Usefulness of FDG-PET (positron emission tomography) in the discrimination between hypoplastic myelodysplastic syndromes and aplastic anemia

Masaaki Takatoku, MD PhD1, Takahiro Nagashima, MD*1, Toshihiko Sato, MD*2, Tadashi Nagai, MD PhD1, Norio Komatsu, MD PhD1, Keiya Ozawa, MD PhD1

1Division of Hematology, Department of Medicine, Jichi Medical School, Minamikawachi, Tochigi, Japan; 2Utsunomiya Central Clinic, Utsunomiya, Tochigi, Japan

It is sometimes difficult to distinguish hypoplastic myelodysplastic syndromes (MDS) from aplastic anemia (AA) using current diagnostic methods, such as bone marrow pathology and chromosome analysis. Although magnetic resonance imaging (MRI) is useful for diagnosis of MDS with hypercellular marrow, it is not easy to discriminate between hypoplastic MDS and AA using this method, because the high intensity pattern on T1 enhanced image is similar in these disorders. Recently, quantitative imaging with fluorine-18 fluorodeoxyglucose (FDG) PET has been recognized as a useful method for the discrimination between benign and malignant regions in various conditions. Because the decrease in the FDG-uptake at the late phase is much slower in malignant region than in benign region, dual time point imaging provides more accurate information than single time point scanning. In this study, we investigated the usefulness of dual time protocol FDG-PET in the differential diagnosis of hypoplastic MDS and AA. Six patients [2 with AA, 4 with MDS (including one hypoplastic MDS)] and 30 healthy adults agreed to participate in this study. Bone marrow biopsy, FDG- PET, MRI, and computed tomography (CT) were carried out, and a PET functional image was integrated into a CT anatomical image. The spine, femur and sternum lesions were detected by their increased 18F-FDG uptake at 60 and 120 min after injection of 0.12 mCi/kg of 18F-FDG. The maximum and mean lesional standardized uptake values (SUVmax and SUVmean) after 60 and 120 min were determined. The median SUVmax and SUVmean values of normal lumbar regions at 60 min were 1.94 ± 0.16 and 1.77 ± 0.11, respectively. In the MDS cases, those values at 60 min were 2.39 (range 2.12–2.72) and 2.06 (range 1.91–2.23), respectively. At 120 min, the median SUVmax and SUVmean values of normal cases were 1.33 ± 0.21 and 1.20 ± 0.16, respectively, whereas those of MDS cases were 2.42 (range 2.08–2.78) and 2.14 (range 1.50–2.26), respectively. Thus, the SUVmax and SUVmean values in MDS cases remained at high levels at 120 min in contrast to the decreased levels in normal cases. It is noteworthy that the SUVmax and SUVmean values of a hypoplastic MDS case were also high (2.21 and 2.01 at 60 min, 2.16 and 1.97 at 120 min), suggesting that bone marrow in MDS has a hyper metabolic state of glucose like other malignant disorders. We also observed patchy hot areas, which may be a visualization of ineffective hematopoiesis, throughout the spine image of hypoplastic MDS. In contrast, the SUVmax and SUVmean values of AA cases were 1.82 and 1.66 at 60 min and 1.31 and 1.19 at 120 min (case 1), 1.69 and 1.61 at 60 min and 1.30 and 1.13 at 120 min (case 2), indicating that there is no difference in the SUVmax and SUVmean values at the both time points between AA and normal cases. These results raised the possibility that the discrimination between hypoplastic MDS and AA, in which MRI shows a common observation, can be made using FDG-PET.

Author notes

Corresponding author

Sign in via your Institution