Aplastic anaemia

• Aplastic anaemia is bone marrow failure: the marrow is empty or nearly empty and cannot make enough red cells, white cells or platelets, so the patient has pancytopenia.
• In most cases the cause is immune: T lymphocytes attack the stem cells that normally rebuild the blood.
• It is rare (about 1 to 3 cases per million people each year) but dangerous, because infection and bleeding are the chief risks.
• Diagnosis needs a bone marrow biopsy to show the empty marrow and, just as importantly, to exclude myelodysplasia, leukaemia and inherited bone marrow failure.
• Treatment depends on age and donor availability: stop any offending drug, support with transfusions, then either an allogeneic stem cell transplant from a matched sibling in younger patients or immunosuppression with horse antithymocyte globulin and ciclosporin, usually with eltrombopag.

By the end of this page you should be able to:

1. Define aplastic anaemia and explain why the marrow is empty and the blood is pancytopenic.
2. List the acquired and inherited causes, and say which are common.
3. Classify severity with the modified Camitta criteria and explain why severity changes urgency.
4. Plan the investigation of pancytopenia, including what the marrow biopsy must exclude.
5. Describe the standard treatment pathway by age and donor status, including drug choices.
6. Recognise a pancytopenia due to something else (myelodysplasia, infection, hypersplenism, PNH) and explain how to separate it.
7. Adapt management where a transplant centre or eltrombopag is not available.

Aplastic anaemia is one of the bone marrow failure syndromes. The problem is not that the blood is being destroyed or lost, but that it is not being made. The marrow cavity, which should be full of developing blood cells, becomes fatty and empty. The result is pancytopenia: anaemia, neutropenia and thrombocytopenia together ^[2].

Modern care has changed the outlook. With timely treatment, five-year survival now exceeds 80 to 90 percent ^[3]. The gains come from better supportive care, better immunosuppression, the addition of eltrombopag, and safer stem cell transplantation. ^[2]

Aplastic anaemia is classified by cause and by severity.

• Acquired (immune) aplastic anaemia is the commonest form and accounts for more than two-thirds of cases ^[3].
• Inherited bone marrow failure syndromes (Fanconi anaemia, dyskeratosis congenita and other telomere biology disorders, Shwachman-Diamond syndrome, and others) account for the rest, mainly in younger patients ^[3].
• Secondary aplasia can follow a drug, toxin or whole-body irradiation.

Severity is graded with the modified Camitta criteria, shown in the severity section below. The grade drives both prognosis and how quickly treatment must be started ^[3].

Aplastic anaemia is rare. In North America and Europe the incidence of acquired disease is about 2 to 3 cases per million people per year, and it is two to three times higher in parts of Asia ^[2]. Another broad review puts the estimate at about 1 to 2 cases per million per year worldwide, again roughly threefold higher in East Asia ^[3].

The age pattern has two peaks: young adults (15 to 25 years) and older adults (over 60 years), with a slight male predominance in some series ^[3]. Most patients present with either an incidental pancytopenia found on a routine blood count, or symptoms of anaemia, bleeding or infection.

Most acquired cases are idiopathic: no single trigger is found. The working model is an antigen-driven, T-cell-mediated attack on haematopoietic stem cells, sometimes kicked off by the agents below ^[3]. Inherited cases come from germline mutations that damage DNA repair, telomere maintenance or ribosomal function, and so slowly destroy the stem cell pool ^[3].

Two practical notes follow. First, Fanconi anaemia has an incidence of about 1 in 160,000 births, and its median marrow failure onset is in childhood ^[2], so a child or young adult who "has aplastic anaemia" deserves a careful search for it. Second, parvovirus B19 and hepatitis viruses are reversible triggers in some cases, which is one reason the viral screen is part of the work-up rather than an afterthought ^[3][4].

Start with normal marrow. It produces every blood cell, and it is continually repaired by a large pool of haematopoietic stem cells. In immune aplastic anaemia, an unidentified antigen leads to a polyclonal expansion of CD4 helper T cells, then to dysregulated, oligoclonal CD8 cytotoxic T cells that infiltrate the marrow ^[3].

Those cytotoxic cells do the damage in three linked ways:

• they kill CD34+ progenitor cells by interferon-gamma and tumour necrosis factor-alpha, mainly through the Fas apoptosis pathway, so the progenitor pool collapses ^[3];
• they increase pro-inflammatory interleukins (IL-17, IL-6, IL-8) and reduce regulatory T cells, so the self-attack is unchecked ^[3];
• the marrow stroma and stem-cell niche are also injured, which makes recovery harder even after the immune attack stops.

