Masafumi Koga

Case 1 presented with severe anemia and received an intrauterine blood cell transfusion at 33 weeks of gestation. The anemia spontaneously improved in early infancy. Case 2, the father of Case 1, had an uneventful birth with no evidence of anemia, though microcytic anemia was observed during childhood. The genetic analysis of the β‐globin gene cluster identified a novel heterozygous deletion of DNA extending from the Gγ‐globin gene downstream to the β‐globin gene, confirming a diagnosis of (^Gγ^Aγδβ)⁰‐thalassemia. In cases where thalassemia is suspected based on blood tests, a genetic diagnosis should be performed for the sake of the offspring.

Background: Hemoglobin A1c (HbA1c) levels are low in patients with hemolytic anemia, as HbA1c reflects mean erythrocyte age (M_RBC). Erythrocyte creatine (EC) is a hemolytic indicator that also reflects M_RBC. We previously reported an equation for estimating M_RBC using EC (EC‐M_RBC).Aims: In this study, EC‐M_RBC was compared to the HbA1c level expressed in the International Federation of Clinical Chemistry and Laboratory Medicine units (iA1c) and to the iA1c/glycated albumin (GA) ratio to estimate M_RBC.Methods: This study included 238 subjects, including patients with hemolytic anemia and/or type 2 diabetes mellitus (T2DM).Results: In non‐diabetic individuals, both iA1c and iA1c/GA showed a strong positive correlation with EC‐M_RBC (p < 0.0001). The equations to estimate iA1c‐M_RBC and iA1c/GA‐M_RBC derived from the regression equations between EC‐M_RBC and iA1c, and EC‐M_RBC and iA1c/GA in nondiabetic individuals were 1.45 × iA1c and 20.0 × iA1c/GA, respectively. iA1c‐MRBC and iA1c/GA‐MRBC in non‐diabetic individuals without hemolytic anemia were 57.6 ± 4.0 and 57.1 ± 6.4 days, respectively, and iA1c/GA‐MRBC in T2DM patients without hemolytic anemia was 56.0 ± 8.8 days.; no significant difference was seen in the comparisons.Conclusions: The M_RBC can be estimated using iA1c or iA1c/GA in non‐diabetic individuals, and iA1c/GA in T2DM patients.

Patients with diabetes mellitus having insulin antibodies (InsAb) with properties of high binding capacity and low affinity, which are observed in insulin autoimmune syndrome (IAS), are known to have greater plasma glucose fluctuations. Glycated albumin (GA) and the GA/HbA1c ratio have been demonstrated to reflect plasma glucose fluctuations. Hence, we hypothesized that GA or the GA/HbA1c ratio in diabetic patients having InsAb with properties of high binding capacity and low affinity may be higher than those in InsAb-negative diabetic patients, and we verified this hypothesis. Subjects were 12 diabetic patients who had InsAb noted while being treated with insulin and were subjected to Scatchard analysis and whose InsAb had properties similar to those of patients with IAS (affinity constant K1 < 0.24 × 1/10^–8 M, number of binding sites R1 ≥ 11.5 × 10^–8 M) [four cases of type 1 diabetes (T1D) and eight cases of type 2 diabetes (T2D)]. The control group consisted of T1D and T2D cases matched to the T1D and T2D cases, respectively, according to sex, age, BMI, and HbA1c. GA and the GA/HbA1c ratio were compared between both groups. GA and the GA/HbA1c ratio in InsAb-positive patients was significantly higher than that in the control group for both T1D and T2D patients. Diabetic patients having InsAb with properties of high binding capacity and low affinity had higher GA and the GA/HbA1c ratio than those of InsAb-negative patients. Greater plasma glucose fluctuations were suggested in InsAb-positive diabetic patients.

