FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2146846747> ?p ?o ?g. }

• W2146846747 endingPage "511" @default.
• W2146846747 startingPage "506" @default.
• W2146846747 abstract "It may seem surprising that in the era of monoclonal antibodies, cytogenetics and molecular genetics, the editors of the British Journal of Haematology perceive a need for an annotation on this topic. The literature on diagnostic pitfalls in leukaemia is scanty and observations tend to be anecdotal rather than evidence-based. This personal perspective is necessarily biased towards the author's and colleagues' experience of clinical problems. Mistakes may arise, albeit infrequently in modern times, either in the diagnosis or classification of leukaemia (Table I). A misdiagnosis of leukaemia subjects patients to the risks of cytotoxic therapy. Choosing the wrong type of treatment for patients who do have acute leukaemia significantly reduces their chance of cure. Both errors are disastrous for the patient and increasingly the subject of litigation. The risk of either type of mistake is reduced by a collaborative approach to diagnosis in which there is systematic clinical and morphological review of all possible cases of leukaemia by senior members of staff and by recognising the circumstances in which diagnostic problems may occur. No article on paediatric leukaemia is complete without a mention of prognostic factors and there are three main risk factors which predispose to diagnostic problems. The first risk factor is undue haste in starting treatment before sufficient good quality diagnostic material has been obtained to enable morphological diagnosis, immunophenotyping and cytogenetics. Well-prepared and stained blood and bone marrow films are essential. Children with possible leukaemia need prompt investigation and treatment for the symptoms of bone marrow failure, but they do not normally need immediate cytotoxic therapy. There should never be any hesitation in repeating bone marrow aspirates or trephines to obtain adequate material. In the few children who may need urgent therapy, for example because of a very high leucocyte count, the diagnosis is usually obvious. The second of these is an inappropriate reliance on modern technology, in particular the interpretation of immunophenotyping results. Clinical examination and morphological review of well-stained blood, marrow films and, if appropriate, biopsy material remain the essential tools for diagnosis of leukaemia and if these are not used properly, mistakes will be made. Other techniques for the diagnosis of leukaemia are merely confirmatory and an immunophenotyping report, however emphatic, cannot per se form a basis for diagnosis and treatment. Lastly, leukaemia in the young infant, particularly in the first few months of life, is very rare and is mimicked by a host of other disorders, particularly infections. The threshold for diagnosis of leukaemia in this age group should be extremely high and, particularly in the newborn infant, the threshold for starting treatment should be even higher. It is impossible to over-emphasize the need to obtain adequate samples before starting treatment. In practice, perhaps one of the most difficult aspects of diagnosis is the distinction between low count acute lymphoblastic leukaemia (ALL) and aplastic anaemia. In this situation, provided good supportive care is initiated, there should be no hesitation in observing the patients and repeating the bone marrow aspirate with a trephine biopsy. Urgent cytotoxic treatment may be indicated in children with a high leucocyte count who are at risk of leucostasis, but in such cases adequate material can readily be obtained for all investigations before starting treatment. Problems may sometimes arise if urgent treatment is needed and the diagnosis cannot be readily confirmed. The most common circumstance leading to this problem is the child with respiratory obstruction, massive anterior mediastinal widening and a normal blood film. The most probable diagnosis in such cases is T-cell non-Hodgkin's lymphoma (NHL) or T-cell leukaemia, an artificial distinction which depends on the degree of bone marrow infiltration. An alternative diagnosis, although usually associated with less acute onset of symptoms, is Hodgkin's disease. Anaesthesia or sedation in such cases may be hazardous but, thanks to close co-operation with anaesthetists and surgeons, a brief period of steroid treatment for 24–48 h or low-dose local irradiation will allow the patient to stabilize before biopsy and bone marrow aspirate. The blood and particularly the bone marrow in the young child have a relatively high proportion of lymphocytes and these may have the phenotype of early B cells, being CD19, CD10 and TdT positive. A young child with enlargement of lymph nodes, liver and spleen, a blood count which is normal or shows lymphocytosis may have a bone marrow full