Acta Pizdiatr Scand 80: 428-435, 199 1
Incidence in Sweden and Clinical Features of Familial Hemophagocytic Lym phohist iocytosis JAN-INGE HENTER,’ GORAN ELINDER,* OLLE SODER’ and AKE OST3 From the Department of Paediatrics. ‘St. Goran ’slKarolinskaand 2Sachs’ Children’s Hospitals and the 3Department of Pathology, Karolinska Hospital, Karolinska Institute, Stockholm, Sweden ABSTRACT. Henter, J.-I., Elinder, G., Siider, 0. and Ost, A. (Department of Paediatrics, St. Giiran’slKarolinska and Sachs’ Children’s Hospitals and Department of Pathology, Karolinska Hospital; Karolinska Institute, Stockholm, Sweden). Incidence in Sweden and clinical features of familial hemophagocytic lymphohistiocytosis. Acta Paediatr Scand 80: 428, 1991. We retrospectively studied the incidence of familial hemophagocytic lymphohistiocytosis (FHL) in children during the 16-year period 1971-86. First, all departments of pediatrics, pathology, and infectious diseases were enquired for children with FHL or disorders resembling FHL. Secondly, the causes of death of all children who died during the study period in Sweden (n=19542) were also investigated. Files and histological specimens were further studied in selected children. By using a set of inclusionlexclusion criteria, we found 32 children with FHL. The incidence was 1.211000000 children per year. One child per 50000 live borns developed FHL during this period. The sex ratio was close to 1 :1. Prominent early clinical signs were fever (91 %), splenomegaly(84 %), hepatomegaly (90 %), rash (43 %), and lymph node enlargement (42 %). Neurological symptoms, which developed in 47 %, could totally dominate the clinical picture and develop prior to other symptoms and signs. Common laboratory findings were pancytopenia, hypertriglyceridemia, hypofibrinogenemia, elevated serum transaminases, hyperbilirubinemia, hyponatremia, hypoalbuminemia, and a moderate spinal fluid pleocytosis. Chest X-ray often revealed mostly discrete pulmonary infiltrates. FHL is an underdiagnosed disease and in only 11/32 children was diagnosis made during their lifetime. It is important to be aware of the disorder as potential therapy now exists. Key words:familid hemophagocytic lymphohistiocytosis (FHL), incidcnce.
Familial hemophagocyticlymphohistiocytosis (FHL) is a rapidly fatal disease clinically characterized by intermittent fever, hepatosplenomegaly, and pancytopenia ( 1). Histological examination shows an accumulation of lymphocytes and macrophages with hemophagocytosis in reticuloendothelial organs. FHL has been considered to be a very rare disorder, but no previous study of the incidence has been reported in the literature. We have, in this retrospective study of a well-defined population, attempted to determine the incidence of FHL in children ‘In Sweden. Clinical features and laboratory findings in the affected children are also presented. In the recently proposed classification of histiocyte syndromes by the Histiocyte Society, FHL is included among “histiocytoses of mononuclear phagocytes other than Lagerhans’ cells” separate from malignant histiocytic diseases and Langerham-cell histiocytosis (LCH), the latter embracing disorders previously included in the entity Histiocytosis X (2). A virus-associated hemophagocytic syndrome (VAHS), which also affects histiocytes other than Langerhans’ cells, was described in 1979 as a potentially self-limited disease mainly seen in immunosuppressed patients (3). Out of the fifteen patients reported with VAHS and a documented viral infection, thirteen (87Yo) were on therapy with azathioprine and prednisone, mostly (12/13) because of a previous renal transplantation (3). It is important to distinguish
Acta Paediatr Scand 80
Incidence in Sweden and clinical features of FHL 429
this entity (with prior immunosuppressive therapy) from FHL, as cytotoxic therapy may be contraindicated in VAHS (3) whereas it is indicated in FHL (4-6). Therefore, we have in this study excluded patients with immunosuppressive therapy prior to onset of their hemophagocytic disorder. On the other hand, a prior or concomitant viral infection-frequently seen in infants-has not been used as a sole exclusion criterion. PATIENTS AND METHODS Patients. The population base consisted of all children under 15 years of age in the Swedish parish records. The period studied was January 1,1971 to December 3 1,1986, with an average child population of 1 627 000/year (7). During this 16-year period 1603 696 children were born alive in Sweden (8). Inclusion criteria. There is still no generally accepted definition of FHL. It was not possible, in this retrospective study, to use the criteria we would propose for a prospective study, as most children are dead. Instead, we evolved a combination of clinical, laboratory and histopathological criteria, with six major and five minor criteria (Table 1). Children who fulfilled the inclusion criteria during the study period and before their 15th birthday were included, whether a proven virus infection was associated or not. Methods. Two independent methods to find children with FHL were used. First, all departments of pediatrics, pathology, and infectious diseases in Sweden were contacted, by a letter which included a comprehensive description of the disease, and requested to report present or previous patients with FHL or with disorders resembling FHL. Reminders by letter and telephone gave a reply rate of 100%. Copies of files, including reports of histological examinations concerning reported cases (n=35) were kindly supplied by each department. Secondly, the causes of death according to death certificates of all deceased children in Sweden during the study period were investigated; altogether 19 542 children (8). In deceased Swedish children, autopsies have to be performed if the cause of death is unknown and they are usually also performed if the cause of death is uncertain. The immediate and contributory
Table 1. Inclusion criteria for retrospective diagnosis of FHLa Major inclusion criteria 1. Prolonged intermittent fever, not caused by bacteria, spirochetes, rickettsiae, protozoa, or helminths 2. Splenomegaly, with or without hepatomegaly 3. Cytopenia in peripheral blood without a hypoplastic bone marrow (hemoglobin 2 mmol/l) 2. Hypofibrinogenemia (
