Evaluation of Clinical and Morphological Practices in the Diagnosis of Transplant-Associated Microangiopathy: A Study on Behalf of the Complications and Quality of Life Working Party of the EBMT

Introduction: Transplant-associated microangiopathy (TAM) is a severe and frequently life-threatening complication of allogeneic stem cell transplantation (SCT). Thus correct laboratory confirmation and differential diagnosis are important, especially in the context of emerging new therapies, like complement inhibitors. However, among the parameters taken into account when establishing diagnosis, schistocyte count seems to be a subjective and variable one (Zini G. et al., 2012). Therefore, a practice of TAM clinical and morphological diagnosis was investigated in 17 EBMT transplant centers.

Methods: The study included two questionnaires, one for transplant physician and one for morphologist, and also a set of 20 blood slide photographs of 20 cases for quantification of proportion of schistocytes. 10 of these cases represented patients with TAM, with diagnosis established based on International Working Group (IWG) criteria at the Pavlov First Saint Petersburg State Medical University and Helsinki university. The remaining 10 were control cases, including diagnosis of myelodysplastic syndrome, autoimmune hemolysis, thalassemia, myelofibrosis and poikilocytosis due to B12 deficiency. 17 EBMT centers participated in the study.

Results: Based on physician's questionnaires, 41% of centers use IWG criteria, 41% - "overall TAM" criteria (Cho et al. Transplantation 2010) and 18% physician's decision for the diagnosis of TAM. 88% of centers stated that the differential diagnosis of TAM is complicated and 65% of centers consider schistocyte count a criterion most commonly not fulfilled in suspected cases. For 65% of physicians the percentage of schistocytes required for the diagnosis of TAM is above 4%, for 18% above 1%, for 12% above 2%, and one center considered the level of schistocytes not important for the diagnosis. Based on morphologists's form, the morphological types of abnormal red blood cells that were counted as schistocytes were triangular cells (in 93% of centers), helmet cells (93% of centers), keratocytes (67%), microspherocytes (53%), degmacytes (47%), stomatocytes (13%), acanthocytes (7%) and codocytes (7%). 27% of centers reported that it is impossible to distinguish TAM morphology form autoimmune hemolysis, 13% - from myelodysplastic syndrome, 20% - from poikilocytosis, 7% - from hereditary anemias. 57% of centers considered automated hematology analyzers not applicable for quantification of schistocytes in patients after allogeneic SCT, 36% stated that manual confirmation of results is required, and only one center reported that automated analyzers are accurate for TAM diagnosis. The analysis of blood slides by morphologists revealed significant variability in schistocyte quantification for all the cases. The median number of schistocytes reported was 3% for TAM cases (range 0-19.6%, coefficient of variation 0.83), and 0.6% for control cases (range 0-8.3%, coefficient of variation 1.06). 36% of centers consistently (in at least 18/20 cases) reported number of schistocytes higher than median (5.1±4.1% for all cases), 36% consistently lower than median (0.9±1.3% for all cases), and 28% of centers reported results around median (2.3±2.5% for all cases). The difference might be in part due to inclusion of microspherocytes in the overall count. Centers that classified microspherocytes as schistocytes reported significantly higher values (mean for all cases 6,1±2,6% vs 2,8±2,4%, p=0.035). No significant differences were observed for the other morphological forms. In 31% of the evaluations on TAM cases the reported proportion of schistocytes was below 2%, while in 21% of control cases the proportion was reported to be higher than 2%.

Conclusion: The survey identified the need for the standardization of TAM morphological diagnosis and clear criteria to distinguish TAM from other post-transplant complications. Several different methodologies for morphological diagnosis were observed. The existing variability in practices of schistocyte assessment among centers might be an obstacle to reliable multicenter interventional studies in TAM.