The invitation to comment on a paper that reports changes which occurred during a period that coincided, within a year or two, with my career as a consultant paediatrician presents me with an irresistible temptation to reminisce. At medical school in the late 1950s I learned that pneumonia, one time captain of the men of death, had responded dramatically first to sulphonamides and then to penicillin, but that some deaths were still “inevitable”—a phrase much used at that time to excuse our inability to manage conditions that we did not fully understand. Asthma was common in children but was considered to be an unusual cause of death, although it was responsible for the first death I encountered as a paediatric senior house officer. In the minds of my teachers asthma was readily distinguished from bronchitis and, to this day, I can replicate a table listing the differences between asthma and wheezy bronchitis, most of which has—to my surprise—turned out to be accurate.¹ In those far off days most children with cystic fibrosis (CF) died in the pre-school years; Pseudomonas species were not the problem they are now and most children died while still colonised with Staphylococcus aureus or Haemophilus influenzae. Bronchiolitis was considered to be rather an esoteric diagnosis for which there was no diagnostic test – it would be a few years before its association with the respiratory syncytial virus (RSV) was appreciated.² It was thought safer to diagnose pneumonia if crackles were prominent and to treat the child with antibiotics, while those with predominant wheeze were labelled “bronchitis” and given ephedrine.

In the mid 1960s I extracted data from the Registrar General’s annual reports when, as a junior, I presented a fatal case of staphylococcal bronchopneumonia to a clinicopathological conference. I was surprised to find not only that there was still considerable childhood mortality from pneumonia but also that, following the rapid decline already mentioned, it had been more or less stable for the previous decade. In my presentation I attributed this to the emergence of penicillin resistant S aureus, an organism that had become ubiquitous despite the widespread presence of gimlet-eyed matrons who, some politicians would have us believe, have the power to stop the more recently emerged methicillin resistant S aureus in its tracks.

By 1968, the first year of the study by Panickar et al³ reported in this issue of Thorax, major changes were afoot. Coincident with—but not necessarily due to—the development of new antibiotics, pneumonia mortality had begun to fall, and when I prepared a figure to illustrate a chapter in a textbook⁴ it was clear that mortality from pneumonia was back on its downward track, on which it has remained despite the emergence of respiratory pathogens of vastly greater virulence and antibiotic resistance than anything we had to deal with 40 years ago. Some children with CF were now entering secondary school, although their survival there was usually brief. Overnight incarceration in mist tents was part of the burden borne by patients with CF; in 1968 I was trying to raise money to purchase tents and was still doing so a few years later when Archie Norman demonstrated their inefficacy.⁵ Antibiotic treatment had improved but nutritional management was woefully inadequate; looking back, it is hard to believe that the era of the Allan diet⁶ had not even arrived, far less departed. However, by 1968 we were becoming more secure in the diagnosis of bronchiolitis, although we had to rely on rising antibody titres to demonstrate the presence of RSV; the immunofluorescent technique for the rapid diagnosis of RSV infection was described in 1968⁷ but would not be in widespread use for several years. In the absence of an organised system of paediatric intensive care units, the availability of artificial ventilation depended on the interest and enthusiasm of local anaesthetists and paediatricians. Asthma was also a continuing problem; salbutamol was announced that year⁸ but was not yet available for clinical use, and inhaled corticosteroids were a few years over the horizon.⁹ The study by Panickar et al³ therefore covers a period during which there were major developments in what we now regard as basic aspects of the management of respiratory tract diseases.

RESPIRATORY DISEASES STUDIED

Pneumonia

The improving mortality from pneumonia reflects many medical advances other than new antibiotics. Panickar et al³ emphasise the importance of immunisations, to which I would add the developments in the organisation and delivery of paediatric intensive care that have occurred in recent years. Nevertheless, there are no grounds for complacency. Children are still dying from pneumonia, and although we have useful guidelines for the management of community acquired pneumonia¹⁰—albeit derived from a somewhat scanty evidence base—the management and, perhaps more importantly, the prevention of nosocomial pneumonia remains problematical.¹¹

