Accurate Detection Of Renal Involvement In Pediatric Patients With Acute Lymphoblastic Leukemia By Utilizing 3D-Reconstruction Models Of Enhanced CT and Its Clinical Implications

Rate of leukemic invasion to kidney varied among reports, and because of it, the clinical significance of renal involvement in pediatric acute lymphoblastic leukemia at diagnosis has remained controversial. Conventionally, renal involvement have been diagnosed by ultrasonography according to the definition of unilateral or bilateral nephromegaly based on the length of major or minor axis >2SD, diffuse abnormal findings, or mass legions. However to detect the leukemic involvement much more accurately, we utilized contrast-enhanced CT to evaluate renal mass legions and to measure renal volume by utilizing 3D reconstruction models of CT images. Leukemic involvement in the kidney was defined only when markedly decrease of renal volume before and after induction chemotherapy and/or mass lesions were observed. The aim of this study was to assess whether our 3D-CT-based criteria determined by utilizing the modern medical technology could be superior to the conventional criteria based on ultrasonography and to reveal significant clinical impact of renal involvement.

From 2006 to 2012, 23 children were diagnosed with acute lymphoblastic leukemia in our hospital. Enhanced CT scan was performed to the all patients before induction therapy, and a follow-up CT scan was performed after induction therapy or later stages on the patients who had abnormal CT findings at diagnosis. Mass legions, length of major and minor axis, and volume changes by utilizing enhanced CT and 3D reconstruction models calculated with the software SYNAPSE VINCENT (FUJIFILM MEDICAL), were evaluated.

　15 out of 23 patients (65.2%) presented with renal involvement when conventional criteria was used (major or minor axis >2SD, or diffuse abnormal findings or mass legions). However when applied to our 3D-CT- based criteria (mass legions or marked decrease of renal volume before and after induction therapy), only 6 patients out of 23 patients (26.1%) showed renal involvement (volume change alone 1, mass lesions alone 3, mass lesion with volume change 2). Quite big difference of renal involvement between criteria was due to the patients who fulfilled conventional criteria did not show the volume decrease of kidney before and after induction chemotherapy, even though these patients reached complete remission in bone marrow and markedly decrease of spleen volume measured by our 3D-CT methods. It suggested that these cases did not have true renal involvement and that the nephromegaly defined by only renal length included many false positive cases. Most importantly, renal mass legions could be detected only by enhanced CT, not by ultrasonography in some cases. Frequency of the presence of central nervous system and extramedullary involvement except lymphadenopathy were significantly higher in the patients with renal mass legions (p=0.005 and P=0.043, respectively). Moreover, the patients with renal mass legions had tendency to relapse.

Our 3D-CT-based criteria for detection of leukemia invasion to kidney is much more reliable than the conventional criteria based on nephromegaly measured by major or minor axis. Renal mass legions strongly correlated with central nervous system and extramedullary involvement except lymphadenopathy.

No relevant conflicts of interest to declare.