The resurgence of tuberculosis in children is greatest in countries with high rates of HIV infection. The diagnosis of tuberculosis is often difficult and resources may be wasted in treating suspected cases who do not have the disease. Now researchers in South Africa (

) have assessed a rapid blood test. The ELISPOT test is an enzyme linked immunospot assay that detects T cells specific for Mycobacterium tuberculosis antigens not present in BCG or most environmental mycobacteria. Among 262 children with suspected tuberculosis 133 were categorised as having confirmed or highly probable tuberculosis. Of these 133, 110 had a positive ELISPOT test (test sensitivity 83%) and 73 (of 116 tested) had a positive Mantoux test (test sensitivity 63%). Among young children (<3 years old), children also infected with HIV, and children with malnutrition the sensitivity of the ELISPOT test was 85%, 73%, and 78% respectively whereas the Mantoux test was only 51%, 36%, and 44% sensitive. Among all children with tuberculosis combined ELISPOT and Mantoux testing was 91% sensitive. Among 13 children categorised as “not tuberculous” four had a positive ELISPOT test. The ELISPOT test is more sensitive than the Mantoux test for the diagnosis of tuberculosis in children. Using both tests gives a sensitivity of over 90%.

A report from Guinea Bissau in 2000 caused concern because it suggested that diphtheria, tetanus, and pertussis (DTP) vaccination of infants might increase overall mortality. Now a study in rural Bangladesh (

; see also Comment, ibid: 2156–7) has shown a 24% reduction in 6 weeks to 9 months mortality associated with DTP vaccination. Other factors associated with reduced infant mortality were maternal education, maternal age, and birth order, and the commentator stresses the health benefits likely to accrue from better education of women in developing countries. The study also confirms that the effects of BCG and measles vaccination on infant mortality may exceed those expected simply from prevention of tuberculosis and measles.

Sudden unexplained death in epilepsy (SUDEP) has been much discussed recently and its causes are debated. Now research in London involving 20 adults with treatment refractory focal epilepsy has shown that asystole during seizures is a probable cause (

; see also Comment, ibid: 2157–8). Each patient had long-term ECG monitoring (average duration 18 months) by means of an implanted loop recorder. Four patients needed permanent insertion of a cardiac pacemaker because of bradycardia or asystole during seizures; three of them had had potentially fatal asystole. Whether the same mechanism occurs in children needs to be established.

The use of supplemental oxygen for children at home has increased but there is a lack of consensus about issues such as when to start home oxygen therapy, target oxygen saturation, and weaning off oxygen. The requirements of children differ from those of adults and the subject has been reviewed (

). The review discusses the ways in which children’s requirements differ from those of adults, the paediatric conditions that lead to a requirement for home oxygen, assessment for home oxygen provision for infants and for older children, criteria for hospital discharge, the equipment that needs to be provided, follow up arrangements, weaning off oxygen, ambulatory oxygen therapy, and the use of oxygen at school.

The Care of Next Infant (CONI) programme of surveillance for infants after a sudden unexpected and unexplained infant death in the family is available in more than 90% of health districts in England, Wales, and Northern Ireland. Between 1988 and the end of 1999 there were 57 deaths among 6373 infants from 5229 families in the programme (

; see also Comment, ibid: 3–4). Forty-eight of these deaths, in 46 families,(the second unexpected infant death in 44 families and the second and third in two) were sudden and unexpected. Of the 46 first unexpected infant deaths in families on the programme 40 were considered natural and six unnatural (homicide), giving a natural to unnatural ratio of 6.7:1.

Avian influenza, when it affects people, is thought of as a purely respiratory infection. Of 45 reported cases of influenza A (H5N1) in 2004 all had respiratory illnesses; 22 died. Now a report from southern Vietnam (

) has suggested that the virus may infect other organs and faecal transmission might be possible. A 9 year old girl and her 4 year old brother presented within 11 days of each other with fever, severe diarrhoea, and encephalopathic symptoms. The girl had no evidence of respiratory disease but the boy developed a cough and rapid breathing within a few days, with pulmonary infiltrates on chest x ray. No virological investigations were done on the girl but influenza A (H5N1) virus was cultured from the boys’ cerebrospinal fluid, stool, throat swab, and serum. Both children died within a week of symptom onset. There was no evidence of child-to-child transmission.

Among a birth cohort of 277 children in Seattle (

; see also perspective article, ibid: 753–5) the cumulative incidence of human herpesvirus 6 (HHV-6) infection, shown by PCR testing of saliva, was 40% by 1 year and 77% by 2 years. Girls and children with older siblings were at greater risk. Almost all children (93%) developed symptoms with infection and the main symptoms were fever, fussiness, diarrhoea, and rash. Twenty-three per cent of infected children developed roseola but none had febrile convulsions.

Parents are more likely to be admitted to hospital with psychiatric illness after the death of a child. National registry data in Denmark (

) have shown that, compared with parents who had not lost a child, bereaved mothers were almost 80% more likely, and bereaved fathers almost 40% more likely, to be admitted with a psychiatric illness over a period of 5 or more years after the child’s death. During the first year after the child’s death the increases in risk of admission for affective disorders were almost sevenfold (mothers) and sixfold (fathers), but there were also significant increases in risk of admission for schizophrenia in both mothers and fathers and for substance abuse in mothers. Among parents who had lost more than one child the risk of hospital admission for psychiatric illness was more than tripled among mothers and more than doubled among fathers. The risks were less for parents with larger families.