of reactive lymphoid cells with the phenotype of immature B cells. Concern about a possible diagnosis of malignancy in such cases may lead to biopsy of an enlarged lymph node. The appearances of the node will be consistent with reaction to infection, whereas the histological appearances would be those of lymphoblastic lymphoma in children with lymphoblastic leukaemia. Bone marrow lymphocytosis may be misinterpreted in congenital and acquired marrow failure syndromes, and may be interpreted as showing relapse in children with ALL who have electively stopped treatment (van Wering et al, 2000). This is perhaps more probable now that routine follow-up bone marrows have become less common in the management of ALL. The use of a wide panel of monoclonal antibodies in classification of leukaemia has also lead to confusion over the diagnosis of biphenotypic leukaemia. True biphenotypic leukaemia is exceptionally rare in childhood and the diagnosis is discussed below. Acute lymphoblastic leukaemia in infants is often associated with a high leucocyte count, organomegaly and the phenotype of early B-cell lineage. It causes no diagnostic problems, apart from sometimes in the interpretation of immunophenotyping. Other features, more frequently seen in infant acute myeloid leukaemia (AML), include skin infiltration, which must be distinguished from congenital infections, and neuroblastoma. Neonatal leukaemia is rare and there are two main types: congenital AML which is usually monoblastic, and transient abnormal myelopoeisis (TAM) which is seen in Down's syndrome (DS). The latter has all the features of megakaryoblastic leukaemia, including additional cytogenetic abnormalities. Most cases of TAM (Lange, 2000) and some other congenital leukaemias resolve without specific treatment. About 1–2% of cases of childhood ALL and a smaller proportion of those with AML are preceded by a period of apparent bone marrow failure, subsequent clinical and haematological remission within weeks, and later development of acute leukaemia, usually within 3 to 9 months. At the time of initial presentation, in contrast to acquired aplastic anaemia in childhood, neutropenia is more pronounced than thrombocytopenia. Bone marrow examination shows a hypocellular marrow, sometimes with increased fibrosis. Usually, the subsequent leukaemia is pre-B ALL (Breatnach et al, 1981; Hasle et al, 1995). Diagnostic confusion in such cases is avoided by prompt supportive care at the time of presentation with bone marrow failure, careful follow up and repeating the bone marrow aspirate or trephine in 4–6 weeks if the patient remains pancytopenic. Cytogenetic analysis of the bone marrow may prove helpful, but it is often difficult to obtain sufficient material for analysis. Such patients should not be given specific treatment until either a confident diagnosis of aplasia is made or there has been recovery and, later, the development of acute leukaemia. Provided such patients have not been inappropriately treated with steroids or immune suppression, their prognosis is as good as that of other children with ALL. The bone marrow in both acquired aplastic anaemia and congenital bone marrow disorders may contain a high proportion of early B lymphocytes and failure to recognize this fact has led to misdiagnosis of congenital neutropenia, Diamond Blackfan anaemia and transient erythroblastopenia of childhood (Foot et al, 1990). Here again, careful consideration of the clinical picture and a period of observation should obviate mistakes in diagnosis. Paediatricians and haematologists in the developed world rarely see megaloblastic anaemia, except in children who have received cytotoxic therapy. Both folate and B12 deficiency are rare in childhood, and overt megaloblastic anaemia in infancy is usually due to metabolic disorders such as transcobalamin II deficiency. The pancytopenia, gross dyserythropoiesis and presence of early abnormal myeloid cells in this rare disease may mimic acute leukaemia (Niebrugge et al, 1982), but the combination of pancytopenia and florid dyserythropoiesis should induce a diagnostic pause. By the time such infants develop marrow failure, neurological symptoms such as developmental delay and poor tone have usually become apparent. Pancytopenia, fever and increasing hepatosplenomegaly are features of infantile haemophagocytic lymphohistiocytosis, a rare condition characterized by infiltration of the spleen, liver, bone marrow and central nervous system with histiocytic cells (Henter et al, 1991). The hallmark of this disease is the finding of haemophagocytosis in the bone marrow or other tissues, but a number of associated diagnostic features include raised fasting triglycerides, raised serum ferritin and low plasma fibrinogen. Repeated bone marrow aspiration and/or liver biopsy may be needed to confirm the diagnosis. A leuco-erythroblastic blood film with blast cells in the blood may