Table 1.
Reports on renal effects of lithium treatment
| Report | Subjects (n) | Age (years) | Lithium exposure (years) | Main findings | |
|---|---|---|---|---|---|
| Li | No Li | ||||
| Hullin et al. (1979) | 30 | 30 | 55 | 6.2 | No difference in eGFR |
| Depaulo et al. (1981) | 99 | 0 | 41 | 2.8 | Creatinine increased with Li |
| Bendz (1985) | 32 | 32 | 49 | 5.7 | No difference in eGFR |
| Bendz et al. (1996) | 13 | 13 | 59 | 18.0 | eGFR fell with Li |
| Coşkunol et al. (1997) | 107 | 29 | 39 | 4.5 | No difference in eGFR |
| Turan et al. (2002) | 10 | 10 | 35 | 1.3 and 6.7 | eGFR fell with long-term Li |
| Bendz et al. (2010) | 3369 | Genl. pop. | 65 | 23.0 | ESRD 6.5-fold more often with Li |
| Rybakowski et al. (2012) | 80 | 0 | 60 | 16.0 | eGFR < 60: 22.5%; 2.4-times more in men |
| Bocchetta et al. (2013) | 139 | 70 | 54 | >1.0 | eGFR < 60: 4.8-fold more often with Li |
| Minay et al. (2013) | 330 | 659 | 48 | – | eGFR < 60: similar with/without Li |
| Aiff et al. (2014) | 1995 | 0 | 66 | 27.0 | ESRD 7.8-fold more often with Li |
| Aprahamian et al. (2014) | 32 | 27 | 74 | 4.0 | No difference in renal function |
| Close et al. (2014) | 2496 | 3864 | 49 | – | eGFR < 60: 3.25-times less with Li |
| Aiff et al. (2015) | 630 | 0 | 66 | ≥10.0 | eGFR < 60: 32%; ESRD: 4.5-fold more with Li |
| Bocchetta et al. (2015) | 1953 | 0 | – | 10 and 25 | eGFR < 50: 12% in 10, 50% in 25 yrs of Li |
| Clos et al. (2015) | 305 | 815 | 43 | 4.6 | No difference in eGFR |
| Shine et al. (2015) | 4678 | 689,228 | 52 | ≤28.0 | eGFR < 60: 1.21-fold more often with Li |
| Castro et al. (2016) | 3850 | 0 | 54 | 1.4 | eGFR < 60: 25.7% lower with multiple doses/day |
| Hayes et al. (2016) | 2148 | 4523 | 46 | 18 | eGFR < 60: ~twofold higher HR with Li |
| Kessing et al. (2015) | Natl. sample | 0 | – | – | Clinical CKD 3.6-times more with Li |
| N = 20 studies | >22,296 | >699,300 | 53.1 ± 10.5 | 10.9 ± 8.9 | Function decreased in 15/20 reports (75.0%) |
Abnormal renal functioning was associated with longer exposure to lithium in these studies (15.3 ± 9.54 vs. 5.00 ± 0.91 years, respectively [t = 2.37, p = 0.035])