Incidence in Sweden and Clinical Features of Familial Hemophagocytic Lym phohist iocytosis JAN-INGE HENTER,’ GORAN ELINDER,* OLLE SODER’ and AKE OST3 From the Department of Paediatrics. ‘St. Goran ’slKarolinskaand 2Sachs’ Children’s Hospitals and the 3Department of Pathology, Karolinska Hospital, Karolinska Institute, Stockholm, Sweden ABSTRACT. Henter, J.-I., Elinder, G., Siider, 0. and Ost, A. (Department of Paediatrics, St. Giiran’slKarolinska and Sachs’ Children’s Hospitals and Department of Pathology, Karolinska Hospital; Karolinska Institute, Stockholm, Sweden). Incidence in Sweden and clinical features of familial hemophagocytic lymphohistiocytosis. Acta Paediatr Scand 80: 428, 1991. We retrospectively studied the incidence of familial hemophagocytic lymphohistiocytosis (FHL) in children during the 16-year period 1971-86. First, all departments of pediatrics, pathology, and infectious diseases were enquired for children with FHL or disorders resembling FHL. Secondly, the causes of death of all children who died during the study period in Sweden (n=19542) were also investigated. Files and histological specimens were further studied in selected children. By using a set of inclusionlexclusion criteria, we found 32 children with FHL. The incidence was 1.211000000 children per year. One child per 50000 live borns developed FHL during this period. The sex ratio was close to 1 :1. Prominent early clinical signs were fever (91 %), splenomegaly(84 %), hepatomegaly (90 %), rash (43 %), and lymph node enlargement (42 %). Neurological symptoms, which developed in 47 %, could totally dominate the clinical picture and develop prior to other symptoms and signs. Common laboratory findings were pancytopenia, hypertriglyceridemia, hypofibrinogenemia, elevated serum transaminases, hyperbilirubinemia, hyponatremia, hypoalbuminemia, and a moderate spinal fluid pleocytosis. Chest X-ray often revealed mostly discrete pulmonary infiltrates. FHL is an underdiagnosed disease and in only 11/32 children was diagnosis made during their lifetime. It is important to be aware of the disorder as potential therapy now exists. Key words:familid hemophagocytic lymphohistiocytosis (FHL), incidcnce.
Familial hemophagocyticlymphohistiocytosis (FHL) is a rapidly fatal disease clinically characterized by intermittent fever, hepatosplenomegaly, and pancytopenia ( 1). Histological examination shows an accumulation of lymphocytes and macrophages with hemophagocytosis in reticuloendothelial organs. FHL has been considered to be a very rare disorder, but no previous study of the incidence has been reported in the literature. We have, in this retrospective study of a well-defined population, attempted to determine the incidence of FHL in children ‘In Sweden. Clinical features and laboratory findings in the affected children are also presented. In the recently proposed classification of histiocyte syndromes by the Histiocyte Society, FHL is included among “histiocytoses of mononuclear phagocytes other than Lagerhans’ cells” separate from malignant histiocytic diseases and Langerham-cell histiocytosis (LCH), the latter embracing disorders previously included in the entity Histiocytosis X (2). A virus-associated hemophagocytic syndrome (VAHS), which also affects histiocytes other than Langerhans’ cells, was described in 1979 as a potentially self-limited disease mainly seen in immunosuppressed patients (3). Out of the fifteen patients reported with VAHS and a documented viral infection, thirteen (87Yo) were on therapy with azathioprine and prednisone, mostly (12/13) because of a previous renal transplantation (3). It is important to distinguish
Acta Paediatr Scand 80
Incidence in Sweden and clinical features of FHL 429
this entity (with prior immunosuppressive therapy) from FHL, as cytotoxic therapy may be contraindicated in VAHS (3) whereas it is indicated in FHL (4-6). Therefore, we have in this study excluded patients with immunosuppressive therapy prior to onset of their hemophagocytic disorder. On the other hand, a prior or concomitant viral infection-frequently seen in infants-has not been used as a sole exclusion criterion. PATIENTS AND METHODS Patients. The population base consisted of all children under 15 years of age in the Swedish parish records. The period studied was January 1,1971 to December 3 1,1986, with an average child population of 1 627 000/year (7). During this 16-year period 1603 696 children were born alive in Sweden (8). Inclusion criteria. There is still no generally accepted definition of FHL. It was not possible, in this retrospective study, to use the criteria we would propose for a prospective study, as most children are dead. Instead, we evolved a combination of clinical, laboratory and histopathological criteria, with six major and five minor criteria (Table 1). Children who fulfilled the inclusion criteria during the study period and before their 15th birthday were included, whether a proven virus infection was associated or not. Methods. Two independent methods to find children with FHL were used. First, all departments of pediatrics, pathology, and infectious diseases in Sweden were contacted, by a letter which included a comprehensive description of the disease, and requested to report present or previous patients with FHL or with disorders resembling FHL. Reminders by letter and telephone gave a reply rate of 100%. Copies of files, including reports of histological examinations concerning reported cases (n=35) were kindly supplied by each department. Secondly, the causes of death according to death certificates of all deceased children in Sweden during the study period were investigated; altogether 19 542 children (8). In deceased Swedish children, autopsies have to be performed if the cause of death is unknown and they are usually also performed if the cause of death is uncertain. The immediate and contributory
Table 1. Inclusion criteria for retrospective diagnosis of FHLa Major inclusion criteria 1. Prolonged intermittent fever, not caused by bacteria, spirochetes, rickettsiae, protozoa, or helminths 2. Splenomegaly, with or without hepatomegaly 3. Cytopenia in peripheral blood without a hypoplastic bone marrow (hemoglobin 2 mmol/l) 2. Hypofibrinogenemia (
