Behind The Medical Headlines

DEATHS FROM ACUTE RESPIRATORY DISEASE IN CHILDREN - INFLUENZA

• Dr TG Marshall, Consultant, Royal Hospital for Sick Children, Edinburgh, Scotland

Summary

Media reports from the US of a number of childhood deaths from influenza in 2003 caused public concern. It was clearly perceived by the public that children in the West should no longer be dying from influenza. How concerned should we be about childhood influenza and does vaccination have a role? Dr Tom Marshall reviews the evidence.

Declaration of interests: No conflict of interests declared

Media reports from the US of deaths in children during the Influenza A outbreak in winter 2003, followed by reports of a few individual deaths in the UK, have caused public alarm. The public perception in the West is that children should not die from common acute respiratory illnesses (ARIs), and these reports have brought such deaths into sharp relief. What should we learn from these recent media reports?

Influenza deaths

Influenza is not generally a reportable condition and it is therefore difficult to identify other than major changes in its incidence, let alone changes in childhood deaths due to the disease. It has been estimated that 92 influenza-associated deaths occurred annually in children under five in the US from 1990-99.¹ The Communicable Diseases Center (CDC) in Atlanta reported 93 influenza-associated deaths under the age of 18 years over approximately a three-month period from October 2003 to January 2004, covering the 2003 influenza outbreak.² Fifty-five of the deaths (60%) occurred in children under four years, and as influenza is a seasonal disease, it is not obvious that this represents an increase in the frequency of death from this disease.

Reports from the CDC give interesting information on the young people who have died of influenzal illnesses.^{2 3} First, the youngest children seem most at risk, as about two-thirds of deaths occurred in those aged less than five years. Second, the progress of fatal disease is rapid, as illustrated by case histories from the Michigan Department of Community Health and as suggested by the finding that a half of deaths occurred at home, in transit to hospital, or in emergency rooms. Third, associated chronic diseases occurred in 35 of 93 cases (38%) and were considered a risk factor for death. Particularly important were chronic pulmonary disease, especially asthma, cardiac disease, endocrine disease, especially diabetes mellitus, renal disease, immunocompromised states, and neurological disease or developmental delay. Finally, bacterial pneumonia and non-respiratory complications (myocarditis, rhabdomyolosis, encephalitis and seizures) requiring high-level hospital care were common.

A further point of interest is that in only one of 45 deaths where the vaccination status was known was there evidence of adequate vaccination. Currently, vaccination is encouraged only for healthy children aged six months to two years.⁴ Future studies are needed to see whether this should be extended.

Deaths from other ARIs

Acute bronchiolitis is the commonest cause of hospital admission in the first year of life. This illness is limited to young children, and some studies indicate that 2-3% of children in this age group will be admitted to hospital. Acute bronchiolitis, caused predominantly by respiratory syncitial virus, is a major burden on health resources for children over the winter months. This is a disease for which there is no cheap effective immunisation available, and there has been no downward trend in admissions over the past 10 years. However, deaths due to bronchiolitis are uncommon and are often associated with underlying chronic neonatal lung disease or congenital abnormality.

Acute asthma admissions have shown a steady downward trend over the past ten years and asthma deaths in children are now uncommon. In 2001 there were 27 deaths under the age of 14 years due to asthma in the UK, with most occurring in teenagers.

Pneumonia has reduced in incidence and fatality in the Western world. The annual incidence in Europe and North America in children under five years is now around 35-40 cases per 100,000. The mortality rate in the 0-14 years age range in the UK has fallen from 30 to 15 per million cases over the last 20 years, and mortality in the US fell by 97% during the 58 years to 1996 when 800 children died.

