Clinical features of vitamin B12 deficiency mimicking thrombotic microangiopathy

Abstract

Vitamin B12 deficiency is one of the most common nutritional deficiencies worldwide.1 Clinically, vitamin B12 deficiency often presents as mild anaemia and can be associated with neurologic deficits. However, in one study of 201 patients with vitamin B12 deficiency, 10% of patients experienced life-threatening haematologic manifestations such as pancytopenia (5%), severe anaemia (2·5%), pseudo-thrombotic microangiopathy (pseudo-TMA) (2·5%), and haemolysis (1·5%).2 The true frequency of pseudo-TMA due to B12 deficiency is not known, as these presentations have been reported in case reports and few case series (Table SI).3-10 There has been confusion in diagnosis and management of these patients because the laboratory parameters often mimic thrombotic thrombocytopenic purpura (TTP). In a 2017 review of all 36 published cases of pseudo-TMA, 38·9% of patients were mistakenly thought to have TTP and treated with plasma product therapy. The median duration of patient symptoms was two weeks before the diagnosis of vitamin B12 deficiency was made.11 It is not well understood if there are typical features of B12 deficiency presenting as pseudo-TMA that would distinguish it on clinical grounds from TTP. We aimed to perform the first systematic investigation into pseudo-TMA from vitamin B12 deficiency and identify unique differences in the presentation of pseudo-TMA and TTP. In this retrospective cohort, we used electronic health record data from Beth Israel Deaconess Medical Center from 1997 to 2019. The protocol was deemed exempt by the institutional review board. During this time period, 2 699 patients had vitamin B12 levels <180 pg/ml. We reviewed the charts of all patients with vitamin B12 <180 pg/ml and lactate dehydrogenase (LDH) >500 IU/l which yielded a total of 27 patients. Of these 27 patients, we identified 16 patients who met criteria for pseudo-TMA due to vitamin B12 deficiency as evidenced by laboratory evidence of haemolysis, schistocytes on peripheral smear, and/or thrombocytopenia. This represented 0·6% of the total cohort of vitamin B12 deficiency. The remaining 11 patients had other concurrent illnesses such as cancer that might explain some of the laboratory abnormalities and were excluded from analysis. The average age of the patients was 54 years (range 30–79 years). Ten out of 16 patients (62·5%) were female. To serve as comparators, we randomly selected 17 patients from the same time period with LDH >500 IU/l and ADAMTS13 <10% consistent with a diagnosis of TTP. Review of the cases was done by two independent reviewers to ensure accuracy. Table I shows the laboratory parameters of our cohort of patients with pseudo-TMA due to vitamin B12 deficiency. We observed that most patients had high mean cell volume (MCV; average 109 fl), markedly elevated LDH (average 3539 IU/l), and lower reticulocyte index (average 0·4%) than expected for the degree of haemolysis. Schistocytes were seen on review of peripheral blood smear in 75% of cases, though often reported as occasional to 1+. Tear drop cells were another common finding on review of the peripheral blood smear in 63% of cases. The high prevalence of tear drop cells in our cohort has not been identified in prior case reports on this phenomenon but can be seen in megaloblastic anaemias.12 Hypersegmented neutrophils were reported in 25% of cases. Intrinsic factor antibodies were detected in 11 of the 16 patients (69%). These patients also had normal kidney function and mildly elevated international normalized ratio (INR) (average 1·3) at presentation. Patient INR When comparing the laboratory parameters of the pseudo-TMA cohort to the TTP cohort as described in Table II, the pseudo-TMA cohort had more significant anaemia (average Hb 61 g/l compared to 94 g/l, P value 0·00016) and less thrombocytopenia (average platelet count 160 K/µl vs. 27 K/µl, P value <0·00001). In addition, the absolute neutrophil count (ANC) was lower in the pseudo-TMA cohort compared to the TTP cohort (average ANC 2 434 cells/µl vs. 7 733 cells/µl, P value 0·00008). Macrocytic anaemia was observed in the pseudo-TMA cohort compared to normocytic anaemia in the TTP cohort (average MCV 109 fl vs. 96 fl, P value <0·00001). There was a statistically significant difference in reticulocyte count between the two groups, with higher reticulocyte index noted in the TTP cohort than the pseudo-TMA cohort (2·4% vs. 0·4%, P value <0·00001), in addition to greater elevation in total bilirubin in the TTP cohort (5·2 mg/dl vs. 1.6 mg/dl, P value 0·008). Despite the higher reticulocyte count and bilirubin levels noted in the TTP cohort, the pseudo-TMA cohort had significantly higher LDH than the TTP cohort (average LDH 3 539 IU/l vs. 1 183 IU/l, P value 0·0002). There were no statistically significant differences in the creatinine and INR of these patients. We also tested the performance of the PLASMIC score, a clinical prediction tool for determining the likelihood of TTP in patients presenting with TMA, to see whether it could distinguish pseudo-TMA from TTP.13 Higher PLASMIC scores were seen in the TTP cohort compared to the pseudo-TMA cohort (6·1 points out of a possible 7 vs. 4·9, P value 0·02), which was driven by the clinically significant differences in MCV and platelet count in these distinct populations. Thus, features of haemolysis with profound elevations in LDH combined with reticulocytopenia and lower than expected bilirubin are a hallmark of pseudo-TMA secondary to vitamin B12 deficiency. Higher LDH and lower than expected bilirubin are seen in pseudo-TMA due to higher concentrations of LDH in immature erythroid precursors and a decreased amount of haemoglobin released during their destruction.14 The relative leukopenia observed in the pseudo-TMA cohort is explained by aberrant DNA synthesis.15 In conclusion, we performed the first systematic investigation of pseudo-TMA due to B12 deficiency, finding it in 0·6% cases of vitamin B12 <180 pg/ml. Our study largely confirmed findings from prior case reports of this phenomenon, while also identifying tear drop cells as a frequent feature. Clinicians should have a high index of suspicion for pseudo-TMA in the presence of reticulocytopenia, markedly elevated LDH, lower than expected bilirubin, and absence of significant thrombocytopenia. Careful review of the peripheral smear will also be helpful as tear drop cells were more common in pseudo-TMA. Distinguishing pseudo-TMA from TTP is crucial to allow for prompt diagnosis and treatment of vitamin B12 deficiency and to avoid unnecessary diagnostic and therapeutic interventions. AGK and JAF designed the research study, analyzed the data, and wrote the manuscript. No sources of funding to disclose. The authors have no conflicts of interest to report. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts