Abstract
Various laboratory parameters are used to guide the treatment of anemic patients; however, their interpretation may be complicated by the etiology of the anemia. We compared anemia-related laboratory values obtained from 261 patients (174 female) screened prior to participation in a clinical trial for cancer and chemotherapy related anemia (ACD), with those from a group of 50 otherwise-healthy women referred for treatment of anemia related to heavy menses (IDA). Pairwise correlations were explored graphically and analyzed using Pearson’s correlation coefficients. Highly skewed data were log transformed. Relationships between Hb and laboratory values associated with anemia were explored using multiple linear regression models; the most parsimonious model was arrived at by stepwise regression. For the ACD group, laboratory values tested in the regression analyses included: ferritin, transferrin saturation (TSAT), reticulocyte Hb content, folate, vitamin B12, transferrin, iron, and albumin. For the IDA group, regression analyses included: ferritin, iron binding capacity (TIBC), iron, albumin, BUN, Creatinine, WBC, C-reactive protein, and endogenous EPO. Complete set of lab values was not available for all patients. In the IDA group, a strong correlation was identified between markers of iron status, serum ferritin and TSAT, but this association was much weaker among the ACD group. In the ACD group, negative correlations were identified between ferritin and TIBC, and between ferritin and albumin, along with a moderately strong positive correlation between albumin and TIBC. These relationships were different in the IDA group. Albumin tended to increase with increasing ferritin, while TIBC tended to decrease. Albumin and TIBC were unrelated.
Lab value, mean, 95% CI ACD group IDA group Hgb g/dL 10.1, 10.0–10.3 8.6, 8.01–9.10 ferritin ng/mL 455, 396–513 10, 6.1–13.0 TSAT% 21.3, 19.7–23.0 12.8, 0.01–25.7 TIBC mg/dL 275, 267–282 422,375–469 iron mcg/dL 61, 57–66 49, 2.5–95.7 albumin g/dL 3.6, 3.55–3.65 4.0, 3.86–4.11 Correlations ferritin vs TSAT r=0.208, p=.001 r=0.793, p<.0001 ferritin vs albumin r=−0.354, p<.0001 r=0.209, p=.144 ferritin vs TIBC r=−0.610, p<.0001 r=−0.269, p=.059 albumin vs TIBC r=0.578, p<.0001 r=−0.037, p=.800
Among the IDA group, both serum iron (coefficient=0.0129, 95% CI=0.00431–0.0214; p=0.008), and EPO (coefficient = −0.002, 95% CI=−0.001 to−0.003; p=0.008) were significant independent predictors of Hgb, such that lower serum iron and higher endogenous EPO levels were associated with lower Hgb. However, among the ACD group, the only consistent and clinically significant independent predictor of Hgb was serum albumin (coefficient = 1.22, 95% CI = 0.97–1.48; p<0.0001), such that lower serum albumin was associated with lower Hgb. Serum ferritin is an acute-phase reactant that increases during inflammatory states, whereas both serum transferrin and serum albumin are negative acute-phase reactants that decrease during inflammatory states. Elevated serum ferritin and decreased transferrin/TIBC may mask an underlying state of iron deficiency in inflamed patients. While traditional biochemical markers ferritin and TSAT can assist in diagnosis of iron deficiency in otherwise-healthy anemic individuals, they are of limited value in the detection of iron deficiency associated with chronic disease.
Biochemical Markers and Hematologic Indices in the Diagnosis of Functional Iron Deficiency