Örebro University School of Medicine Degree Project, 15 ECTS June 2017
Characteristics, treatment and outcome of
warm autoimmune hemolytic anemia
Version 2
Author: Rickard Hedberg Supervisor: Bertil Uggla, MD, PhD Örebro, Sweden
Rickard Hedberg Page 2 Abstract
Background
Warm autoimmune hemolytic anemia (wAIHA) is an acquired, rare autoimmune hemolytic disease in which antibodies are created against the erythrocytes produced by the body.
Aim
In order to create an overview of characteristics, adherence to guidelines regarding treatment and frequency of remission, this study was conducted.
Methods
This study is a single-center retrospective study of patients who received the diagnosis of wAIHA at the Division of Hematology, Department of Medicine, Örebro University Hospital, Örebro, between 2006 and 2015. The study included data from both patient journals and laboratory test results. Patient characteristics and adherence to guidelines regarding treatment were studied by documenting the treatment given for wAIHA.
Results
Fifty patients (19 females, 31 males) were included; the mean age at time of diagnosis was 67 ± 16 years. Twenty-nine patients were considered having primary wAIHA. The remaining 21 patients were considered having secondary wAIHA, 15 of which had wAIHA secondary to chronic lymphocytic leukemia (CLL). Every patient except for three needed and received treatment with corticosteroids. 87 % of patients who received corticosteroids responded to corticosteroids by reaching either complete remission (CR) or partial remission (PR). 14 (30 %) of those who received first line treatment, received at least one second line treatment, 11 of which received rituximab.
Conclusions
Adherence to guidelines regarding first line treatment and second line treatments were higher than adherence regarding for folic acid supplementation, calcium-D-vitamin supplements, bisphosphonates and specific thromboprophylaxis.
Rickard Hedberg Page 3 Background
Autoimmune hemolytic anemia (AIHA) is a form of acquired hemolytic anemia in which the host’s immune system gives rise to autoantibodies that target the host’s erythrocytes [1-2]. AIHA is serologically divided into warm autoimmune hemolytic anemia (wAIHA), that make up 65 % of cases, and cold autoimmune hemolytic anemia, which accounts for 30 % of cases. There is as well a mixed type that accounts for 5 % of cases [1]. AIHA can be either primary or secondary to an underlying disorder, which may be malignancy (often CLL), infection or another autoimmune disease. AIHA can arise in both adults and children, among children most commonly before the age of 5. The estimated annual incidence of AIHA among Caucasians is 1-3 per 100 000 individuals with a rising incidence with increasing age [1-3].
The hemolysis initiates with erythrocytes being opsonized by the autoantibody. The degree of hemolysis is determined by antibody related factors such as: its ability to bind to tissue macrophages, its ability to fix complement factors, quantity of antibodies, the specificity of the antibodies, its thermal amplitude. Other factors include density and expression of the antigen as well as the age of the patient [2].
Patients with AIHA can exhibit symptoms of anemia (dizziness 50 %, dyspnea 9 % and fatigue 88 %), chest pain and hemolysis (dark urine 3 %, jaundice 21 %) as well as symptoms of underlying disorders [2-3]. Symptoms may have an insidious onset which may be subacute or acute [4]. An increased risk for venous thrombosis in patients with wAIHA has been described [5].
Therapy for remission induction is necessary for most patients with AIHA. In cases with wAIHA, glucocorticoids with or without supplementary high dose immunoglobulins remain the first line therapy. A common second line therapy is splenectomy. Primary wAIHA has a high response rate to corticosteroid therapy. Rituximab (anti-CD20) is another immunosuppressive treatment that has emerged as a preferred second line treatment as it has proved to be very effective in patients with secondary wAIHA that came to be refractory to first line therapy [3, 4, 6, 7]. The response rate of rituximab is approximately 60 % [6].
Besides from remission-inducing treatments, such as corticosteroids, rituximab and immunoglobulins, there are often a need for supplementing with treatment of the anemia itself. In cases of severe anemia blood transfusion is necessary [8].
Rickard Hedberg Page 4 An elevated risk of venous thrombosis is a significant cause of morbidity and mortality among patients with wAIHA, the greater the hemolysis is, the greater the risk of thrombosis is found to be. Two studies have found that in patients with severe AIHA (defined as Hb less than 85 g/l), venous thrombosis occurred in 20-21 % of all cases, in patients without thromboprophylaxis as many as one of three had thrombosis, in patients with thromboprophylaxis as few as one of 21 had thrombosis [5, 9].
