Objective:To investigate the accuracy of Positron Emission Tomography/Computed Tomography(PET/CT)in diagnosing bone marrow infiltration in patients with first pathologically confirmed Diffuse Large B cell Lymphoma(DLBCL)and the correlation between the different uptake patterns of bone marrow 18F-FDG and the prognosis of DLBCL patients.Method:Clinical data of 156 patients with DLBCL diagnosed in Shanxi Cancer Hospital were retrospectively analyzed.All patients underwent bone marrow aspiration biopsy(BMB),bone marrow smear,Flow Cyto Metry(FCM)and 18F-Fluorodeoxyglucose(18F-FDG)PET/CT scan before treatment.Using the normal liver right lobe S6 segment 18F-FDG uptake value as the standard,patients were classified as having focal increased bone marrow uptake(focal PET,f PET+),diffuse increased bone marrow uptake(diffuse PET,d PET+)and normal bone marrow uptake(normal PET,n PET)patterns.Differences in clinicopathological factors between groups were analyzed by t-test,Mann-Whitney U test,χ2test,or Fisher’s exact test,depending on the type of variables.Kappa test was used to determine the consistency of PET/CT and bone marrow examination in diagnosing BMI in DLBCL patients.Subjects’Operating Characteristic(ROC)curves were plotted using the Maximun standardized uptake value(SUVmax)ratio of bone marrow/liver,and the Area Under Curve(AUC)to find the optimal threshold,and the difference was considered statistically significant at P<0.05.The prognostic significance of different uptake patterns of bone marrow was derived by the Kaplan-Meier method,and Cox regression models were used to identify relevant risk factors from clinical,pathological,and imaging variables.Results:There were 45 PET+patients,including 17 f PET+patients,28 d PET+patients and 111n PET patients.There were 21 patients with positive Bone Marrow Infiltration+(BMI+)and135 patients with BMI-.In f PET+patients,BMI+accounted for 41.2%and in d PET+patients,BMI+accounted for 42.9%.There were 12 patients with d PET+in patients with BMI+and 16 patients with d PET+in patients with BMI-.The BMI+condition assessed by bone marrow examination was present in 2 cases in n PET patients.The sensitivity,specificity,positive predictive value and negative predictive value of f PET+were 33.3%,92.6%,41.2%and 89.0%respectively.The sensitivity,specificity,positive predictive value and negative predictive value of d PET+were 57.1%,88.1%,42.9%and 93.0%,respectively.ROC working curve of d PET+vs n PET patients with bone marrow/liver SUVmax ratio to determine BMI:AUC=92.3%(P<0.001),with an optimal threshold of bone marrow/liver SUVmax=1.19,sensitivity of 86.7%,and specificity of 91.1%.d PET+vs bone marrow SUVmax plotted ROC curve:AUC=94.8%(P<0.001),with an optimal threshold of bone marrow SUVmax=3.60,sensitivity of 86.7%,and specificity of91.1%.Univariate analysis showed that Ann Arbor stage,B symptoms,National Comprehensive Cancer Network-International Prognostic Index,NCCN-IPI)score,Lactate Dehydrogenase(LDH),BMI+and bone marrow uptake pattern were correlated with Progression-Free Survival(all P<0.05).Ann Arbor staging,NCCN-IPI score,LDH,BMI+and bone marrow uptake pattern were correlated with Overall Survival(OS)(all P<0.05).Multivariate analysis showed that stage III/IV,LDH and f PET+(all P<0.05)was an independent predictor of PFS.There were no independent predictors in the multivariate analysis of OS.Conclusion:1.18F-FDG PET/CT as a means of diagnosing BMI in DLBCL patients;f PET+is a typical manifestation of lymphoma BMI,and d PET+indicates suspicious BMI in DLBCL patients.d PET+patients,bone marrow/liver SUVmax>1.19,bone marrow;3.60 May serve as a diagnostic critical point for predicting the outcome of bone marrow infiltration;bone marrow SUVmax>3.60 may be diagnostic of BMI critical point;2.The bone marrow uptake pattern in PET/CT imaging was associted with the prognosis of DLBCL patients,and f PET+correlated with the PFS of patients.f PET+may provide a basis for patient BMI diagnosis,treatment and prognosis aspect. |