IntroductionThyroid hormones can directly affect kidney function; elevated levels of thyroid-stimulating hormone (TSH) and chronic kidney disease (CKD) are associated with proteinuria, decreased estimated glomerular filtration rate (eGFR), and progression to end-stage renal disease. Our hypothesis is that in patients with CKD and TSH at levels considered to be in the low subclinical hypothyroidism (SCH) range, lowering TSH with levothyroxine (LVX) improves the clinical parameters of renal function.MethodsThis was a double-blind, randomized, pilot clinical trial in patients with proteinuric CKD (eGFR <60 ml/min per 1.73 m2 and proteinuria >150 mg/d) performed at the Hospital Civil de Guadalajara, with the intention of lowering TSH (levels of 1.25−2.5 μIU/l) in patients with TSH (levels of 2.6−9.9 μIU/ml with FT4 in the range of 0.7−1.8 ng/dl). Patients were randomized 1:1 to receive LVX or placebo for 12 weeks. The primary objective was to evaluate absolute levels of proteinuria at the beginning compared to the end of the study and, as a secondary objective, the changes in serum creatinine (sCr), eGFR, cholesterol, triglycerides, low-density lipoprotein (LDL), and blood pressure, and to assess the tolerability and safety of LVX.ResultsBetween March and November 2018, a total of 163 patients were assessed for eligibility; 119 patients did not meet the inclusion criteria or were excluded, and 32 patients were randomized. The demographic and clinical characteristics of the 2 study groups were essentially not different. Subjects were 66.87 (SD 12.19) years of age, 62.5% were female, 75% were diabetes mellitus, eGFR was 23.55 (±12.91) ml/min per 1.73 m2, TSH was 5.37 ± 2.13 μIU/ml, proteinuria in 24-hour urine collection was 1.52 ± 1.12, and all of them were taking angiotensin-converting enzyme inhibitors (ACEIs) or angiotensin II receptor blockers (ARBs). Proteinuria at 12 weeks in the LVX group was 0.89 SD ± 1.28 g/d, and in the placebo group it was 1.35 SD ± 0.85 g/d; when compared to placebo, LVX showed a significant decrease in proteinuria of 1.1 g/d (P = 0.0011). The eGFR in the LVX group showed an improvement of 4 ml/min/1.73 m2 (P = 0.049); in the placebo group, there was a decrease of 1.98 ml/min per 1.73 m2. The sCr, cholesterol, triglycerides, low-density lipoprotein, systolic blood pressure, and diastolic blood pressure were not different between groups. Adverse events were reported in the LVX group in 7.14% of patients and in 11.11% of patients in the placebo group; none left the study because of adverse effects, and there were no serious adverse events.ConclusionThis single-center, randomized, double-blind, placebo-controlled pilot clinical trial in patients with advanced proteinuric CKD who already used ACEIs or ARBs demonstrated that administering LVX to obtain a TSH range close to 2.5 μIU/ml decreased proteinuria and improved eGFR. Future research is needed to confirm our results and to determine whether our findings generalize to patient groups not explicitly enrolled in this small pilot trial.