Pediatric Hematology and Oncology, 24:309–315, 2007

Copyright
C Informa Healthcare

ISSN: 0888-0018 print / 1521-0669 online

DOI: 10.1080/08880010701360783

Letter to the Editor

AUTOIMMUNE HEMOLYTIC ANEMIA IN CHILDREN

Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

Rahul Naithani, MD, Neerja Agrawal, MD, M. Mahapatra, MD, Rajat Kumar,
MD, FRCP, H. P. Pati, MD, and V. P. Choudhry, MD, MNAMS 2 Department of
Hematology, All India Institute of Medical Sciences, New Delhi, India

2 The clinical and hematological profile and treatment outcome of children with warm autoim-
mune hemolytic anemia (AIHA) were assessed using retrospective case record analysis. There were 26
(17 idiopathic; 9 secondary) patients with a median age of 11 years. Pallor (100%), fever (39%),
and jaundice (59%) were the main presenting complaints. Jaundice was much more common
in idiopathic (70%) compared to secondary (44%). Direct antiglobulin test was negative in 3
patients. Oral prednisolone produced remission in 81% patients. Four patients relapsed after a
For personal use only.

median period of 7 months (2 months to 2 year) after response. All responded to a second course
of steroids in median 14 days. One child required cyclosporin A in addition. No correlation was
found between response and parameters such as age, sex, jaundice, low pretreatment hemoglobin,
reticulocyte count, total leukocyte count, platelet count, subtype of AIHA, and hepatosplenomegaly.
Relapse correlated with increased duration between the onset of symptoms and treatment. This
study indicates that oral prednisolone is an effective therapy for autoimmune hemolytic anemia.
In refractory cases cyclosporine A may be useful.

Keywords children, immune hemolysis, coticosteroids, cyclosporin

Autoimmune hemolytic anemia (AIHA) has a prevalence of 1 per

100,000 and may be life-threatening. In children it often presents as an
acute, self-limited illness, with good response to short-term steroid therapy
in nearly 80% of patients [1–3]. In children younger than 2 years of age or
in teenagers, the clinical course of the disease may show either resistance
to steroids or dependence on high-dose steroids [4]. The mortality rate in
these children with idiopathic AIHA has been reported to be on the order
of 10% [3].
We are presenting a series of 26 cases to analyze the etiopathogenesis,
clinical, and hematological profile and efficacy of the treatment.

Received 16 August 2006; accepted 2 March 2007.

Address correspondence to Dr Rahul Naithani, Department of Hematology, All India Institute of
Medical Sciences, New Delhi 110029, India E-mail: dr rahul6@hotmail.com

309
 310 R. Naithani et al.

MATERIAL AND METHODS

All children diagnosed between January 2001 and December 2005 were
included in the study. We, at our institute, examine a peripheral smear of
anemic children. Hemolysis is established with peripheral blood spherocy-
tosis and fragmented cells with elevated reticulocyte count, unconjugated
hyperbilirubinemia, and a positive direct antiglobulin test (DAT, Coombs’
test) with or without indirect antiglobulin test (IAT). Initially we were using
the conventional tube methods. Now we use a gel card using Sephadex
columns for antiglobulin test as it has a higher sensitivity. Coombs’ test
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

was done using polyspecific sera. Patients who had a negative Coombs’ test
were not tested with a monospecific serum. Flow cytometric analysis of CD55
and CD59 is done to rule out paroxysmal nocturnal hemoglobinuria, before
labeling a Coombs-negative AIHA.
A detailed drug history was taken to exclude drug-induced AIHA.
Depending on the history, area of origin, clinical presentation, and pe-
ripheral blood picture, exclusion of other causes was done. All Coombs’-
positive patients were investigated with high-performance liquid chro-
matography, G6PD screening, tests for CD55 and CD59 (before labeling
Coombs-negative AIHA). All infants who present with AIHA are screened
For personal use only.