cause confusion with acute leukaemia and is a feature of many disorders, including severe infection, bone marrow infiltration with solid tumours (see below) and thalassaemia major. All these conditions may be associated with hepatosplenomegaly. Careful examination of the blood film and, when appropriate, the bone marrow should clarify the diagnosis. Infantile osteopetrosis is also associated with a leuco-erythroblastic anaemia, a bone marrow that is difficult to aspirate or may yield an inadequate sample. It is important to try and make this diagnosis promptly before visual deterioration as bone marrow transplantation provides effective therapy (Wilson & Vellodi, 2000). Radiographs of the long bones are normally diagnostic and should be reviewed with a paediatric radiologist. The clinical picture of idiopathic thrombocytopenic purpura (ITP) may superficially resemble acute leukaemia but the blood count is usually normal apart from reduced platelets. Bone marrow examination may not be necessary in the classical case of acute ITP, but is recommended if steroid therapy is contemplated to avoid masking a diagnosis of ALL. Bone pains, joint pains and swelling are common presenting features of lymphoblastic leukaemia and skeletal radiographs (rarely clinically indicated) show lytic lesions, periosteal reactions or metaphyseal changes in many children with ALL. Such patients frequently have low leucocyte counts and the diagnosis of leukaemia may not be apparent on the blood film (Hughes & Kay, 1982). A common diagnostic error, fortunately now rare, was to diagnose children with such symptoms as having juvenile rheumatoid arthritis, treat them with steroids and only recognize leukaemia after relapse of a steroid-induced remission. Juvenile rheumatoid arthritis is a diagnosis of exclusion and it could be argued that a routine bone marrow should precede steroid therapy in most cases. Most other diagnostic problems involve possible infection, particularly in the young infant. A large number of infections may mimic leukaemia – these include acute bacterial infection, cytomegalovirus, parvovirus, rubella and toxoplasmosis. These may cause confusion with possible acute lymphoblastic leukaemia or myelomonocytic leukaemia. To this list must now be added congenital human immunodeficiency virus (HIV) infection, which we have seen mimic acute leukaemia in infancy and which in older children is associated with lymphomas and a variety of other malignancies (Granovsky et al, 1998). More unusually, infantile leishmaniasis is associated with fever, progressive pancytopenia and an enlarging liver and spleen (Smith et al, 1995). Epstein-Barr virus (EBV) infection is well recognized as superficially resembling leukaemia, but in practice appropriate investigations exclude it. The most common paediatric solid tumour involving the bone marrow at the time of diagnosis is neuroblastoma. The clinical picture of disseminated disease with bone pain and anaemia is very similar to that of acute leukaemia. Abnormal cells are rarely seen in the blood film, which may be leuco-erthryoblastic, but there may be almost complete replacement of the bone marrow by syncytial clumps or even sheets of malignant cells. Immunophenotyping and immunocytochemistry may be helpful in distinguishing solid tumours from leukaemic blasts. The diagnosis of neuroblastoma is usually confirmed by the finding of raised catecholamines and a primary mass in the abdomen or posterior thorax. Other disseminated cancers which may mimic leukaemia include soft tissue sarcomas, germ cell tumours and, in children with a ventriculo-peritoneal shunt, medulloblastoma. Occasionally, children present with bone pain, bone marrow involvement with tumour cells and no obvious primary lesion. This diagnosis of a solid tumour should be considered in all children with large atypical blasts with no indication of haemopoeitic lineage. Cytogenetic analysis may be helpful in showing t(11;22) suggestive of primitive neuroectodermal tumour, or t(1;13)/(2;13) in alveolar rhabdomyosarcoma. Most children with leukaemia have only one good chance of cure and that is at the time of diagnosis. It is imperative that appropriate first-line treatment is given and this means not only distinguishing AML from ALL, but also identifying the 1–2% of children with ALL and surface membrane immunoglobulin-positive B cells who are highly curable with short-term high-dose therapy, as used in B-cell NHL (Patte et al, 1994). This distinction is easily made when L3 blasts in the marrow are found in a child with a large abdominal mass or nasopharyngeal tumour, but may be overlooked in an otherwise typical case of acute leukaemia. The basophilic cytoplasm in such cases may raise a suspicion of M6 (erythroid) AML, but the diagnosis is confirmed by the results of immunophenotyping and the characteristic cytogenetic abnormality of t(8;14) or related translocation. Occasionally, blasts