Probability scoring system of intravascular large B-cell lymphoma for the application of random skin biopsy: A retrospective cohort study

BACKGROUND: Intravascular large B-cell lymphoma (IVLBCL) is rare and fatal when diagnosed late in the disease course. Random skin biopsy (RSB) is useful for early diagnosis, but criteria for its application are not well established. OBJECTIVE: To develop an IVLBCL-probability scoring system for stratifying patients and investigate its feasibility and capability for RSB application. METHODS: A retrospective cohort of 77 consecutive patients with suspected IVLBCL who underwent RSB was included in this study. All patients were classified into 3 IVLBCL-probability groups according to the IVLBCL-probability scoring system comprising the following 4 components: general symptoms, organ-specific symptoms, serum soluble-interleukin-2 receptor levels, and serum lactate-dehydrogenase levels. RESULTS: The high (score 7-10), intermediate (score 4-6) and low (score 1-3) IVLBCL-probability groups contained 32, 30, and 15 patients, respectively. All 5 patients with IVLBCL were stratified into the high IVLBCL probability group. Accuracies in the diagnosis of IVLBCL were 100%, 100%, and 93.8% for the low, intermediate, and high IVLBCL-probability groups. The positive detection rate in the high IVLBCL-probability group increased to 9.4% from 3.9% across all groups. CONCLUSIONS: The newly-developed IVLBCL-probability scoring system has good capability for stratification of patients and could allow limiting application of RSB for diagnosis only to high-probability groups.