for TORCH infections. Autoimmune markers for antinuclear antibody,

rheumatoid factor, and double-stranded DNA were done in patients with
joint or musculoskeletal complaints. Anti-LKM1 and smooth muscle anti-
body were done in patients with significant hyperbilirubinemia. CT scan,
ultrasonography, or bone marrow aspirate was done in those patients where
leucopenia and/or thrombocytopenia was present to rule out underlying
malignancies. Children who had other features consistent with AIHA but
negative Coombs’ test were given a trial of corticosteroids and diagnosed as
Coombs-negative AIHA when they showed response.
Response criteria were transfusion independence along with Hb >
2 g/dL over baseline and absolute value more than 8 g/dL. All patients
were given a course of steroid in a daily dose of 2 mg/kg till hemoglobin
reached to normal level or 6 weeks, whichever was earlier. Prednisolone
was very slowly tapered in patients who responded. Dose was reduced by
10% of initial dose and patients were kept at this dose for 3–4 weeks
before it was tapered further. The dose was reverted to initial dosage if
hemoglobin fell on reducing the dose of prednisolone. Children were also
supplemented with folic acid supplementation. In children not responding
to prednisolone, cyclosporin A is added at a dose of 5 mg/kg/day in
children less than 5 years awaiting splenectomy. Blood transfusions are
avoided as far as possible. No definite hemoglobin cutoff value is used
for transfusions, which are given only in children with impending CCF or
depending on the rate of fall of hemoglobin. IVIg is not used because
of poor and transient response rates at our hospital earlier. Splenectomy
 Autoimmune Hemolytic Anemia 311

with vaccination against pneumococcus, H. influenzae, and meningococcus

is considered if children fail to respond to 6 weeks of steroids.
Data analysis was done using the computer software Epi Info version 3.2
(Chicago, IL, USA).

RESULTS
Twenty-six (11 M: 15 F) patients between 2 months and 17 years (me-
dian: 11 years) were studied. Table 1 shows the presenting manifestations
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

of idiopathic and secondary AIHA. Prevalence of secondary AIHA in our

patients was found more in females (7/9) than in idiopathic AIHA (9/17).
Fever was seen in 39% children. There was no bleeding manifestation or
lymphadenopathy. Unconjugated hyperbilirubinemia was seen in 18 (69%)
patients. Prevalence of jaundice was much more in idiopathic cases (70%) as
compared to secondary (44%). Seven (27%) patients had splenomegaly of
>5 cm, with 6 in the idiopathic and 1 in the secondary group. Seven (27%)
patients had anemia of >6 months duration. One patient had associated Hb
H disease.
Thrombocytopenia was noted in 6 (23%) patients and all had platelet
For personal use only.

counts less than 100,000/mm3 and pancytopenia was seen in 1 (4%) child.
Bone marrow examination in such cases showed erythroid hyperplasia with
megalobastic changes but no visible change in myeloid cells or megakaryo-
cytes. Peripheral smear examination showed spherocytosis in 20 (77%)
patients. DAT and ICT were negative in 3 (11.5%) and 8 (31%) children,
respectively. Antinuclear antibody was positive in 3, rheumatoid factor in
2, and ds DNA in 2 (3.8%) patients. Significant correlation was found
between degree of anemia with jaundice (r = .93, p = .0001) and also with
reticulocyte count (r = .40, p = .002). No drug was incriminated as the
cause of hemolysis in our series. Table 2 shows the conditions associated
with AIHA.

TABLE 1 Patients Characteristics

Idiopathic AIHA (n = 17) Secondary AIHA (n = 9)

Median (range) Median (range)

Age (years) 7.5 (1.2–17) 11.5 (2.5–18)

Pallor duration (months) 3.0 (0.5–18) 6 (0.6–48)
Hemoglobin (g/dL) 5.0 (2.4–8.9) 4.5 (2.1–6.4)
Duration of jaundice (months) 4.0 (0–12) 1 (0–48)
Liver (cm below costal margin) 2 (0–5) 2 (0.5–9)
Spleen (cm below costal margin) 4 (0–7) 0.5 (0–8)
Reticulocyte count (%) 28 (2–70) 15 (8–50)
Symptom to T/t gap (months) 4 (0.3–19.5) 2 (0.1–15)
Time to response (days) 9 (6–28) 18 (4–63)
 312 R. Naithani et al.

TABLE 2 Distribution of Secondary Conditions Associated with AIHA

Number of patients
Conditions (n = 9)

Autoimmune conditions 5 (56%)

Evan syndrome 2 (22%)
ESLE 2 (22%)
EJRA 1 (12%)
Infections 2 (22%)
ECMV 1 (11%)
ERubella 1 (11%)
Others 2 (22%)
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

EEndocrine disease 1(11%)

ENephritis 1 (11%)

One secondary AIHA (rubella) was followed up for resolution of

anemia without treatment. Steroids were prescribed in all the other cases.
Three patients were lost to follow-up. Response was achieved in 21 (81%)
patients. Median response time in our patients was 14 days (6–29 days).
Steroids were continued for a period varying from 6 to 48 months.
One child with idiopathic AIHA required splenectomy. No correlation
For personal use only.

was found between response and parameters like age, sex, jaundice, low
pretreatment hemoglobin, reticulocyte count, TLC, platelet count, and
hepatosplenomegaly. Four patients of idiopathic AIHA relapsed after a
median period of 7 months (2 months to 2 years) after response. There
was no relapse in secondary cases. Three responded to a second course
of steroids in response in 14 days to 2 months (median 14 days). Another
child required cyclosporin A (5 mg/kg/day) in addition to prednisolone
and response was seen in the 3rd week. All were DAT positive and 3 were
ICT positive. Relapse rate correlated with increasing duration between the
onset of symptoms and treatment (r = .996; p = .0001).