with L1 or L2 morphology show surface membrane immunoglobulin, or blasts may appear L3 but lack surface membrane immunoglobulin. Cytogenetic analysis is crucial in such cases and would determine the approach to treatment. In clinical paediatric practice, most morphological difficulties arise in distinguishing between L2 ALL and undifferentiated M0 AML or M5 AML, and in recognition of M6 and M7 AML. The distinctions are facilitated by the use of appropriate cytochemistry and a panel of monoclonal antibodies, for example those recommended by the British Committee for Standards in General Haematology (1994), see Table II. Such panels combine antibodies which are highly specific, for example cytoplasmic (cy) CD22 (or CD79a) for B-lineage, cy CD3 for T lineage, anti-myeloperoxidase (MPO) for myeloid lineage, and those which are highly sensitive such as CD19, CD7 and CD 13/33. Additional antibodies are required for confirmation of the diagnosis of M6 AML and the rare M7, particularly when it occurs in infants without DS. The early literature on childhood leukaemia was replete with mentions of acute undifferentiated leukaemia (AUL), but since the advent of cytochemistry and a panel of monoclonal antibodies, virtually all acute leukaemia can be classified. It is probable that most of these cases of AUL were in fact lymphoblastic leukaemia. Clinical confusion has arisen in assigning treatment to patients with acute leukaemia and co-expression of lymphoid and myeloid antigens. These have variously been referred to as mixed-lineage, hybrid or biphenotypic leukaemias. It has become clear that aberrant antigen expression is a feature of both ALL and AML and attempts have been made to develop more stringent criteria for the diagnosis of true biphenotypic acute leukaemia. One proposed scoring system is illustrated in Table III (Matutes et al, 1997). It is apparent from Table III that most emphasis is given to the presence of highly specific antibodies and, for example, the expression of both CD13 and CD33 does not render ALL a biphenotypic leukaemia. The prognosis of ‘true’ biphenotypic leukaemia is related to the accompanying genetic abnormalities and many cases are associated with the presence of the Ph′ chromosome (Carbonell et al, 1996). Myeloid-associated antigen expression, in particular CD15 and CD65 is frequently seen in acute leukaemia of infancy associated with MLL gene rearrangements (Behm et al, 1996). Such infants with ALL are now usually treated on protocols which include elements active in AML in an attempt to improve their prognosis. A number of studies have now confirmed that aberrant antigen expression which does not fulfil the criteria for mixed-lineage leukaemia has no prognostic significance in children. In the Medical Research Council (MRC) UKALL XI trial, 17% of children had leukaemia with either CD13 and/or CD33 positivity (Hann et al, 1998). It was originally suggested that such cases in children had a poor prognosis. However, recent reports including that from UKALL XI confirm that such findings have no adverse prognostic significance and there is no indication to change treatment. Similarly, the expression of lymphoid-associated antigens is of no prognostic significance in paediatric AML (Smith et al, 1992). For many years it was standard practice to perform regular bone marrows during therapy of acute leukaemia and for some time thereafter. When it became apparent that this practice did not influence the prognosis (Rogers et al, 1984), it was widely abandoned. There is no doubt that problems did arise in interpretation of bone marrows, in particular the lymphocytic rebound seen after interruption of therapy could be interpreted as relapse. The distinction between rebound and relapse can be facilitated by the use of a panel of monoclonal antibodies or molecular measurement of minimal residual disease. It is, however, extremely important that such techniques are rigorously standardized (Campana & Behm, 2000) and that treatment decisions based on minimal residual disease are instigated as part of rigid prospective studies with adequate technical supervision rather than performed on an ad hoc basis. An occasional difficulty arises with interpretation of marked abnormalities of erythroid and myeloid morphology induced by cytotoxic therapy. Similar problems may be encountered in patients who have been given granulocyte colony-stimulating factor (G-CSF) during chemotherapy. These appearances may cause concern about the possibility of relapse or secondary leukaemia. A conservative approach to management is indicated unless objective evidence in the form of cytogenetic change supports such a diagnosis. The interpretation of cytospins of the cerebrospinal fluid (CSF) is more difficult than is widely admitted and the use of a barrage of monoclonal antibodies does not really address this problem. Children with ALL virtually all