the patient. In cystic renal disease, the more space occupied by the cysts, the lower the GFR. It is likely that a similar relationship exists in lithium-associated microcystic changes. 29 This provides a potential mechanism by which microcyst formation may lead to reduced GFR. The relationship renal parenchyma and the renal capsule is analogous to that between brain tissue and the skull. Just as an enlarging brain mass compromises healthy brain tissue by “squeezing it” against the hard skull, so will enlarging renal cyst volume compromise healthy renal tissue by “squeezing it” against the unyielding capsule. If this mechanism is accurate, it provides an easy and reliable way to monitor the decline of renal function even before an observed severe decline in GFR; specifically, the volume of microcysts may provide a clue as to the extent of compromised renal tissue.
Mechanism of nephropathology
Similar changes to renal function occur in rodents exposed to “therapeutic” lithium levels postnatally but not prenatally.24,25,38 Availability of this animal model allows for examination of potential mechanisms. While the absolute mechanism of lithium-related renal damage is still not known, glycogen synthase kinase-3beta (GSK-3β) appears to be important. GSK-3β plays a role in many lithium-related processes. 39 GSK-3β is an enzyme that phosphorylates multiple targets, usually inhibiting their activity. While glycogen synthase was the first target described, GSK-3β is involved in multiple cellular processes including energy metabolism, apoptosis, cellular proliferation, and cytokine production. Inhibition of this enzyme by lithium at therapeutic levels has been previously demonstrated. 40 This action of lithium is believed to be important in the neuroprotective role of lithium. 41 Specifically, lithium may prevent the proapoptotic effect of GSK-3β. 42 This same antiapoptotic mechanism may be the reason for renal dysfunction.
Specifically, the epithelial cells in the tubules of the kidney are replaced on a regular basis, and that replacement process involves apoptosis of the older cells. 43 However, if lithium prevents these older cells from dying and disappearing to make room for the new replacements, then the excess epithelial surface area begins to invaginate and ultimately forms cysts. 44 The expansion of the volume of cysts ultimately reduces the volume of healthy renal tissue, and as this process continues to a great extent, a decline in GFR becomes evident. Thus, ironically, it is the somewhat desired antiapoptotic action of lithium that may lead to renal dysfunction.
If this is the actual mechanism of lithium-induced renal dysfunction, then one might predict that other antiapoptotic medications may have similar effects over long period of time. Specifically, since valproic acid may also have antiapoptotic effects, 45 long-term treatment with this anticonvulsant may also predispose to formation of renal microcysts. This phenomenon may underlie the frequently observed problem of progression of renal dysfunction in patients with bipolar illness even after the discontinuation of lithium. 23 It is clear that studies of high-resolution renal imaging of patients treated for long durations with neuroprotective, antiapoptotic anticonvulsants are indicated.
Clinical relevance
The discovery that microcysts can be easily and reliably imaged with noninvasive techniques provides a potentially potent technique for physicians to screen their patients for lithium-associated renal dysfunction. Microcysts appear prior to the decline in GFR or elevation in serum creatinine. Changes in microcyst number can be followed over time with repeated imaging. It is likely that T2-weighted MRI imaging, with specific attention to kidney microcysts, is more sensitive and more specific for lithium-associated renal injury than laboratory monitoring of creatinine and estimated GFR. Additionally, the utilization of MRI has the added benefit of reducing radiation exposure that would be associated with repeated computed tomography scanning. Future research is required to determine the optimal interval of imaging, whether baseline (pre-lithium initiation) imaging is required, the optimal manner of quantifying the microcyst volume or the number of cysts per kidney, and the true numeric relationship between elaboration of microcysts and GFR decline. In the meantime, mental health clinicians need to become familiar with this phenomenon and utilize it to provide better care for their patients.
Footnotes
Declaration of Conflicting Interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.