Primary Versus Secondary Autoimmune Hemolytic Anemia: A Retrospective Analysis of Diagnostic Patterns and Treatment Outcomes

Introduction: Autoimmune hemolytic anemia (AIHA) is an acquired autoimmune disorder caused by development of antibodies against antigens on erythrocytes. AIHA is divided into primary (idiopathic) and secondary type, the latter of which is often associated with hematologic malignancy (HM). Previous studies estimate that between 5-10% of people with HM develop AIHA at some point, but the temporal relationship between HM and AIHA has, to our knowledge, not been explicitly studied. Furthermore, only one small study has compared the treatment protocols and complication rates for primary AIHA to those for secondary AIHA. This study suggested that treating AIHA secondary to HM with corticosteroids may not be the most effective first-line strategy due to complications and frequent treatment failure. Instead, treating the underlying malignancy may be more beneficial. This study aims to shed light on the temporal relationship between diagnosis of AIHA and HM, and to compare and contrast the treatments and treatment outcomes for primary and secondary AIHA.

Methods: Patients with AIHA diagnosed between January 2000 and December 2010 were retrospectively identified by a search of ICD codes. Patients were excluded if they did not have a positive direct antibody test (DAT), anemia, or laboratory findings of hemolysis. We recorded the date of AIHA diagnosis, lab findings at diagnosis, the treatments and duration of each therapy, and number and frequency of blood transfusions. The medical record was then analyzed for any diagnosis of HM during the patient's life. If HM was diagnosed, the date of diagnosis and type of HM were documented. The means or medians of continuous variables were compared using Student's t tests, Wilcoxon's rank-sum tests, ANOVA, or Kruskal-Wallis tests, depending on the distribution of the data. Categorical variables were analyzed using chi-square or Fisher's exact tests. Univariate and multivariate Cox proportional-hazards models were used to assess the effect of different first AIHA treatment types on PFS and OS, adjusting for baseline characteristics. Statistical significance was considered to be P <0.05.

Results: A total of 67 patients were included in this study. Of these patients, 38 (56.7%) did not have HM, while 29 (43.3%) did. Within the HM group, 14 (48.3%) were diagnosed before developing AIHA, 6 (20.7%) were diagnosed after, and 9 (31%) were diagnosed within the same encounter. Demographically, the patients with HM were significantly older when diagnosed with AIHA (70.8 years of age for HM, compared to 60.7 years of age for AIHA only) and were more likely to be male. Between the two groups, there were no significant differences in hemoglobin, bilirubin, haptoglobin, lactate dehydrogenase, or reticulocyte count at AIHA diagnosis. The volume of blood transfusions was significantly higher in the HM group, with a mean of 26.1 units of packed red blood cells compared to 7.4 units in the AIHA only group. Regarding first treatment for AIHA, 38 (56.7%) of patients were treated with steroids alone, 9 (13.4%) were treated with intravenous immunoglobulin (IV Ig), and 16 (23.8%) were treated with observation or blood transfusions only. Only 2 patients were treated with rituximab before steroid or IV Ig therapy as a first treatment. There was no significant difference in first treatment for AIHA between the HM and AIHA only groups.

Discussion: Nearly half of our study population had AIHA associated with HM. Most of the patients with HM were diagnosed with AIHA months to years after diagnosis of their malignancy. However, of the 44 patients presenting with AIHA without a known diagnosis of HM, 15 (34.1%) were found to have HM during that encounter, or in the years following. This reinforces the need to search for underlying malignancy in a patient with newly diagnosed AIHA. This study was unable to identify specific markers that were associated with an increased risk of HM. Regarding treatment, there was a great deal of variety in first treatment for AIHA, with a similar variety in response. Treatment that included IV Ig was associated with a statistically significant increase in relapse rate than treatment with steroids alone. This finding is likely confounded to some degree, the patients who received IV Ig as a first treatment had slightly, although non-significantly, lower presenting reticulocyte counts and required more transfusions, indicating they may have had more severe disease.