The use of folic acid supplementation has reduced the amount of cases of folic acid deficiency and megaloblastic anemia in patients with chronic hemolysis [1].
Long term (≥ 3 months) treatment with glucocorticoids has been found to cause osteoporotic fractures in up to 30-50 % [10]. In order to counteract the osteoporosis-inducing effect of long term treatment with glucocorticoids, calcium and vitamin D supplements are recommended for all patients receiving glucocorticoids [10-11]. Bone mineral density has been found to be increased by bisphosphonates and is therefore along with calcium and vitamin D supplements recommended for postmenopausal women and men above 50 years age who receive long term treatment with glucocorticoids [12-14].
The purpose of this study is to make a single-center retrospective study of diagnostics, treatment and complications regarding patients with a first episode of AIHA at the Section for Hematology at USÖ. In Sweden, there are no national guidelines for wAIHA, instead British guidelines (British Society for Haematology, 2016) will be used.
Key abbreviations: CLL = Chronic Lymphocytic Leukemia. DAT = Direct Antiglobulin Test. IViG = IntraVenous ImmunoGlobulin. LD = Lactate Dehydrogenase. Remission = a state in a chronic disease in which the symptoms partially or completely are absent. wAIHA = warm AutoImmune Hemolytic Anemia. WBC = White blood cell count.
Rickard Hedberg Page 5 Methods
This study was a single-center retrospective cohort study conducted at the Örebro University Hospital (Örebro, Sweden).
The criteria for including patients were:
1. The patient received a definite diagnosis of wAIHA.
2. The diagnosis and treatment occurred at the Division of Hematology, Department of Medicine, Örebro University Hospital between January 1st 2006 and December 31st 2015.
3. The patients had a positive DAT-pattern with an isolated IgG, isolated C3d or IgG + C3d pattern.
In order to study the state in which the patients were at the time diagnosis as well as the severity of their state, several blood levels as well as other data was included. The type of data that was included and the reason for inclusion will be presented below.
Presence of blood malignancies is a common cause to secondary wAIHA, therefore this was important to find in order to determine whether the patient had primary or secondary wAIHA [4].
Presence of other autoimmune diseases is another cause to secondary wAIHA, for the same reason as above this data was included [4].
Hb level is very important in order to detect anemia and determine its severity as well as the efficacy of treatment and whether or not the patient achieves complete or partial remission [2-4].
Bilirubin will typically be increased in cases of hemolysis since unconjugated bilirubin is a rest product from destroyed erythrocytes. If bilirubin levels are above 50 μmol/l it will give rise to prehepatic icterus that can be seen by a yellow color in the sclera [1, 4, 15].
LD (Lactate dehydrogenase) is an enzyme that is set free in when damage is made to the heart, liver, skeletal muscles and erythrocytes. In cases of hemolysis LD may be either normal or elevated [1, 4, 16].
Haptoglobin binds to free hemoglobin and is thereby consumed, therefore haptoglobin is a marker for hemolysis that in case of hemolysis usually become reduced [1, 4, 17].
Rickard Hedberg Page 6 Reticulocyte count is elevated in cases of hemolysis since the increased loss of mature erythrocytes is being compensated by an increased production of new ones, which gives rise to higher count of immature erythrocytes called reticulocytes [1, 4, 18].
WBC (White blood cell count) may be elevated as a result of malignancy, however, there are many other conditions in which WBC may be elevated, such as infection or inflammatory disorders [19].
Platelet level, which if elevated is linked to increased risk of thrombosis [20].
In order to tell whether the hemolysis is autoimmune a direct antiglobulin test (DAT) is made, a positive result indicates an existence of autoreactive immunoglobulin IgA, IgG, IgM or complement (commonly C3d) that is bound to the erythrocytes cell membrane. If the patient has a positive DAT for either IgG or IgG + C3d, the test indicates that the patient has wAIHA [1-4].
A positive DAT result does not necessary mean that the patient has AIHA, since the test is not entirely specific. A positive DAT may occur in many non-hemolytic diseases, such as chronic infection, liver disease, malignancy and SLE [1].
There are however patients who have AIHA and a negative DAT, some may have a false negative DAT, others have a false negative due to treatment with immunoglobulins, etc. Only DAT positive patients were included in the study [1-4].
Need for transfusion is an important way of telling the percentage of patients who had a severe enough anemia during their disease progression to receive transfusion.
Number of units transfused is a further way of telling the severity of the disease.
