OBJECTIVES: To determine the diagnostic values of serum ferritin and other conventional laboratory tests in the diagnosis of iron deficiency anaemia in patients with liver cirrhosis. DESIGN: Cross-sectional study for diagnostic tests. SETTING: University hospital. SUBJECTS: Seventy-two consecutive patients with liver cirrhosis in whom the haemoglobin level was less than 13.0 g dL(-1) for men and 12.0 g dL(-1) for women. The diagnosis of liver cirrhosis was based on characteristic clinical and hepatic ultrasonographic findings. MAIN OUTCOME MEASURES: By using absence of bone marrow iron as the standard criterion, the diagnostic powers of mean corpuscular volume, transferrin saturation, serum ferritin and the presence of hypochromic red cells in the diagnosis of iron deficiency were compared by analysing the likelihood ratios, the area under the receiver operating curves (ROC) and the stepwise logistic regression associated with each test. RESULTS: Twenty-nine patients (40.3%) demonstrated no stainable iron in the bone marrow. The likelihood ratios, the area under the ROC and the stepwise logistic regression indicated that serum ferritin was the most powerful test predictive of iron deficiency. Other tests added little further diagnostic values. The likelihood ratios associated with the serum ferritin levels were as follows: <50 microg L(-1), 22.3; 51-200 microg L(-1), 1.5-1.8; 201-400 microg L(-1), 1.0; >400 microg L(-1), <1. These results indicate that serum ferritin level <50 microg L(-1) depict a very high probability of iron deficiency anaemia (0.83-0.99) irrespective of the patient's pre-test probability. CONCLUSION: Serum ferritin is the most powerful noninvasive test for the diagnosis of iron deficiency anaemia in patients with liver cirrhosis. It should be the primary test of choice in patients suspected of having the disease. When the level was less than 50 microg L(-1), iron supplement may be prescribed without necessitating bone marrow aspiration.
OBJECTIVES: To determine the diagnostic values of serum ferritin and other conventional laboratory tests in the diagnosis of iron deficiency anaemia in patients with liver cirrhosis. DESIGN: Cross-sectional study for diagnostic tests. SETTING: University hospital. SUBJECTS: Seventy-two consecutive patients with liver cirrhosis in whom the haemoglobin level was less than 13.0 g dL(-1) for men and 12.0 g dL(-1) for women. The diagnosis of liver cirrhosis was based on characteristic clinical and hepatic ultrasonographic findings. MAIN OUTCOME MEASURES: By using absence of bone marrow iron as the standard criterion, the diagnostic powers of mean corpuscular volume, transferrin saturation, serum ferritin and the presence of hypochromic red cells in the diagnosis of iron deficiency were compared by analysing the likelihood ratios, the area under the receiver operating curves (ROC) and the stepwise logistic regression associated with each test. RESULTS: Twenty-nine patients (40.3%) demonstrated no stainable iron in the bone marrow. The likelihood ratios, the area under the ROC and the stepwise logistic regression indicated that serum ferritin was the most powerful test predictive of iron deficiency. Other tests added little further diagnostic values. The likelihood ratios associated with the serum ferritin levels were as follows: <50 microg L(-1), 22.3; 51-200 microg L(-1), 1.5-1.8; 201-400 microg L(-1), 1.0; >400 microg L(-1), <1. These results indicate that serum ferritin level <50 microg L(-1) depict a very high probability of iron deficiency anaemia (0.83-0.99) irrespective of the patient's pre-test probability. CONCLUSION: Serum ferritin is the most powerful noninvasive test for the diagnosis of iron deficiency anaemia in patients with liver cirrhosis. It should be the primary test of choice in patients suspected of having the disease. When the level was less than 50 microg L(-1), iron supplement may be prescribed without necessitating bone marrow aspiration.
Authors: K S Phiri; J C J Calis; D Kachala; E Borgstein; J Waluza; I Bates; B Brabin; M Boele van Hensbroek Journal: J Clin Pathol Date: 2009-08 Impact factor: 3.411