It has already been reported that HbA1c levels measured by immunoassay (IA) (IA-HbA1c) during off-site health checkups present falsely lower results. We also reported that HbA1c levels measured by enzymatic assay (EA) (EA-HbA1c) during off-site health checkups are lower. In the present study, we compared IA-HbA1c levels or EA-HbA1c levels during off-site health checkups with on-site high-performance liquid chromatography (HPLC)-HbA1c levels using the same samples. Subjects were 88 non-diabetic individuals who had health checkups in Nishinomiya Municipal Central Hospital. Subjects with a history of diabetes mellitus and those with HPLC-HbA1c ≥ 6.5% were excluded. IA-HbA1c levels (Study 1) or EA-HbA1c levels (Study 2) in the health checkups were compared with on-site HPLC-HbA1c levels using the same samples. Both IA-HbA1c levels and EA-HbA1c levels had positive correlations with HPLC-HbA1c levels (p <0.0001 for both), although both were significantly lower than HPLC-HbA1c levels (p <0.0001 for both). The degrees of reductions in the IA-HbA1c levels and EA-HbA1c levels compared with HPLC-HbA1c levels were almost same to each other. Similarly to IA-HbA1c levels, EA-HbA1c levels during the health checkups were lower than HPLC-HbA1c levels. It was demonstrated that HbA1c levels decrease similarly if measured by either EA or IA during off-site health checkups.

We herein report a 50-year-old Chinese woman with Hb Phnom Penh (α117Phe-Ile-α118Thr] showing high or reasonable HbA1c values depending on the type of high-performance liquid chromatography (HPLC) system. A high HbA1c value of 7.5% (HPLC assay: G9) and a reasonable HbA1c value of 5.2% (assay unknown) were observed. Therefore, the patient was refereed to our hospital; the oral glucose tolerance test showed normal glucose tolerance. The HbA1c values measured by an enzymatic assay, immunoassay, and affinity assay, as well as most HPLC assays were within the reference range, whereas those measured by the Tosoh HPLC systems were high.

Labile HbA1c migrates in the #C fraction with modified hemoglobin (Hb) (such as carbamylated Hb, acetaldehyde Hb, and acetylated Hb) when HbA1c is measured by Arkray's high-performance liquid chromatography (HPLC). We previously reported the usefulness of #C levels for the screening of variant Hb without diabetes mellitus. Because the #C levels are affected by plasma glucose levels, we investigated the usefulness of plasma glucose adjusted #C (PGa#C) for the screening of variant Hb complicated with diabetes mellitus. In this study, nine types of variant Hb in nine diabetic patients were included. HbA1c and the #C fraction were measured by Arkray's HPLC. Furthermore, we established a calculation formula for PGa#C by using the regression equation of #C and plasma glucose of 2,299 diabetic patients without variant Hb. If the cutoff value of PGa#C for the screening of variant Hb with diabetes mellitus was set at 1.3% or lower and 2.3% or higher, sensitivity and specificity were 89% and 99.8%, respectively. The PGa#C levels in all four slow moving variant Hb with diabetes were less than 1.3%, while the PGa#C levels of fast moving variant Hb with diabetes were abnormal values in four out of five patients [high #C level in one and low #C levels in three patients]. The screening of variant Hb with diabetes with high sensitivity and high specificity was possible by using the same cutoff values for the reference range of PGa#C as the #C values reported in non-diabetic subjects.

Although HbA1c measurement by enzymatic assay (EA-HbA1c) is widely used in health-screening settings in Japan, recent studies have suggested lower EA-HbA1c levels as compared with HbA1c levels measured by high-performance liquid chromatography (HPLC-HbA1c). Hypothesizing that falsely low levels of EA-HbA1c are attributable to hemolysis caused by sample transport and/or storage, we measured EA-HbA1c in blood cells and whole blood after sample transport and compared them with HPLC-HbA1c levels. Blood samples were collected from ten non-diabetic individuals into sodium fluoride-containing blood collection tubes and immediately measured for EA-HbA1c in blood cells. After transport, the blood samples were again subjected to measurement of EA-HbA1c levels in blood cells and whole blood the following day. These EA-HbA1c levels were compared with HPLC-HbA1c levels. EA-HbA1c levels in blood cells measured immediately after sample collection did not significantly differ from HPLC-HbA1c levels. Transported blood samples showed hemolysis and significantly lower EA-HbA1c levels in blood cells, as compared with HPLC-HbA1c levels, whereas no significant difference was observed between EA-HbA1c levels in whole blood and HPLC-HbA1c levels. Transported blood samples showed hemolysis and falsely low EA-HbA1c levels in blood cells. Hemolysis caused by sample transport and/or storage might be responsible for the falsely low EA-HbA1c levels. This should be kept in mind, because falsely low HbA1c levels may lead to a false-negative diagnosis of diabetes.