Use of Respiratory Syncytial Virus Testing Could Safely Eliminate Many Sepsis Evaluations—Reply

We read with interest the recent article on diagnostic testing for serious bacterial infections in infants with bronchiolitis.¹ The authors suggest that previously healthy, well-appearing infants with wheezing and other signs of an upper respiratory tract infection may not need laboratory testing for bacterial infections even in the presence of fever. We describe an infant with pneumococcal bacteremia who presented with the symptoms and signs of bronchiolitis. If the aforementioned recommendations were followed, we would have failed to make an early diagnosis in this infant.

A 7-month-old boy was seen in our office in the month of November for nasal congestion, progressively worsening cough, and a decreased appetite of 2 days' duration. A temperature of 37.8°C had been noted. On physical examination, he was afebrile and in mild respiratory distress with bilateral expiratory wheezes. Nebulized albuterol was administered with improvement in his wheezing. Acute bronchiolitis was diagnosed, and oral albuterol was prescribed as outpatient therapy. Four hours later, the patient was admitted to the hospital with increasing respiratory distress.

Physical examination at admission revealed an alert, well-appearing infant. His vital signs were temperature 36.7°C, respiratory rate 80/min, heart rate 140/min, and pulse oximetry 91% in room air. Both tympanic membranes were inflamed. Mild to moderate subcostal and intercostal retractions were evident with bilateral expiratory wheezes.

Initial laboratory tests included a complete blood cell count, which showed a hemoglobin level of 116 g/L, hematocrit of 0.34, white blood cell count of 21.0 × 10⁹/L, and platelet count of 795 ×10⁹/L. The differential cell count showed 0.21 neutrophils, 0.25 band cells, 0.46 lymphocytes, 0.07 monocytes, and 0.01 eosinophils. Results of an enzyme-linked immunosorbent assay for respiratory syncytial virus (RSV) performed on nasopharyngeal secretions were negative. The chest roentgenogram showed bilateral upper lobe infiltrates.

The infant was initially treated with nebulized albuterol and oxygen. Two hours after admission, receipt of the white blood cell count results prompted the immediate commencememt of a daily dose of ceftriaxone at 75 mg/kg. Within 24 hours of admission the blood culture was positive for Streptococcus pneumoniae, with intermediate resistance to penicillin (minimal inhibitory concentration, 1.0 µg/mL) but sensitivity to ceftriaxone. Intravenous ceftriaxone was given for 7 days. He was asymptomatic at hospital discharge.

As referenced in the article by Liebelt et al,¹ several recent publications in the medical literature have decried the need for further investigations in infants with bronchiolitis. A review article² stated that bacterial infection is almost never simultaneously present in bronchiolitis. Our patient's illness, however, adds a timely caveat to these recent recommendations. We feel that the wheezing infant sick enough to be hospitalized should have a complete blood cell count, blood culture, and chest radiograph done on admission, even if the infant does not have a toxic appearance. Leukocytosis, especially with a left shift in the differential cell count, should raise the suspicion of a bacterial infection. The results of the RSV antigen test should not influence therapeutic decision making since the test is known to be only 70% to 90% sensitive.³ Antibiotic therapy should be promptly instituted until the clinical course is further evaluated and the results of cultures are obtained.

Coinfection with S pneumoniae has been previously reported in very few infants in the United States with documented RSV infections. Tristram et al,⁴ Timmons et al,⁵ and Antonow et al⁶ described a few patients with such infections. However, these infants were all very seriously ill clinically.⁴ Our patient did not have a toxic appearance.

An issue raised in the management of bronchiolitis is the indiscriminate use of antibiotics in many infants with an obvious viral infection. We feel that antibiotics are not indicated in infants with wheezing who have normal white blood cell counts, unless there are compelling clinical reasons for administering them. In fact, Hall et al⁷ showed an increased incidence of subsequent bacterial infections (1.2%) in infants with RSV who received antibiotic therapy.

In studies from developing countries, isolation of a bacterial pathogen in bronchiolitis is rare,^{8 - 12} except for a study from Pakistan that showed a 26% incidence of a combined infection with bacteria and RSV.¹³ In this study, wheezing was noted in 36% of children with bacteremia due to S pneumoniae, and in 44% of those with bacteremia due to Haemophilus influenzae.

It is important to note that in developing countries, where laboratory and radiologic facilities are not readily available, World Health Organization guidelines recommend the use of antibiotics in wheezing children with a respiratory rate of more than 50/min.¹⁴ This fact needs emphasis, given the fact that medical journal articles are avidly read in many parts of the world but often give recommendations as they pertain to developed countries only.

In conclusion, we feel that the evaluation of the hospitalized infant with bronchiolitis must include a complete blood cell count, blood culture, and chest roentgenogram even if the infant does not have a toxic appearance. Leukocytosis and/or unusual pulmonary findings should alert the clinician to the possibility of a concomitant bacterial infection.