Complete and partial remission was used to assess treatment efficacy. Complete remission was defined as an Hb ≥ 120 g/l. Partial remission was defined as an Hb ≥ 100 g/l with an increase of ≥ 20 g/l from Hb at diagnosis.
First, second, third and fourth line treatment was included in order to describe the percentage who needed different treatment lines as well as different treatments.
Time until second treatment line was of interest in order to measure the time it took until the need to complement the first line treatment with a second line treatment arose.
Rickard Hedberg Page 7 Patients with corticosteroids were measured as the amount and percentage of the study population who received corticosteroids. It is included in order to tell how many patients who received the recommended first line treatment.
Corticosteroid response was defined as the amount and percentage of the study population who received corticosteroids and had a complete or partial remission.
Patients with folic acid treatment, calcium-D-vitamin treatment, bisphosphonate treatment and thromboprophylaxis treatment are being assessed in order to tell how many patients who received these treatments according to recommendations.
Descriptive statistics comprised mean ± SD for quota variables where 1 SD < mean, median (range) was used for quota variables in cases where 1 SD > mean. Frequency (amount and percentage) was used to describe nominal variables such as DAT pattern. Frequency was also used to describe variables otherwise considered quota variables that given uncertainty of measurement was better described in frequencies, such as haptoglobin, which in most cases was measured as <0.1 g/l. Fisher exact test or Chi2 test was used to retrieve P values when comparing two nominal variables. Fisher exact test was used if at least one cell contained less than 5 observations. Chi2 test was used if all cells contain at least 5 observations each. Unpaired T test was used to retrieve P values when comparing one nominal variable with one quota variable, for example unpaired T test was used to retrieve P value when comparing Hb levels between patients with primary and secondary wAIHA. All observations were unpaired, therefore was unpaired and not paired T test used. Differences were considered of significance at P < 0.05.
Ethics
This study has been conducted as a part of regularly quality assessment at the Division of Hematology, Department of Medicine, Örebro University Hospital, however it was conducted in agreement with the ethical standards of the Helsinki Declaration.
This study comprises a thorough review of patients’ journals and laboratory tests. The large amount of patients combined with the fact that single cases are not depicted in detail, means that is highly unlikely, if not impossible, to identify any single patient from this report. Furthermore, any information given about patients in the study population will be presented in a manner that obstructs recognition of single patients.
Rickard Hedberg Page 8 Aim
The purpose of this study was to create an overview of baseline characteristics and treatment of patients with wAIHA at USÖ and to assess whether the patients are treated in accordance with British guidelines (British Society for Haematology, 2016).
Rickard Hedberg Page 9 Results
Patient characteristics at diagnosis
Fifty patients (19 females and 31 males), who fulfilled the inclusion criteria were included in this study. 29 patients had primary wAIHA,
the remaining 21 patients had secondary wAIHA.
The mean age of the entire study population at the time of wAIHA diagnosis was 67.2 ± 15.8 years (range 25-88 years). Figure 1 displays the amount of patients in different age groups at the time of diagnosis.
The mean Hb level for the entire study population at the time of diagnosis was 75.9 ± 21.9 g/l (reference: 117-153 g/l (F), 134-170 g/l (M)).
Reticulocyte count was at diagnosis measured in 28/29 with primary wAIHA and in 20/21 with secondary wAIHA. Reticulocyte count was 172.7 ± 131.1 109/l (ref: 30-105 109/l) in the study population, 67 % had a reticulocyte count above reference interval.
Haptoglobin was measured in all patients except for one patient with primary wAIHA. Haptoglobin level below reference interval was measured in 84 % of the study population. A bilirubin level above reference interval was measured in 64 % of the study population.
DAT showed an IgG + C3 pattern in 60 % of cases and an isolated IgG pattern in 40 % of cases. WBC above reference interval was found in 22 (44 %) of cases. Mean platelet count was found to be 218 ± 114 109/l (ref: 165-387 (F), 145-348 (M)). Mean LD was 7.3 ± 4.0 μkat/l (ref: < 3.5 μkat/l).
Secondary wAIHA
The most common cause of secondary wAIHA in the study population was chronic lymphocytic leukemia (CLL), 15 of 21 (71 %) cases.
Other causes included myeloproliferative disorder (n = 1), monoclonal gammopathy of undetermined significance (n = 1), myeloma (n = 1), myelodysplastic syndrome (n = 1), splenic lymphoma with villous lymphocytes (n = 1) and autoimmune hepatitis (n = 1).