Asthma

However, pneumonia is not the only important cause of respiratory mortality in children. Although fewer than 5% of asthma deaths occur in childhood,¹² asthma is an eminently treatable condition and our objective should be to reduce this figure to zero. Most asthmatic children who reach hospital alive will survive, the majority of deaths occurring outside hospital.¹³ Management in the community is therefore of prime importance. Although there is controversy about the role of written management plans in improving asthma care,¹⁴ there is no doubt that comprehensive asthma management programmes are effective in reducing asthma morbidity and hospital admissions.¹⁵ The problem is not that we don’t know what to do, but that the very patients who need to be included in such programmes—and in whom much of the childhood asthma morbidity and mortality occurs—are hard to reach because a combination of socioeconomic deprivation,¹⁶ family problems, and psychological factors¹⁷ leads to failure to recognise the severity of an attack and failure to seek medical help in time.¹⁸

Cystic fibrosis

It is already well known that the mortality from CF has fallen to the extent that it is no longer an important cause of childhood death.¹⁹ This improvement has, however, been bought at a considerable price in terms of parental stress²⁰ and family disruption²¹—stress that was described at a recent meeting as “the daily grind.”²² Moreover, socioeconomic deprivation has a substantial adverse impact on the outcome for patients with CF,²³ as evidenced for instance by the poorer clinical progress made by CF children of single parents.²⁴

Panickar et al found that, for most respiratory conditions, mortality is still higher in boys than in girls. However, the reverse is true in CF. Girls, who are so often model patients during childhood and comply conscientiously with the rigours of treatment, commonly deteriorate during adolescence, creating a gender gap in mortality that was demonstrated clearly in data from the UK CF Survey.²⁵ An examination of the large database maintained by the US CF Foundation showed that, after adjustment for other risk factors, female sex was a significant risk factor for death in patients <21 years of age, with a relative risk compared to males of 1.6.²⁶ Although the added disadvantage suffered by adolescent girls with CF is widely recognised, it has never been adequately explained. Studies have demonstrated differences between the sexes in energy intake²⁷ and resting energy expenditure,²⁸ in everyday activity levels,²⁹ and in some psychological traits,^30,31 but none of these offers a convincing explanation for the large excess in female mortality.

In addition to applying the medical advances that must sooner or later flow from our increased understanding of the molecular biology of CF, future challenges include narrowing the gender gap in mortality, improved delivery of care to economically disadvantaged families, and reduction of the psychosocial impact of the CF child on other family members.

Acute viral bronchiolitis

The last major cause of paediatric respiratory death discussed by Panickar et al is acute viral bronchiolitis. The antiviral drug ribavirin, given as a small particle aerosol, has failed to live up to early expectations³² and is unlikely to have had much influence on mortality. The use of palivizumab, an effective monoclonal antibody directed against RSV, is reserved for high risk cases and the extent of its use varies widely. Its introduction was too recent to have affected mortality in all but the most recent years of the study, and the dramatic falls in mortality must therefore be attributed to improvements in general paediatric care together with improved access to paediatric intensive care.

IMPLICATIONS OF STUDY FINDINGS

So what are the lessons to be learned from this report on childhood mortality? Clearly, paediatric respirologists are getting results at least as satisfactory as those of their colleagues in other paediatric specialties. Deaths are at an all-time low, and for this we can allow ourselves a moment’s satisfaction. However, much remains to be done. One obvious need is for well designed clinical trials in almost every aspect of paediatric respirology, to allow us to fine-hone our management and to discard useless treatments. In particular, although the excellent results obtained in CF in the absence of a satisfactory evidence base testify to the value of clinical experience and common sense in directing patient care, our use of ever more expensive and complex treatments must in future be more rationally based.

However, one challenge that will tax our ingenuity and for which no obvious solution is in sight is the delivery of health education and medical care to families from socially disadvantaged backgrounds. Many governments have pledged to reduce social inequalities in health, but the gap persists obstinately.³³ These differences are seen throughout the world, between as well as within nations,³⁴ and a solution to the problem has so far proved elusive. As we have seen, paediatric respirology is by no means immune to these problems; there will be little benefit from screening for CF or improving asthma management plans if our attempts to deliver appropriate care are frustrated by socioeconomic and psychological barriers. Paediatricians are accustomed to working in multidisciplinary teams, but perhaps it is time to reconsider the composition of such teams which, in the past, have tended to be disease-specific. Should we be making better use of the more generic facilities of the school health service,³⁵ the social work department,³⁶ or the clinical psychology department? Whatever the ultimate solution, it is clear that current practice is inadequate and that radical new ideas are needed.