DISCUSSION
The clinical pattern of AIHA in children is predominantly an acute,
transient type lasting for 3–6 months in 70–80% of cases. Chronic hemolysis
develops in 30% of patients with an underlying secondary cause [5].
Autoantibodies in AIHA (predominantly IgG) are antigen specific against
substances on RBC, i.e., agglutinogen like A-B-O, H-N, and Rh − Hr system
[6]. Rh erythrocyte antigen is involved in more than 70% of cases. Coombs’
test has false-negative and false-positive rates in about 2–4% and 8% of all
cases, respectively [7].
Nine (35%) children had secondary AIHA in this series. Median age
at presentation was 11 years (range 0.2–17), which is in contrast to the
data published from other centers with a median age of 2 years [2] and 8
 Autoimmune Hemolytic Anemia 313

years [10]. Buchanan et al. [8] found a 32% incidence of secondary AIHA.
Median hemoglobin in our series is 6.0 g/dL (2.4–14.3). The hemoglobin
level in our patients did not correlate with the strength of DAT positivity.
This is because the severity of hemolysis may depend on avidity and subtype
of antibody, quantity of the complement bound, characteristics of RBC
antigen, number of IgG Fc and complement receptor on macrophages,
and functional status of mononuclear phagocytic system [11,12]. Hep-
atosplenomegaly did not correlate with degree of anemia or jaundice.
One child had dark urine at presentation. Dark urine was reported in
4/8 children in one study [13]. One child had CMV infection, as reported
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

earlier also [l3]. One girl had SLE at diagnosis and another developed SLE
on follow-up after 3 years. In 8 (31%) patients reticulocyte counts were low
due to hemolytic process affecting the reticulocytes circulating in peripheral
blood also [8]. DAT was negative in 3 patients (2 M: 1 F), but peripheral
blood smear showed a hemolytic picture. Reticulocyte count was raised in
2 and ICT was positive in 1 of these children. All had hepatosplenomegaly,
and jaundice was present 2 of them. All these patients responded well to the
steroid therapy without any relapse. Negative DAT in patients with clinical
picture of AIHA could be due to a low level of antibodies on the red blood
cells or lower sensitivity of DAT [14].
For personal use only.

Eighty-one percent of our patients responded to oral corticosteroid

therapy. The mechanism of action of steroids is probably downregulation of
Fc receptors on phagocytes and reduced IL-2 production [14]. The most im-
portant early effect of steroids is to suppress sequestration of opsonized red
cells by splenic macrophages. Another proposed early effect could be reduc-
tion in the binding affinity of autoantibodies for the patient’s red cells [13].
Four children relapsed and all belonged to idiopathic group. Relapse
correlated with increased duration between the onset of symptoms and
treatment (p = .0001). One child required the addition of cyclosporin A
after a retrial of 2 weeks of steroid treatment failed to elicit a response.
A 6-month-old boy with severe AIHA who initially responded to high-
dose methylprednisolone and intravenous immunoglobulin G therapies but
eventually became refractory was then treated with low-dose cyclosporine
and prednisone successfully [15]. In one series of 164 adult cases the
response rate of adrenocortical hormones and cyclosporin A (CsA) was
90.9% [16]. Another series found that patients treated with prednisolone
and cyclosporin relapsed less than those treated with prednisolone alone
[17]. Administration of cyclosporine in 2 of the 4 patients did not result in
any amelioration of the symptoms in another series [10].
A series of 51 children (18 acute anemia: 33 chronic anemia) showed
that corticosteroids were effective in all acute cases but the results were
variable in the chronic cases. The acute form was more frequent in young
children, while chronic AIHA occurred mainly among children at puberty.
 314 R. Naithani et al.

In chronic form, it was sometimes necessary to add immunosuppressive

drugs and in 2 cases to perform a splenectomy [18].
In the present series transfusions were given in 6 (23%) patients who
were symptomatically severely anemic. Transfusion of red cells in AIHA
poses 2 difficulties. Owing to the presence of autoantibodies, it is nearly
impossible to find truly serocompatible donor blood. In such circumstances
it is best to use donor blood of the same ABO and Rh group as the
patient, which is least incompatible, along with corticosteroids [19]. The
second problem is the rapid in vivo destruction of transfused red cells.
Once selected, 20–50 mL of packed RBCs (not whole blood) should be
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14

transfused very slowly while the child is closely monitored for signs of
hemolytic reactions [13].
The response rates were similar between secondary and idiopathic
AIHA, as evident from other studies [20].
Initial intensive inpatient monitoring may be required, followed by
outpatient therapy. There is a high rate of remission with corticosteroids.
In refractory cases cyclosporine can be tried before other options such as
splenectomy, cyclophosphamide [2], and rituximab [9, 21] are pursued.
For personal use only.