0 5 10 15 20 25 <50 50-69 ≥70 Amo un t o f p at ie nt s Age group
Rickard Hedberg Page 10
Treatment
Need for transfusion
Overall, 22/50 (44 %) of patients received transfusion at least once because of severe anemia. Mean Hb levels (± 1 SD) at diagnosis among patients who received transfusion was 64.5 ± 18.4 g/l. Of those who received transfusions, 18 (82 %) received less than 10 erythrocyte units in total. The remaining four patients received 11, 13,
20 and 43 units of erythrocytes. Patients received a median of 4.5 units erythrocytes. First line treatment
In all, 47/50 (94 %) patients received the recommended first line treatment of corticosteroids (prednisolone and/or deltisone), while two patients with primary and one patient with secondary wAIHA were observed instead of receiving first line treatment since their conditions were mild and did not have any mentionable impact on their everyday life.
Of the 47 patients who received corticosteroids, the response was possible to determine in 46 patients. One patient had a mild condition with Hb above 100 g/l at diagnosis and Hb rarely measured thereafter, making it not possible to determine whether the patient achieved complete remission. In 40/46 patients (87 %) complete remission or partial remission was achieved. Of these 40 patients, 34 (85 %) achieved complete remission. Twenty of 46 patients (43 %) achieved complete or partial remission within day 21.
Second line treatment
Of the 47 patients who received first line treatment, 14 (30 %) received at least one second line treatment, at a median of 21 days after starting corticosteroid therapy.
Figure 3. Boxplot over Days Until 2nd Line Treatment. Figure 4. Overview of time until second line treatment.
Figure 2. Number of Units Erythrocytes Transfused.
Units Transfused
Rickard Hedberg Page 11
Rickard Hedberg Page 12 Rituximab. 11 of 50 (22 %) received at least one course of rituximab. Of the eleven patients who received rituximab, 5 (45 %) had primary wAIHA and 6 (55 %) had secondary wAIHA. Splenectomy. 2 of 50 (4 %) patients underwent splenectomy. One of which had a splenectomy after treatment with corticosteroids and IViG had a poor effect on the hemolysis caused by its primary wAIHA.
IViG. 3 of 50 (6 %) patients received treatment with IViG after 7, 10 and 11 days after start of corticosteroid therapy. One of the patients who received IViG treatment afterwards was treated with rituximab. The second patient only received IViG as second treatment line and achieved neither partial nor complete remission. The third patient with IViG was reported to have a poor response to both corticosteroids and IViG.
Other treatments. Other treatments included folic acid, calcium-D-vitamin, bisphosphonates and thromboprophylaxis. Folic acid treatment was given to 26 of 50 patients (52 %) at any time after diagnosis. Calcium-D-vitamin treatment was given to 20 of 50 patients (40 %) at any time after diagnosis.
Bisphosphonate treatment was given to four patients at any time after diagnosis. One patient (4 %) of the 23 patients in the age group ≥70 years received bisphosphonates and three patients (16 %) of the 19 patients in the age group 50-69 years received bisphosphonates. None received specific thromboprophylaxis, while 20 patients received anticoagulants due to other conditions.
Rickard Hedberg Page 13
Comparison of primary and secondary cases
Table I compares the main characteristics of patients with primary and secondary wAIHA. At the time of diagnosis, the biological characteristics did not differ, except for in the frequency of increased bilirubin level (P = 0.04) and the frequency of increased WBC (P = 0.03). Bilirubin levels were increased in 22 (76 %) of primary cases and in 10 (48 %) of secondary cases. WBC was increased in 9 (31 %) of primary cases and 13 (62 %) of secondary cases.
Differences in DAT pattern between the two groups was found to be of significance (P = 0.047). In the group with primary wAIHA an isolated IgG pattern was found in 15 of 29 (52 %) and an IgG + C3 pattern in 14 of 29 (48 %), whereas in the group with secondary wAIHA 5 of 21 (24 %) had an isolated IgG pattern and the remaining 16 of 21 (76 %) was found to have an IgG + C3 pattern. One patient with wAIHA secondary to CLL who did not receive corticosteroids achieved a spontaneous complete remission. 13 primary and 7 secondary cases received anticoagulants prior to diagnosis due to pre-existing conditions.
TABLE I. Main Characterstics of Patients with Primary and Secondary wAIHA.