Telomerase impairment can contribute in some patients: recurrent mutations in TERT and TERC, and shortened telomeres in granulocytes or lymphocytes, point to telomere biology disorders. Other recurrent changes include mutations in BCOR/BCORL1, PIGA, DNMT3A and ASXL1, and cytogenetic abnormalities such as monosomy 7 or trisomy 8 ^[3].

Horse antithymocyte globulin (ATG) plus ciclosporin reverses the disease in up to about 80 percent of patients, which is strong evidence that the immune mechanism is the driver ^[3].

The trephine biopsy shows a markedly hypocellular marrow, often under 25 percent cellularity, with fatty replacement of the normal cellularity and no abnormal infiltrate or fibrosis ^[3][4]. The residual cells are mostly lymphocytes, plasma cells and stromal cells, with scant or absent megakaryocytes.

The aspiration is often a "dry" or dilute tap, and what cells are present are morphologically unremarkable. Milder, acquired marrow failure can still be cellular, which is why the diagnosis is a combination of blood counts and marrow histology rather than a single finding.

Margins matter. Mild dyserythropoiesis can occur, but granulocytic or megakaryocytic dysplasia should be absent in true aplastic anaemia; micromegakaryocytes and marked dysplasia point to myelodysplasia. Monosomy 7 or 5q loss argues for myelodysplasia rather than aplasia ^[3]; cytogenetic abnormalities occur in about 10 percent of AA patients.

The presentation follows the cytopenias. There are no disease-specific findings; instead there are the ordinary consequences of too few red cells, white cells and platelets.

History. Fatigue, breathlessness and pallor from anaemia; easy bruising, petechiae, epistaxis, gum bleeding or menorrhagia from thrombocytopenia; recurrent or severe infection from neutropenia. Some patients are entirely asymptomatic and are found when a routine full blood count is taken ^[3]. The history must cover recent medicines, chemicals and radiation; family history of cytopenia or premature greying, lung or liver disease; congenital anomalies; and any previous atypical infection ^[1][3].

Examination. Pallor, purpura and petechiae, sometimes thrush or other infection. Hepatosplenomegaly and lymphadenopathy are not features of aplastic anaemia. Their presence should make you look for leukemia, infection, or an infiltrative disorder instead ^[4]. In children, subtle congenital anomalies (short stature, skeletal or urogenital abnormalities, skin or nail changes, premature greying) may be the only clue to an inherited syndrome ^[3].

Pancytopenia is a syndrome, not a diagnosis. Before settling on aplastic anaemia, the causes below must be considered. The table is ordered by mechanism and explains what separates each one.

The aim of investigation is to confirm bone marrow failure, grade severity, exclude inherited disease, and separate aplastic anaemia from its mimics.

First-line and screening bloods:

• full blood count with differential and a reticulocyte count (expected to be low) ^[4];
• blood film (blasts, teardrop cells, macrocytosis);
• haematinics (B12, folate, iron studies, copper and ceruloplasmin if risk factors) ^[4];
• liver, renal and thyroid function;
• haemolysis screen (bilirubin, lactate dehydrogenase, haptoglobin, direct antiglobulin test);
• viral serology/PCR for HIV, hepatitis B and C, Epstein-Barr virus, cytomegalovirus, parvovirus B19 and others as clinically indicated ^[3][4];
• autoimmune screen when features suggest it ^[4];
• abdominal ultrasound if the liver or spleen is enlarged ^[4].

Confirmatory tests:

• bone marrow aspiration and trephine biopsy, with cellularity assessment, morphology, and cytogenetics (including FISH for chromosome 7 and 5q) ^[2][3];
• flow cytometry for a PNH clone ^[1][3];
• germ-line genetic testing (Fanconi chromosomal breakage testing, telomere length) when an inherited syndrome is suspected, particularly in those under 40 years ^[1][2];
• HLA typing as soon as severe or very severe disease is suspected in a transplant-eligible patient ^[1].

Interpretation. Aplastic anaemia shows low reticulocytes and empty marrow, whereas haemolysis shows high reticulocytes and a cellular marrow; megaloblastic anaemia shows a cellular, megaloblastic marrow; and leukaemia shows blasts. Fundamentally, no single blood test makes the diagnosis: the biopsy is required.

A practical diagnostic framework uses at least two suppressed lineages plus a hypocellular marrow.

