Bronchiolitis rates increase, treatment has changed little over the last few decades

February 2006

If you live in the north, it is about that time of year for bronchiolitis. If you live in the southeast, you are already well into the season.

It has been estimated that as many as 3% of infants are hospitalized each year with bronchiolitis, and the incidence of this disease has almost doubled in the past few decades. The increase has been most marked in those younger than 6 months of age. It is also estimated that there are, on average, 123,471 bronchiolitis hospitalizations annually among children younger than 1 year and 154,365 among children younger than 5 years; 50,000 to 80,000 were due to respiratory syncytial virus (RSV) (JAMA. 1999;282:1440-1446).

If one adds the hospitalization rates for this and gastroenteritis, it would appear that almost one in 25 infants born each year are hospitalized for these two conditions alone, assuming no one infant is unfortunate enough to be hospitalized for both. I find these figures incredible. Part of the increase in bronchiolitis probably is due to the survival of more and more low- and extremely low-birth weight infants who seem to be particularly prone to bronchiolitis. Low-birth weights increased from 6.7% in 1984 to 8.1% in 2004, a 21% increase (Pediatrics. 2006;117:168-183). However, this increase could not in itself account for the rise in cases of bronchiolitis.

When I was a resident assigned to newborn service before the neonatal ICU with lots of time on my hands, I decided to review the pathology reports of children dying of bronchiolitis. In spite of the fact that the wards were filled with these children every winter, very few made it to the pathology department despite the fact that the hospital had a 90% autopsy rate. In fact, most writers referring to the pathology of bronchiolitis still cite the descriptions of decades ago. Indeed, examination of material collected in more recent years may be complicated by therapeutic interventions. Although classical pathologic descriptions do not change, we now have sophisticated tools to elucidate the pathogenesis of this disease. In recent years, there has been an increasing interest in studying these (Arch Ped Adol Med.2004;158:111-112).

Unfortunately, not only has the basic pathology changed little, the same can be said of our treatment of this condition. In the 1950s, during my residency, we designed a trial to determine whether steroids would be beneficial in babies with bronchiolitis. This involved the “high-tech” use of cameras to evaluate breathing in effort not to introduce an artifact and disturb the infant.

The tapes could have been evaluated blindly to maintain objectivity. We are all guilty of bias when we use a treatment. Since this time, there have been countless studies — some of which have been reported — to evaluate several modalities for treatment. Most of these have been used in the management of asthma, which is a poor model on which to base interventions. These include steroids, ß agonists, epinephrine given by various routes and, most recently, helium. The endpoints have varied from improvement in physiologic parameters and more relevant ones (eg, earlier discharge or need for intubations or intensive care).

A number of meta-analyses (Arch Ped Adol Med. 2004;158:127-137) have come to essentially the same conclusion for most of these treatments. That is, there may be a transitory improvement in physiologic parameters, but in achieving the critical objectives, there has been little perceptible difference. (Brunell rule #46.32 — If it takes half a century to determine whether an intervention is effective, it probably is not.) Many try something, usually epinephrine, in the hope that it may “help a subset.”

My question is, will it harm a subset? Meta-analysis has shown that some patients may be harmed by these interventions (Arch Pediatr Adolesc Med. 2004;158:127-137). Ribavirin, for which we had hopes a few seasons ago does have potential toxicity. One of the caveats applying the lessons from these trials is that most have been done in hospitalized severely ill patients rather than patients likely to be seen in the office. Clearly, we need a radically different approach to treatment.

The greatest advancement has been the use of passive immunization to prevent RSV-caused bronchiolitis in high-risk infants. Palivizumab (Synagis, MedImmune) is the standard of care for certain infants. This preparation is a humanized monoclonal antibody that binds to the F antigen of RSV. It has shown to be effective in reducing bronchiolitis hospitalizations in those infected with RSV by 55% as well as decreasing hospital stays and the need for supplemental oxygen (Pediatrics. 1998; 102:531-537). It will probably be more effective during the RSV season than other times of the year when parainfluenza and other viruses are responsible for a significant number of cases of bronchiolitis. It is not effective against human metapneumoviruses, which cause about 10% of cases of bronchiolitis; their season usually overlaps with the RSV season. The recommendations for palivizumab use have been updated to include patients with significant cardiac disease (Pediatrics. 2003;112:1442-1446). Palivizumab is very expensive and labile, so infants should be cohorted to use the material as efficiently as possible once the vial has been opened. Cost is a major consideration in infants born between 32 and 35 weeks gestation. Additional factors, such as impairment of ventilation and exposure to tobacco smoke, should be considered when weighing the decision for use in this group. It is necessary to emphasize the need to eliminate the exposure to tobacco smoke. Spread of RSV occurs readily and attention to hand washing is applicable in every situation.

Since RSV first was isolated many decades ago, there have been multiple attempts to prevent infection by active immunization. The initial attempt — using a killed vaccine — was an utter failure as RSV vaccine recipients had more severe disease following exposure to RSV than those who had been given parainfluenza vaccines; the attack rates were similar. (Am J Epidemiol. 1969;89:422-434). This experience was, needless to say, very discouraging and probably a retarded progress for some time. However, there have been vaccines since that time (eg, subunit F and G components and intranasal live vaccines that have shown some promise). As the most severe disease occurs early in infancy, immunization must occur as soon after birth as possible. Parenterally administered live vaccines are inactivated by maternal antibodies, which are highest postpartum. The intranasal candidates have shown promise, but the early generations have caused nasal stuffiness in some young infants, which is a major problem in this age group as they are obligatory nose breathers.

There are two areas that I think bear looking into. We are continuing to learn more about the innate immune responses (ie, the immediate response to infection that is manifested mainly by inflammation). We need to identify the specific factors involved in the pathogenesis of bronchiolitis and how they might be modified by therapy. Steroids are not doing it. In addition, we need to study the differences in responses of those infants who are infected with RSV and contract bronchiolitis and those who just get an upper respiratory infection or a fever when infected with the same virus. By the second year of life, most infants already have been infected, yet only a fraction presents with significant illness. Clearly, some of the differences in clinical expression are due to size of the airways. That, however, does not explain the occurrence of bronchiolitis in full-term or older infants.

RSV does not pick out the bronchioles; it involves the entire respiratory system, hence the high rate of otitis in children with RSV bronchiolitis. Finally, we should not forget that the nasal passages are part of the affected respiratory tract and, even when unaffected, produce a significant portion of airway resistance. I am impressed how many writings on this subject totally ignore the nose and proceed directly to the lungs.

In addition to inflammatory cells and debris, which occlude the airways, diminish their size and increase resistance to air flow, turbulence also is increased. Nasal exudates, which are present in almost all infants with bronchiolitis, are particularly important in obligatory nose breathers with a small nasal airway diameter. One of my “low-tech” approaches is to clean the nose of an infant who has bronchiolitis with an infant nasal aspirator, not an ear aspirator, which does not enter the nasal cavity and causes more inflammation and slides right past much of the obstructing secretions. I have been impressed how frequently, when I have returned a few minutes after this procedure, how the respiratory rate has decreased. You might want to try it on infants whom you are uncertain as to whether hospital admission is indicated. It is cheap, innocuous and likely to work. I have included a table that appeared in a British publication on criteria for office guidance on the management of infants with bronchiolitis (Arch Dis Child. 2005;90:ep81-ep86).