Characteristics Primary wAIHA Secondary wAIHA P value
Amount of patients 29 21
Mean age at diagnosis (years) 66.1 ± 18.6 68.8 ± 11.1 0.56
Sex ratio (Females/Males) 14 / 15 5 / 16 0.08
Biological characteristics at diagnosis Mean Hb (g/L) Reference: 117-153 g/L (F), 134-170 g/L (M) 77.7 ± 17.4 72.2 ± 22 0.34 High WBC n (%) Reference: 3.5-8.8 x109/L 9 (31 %) 13 (62 %) 0.03
Mean Platelet count (109/L)
Reference: 165-387 x109/L (F), 145-348 x109/L
(M)
228.8 ± 127.4 202.3 ± 92.9 0.42
Mean Reticulocyte count (109/L)
Reference: 30-105 x109/L
194 ± 139.5 143 ± 109.9 0.18
High Reticulocytes n (%) 21 (75 %) 11 (55 %) 0.15
Low Haptoglobin n (%) 26 (93 %) 15 (71 %) 0.06
Elevated Bilirubin level n (%) 22 (76 %) 10 (48 %) 0.04 LD Reference: < 3.5 μkat/L 7.6 ± 4.3 7.0 ± 3.8 0.59 DAT pattern IgG n (%) 15 (52 %) 5 (24 %) 0.047 IgG + C3 n (%) 14 (48 %) 16 (76 %) 0.047 C3 n (%) 0 (0 %) 0 (0 %) Treatment
Need for transfusion n (%) 12 (41 %) 10 (48 %) 0.79
Patients with corticosteroids n (%) 27 (93 %) 20 (95 %) 1
Corticosteroid response n (%) 24 (89 %) 16 (80 %) 0.44
Second treatment lines n (%) 7 (26 %) 7 (35 %) 0.50
Complete remission of AIHA n (%) 21 (78 %) 14 (70 %) 0.55
Partial remission of AIHA n (%) 3 (11 %) 3 (15 %) 1
No response n (%) 3 (11 %) 3 (15 %) 1
Rituximab n (%) 5 (17 %) 6 (29 %) 0.34
Splenectomy n (%) 1 (3 %) 1 (5 %) 1
IViG n (%) 2 (7 %) 1 (5 %) 1
Folic acid treatment n (%) 18 (62 %) 8 (38 %) 0.09
Calcium-D-vitamin treatment n (%) 9 (31 %) 11 (52 %) 0.13
Bisphosphonate treatment n (%) 2 (7 %) 2 (10 %) 1
Rickard Hedberg Page 14
Characteristics of patients with associated CLL
Table II illustrates the main characteristics of patients with wAIHA secondary to CLL (n = 15) and the rest of the study population (n = 35). At diagnosis, there was a significant difference (P = 0.002) regarding WBC, showing that 80 % of patients with CLL have an increased WBC, whereas 31 % of other patients had a high WBC. Mean WBC among CLL patients was 68 x109/l, while mean WBC among the rest of the study population was 9 x109/l.
Six of the 14 (43 %) with CLL who had reticulocytes measured had high reticulocytes. In the rest of the study population, 26 of the 34 (76 %) who had reticulocytes measured exhibited high reticulocytes. This was found to be a difference of significance (P = 0.02). A significant difference in DAT pattern (P = 0.01) presented itself between the two groups, displaying a DAT pattern of IgG + C3 in 87 % of cases in the group with CLL, whereas the rest of the study population had an IgG + C3 pattern in 49 % of cases and an isolated IgG pattern in 51 % of cases.
TABLE II. Main Characteristics of patients with wAIHA secondary to CLL compared to the rest.
