A 68-year-old man was admitted to our hospital with swelling of his left leg. We administered antibiotics following a diagnosis of cellulitis, and the symptoms improved. However, the cellulitis recurred at the same site and became severe, and he was hospitalized for treatment and an examination. His plasma glucose was 341 mg/dL, but his hemoglobin A1c (HbA1c) was only 4.3 % on high-performance liquid chromatography (HPLC; HLC^®-273G9; Tosho. We therefore suspected the presence of variant hemoglobin and performed isoelectric focusing followed by a genetic analysis. A globin gene analysis revealed a heterozygous mutation, and we identified variant hemoglobin HbE-Saskatoon [β22 Glu (GAA) →Lys (AAA) ]. We considered the HbA1c value measured by HPLC to have been a falsely low value. We then treated him with no reference to the HbA1c, and he has since shown a good course with no recurrence of cellulitis. We encountered a case of HbA1c on HPLC showing a falsely low value due to HbE-Saskatoon. Treatment of the patient's infectious disease resulted in a good outcome following the identification of the cause of the false HbA1c value and appropriate glycemic control.

A 53-year-old woman was admitted to our hospital because of central nervous system lupus. She had been diagnosed with systemic lupus erythematosus and begun treatment with immunosuppressive therapy at 46 years old. Because her glucose tolerance had become exacerbated, DPP4 inhibitor therapy was started at 52 years old. After admission, we began steroid pulse therapy and subcutaneous insulin infusion therapy. Insulin resistance was an initial problem (CPR 8.1 ng/mL) hampering her glucose intolerance, but her glycemic control improved gradually. A month after the admission, reactivation of cytomegalovirus occurred. Thereafter, her pancreatic exocrine enzyme level increased, and her pancreatic endocrine function decreased (Amy 1796 U/L, CPR 2.6 ng/mL) concomitant with swelling of the pancreas on computed tomography. Her glycemic control became disturbed once again, so we started intravenous insulin infusion therapy. She had also been receiving antiviral therapy and steroid therapy. Once the reactivation of cytomegalovirus had disappeared, her pancreatic exocrine enzyme levels normalized, and her endocrine function returned to its previous level (Amy 179 U/L, CPR 7.4 ng/mL). We suspected that the infection had spread to her pancreas, but antiviral and continued steroid therapy may have helped reduce the inflammation in her pancreas.