• Pancytopenia threshold used to recognise the syndrome: haemoglobin under 100 g/L (under 10 g/dL), neutrophils under 1.5 x 10^9/L and platelets under 100 x 10^9/L ^[4]; another diagnostic framework for aplastic anaemia uses haemoglobin under 100 g/L, neutrophils under 1.5 x 10^9/L and platelets under 50 x 10^9/L ^[3]. The two overlap; both are quoted because textbooks and guidelines differ.
• Marrow: typical hypocellularity under 25 percent (or under 50 percent with under 30 percent haematopoietic cells) ^[2].
• Absence of abnormal infiltration, dysplasia or fibrosis.

Severity is then graded with the modified Camitta criteria ^[3].

VSAA is tied to a lower overall survival than SAA, and it also changes the urgency of second-line treatment ^[1][3]. Neutrophils under 0.2 x 10^9/L mean the patient is at very high risk of severe infection and usually needs admission, prompt therapy and antimicrobial prophylaxis ^[1][3].

Management should take place where there is expertise in bone marrow failure ^[3]. The first step in everyone is support and safety: stabilise bleeding and infection, give irradiated, leukodepleted red cells and platelets, and avoid aspirin or other antiplatelets. Only after that comes definitive treatment.

• Red cells: transfuse packed cells to relieve symptoms, not to reach a normal haemoglobin. A restricted strategy is usual, commonly keeping haemoglobin above 70 g/L (7 g/dL) ^[4].
• Platelets: platelet concentrates only if the count is under 20 x 10^9/L or there is active bleeding ^[3].
• Infection prevention: all blood products should be leukodepleted and irradiated; Pneumocystis prophylaxis is recommended for lymphopenic patients; antifungal prophylaxis may be needed if neutrophils remain under 0.2 x 10^9/L by one review ^[3], whereas current international guidance suggests mold-active antifungal prophylaxis for severe disease with neutrophils under 0.5 x 10^9/L ^[1].
• Antimicrobials: in the severe-neutropenia period, antibiotics and antifungal prophylaxis should be considered, as current international guidelines recommend for high-risk patients ^[1].
• Stop the cause: withdraw any suspect drug (chloramphenicol, sulfonamides, carbamazepine, others) and remove any chemical or radiation exposure. When a reversible trigger is responsible, withdrawal can be enough; improvement may take weeks ^[3][4].
• Steroids are not recommended: they increase infection risk and do not improve marrow function ^[3].

The two definitive treatments are allogeneic haematopoietic stem cell transplantation (HSCT) and immunosuppressive therapy (IST). The choice follows age, donor availability and fitness rather than the severity alone ^[3]:

• under 40 years with a matched sibling donor: HSCT is the primary treatment, with curative potential and event-free survival above 85 percent in children ^[3];
• under 40 years without a matched sibling: matched unrelated donor HSCT is now a reasonable first-line option when a 10/10 or 8/8 donor can be found within 2 to 3 months of diagnosis, to avoid prolonged neutropenia and clonal change; otherwise IST is used ^[1][3];
• 40 to 60 years: HSCT can be used in good health without major comorbidity; otherwise IST ^[3];
• over 40 (or over 60) years: IST is the preferred first treatment, with HSCT reserved for non-response ^[1][3].

The conditioning matters. For patients under 40 years, the predominant regimen combines cyclophosphamide 6 g/m^2 with antithymocyte globulin; for older patients a reduced-intensity regimen of fludarabine 120 mg/m^2, cyclophosphamide 480 mg/kg and ATG is used ^[2]. ATG lowers both acute and chronic graft-versus-host disease ^[2].

IST is the combination of horse ATG plus ciclosporin. It works better than ATG alone, and adding a third agent such as mycophenolate or sirolimus has shown no benefit for response or relapse, so the two-drug (plus short-course steroid for serum sickness) regimen stays standard ^[3]. Ciclosporin is dosed to blood levels and continued for months, with slow tapering once counts recover ^[3]. G-CSF shortens the infection window during the response wait but does not improve survival or final response ^[3].

Eltrombopag is added. In studies combining eltrombopag with classical IST, overall response at 6 months reached 86 percent, and with eltrombopag from the first ATG dose 94 percent, of whom 58 percent achieved complete response ^[3]. Current international guidelines therefore suggest adding eltrombopag to IST for adults and children with severe or very severe disease ^[1]. It is worth noting that children already respond well to IST, with 80 to 90 percent overall survival, so the additional benefit in children should be interpreted with caution ^[1].

Timing is itself a treatment issue. One open-access review links a delay of more than 2 to 3 months between diagnosis and treatment in severe disease to a worse prognosis ^[3]; another review uses delays beyond 3 to 6 months ^[2]. Current international guidance frames the second-line decision point at no later than 6 months after ATG administration in either severe or very severe disease ^[1]. For non-responders under 40 to 60 years, HSCT is generally preferred over a second ATG course; over 60 years, either option is acceptable ^[1]. Matched unrelated or haploidentical donors can be used when a sibling is unavailable, but ATG-based IST remains an alternative in the older or less fit ^[1].