have presymptomatic leukaemia at diagnosis, as shown by the inevitability of CNS relapse in those who do not receive CNS-directed therapy. Therefore, the presence of cells in the CSF is indicative of the tip of the iceberg. The widely accepted definition of overt CNS leukaemia is the presence of either cranial nerve infiltration or a CSF pleocytosis of more than 5 cells per mm3 with recognizable blast cells. Children with these findings at diagnosis or the time of relapse usually receive more intensive intrathecal chemotherapy and, at least in ALL, cranial or craniospinal irradiation. Most protocols specify that children with a lower cell count are treated as having a ‘clear’ CSF and given no additional treatment. It has recently been suggested that the presence of equivocal CNS findings or a bloody CSF at diagnosis may be associated with a worse prognosis and benefit from additional treatment. This has led to a proposed amendment for the staging of CNS leukaemia at diagnosis: a clear CSF, unequivocal CNS disease as described above and an ‘intermediate’ category with a low number of blasts in the CSF (Smith et al, 1996). There is no doubt that these equivocal cases cause problems in management and some have advocated that they should receive additional intrathecal therapy (Pui et al, 1998). A second difficult area is the interpretation of CSF samples in children who are undergoing regular intrathecal chemotherapy when a low number of blast cells is found on a routine CSF cytospin. There is little point in promptly repeating the lumbar puncture as the blast count will have been reduced by the intrathecal therapy. Despite the report that children with these findings do not have an adverse prognosis (Tubergen et al, 1994), we have found that recurrent low-grade pleocytosis may be a prelude to CNS relapse, sometimes with focal rather than generalized disease. We normally repeat the CSF after 4–6 weeks without giving more intrathecal therapy, to avoid masking the diagnosis. Immunophenotyping (Hooijkaas et al, 1989) may be helpful in this situation, and cytogenetics or fluorescence in situ hybridization (FISH) may prove useful if adequate material can be obtained. These have recently been comprehensively reviewed in this journal (Emanuel, 1999) and perhaps the main emphasis in terms of differential diagnosis is the extreme rarity of both primary polycythaemia and essential thrombocythaemia – strenuous efforts are required to identify an underlying cause. The main pitfalls in the diagnosis of myelodysplastic syndrome (MDS) concern recognition of sideroblastic anaemia and chronic myelomonocytic leukaemia (CMML). Refractory anaemia with ringed sideroblasts is an exceptionally rare type of MDS in childhood. Sideroblastic anaemia is either due to congenital disorders of haem synthesis or mitochondrial cytopathies, in particular Pearson's syndrome, in which failure to thrive, acidosis and anaemia are associated with vacuolated myeloid and erythroid precursors. Myelodysplasia in children, as in adults, is usually associated with cytopenia in the blood and a cellular bone marrow. Diagnostic problems may arise when the marrow is hypocellular but shows marked dysplasia. These features are most often seen in chronic aplastic anaemia and our practice has normally been to treat such patients as aplastic in the absence of clonal cytogenetic abnormalities. Chronic myelomonocytic leukaemia, now renamed juvenile myelomonocytic leukaemia, usually presents with hepatosplenomegaly, thrombocytopenia, a blood film characterized by dysplasia, abnormal monocytes and blast cells, and a relatively uninformative bone marrow (Emanuel, 1999). Other features include a raised fetal haemoglobin and, sometimes, monosomy 7. A non-specific skin rash bears a superficial resemblance to that of Langerhans cell histiocytosis. Patients may occasionally present with features of an immune regulatory disorder. The clinical picture is normally very aggressive, but may be indolent in the infant and mimic viral infections. The blood film appearances are normally quite characteristic, but occasionally diagnostic confusion may arise with M5 AML or erythroleukaemia. This distinction is important because CMML is refractory to chemotherapy and a bone marrow transplant, although associated with a high relapse rate, affords the only hope of cure. A variety of conditions may mimic acute leukaemia and related disorders, particularly in the young child. Immunophenotyping, cytogenetics and molecular genetics allow a rational approach to diagnosis and management, but are not a substitute for traditional morphology. Most problems in diagnosis arise from undue haste and failure to recognize that these tools are our servants and not our masters. I would like to thank the Leukaemia Research Fund for their continued support and my colleagues, especially Dr Ian Hann, for constant and stimulating discussion." @default.
