Diagnosing Pneumonia Under Low-resource Conditions

Study organisation. This is a prospective observational study run simultaneously in four Indian public hospitals (King George Medical University, Lucknow; Regency Hospital, Kanpur; Vanivilas Hospital, Bangalore; Bowring and Lady Curzon Hospital, Bangalore). The study started in Oct 2012 to cover the Indian respiratory viral season. In order to maintain high standards of data collection, a post-graduate research coordinator is employed at each hospital. Because of the use of standardised scoring systems and the need for accurate clinical data collection, a member from the Canadian team spent a week at each centre familiarising local research team members with the study protocol and standardised scoring systems. This was followed by a one week trial period of data collection and electronic transmission of files to Canada.

Diagnostic definitions and standardised scores. The primary problem facing any study of pneumonia is accurate diagnosis. The overlap of clinical and radiological findings between severe viral infections, asthma and bacterial pneumonia can make it difficult to determine which febrile tachypneic children have wheezy diseases and which ones would benefit from antibiotic treatment. In many low-resource areas, this is further complicated by infectious diseases, such as malaria and dengue, which can have similar presentations. Early WHO tachypnea-based diagnostic protocols were intentionally over-sensitive to ensure that all children with bacterial pneumonia received antibiotics. Later attempts were made to improve the detection of wheezy diseases, by adding audible wheeze or acute bronchodilator response to the basic criteria. However, these were shown to be imprecise, particularly amongst the sickest children.

Investigators chose to formalize the diagnostic method described by Sachdev et al who classified patients into four groups (pneumonia, wheezy disease, mixed and non-respiratory) based on consultant review of a detailed history and examination plus a chest radiograph (CXR). Investigators recorded 29 items from a protocol that included history, examination, CXR and oximetry (see table). In order to combine results from data collection between centres, standardised scoring systems for conscious level and auscultation findings were used. For chest radiographs, a modification of the recently updated system recommended by the WHO was used. During the one week preparatory period, the system was explained to all involved ER physicians and pediatricians using examples and practice interpretation.

Study protocol. All children below 5 years age who present to the emergency rooms of the study hospitals with cough or difficulty breathing of less than 5 days, are identified. If their initial respiratory rate met WHO criteria for pneumonia, the study is explained by a native speaker of their primary language and they are invited to enter the study. Families are not paid to enrol but the study covers the cost of a CXR for every child plus travel expenses for outpatients to return for review on day four. After enrolment, twenty nine features of the child's history, examination and CXR are assessed by the ER physician and recorded by the study coordinator(Table 2).

After reviewing the data, the ER physician is asked to place the child into one of four diagnostic categories: pneumonia, wheezy disease (asthma and bronchiolitis), mixed (evidence of pneumonia and wheeze) and non-respiratory (malaria, dengue etc). The ER physician is solely responsible for subsequent management decisions. All study patients are reviewed four days later by a qualified pediatrician who is blinded to the ER physician's CXR interpretation and diagnosis. Based on a review of the clinical data at presentation, plus subsequent course over 4 days and a second examination, the pediatrician places the patient into one of the four diagnostic categories. This is considered the child's final diagnosis for analysis.

Statistical analysis. Logistic regression analysis will be used to determine which of the initial clinical variables had the best predictive power for pneumonia and asthma. The best cut-off values for continuous variables were established using receiver operating curve analysis. For each predictive variable, sensitivity, specificity plus positive and negative predictive value will be calculated from conventional tables using standard equations. When the predictive value of combining variables is tested, investigators will link them as 'A and/or B'. This increases sensitivity but decreases specificity, compared to using 'A and B.' Continuous variables will be displayed as mean +/- one standard deviation. Categorical variables will be displayed with box and whisker plots where the whiskers represent full range.

Table of investigations performed in the ER at presentation.

History Recorded details Cough yes/no Difficulty Breathing yes/no Lethargy yes/no Reduced feeding yes/no Fever yes/no Previous similar episodes number Vaccinations number and type

Examination Recorded details Age months Weight weight for age 'z' score Temperature ⁰ Celsius Heart rate beats/minute Respiratory rate breaths/minute Indrawing present/absent

Responsiveness score:

A fully alert V responds to voice P responds to pain U unconscious

Auscultation score:

Chest clear normal vesicular breath sounds Crackles coarse or fine inspiratory crackles/rattles Wheeze high pitched whistling noise, inspiratory or expiratory Crackles and wheeze both sounds present Bronchial breathing tracheal breath sounds heard over the lungs

Investigations Recorded details

CXR score: one or more of:

Normal ) Hyperinflation ) Minor patchy changes ) definitions, see table 3 Major patchy changes ) Lobar changes ) Pleural fluid ) Oximetry % oxygen saturation