Characteristics Secondary to CLL Other forms P value
Amount of patients 15 35
Mean age at diagnosis (years) 70.5 ± 8.8 65.9 ± 17.9 0.35
Sex ratio (Females/Males) 3 / 12 16 / 19 0.12
Biological characteristics at diagnosis Mean Hb (g/L) Reference: 117-153 g/L (F), 134-170 g/L (M) 74.9 ± 24.5 76.4 ± 21.1 0.90 High WBC n (%) Reference: 3.5-8.8 x109/L 12 (80 %) 11 (31 %) 0.002
Mean Platelet count (109/L)
Reference: 165-387 x109/L (F), 145-348 x109/L (M)
188.1 ± 96.9 230.3 ± 119.5 0.23
Mean Reticulocyte count (109/L)
Reference: 30-105 x109/L
140.1 ± 119.4 186.2 ± 132.5 0.27
High Reticulocytes n (%) 6 (43 %) 26 (76 %) 0.02 Low Haptoglobin n (%) 11 (73 %) 30 (88 %) 0.23
Elevated Bilirubin level n (%) 7 (47 %) 25 (71 %) 0.09
Mean LD (μkat/L) Reference: < 3.5 μkat/L 7.1 ± 3.8 7.4 ± 4.2 0.81 DAT pattern IgG n (%) 2 (13 %) 18 (51 %) 0.01 IgG + C3 n (%) 13 (87 %) 17 (49 %) 0.01 C3 n (%) 0 0 Treatment
Need for transfusion n (%) 6 (40 %) 16 (46 %) 0.71
Patients with corticosteroids n (%) 14 (93 %) 33 (94 %) 1
Corticosteroid response n (%) 10 (71 %) 30 (94 %) 0.06
Second treatment lines n (%) 6 (43 %) 8 (24 %) 0.20
Complete remission of AIHA n (%) 9 (64 %) 26 (79 %) 0.30
Partial remission of AIHA n (%) 2 (14 %) 4 (12 %) 1
No response n (%) 3 (21 %) 3 (9 %) 0.34
Rituximab n (%) 5 (33 %) 6 (17 %) 0.21
Splenectomy n (%) 1 (7 %) 1 (3 %) 0.51
IViG n (%) 0 (0 %) 3 (9 %)
Folic acid treatment n (%) 6 (40 %) 20 (57 %) 0.27
Calcium-D-vitamin treatment n (%) 7 (47 %) 13 (37 %) 0.53
Bisphosphonate treatment n (%) 1 (7 %) 3 (9 %) 1
Rickard Hedberg Page 15 Discussion & Conclusion
Comparison of baseline characteristics
An earlier, more comprehensive, study of similar nature was conducted by Roumier et al in a French center for hematology. In that study, baseline characteristics were presented as well. Neither our nor Roumier’s study found any significant differences regarding mean age at diagnosis or sex ratio between patients with primary and secondary wAIHA. However, the mean age was roughly 12 and 16 years higher in primary and secondary groups respectively in our study than in Roumier’s study [3].
The age at diagnosis had a peak frequency above the age of 70 as figure 1 displays, which has been shown in earlier data [1-3, 21].
Roumier’s study found a significant difference (P = 0.03) in mean Hb level at onset between patients primary and secondary wAIHA, while we did not detect such a difference (P = 0.34) [3].
The DAT pattern in patients with primary wAIHA was exactly the same in this study as in Roumier’s study, showing an isolated IgG pattern in 52 % of cases and an IgG + C3 pattern in 48 % of cases. The DAT pattern in patients with secondary wAIHA differed, showing a higher prevalence of an IgG + C3 pattern in this study. Also, among patients with secondary wAIHA in Roumier’s study, 2.5 % of patients tested positive for an isolated C3 pattern and 2.5 % for IgA. Because of this difference between studies, the difference in DAT patterns between primary and secondary wAIHA was found to be of significance in this study (P = 0.047), but not in Roumier’s study (P = 0.2 for isolated IgG and P = 0.3 for IgG + C3) [3].
Elevated bilirubin levels was found in 89 % of primary cases and 79 % of secondary cases in Roumier’s study and was therefore not a difference of significance (P = 0.5) [3]. In this study however, a significant difference was found (P = 0.04). The large difference in prevalence of elevated bilirubin between patients with secondary wAIHA in our study and the study by Roumier et al may be a result of less severe hemolysis among our secondary cases.
Signs of this being the case are that mean Hb at diagnosis was 12 g/l higher among secondary cases in this study and that mean reticulocytes at diagnosis was 122 x109/l lower among secondary cases in this study, however lower reticulotcytes could also be the result of bone marrow inhibition in CLL patients [3, 22].
Rickard Hedberg Page 16 Low haptoglobin was found in 93 % of primary cases, but only in 71 % of secondary cases, this may be due to the fact that haptoglobin is an acute phase protein, which may be normal or elevated in cases of hemolysis if a malignancy or chronic inflammation is present [4]. Also, this study included a higher amount of patients with wAIHA secondary to CLL than Roumier’s study (Ours 15 vs. Roumier’s 3), which may influence the results [3].
Comparison of CLL cases and other cases
This study found a significant difference (P = 0.03) in the frequency of high WBC among patients with primary and secondary wAIHA. The same significant difference (P = 0.002) was found between patients with wAIHA secondary to CLL and other forms of wAIHA. This is expected, since CLL patients are characterized by lymphocytosis.
