Abstract
BACKGROUND: Fatigue, although common and associated with outcomes in dialysis-dependent chronic kidney disease (CKD), has not been studied in nondialysis chronic kidney disease (CKD-ND) patients. METHODS: In this longitudinal cohort of 266 outpatients with CKD-ND stages 2-5, we measured self-reported fatigue on 3 scales-Quick Inventory of Depression Symptomatology-Self Report (QIDS-SR16), Beck Depression Inventory-I (BDI-I), and short form 12 health survey (SF-12) questionnaires and evaluated the prespecified composite of progression to dialysis initiation, death, or hospitalization after 12 months. Logistic and linear regression assessed characteristics associated with fatigue. Survival analysis measured associations of fatigue with outcomes. RESULTS: Mean age was 64.4 ± 12.0 years, and mean estimated glomerular filtration rate (eGFR) was 31.6 ± 16.7 mL/min/1.73 m2. Fatigue was common, with 69.2% reporting fatigue on QIDS-SR16 and 77.7% on BDI-I. Unemployment, comorbidities, use of antidepressant medications, and lower hemoglobin correlated with fatigue. There were 126 outcome events. Participants that reported any versus no fatigue on QIDS-SR16 were more likely to reach the composite, hazard ratio (HR) 1.70 (95% CI 1.11-2.59), which persisted after adjusting for demographics, comorbidities, substance abuse, hemoglobin, albumin, eGFR, and calcium-phosphorus product, HR 1.63 (1.05-2.55). Fatigue severity by the SF-12 was also associated with outcomes independent of demographics, comorbidities, and substance abuse, HR per unit increase 1.18 (1.03-1.35). No association was observed with fatigue on the BDI-I. CONCLUSION: Fatigue affected about 2/3 of CKD-ND patients and associated with unemployment, comorbidities, antidepressant medication use, and anemia. Fatigue measured by the QIDS-SR16 and SF-12 independently predicted outcomes in CKD patients. Eliciting the presence of fatigue may be a clinically significant prognostic assessment in CKD patients.