REFERENCES
[1] Schwartz RS, Berkman EM, Silberstein LE. Autoimmune hemolytic anemias. In: Hoffman RH, Benz
EJ Jr, Shattil SJ, et al., eds. Hematology: Basic Principles and Practice, ed 3. Philadelphia, PA: Churchill
Livingstone; 2000:611–630.
[2] Moyo VM, Smith D, Brodsky I, Crilley P, Jones RJ, Brodsky RA. High-dose cyclophosphamide for
refractory autoimmune hemolytic anemia. Blood. 2002;100:704–706.
[3] Ware RE, Rose WF. Autoimmune hemolytic anemia. In: Nathan DG, Orkin SH, eds. Nathan and
Oski’s Hematology of Infancy and Childhood. Philadelphia, PA: Saunders; 1998:499–522.
[4] Heisel MA, Ortega JA. Factors influencing prognosis in childhood autoimmune hemolytic anemia.
Am J Pediatr Hematol Oncol. 1983;5:147–152.
[5] Beutler E, Luzzatto L. Hemolytic anemia. Semin Hematol. 1999; 36(4 Suppl 7):38–47.
[6] Dacie J. The immune hemolytic anemias: a century of exciting progress in understanding:
historical review. Br J Haematol. 2001;114:770–785.
[7] Agarwal B. Autoimmune hemolytic anemia. Indian J Pediatr . 1998;65:663–668.
[8] Buchanan GR, Boxer LA, Nathan DG. The acute and transient nature of idiopathic immune
hemolytic anemia in childhood. J Pediatr. 1976;88:780–783.
[9] Zecca M, Nobili B, Ramenghi U, et al. Rituximab for the treatment of refractory autoimmune
hemolytic anemia in children. Blood. 2003;101:3857–3861.
[10] Duru F, Gurgey A, Cetin M, Kanra T, Altay C. Chronic autoimmune hemolytic anemia in children:
a report of four patients. J Med. 1994;25:231–40.
[11] Gibson B. Autoimmune hemolytic anemia: current concepts. Aust NZ J Med. 1988(18):625–637.
[12] Pirofsky B. Clinical aspects of autoimmune hemolytic anemia. Semin Hematol. 1976;13:251–265.
[13] Park DC, Yang CH, Kim KY. Autoimmune hemolytic anemia in children. Yonsei Med J
1987;28:313–321.
[14] Collins PW, Newland AC. Treatment modalities of autoimmune blood disorder. Semin Hematol.
1992;29:64–74.
[15] Yarali N, Fisgin T, Kara A, Duru F. Successful management of severe chronic autoimmune
hemolytic anemia with low dose cyclosporine and prednisone in an infant. Turk J Pediatr .
2003;45:335–337.
 Autoimmune Hemolytic Anemia 315

[16] Zhang Y, Chu Y, Chen G. Clinical analysis of 164 cases Coombs test positive autoimmune hemolytic
anemia. Zhonghua Xue Ye Xue Za Zhi. 1998;19:573–575.
[17] Liu H, Shao ZH, Cui ZZ, et al. An analysis of relapse and risk factors of autoimmune hemolytic
anemia and Evans syndrome. Zhonghua Xue Ye Xue Za Zhi. 2003;24:534–537.
[18] Gurgey A, Yenicesu I, Kanra T, et al. Autoimmune hemolytic anemia with warm antibodies in
children: retrospective analysis of 51 cases. Turk J Pediatr. 1999;41:467–471.
[19] Shah A. Acquired hemolytic anemia. Indian J Med Sci. 2004;58:533–536.
[20] Genty I, Michel M, Hermine O, Schaeffer A, Godeau B, Rochant H. Characteristics of autoimmune
hemolytic anemia in adults: retrospective analysis of 83 cases. Rev Med Intern. 2002; 23:901–909.
[21] Quartier P, Brethon B, Philippet P, Landman-Parker J, Le Deist F, Fischer. Treatment of childhood
autoimmune haemolytic anaemia with rituximab. Lancet. 2001; 358(9292):1511–1513.
Pediatr Hematol Oncol Downloaded from informahealthcare.com by Mcgill University on 11/06/14
For personal use only.