Mortality in developing countries

Regrettably, falling mortality in Europe and North America contrasts starkly with mortality worldwide, where pneumonia kills 3-4 million children each year. In developing countries, pneumonia affects 10-20% of children under five years of age each year. The World Health Organization estimates that in developing countries one in three deaths are due to, or associated with, ARIs including pneumonia. Acute respiratory illnesses with or without pneumonia are common complications of measles and pertussis and are often associated with severe malnutrition. There has been some evidence of a fall in the number of deaths due to pneumonia over the last decade in developing countries due to the more widespread use of antibiotics, but in Africa the increasing prevalence of HIV infection has led to an increase in the incidence of bacterial pneumonia.

Recognition of severe childhood respiratory illness

The signs and symptoms of the common causes of severe respiratory illness are well known and are important in view of the speed with which ARIs can progress. The main signs are shown in Figure 1, and in addition, lethargy, convulsions, or coma indicate very severe pneumonia. Significant upper airway obstruction, as in viral croup, is suggested by stridor at rest and biphasic stridor implies particularly severe airway obstruction.

The new BTS/SIGN guidelines for severe asthma in children are shown in Figure 2.⁵

Referral to hospital

The essential step for health professionals is to make a careful physical examination of children with ARIs and to document the important physical signs and physiological data (Figure 1 and Figure 2). This information should enable a determination to be made as to whether a child is unwell enough to need urgent referral to hospital or requires repeated examinations to determine whether improvement or worsening is occurring. Rapid progression of ARIs means that repeat assessments should be at short intervals.

It is important for the current illness to be placed in the context of the recent medical history and any co-morbidity taken into account. For example, the child with moderate asthma who has had hospital admissions and received oral corticosteroids for acute severe asthma in the past year should be referred to hospital at an earlier stage than a child experiencing moderate, viral-induced wheezing for the first time. The child with chronic neonatal lung disease on home oxygen should always be referred for early assessment and monitoring on developing an acute respiratory infection such as bronchiolitis.

The recommendation in the new asthma BTS/SIGN guidelines that oxygen saturation monitoring should be available routinely to those assessing acute asthma in the community has not so far been widely taken up. In a recent survey in Edinburgh, only one-third of general practitioner (GP) ‘out-of-hours cooperatives’ had access to an oxygen saturation monitor. Fewer than 10% of GPs had a saturation monitor available in their practices, and fewer than 20% of referral letters for children with acute illness made mention of an oxygen saturation recording. Oxygen saturation assessment should be a useful additional tool that will enable doctors in primary care to make a quick and more accurate assessment of the severity of acute respiratory disease in children in the community in the future.

Prevention

The burden of respiratory disease worldwide would be reduced dramatically by improvement in immunisation rates. Immunisation against pertussis and measles in developing countries could reduce the incidence of pneumonia by 15-20%. In the UK, HiB immunisation has dramatically reduced the incidence of invasive haemophilus disease. The introduction of routine pneumococcal immunisation in the future may further reduce the burden of illness due to acute pneumonia. Current recommendations for children regarding pneumococcal immunisation and influenza vaccination⁴ are unclear and not sufficiently specific.

Conclusion

Press reports of deaths from influenza A in children have alarmed the public but have also highlighted the importance of all doctors involved in the acute care of children having a clear understanding of the signs and symptoms of acute severe respiratory infection in children. It is important for the future that clearer and more specific recommendations for influenza and pneumococcal immunisation in children are produced. In addition, it is important to recognise that the prevalence of chronic respiratory illness in children exceeds that of all other disease conditions combined and causes 8% of all childhood deaths in the UK.

References

1. Thompson W, Shay DK, Weintraub E et al. Mortality associated with influenza and respiratory syncitial virus in the United States. JAMA 2003; 289:179-86.
2. CDC 2004 Update: Influenza-associated deaths reported among children aged <18 years - United States, 2003-04 influenza season. MMWR 2004; 52:1286-8.
3. CDC 2003 Severe morbidity and mortality associated with influenza in children and young adults - Michigan, 2003. MMWR 2003; 52:837-40.
4. CDC 2003 Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 2003; 52:(No. RR-8).
5. BTS/SIGN British Guideline on the Management of Asthma. Thorax 2003; 58:Supplement 1.