OBJECTIVE: Anemia and iron deficiency are common complications in patients with chronic kidney disease (CKD). However, information about the diagnostic indicators of bleeding-related upper gastrointestinal (GI) tract lesions is sparse and few studies have investigated anemic upper GI tract lesions. METHODS: We included 165 anemic patients with non-dialysis-dependent CKD stages 3 to 5 (44 patients at stage 3, 52 patients at stage 4, and 69 patients at stage 5). Transferrin saturation (TSAT), serum ferritin, mean corpuscular volume, and corrected reticulocyte count data were collected to evaluate their diagnostic use for bleeding-related upper GI tract lesions, which were identified by esophagogastroduodenoscopy. RESULTS: Bleeding-related GI tract lesions were found in 57 patients (34.5%). The area under the receiver-operating characteristic curve used to predict bleeding-related lesions was 0.63 for TSAT (P = 0.007), and the best cutoff value was 19.7% (sensitivity, 0.53; specificity, 0.76). The combination of cutoffs TSAT less than 20% or serum ferritin less than 100 ng/mL produced a 17% increment in sensitivity compared with that of TSAT less than 20% alone. The corrected reticulocyte levels and mean corpuscular volume had no significant diagnostic use. In patients with CKD stage 5, the sensitivity of TSAT or its combination with serum ferritin less than 100 ng/mL was significantly lower than in patients with CKD stage 3 (all P < 0.05). CONCLUSIONS: Transferrin saturation is a significant predictor of anemic lesions in the upper GI tract, and serum ferritin can increase the sensitivity of TSAT. However, these indicators should be used with caution in patients with CKD stage 5 because their sensitivity is poor in this context.
OBJECTIVE:Anemia and iron deficiency are common complications in patients with chronic kidney disease (CKD). However, information about the diagnostic indicators of bleeding-related upper gastrointestinal (GI) tract lesions is sparse and few studies have investigated anemic upper GI tract lesions. METHODS: We included 165 anemicpatients with non-dialysis-dependent CKD stages 3 to 5 (44 patients at stage 3, 52 patients at stage 4, and 69 patients at stage 5). Transferrin saturation (TSAT), serum ferritin, mean corpuscular volume, and corrected reticulocyte count data were collected to evaluate their diagnostic use for bleeding-related upper GI tract lesions, which were identified by esophagogastroduodenoscopy. RESULTS: Bleeding-related GI tract lesions were found in 57 patients (34.5%). The area under the receiver-operating characteristic curve used to predict bleeding-related lesions was 0.63 for TSAT (P = 0.007), and the best cutoff value was 19.7% (sensitivity, 0.53; specificity, 0.76). The combination of cutoffs TSAT less than 20% or serum ferritin less than 100 ng/mL produced a 17% increment in sensitivity compared with that of TSAT less than 20% alone. The corrected reticulocyte levels and mean corpuscular volume had no significant diagnostic use. In patients with CKD stage 5, the sensitivity of TSAT or its combination with serum ferritin less than 100 ng/mL was significantly lower than in patients with CKD stage 3 (all P < 0.05). CONCLUSIONS:Transferrin saturation is a significant predictor of anemic lesions in the upper GI tract, and serum ferritin can increase the sensitivity of TSAT. However, these indicators should be used with caution in patients with CKD stage 5 because their sensitivity is poor in this context.