Even though mean WBC in primary cases is low, the presence of high WBC in some cases of primary wAIHA could be explained by the fact that wAIHA may precede the onset of blood malignancy, as confirmed by Roumier et al [3]. Just like previous studies, this study confirms that CLL is the leading cause of secondary wAIHA [1-3, 23, 24].
CLL-induced cases of wAIHA were found to overwhelmingly have a DAT pattern of IgG + C3 (87 % of cases, P = 0.01), compared to the rest of the study population, which exhibited an isolated IgG pattern in 51 % of cases and IgG + C3 pattern in 49 % of cases. It is unknown why this difference presented itself, however a similar difference was found by Roumier et al [3]. There may be a connection between CLL-induced wAIHA and complement activation, however, more studies need to be done in order to determine whether there is such a connection.
Adherence to guidelines
As guidelines recommend, corticosteroids should constitute the cornerstone of first-line treatment [1, 22, 25]. In this study, all patients received corticosteroids, except for two patients with primary wAIHA and one patient with secondary wAIHA who had mild conditions and therefore were observed only. Therefore, corticosteroids were given according to guidelines. Rituximab proved to be the preferred second line treatment in this study, as in the studies by Hill et al (2016) and Valent and Lechner (2008) [1, 4]. The formerly preferred second line treatment of splenectomy was used in only two cases.
According to guidelines, all patients with AIHA should be given folic acid supplementation [1]. Overall, 26 of 50 patients (52 %) were given folic acid supplementation, giving adherence to
Rickard Hedberg Page 17 guidelines regarding folic acid in 52 % of cases. Guidelines regarding Calcium-D-vitamin supplementation say that all patients on corticosteroids should be given Calcium-D-vitamin [1]. Of the 47 patients who received corticosteroids, 20 patients (43 %) were given Calcium-D-vitamin, giving adherence to guidelines in 43 % of cases.
Bisphosphonate treatment should according to guidelines be given to postmenopausal women and men ≥ 50 years of age who receive corticosteroids [1]. Four patients received bisphosphonates, meaning that more patients should receive bisphosphonates. Given the uncertainty of how many of the women who should have received bisphosphonates, it is difficult to state the exact number of patients who should have received this treatment. However, given the fact that 4 of the 42 patients of age 50 years or more received this treatment, it is safe to assume that more patients should have received this treatment.
None of the patients received specific thromboprophylaxis, however, 20 patients were treated with anticoagulant treatments, such as aspirin or warfarin, for other conditions.
Remission
Treatment of wAIHA was more successful in this study than in Roumier’s study, leading to complete or partial remission in 87 % of all cases in our study (72 % in Roumier’s). 74 % of cases led to complete remission and 13 % led to partial remission, but not complete remission. In Roumier’s study, 47 % achieved complete remission and 25 % achieved partial remission [3]. This difference may in part be due to a less severe hemolysis and thereby also less severe anemia in our study population than in Roumier’s study population.
Conclusions regarding treatment
Adherence to guidelines regarding first line treatment and second line treatments were high in the cases that comprise the study population. However, a lower adherence to guidelines was found regarding folic acid supplementation, calcium-D-vitamin supplementation, bisphosphonate supplementation and thromboprophylaxis.
Weaknesses
The weaknesses of this study were that data regarding prevalence of thrombosis, relapse frequency, time to relapse and corticosteroid dependency was not collected.
Rickard Hedberg Page 18 References
1. Hill QA, Stamps R, Massey E, Grainger JD, Provan D, Hill A. British Society for Haematology (2016). The diagnosis and management of primary autoimmune haemolytic anaemia. Br J Haematol. Doi: 10. 1111/bjh. 14478.
2. R. K. Chaudhary and Sudipta Sekhar Das. Autoimmune hemolytic anemia: From lab to bedside. Asian Journal of Transfusion Science 2014;8:5-12.
3. Roumier M, Loustau V, Guillaud C, Languille L, Mahevas, M, Khellaf M, et al. Characteristics and outcome of warm autoimmune hemolytic anemia in adults: new insights based on a single-center experience with 60 patients. American Journal of Hematology, Vol. 89, No. 9, September 2014.
4. Valent P, Lechner K. Diagnose und Therapie von autoimmunhämolytischen Anämien Im Erwachsenen: Eine Klinische Übersicht. Wien Klin Wochenschr (2008) 120/5-6: 136-151.