• W2146846747 created "2016-06-24" @default.
• W2146846747 creator A5072052288 @default.
• W2146846747 date "2001-09-01" @default.
• W2146846747 modified "2023-10-17" @default.
• W2146846747 title "Pitfalls in the diagnosis of childhood leukaemia" @default.
• W2146846747 cites W152381011 @default.
• W2146846747 cites W1777458938 @default.
• W2146846747 cites W1982187522 @default.
• W2146846747 cites W2037243028 @default.
• W2146846747 cites W2045394616 @default.
• W2146846747 cites W2046381186 @default.
• W2146846747 cites W2051365008 @default.
• W2146846747 cites W2053786772 @default.
• W2146846747 cites W2067895408 @default.
• W2146846747 cites W2069455563 @default.
• W2146846747 cites W2081443304 @default.
• W2146846747 cites W2110915470 @default.
• W2146846747 cites W2119685895 @default.
• W2146846747 cites W2141160670 @default.
• W2146846747 cites W2150674676 @default.
• W2146846747 cites W2238647955 @default.
• W2146846747 cites W2239688246 @default.
• W2146846747 cites W2256419275 @default.
• W2146846747 cites W2408524419 @default.
• W2146846747 cites W4251505416 @default.
• W2146846747 cites W760311 @default.
• W2146846747 doi "https://doi.org/10.1046/j.1365-2141.2001.02994.x" @default.
• W2146846747 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/11552974" @default.
• W2146846747 hasPublicationYear "2001" @default.
• W2146846747 type Work @default.
• W2146846747 sameAs 2146846747 @default.
• W2146846747 citedByCount "23" @default.
• W2146846747 countsByYear W21468467472013 @default.
• W2146846747 countsByYear W21468467472014 @default.
• W2146846747 countsByYear W21468467472015 @default.
• W2146846747 countsByYear W21468467472016 @default.
• W2146846747 countsByYear W21468467472017 @default.
• W2146846747 countsByYear W21468467472018 @default.
• W2146846747 countsByYear W21468467472019 @default.
• W2146846747 countsByYear W21468467472020 @default.
• W2146846747 crossrefType "journal-article" @default.
• W2146846747 hasAuthorship W2146846747A5072052288 @default.
• W2146846747 hasBestOaLocation W21468467471 @default.
• W2146846747 hasConcept C177713679 @default.
• W2146846747 hasConcept C187212893 @default.
• W2146846747 hasConcept C2991856797 @default.
• W2146846747 hasConcept C71924100 @default.
• W2146846747 hasConceptScore W2146846747C177713679 @default.
• W2146846747 hasConceptScore W2146846747C187212893 @default.
• W2146846747 hasConceptScore W2146846747C2991856797 @default.
• W2146846747 hasConceptScore W2146846747C71924100 @default.
• W2146846747 hasIssue "3" @default.
• W2146846747 hasLocation W21468467471 @default.
• W2146846747 hasLocation W21468467472 @default.
• W2146846747 hasOpenAccess W2146846747 @default.
• W2146846747 hasPrimaryLocation W21468467471 @default.
• W2146846747 hasRelatedWork W1506200166 @default.
• W2146846747 hasRelatedWork W1995515455 @default.
• W2146846747 hasRelatedWork W2048182022 @default.
• W2146846747 hasRelatedWork W2080531066 @default.
• W2146846747 hasRelatedWork W2604872355 @default.
• W2146846747 hasRelatedWork W2748952813 @default.
• W2146846747 hasRelatedWork W2899084033 @default.
• W2146846747 hasRelatedWork W3031052312 @default.
• W2146846747 hasRelatedWork W3032375762 @default.
• W2146846747 hasRelatedWork W3108674512 @default.
• W2146846747 hasVolume "114" @default.
• W2146846747 isParatext "false" @default.
• W2146846747 isRetracted "false" @default.
• W2146846747 magId "2146846747" @default.
• W2146846747 workType "article" @default.