Special Feature

Are X-rays Necessary in a Child With Pneumonia?

By Thomas F. Dolan, MD, FAAP

The gold standard for diagnosing pneumonia is an aspirate from the lower respiratory tract obtained by either a lung puncture or biopsy or bronchoalveolar lavage. However, these procedures are usually performed only in life-threatening circumstances or in immune suppressed patients. In practice, a chest x-ray is often used as a reference standard for the diagnosis of pneumonia. The practitioner may wonder whether it is always necessary to order a chest x-ray in the evaluation of a patient with possible lower respiratory tract infection, or whether history and physical examination are sufficient.

There are two important goals of history and physical examination in a child presenting with a cough. The first is to try to identify the clinical syndrome producing the child's illness. The second goal is to make an enlightened estimate about the severity of the child's illness.

It is important to maximize the observer's skills in the examination of a frightened infant or toddler. The child should be initially examined while being held by a parent. Of utmost importance is counting of the respiratory rate for one full minute or for two 30-second periods when the patient is not stressed. Tachypnea is the single most important finding that suggests lower respiratory tract disease. Signs of increased efforts of breathing (i.e., flaring of alae nasi, grunting, use of accessory muscles of respiration, or indrawing of the lower chest) increase the likelihood of pneumonia being present. There is general agreement on the significance of these signs as indications of pneumonia in many reviews. Surprisingly, there is less agreement concerning the diagnostic significance of auscultatory findings (crackles rales) and the, sometimes, difficulty in differentiating these from adventitious sounds.

In a recent review and meta-analysis of the pediatric literature from 1982 to 1995, Margolis and Gagonski from the University of North Carolina conclude that in the absence of tachypnea and other signs of increased work of breathing, a chest x-ray is unlikely to be positive.1 Observation and history are probably the most important factors in reaching a diagnosis of pneumonia.

Another report by Swingler and Hussey from Cape Town, South Africa, presents the results of a randomized study of clinical outcome in ambulatory children with possible lower respiratory tract infection. Half of these children had chest x-rays and half did not.2 The study used the World Health Organization (WHO) guidelines for the diagnosis and management of acute lower respiratory infections in developing countries.3 The WHO criteria do not include chest x-ray or other laboratory work. Tachypnea was defined as a respiratory rate more than 50/min in children 2-11 months and more than 40/min in children 12 months or older. Exclusion criteria were a cough of more than 14 days duration, a history of household contact with an active case of tuberculosis, a localized wheeze, clinical signs of cardiac failure, or a clinician's assessment that a chest x-ray was mandatory.

Two hundred fifty-seven patients had a chest x-ray and 261 had no x-ray. Of those who had x-rays, only 37 (14.6) were diagnosed as pneumonia, and, of those with no x-ray, 22 (8.4%) were diagnosed as having pneumonia. The majority of patients had a final diagnosis of bronchiolitis (43.9% x-ray, 8.4% no x-ray), asthma (5.8% x-ray, 7.7% no x-ray), and nonspecific lower respiratory infection (12.1% x-ray to 3% no x-ray).

About 50% of patients received antibiotics regardless of final diagnosis. However, this study from South Africa followed WHO recommendations for the use of antibiotics on all ambulatory children who met the clinical criteria for pneumonia.

Of note is the fact that chest x-ray findings did not affect clinical outcome in patients with lower respiratory tract infections.

It is evident to me that by careful history and observation, a clinical diagnosis of pneumonia or other respiratory syndrome can often be reached. For example, if two or three members of a family have similar respiratory symptoms, the diagnosis is almost always a viral infection. An infant with fever, rhinorrhea, tachypnea, and wheezes in the winter months almost always has bronchiolitis. Recurrent bouts of wheezing suggest a diagnosis of asthma or reactive airway disease. X-rays are not indicated in patients with bronchiolitis, even though about 30% of such patients will have pulmonary infiltrates. However, this will not change the management.

Patients who appear toxic or cyanotic, who have an altered sensorium or are extremely febrile and tachypneic, should have all of the appropriate studies, including chest x-ray, CBCs, blood cultures, and, usually, admission to the hospital. Similarly, any patient who is suspected of having aspirated a foreign body, because of unequal breath sounds and/or localized wheezing, should have a chest x-ray.

A much lower threshold for ordering chest x-ray should be used if a patient has an underlying problem such as sickle cell diseases, cystic fibrosis, or immune suppression.

There are several specific pneumonias that should be mentioned. After 2 months of age, the bacterial pathogen of major concern is Streptococcus pneumoniae, which accounts for more than 90% of bacterial pneumonias. The diagnosis of pneumococcal pneumonia should be suspected in a toxic child particularly before 2 years of age. An elevated WBC should increase one's suspicion. A chest x-ray is indicated only if lobar consolidation or a pleural effusion are detected on physical examination.

Chlamydia trachomatous pneumonitis typically presents in a 2- to 3- month infant as an afebrile pertussis-like disease. A history of maternal perinatal chlamydial infection or neonatal purulent conjunctivitis is helpful. Routine use of HIB vaccine has caused the virtual disappearance of H. influenzae pneumonia in infancy. We rarely see Staph aureus pneumonia in immunologically competent infants and children.

Another treatable pneumonia is that caused by Mycoplasma pneumoniae. This occurs almost exclusively in children older than 5 years of age who have a dry cough and are not very toxic. They will usually have a good response to macrolide therapy. Because there are no good spot tests to confirm a diagnosis of mycoplasma pneumonitis, I think it reasonable to use a macrolide in a relatively non-toxic 5-18-year-old patient with auscultatory crackles, fever, and a non-productive cough.

Using simple history, observation, and auscultation of the chest, all previously recommended by Leventhal, the physician should usually be able to make a probable diagnosis, as well as make an assessment of need for chest x-ray, the need for antibiotics, and/or the need for hospitalization.4 There is an increasing consensus that we do too many x-rays in our work up of previously healthy, febrile children with cough, and the results of routine chest x-rays in this situation rarely has a significant effect on clinical outcome. (Dr. Dolan is Professor of Pediatrics [Pulmonary Medicine], Yale University School of Medicine).


    1. Margolis P, Gadomski A. Does this infant have pneumonia? JAMA 1998;279:308-313.

    2. Swingler EH, Hussey GD. Randomized controlled trial of clinical outcome after chest radiograph in ambulatory acute respiratory infection in children. Lancet 1998;351:404-408.

    3. World Health Organization. The management of acute respiratory infections in children. Practical Guidelines for Outpatient Care. Geneva: WHO, 1995.

    4. Leventhal J. Clinical predictors for pneumonia as a guide to ordering chest roentgenogram. Clin Pediatr 1992;21:730-734.