5. Hendrick AM. Auto-immune haemolytic anaemia – a high risk disorder for thromboembolism? Hematology 2003 Feb; 8(1):53-6.
6. Barros MM, Blajchman MA, Bordin JO. Warm autoimmune hemolytic anemia: recent progress in understanding the immunobiology and the treatment. Transfus Med Rev. 2010 Jul; 24(3):195-210.
7. Chaturvedi S, Chang A, McCrae KR. Long Term Outcomes of Warm and Cold Antibody-Mediated Autoimmune Hemolytic Anemia Treated with Rituximab from a 13 Year Retrospective Cohort. Blood 2014 124:4009.
8. Petz LD, Garraty G. Immune Haemolytic Anemias, 2nd Ed. Philadelphia, PA: Churchill Livingstone; 2004.
9. Lecouffe-Desprets M, Neel A, Graveleau J, Leux C, Perrin F, Visomblain B, Artifoni M, Masseau A, et al. Venous thromboemobolism related to warm autoimmune hemolytic anemia: a case-control study. Autoimmunity reviews, 14, 1023-1028. (2015) 10. Rizzoli R, Adachi JD, Cooper C, Dere W, Devogelaer JP, Dies-Perez A, et al.
Management of glucocorticoid-induced osteoporosis. Calcified Tissue International, 91, 225-243. (2012)
11. Weinstein RS. Clinical practice. Glucocorticoid-induced bone disease. New England Journal of Medicine, 365, 62-70. (2011)
12. Grossman JM, Gordon R, Ranganath VK, Deal C, Caplan L, Chen W, et al. American college of Rheumatology 2010 recommendations for prevention and treatment of
Rickard Hedberg Page 19 glucocorticoid-induced osteoporosis. Arthritis Care and Research, 62, 1515-1516. (2010)
13. Hansen KE, Wilson HA, Zapalowski C, Fink HA, Minisola S, Adler RA, et al. Uncertainties in the prevention and treatment of glucocorticoid-induced osteoporosis. Journal of Bone and Mineral Research, 25, 1989-1996. (2011)
14. Lekamwasam S, Adachi JD, Angusdei D, Bilezikian J, Boonen S, Borgström F, et al. A framework for the development of guidelines for the management of glucocorticoid-induced osteoporosis. Osteoporosis International, 23, 2257-2276. (2012)
15. I.J Beckingham, S.D Ryder. Investigation of liver an biliary disease. BMJ. 2001 Jan 6; 322(7277): 33-36.
16. L Ryding. Laktatdehydrogenas, LD [Internet]. Sweden: Blodkollen; 2015 [cited 2017-05-03]. Available from: https://www.blodkollen.se/sv/analys/laktatdehydrogenas-ld 17. C Wilczynski. Haptoglobin [Internet]. Loyola: Medscape; 2014 [updated 2014-02-12;
cited 2017-05-03]. Available from: http://emedicine.medscape.com/article/2085592-overview#a1
18. J. David Bessman. Reticulocytes. Clinical Methods: The History, Physical, and Laboratory Examination. 3rd edition. Chapter 156. Boston: Butterworths; 1990.
19. H Naushad. Leukocyte Count (WBC) [Internet]. Creighton: Medscape; 2015 [updated
2015-07-15; cited 2017-05-04]. Available
from: http://emedicine.medscape.com/article/2054452-overview#a1
20. Skoda RC. Thrombocytosis Hematology. Am Soc Hematol Educ Program. 2009:159-67.
21. Gehrs BC, Friedberg RC. Autoimmune haemolytic anemia. Am J Hematol 2002;69:258-271.
22. Zent CS, Ding W, Schwager S, Reinalada MS, Hoyer JD, Jelinek DF, et al. The Prognostic Significance of Cytopenia in Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma (CLL). Br J Haematol. 2008 May; 141(5): 615-621.
23. Lechner K, Jäger U. How I treat autoimmune hemolytic anemias in adults. Blood 2010;116:1831-1838.
24. Moreno C, Hodgson K, Ferrer G, Elena M, Filella X, Pereira A, et al. Autoimmune cytopenia in chronic lymphocytic leukemia: Prevalence, clinical associations, and prognostic significance. Blood 2010;116:4771-4776.
Rickard Hedberg Page 20 25. Crowther M, Chan YL, Garbett IK, Lim W, Vickers MA, Crowther MA.
Evidence-based focused review of the treatment of idiopathic warm immune hemolytic anemia in adults. Blood 2011;118:4036-4040.