Relapses occur in a minority and can appear months or years later. Clonal change is the real long-term risk: one review estimates transformation to myelodysplastic syndrome at 10 to 15 percent over 10 to 15 years and acute myeloid leukaemia at 5 to 10 percent ^[3], while another reports a combined myelodysplastic syndrome and acute myeloid leukaemia figure of about 11 to 20 percent ^[2], mainly in older patients or those with persisting cytopenias. A PNH clone is detected in 20 to 50 percent of patients on standard flow cytometry ^[3], and in roughly 40 to 60 percent when high-sensitivity flow cytometry is used ^[2]; small clones are not haemolytic and may predict a good response to IST, whereas larger clones behave as overt PNH ^[3]. Cytogenetic and clone monitoring during follow-up is therefore part of good care.

• infection and sepsis during neutropenia, the leading early cause of death ^[3];
• haemorrhage from thrombocytopenia;
• transfusion reactions and iron overload in heavily transfused patients;
• failure to respond or relapse after IST;
• clonal evolution to myelodysplasia, acute myeloid leukaemia or PNH ^[3];
• treatment-related toxicity of transplantation (graft-versus-host disease) and conditioning.

Untreated severe aplastic anaemia carries high early mortality from infection or bleeding. With modern therapy, five-year survival exceeds 80 to 90 percent ^[2]. Outcomes are driven by age, response, relapse and clonal evolution rather than by the presenting count alone ^[2][3]. HLA-matched family donor transplantation gives overall survival of roughly 76 to 90 percent, with event-free survival over 85 percent in children ^[3].

There is no population screening. The useful preventive act is awareness: avoidance of chloramphenicol, needless sulfonamides and carbamazepine where alternatives exist, and of benzene exposure ^[3][4]. In the inherited syndromes, early recognition allows donor planning before clonal progression and avoids giving IST which is generally ineffective in inherited bone marrow failure ^[2].

Children. Onset in childhood should trigger a search for an inherited syndrome (Fanconi anaemia, dyskeratosis congenita, Shwachman-Diamond). About one-third of patients may have no obvious physical abnormality ^[3]. Transplantation from a matched family donor is the primary treatment, with excellent event-free survival ^[3].

Older adults. Clonal disorders (myelodysplasia, clonal cytopenia) are more likely. Performance status and comorbidity, not age alone, should guide whether transplantation is offered ^[1][2].

Pregnancy. IST and transplantation decisions in pregnancy need haematology, obstetric and transplant teams. Severe disease should be treated on clinical grounds; some cases improve with delivery.

Where a transplant programme or newer agents are not available, these steps preserve outcomes:

• Diagnose first. Confirm pancytopenia (haemoglobin, white cells, platelets) and take a full drug, toxin and family history. A single trephine biopsy, widely available even in district hospitals, is what separates aplastic anaemia from leukaemia and myelodysplasia ^[1][4].
• Screen for the inherited syndromes as cheaply as possible. Look for congenital anomalies, short stature, and skin or nail changes. Fanconi anaemia testing (chromosomal breakage studies) may require referral to a tertiary centre ^[2][3].
• Support safely. Use restricted red cell transfusion and platelet support under 20 x 10^9/L; leukodeplete and irradiate products if possible; treat infection promptly. Blood transfusion alone will not cure marrow failure ^[3][4].
• Use the IST that exists. Horse or rabbit ATG may be unavailable, but ciclosporin monotherapy is sometimes used where ATG is not affordable; outcomes are inferior but better than no treatment. Check local guidelines and costs. Eltrombopag access is limited by cost and supply in many settings ^[1].
• Refer early. Any young patient with a matched sibling should be identified and referred before irreversible infection, because transplantation is the curative option and delay worsens prognosis ^[1][3].
• Antifungal prophylaxis should be considered where neutropenia persists, at thresholds from 0.2 x 10^9/L in older reviews to 0.5 x 10^9/L in ASH 2026, using whichever agent is locally available and licensed ^[1][3].

Aplastic anaemia sits among the bone marrow failure syndromes and links to anaemia (approach and iron deficiency), immune thrombocytopenia, haematological malignancies, transplantation topics, Fanconi anaemia, telomere biology disorders, paroxysmal nocturnal haemoglobinuria, and antithymocyte globulin treatment. These are named here in plain text because their own MedGrasp pages are written separately.