Procalcitonin and lung ultrasound algorithm to diagnose severe pneumonia in critical paediatric patients (PROLUSP study). A randomised clinical trial

Pneumonia is responsible for more than 2 million child deaths around the world; it is the leading cause of childhood morbidity and mortality [1]. Moreover, severe lower respiratory infection is the most common cause of admission to paediatric intensive care units (PICUs) [2]. Around 60–70% of these have a viral etiology [3]. For physicians, it is important to differentiate between viral and bacterial infection because each entails a different therapeutic approach (whether antibiotics are indicated and the treatment duration), a different prognosis in the short and long term, and requires a different hospital isolation policy [4, 5]. It is challenging to rule whether a case of pneumonia is viral or bacterial in origin [6‐8]. Clinical symptoms are similar and although conventional chest radiographs (CXR) have traditionally been considered the best diagnostic option [9], the specificity of CXR for determining bacterial etology remains low. Chest computed tomography may detect smaller pulmonary consolidations not visible via CXR, but it cannot be routinely performed on critically ill patients due to cost, exposure to radiation, and transportation risk [9]. Therefore, there is a need to have an examination tool that can be used at the patient’s bedside and which is easily reproducible to help detect lung consolidations. LUS has recently emerged as a radiation-free technique; it is non-invasive, with a high interobserver [10] agreement for lung pathologies such as consolidation [11], pleural effusion [12], interstitial syndrome [13], and pneumothorax [14]. To determine the etiology, conventional microbiology tests such as blood culture, pleural aspiration, and bronchoalveolar lavage are usual practices, but some of these are invasive, may not detect all etiologies [15], and the results may not be immediate.

Besides this, the use of biomarkers such as procalcitonin (PCT) has become more widespread during the past 10 years, helping clinicians diagnose bacterial etiology, especially in patients who have only had a fever for a few hours or those admitted to intensive care units [16‐20]. The use of PCT has allowed for a decrease in antibiotic prescription [21, 22], even in nosocomial and community pneumonia [23, 24]. Despite its role in the diagnosis of pneumonia, PCT values without any other tests may not be a complete diagnostic biomarker for pneumonia.

Quality of care is defined by the World Health Organization as medical care in which the patient is diagnosed and treated correctly, according to current medical knowledge (scientific and technical quality), and to their biological factors (optimal state of health to attain), with a minimum cost (efficiency), minimum possible exposure to risk for more harm, and maximum patient satisfaction [25]. At the hospital level, a quality assurance plan should include different levels of action. The first step is level 1 of quality promotion, which requires institutional support and the availability of clinical and quality protocols. The second step is the research level: descriptive studies for detecting and quantifying a specific situation or health issue, and the use of databases or specific studies to evaluate health services, among others.

Therefore, we propose this clinical trial, based on combining LUS and PCT in an algorithm with the aim to improve quality of care in children with pneumonia in a PICU. We hypothesize that the diagnostic performance of LUS and PCT will be better than conventional CXR.

The study is designed as a randomized, blinded clinical trial of children with severe community or nosocomial pneumonia. It will be conducted at a single PICU at Sant Joan de Déu Hospital, a tertiary children’s hospital in Barcelona. Period of recruitment and follow-up was from September 2017 to December 2019.

Nowadays, care standards are focused on the quality of care. As the World Health Organization stipulates, the patient must be diagnosed and treated correctly. This clinical trial is focused on improving the quality of care for paediatric patients with suspected bacterial pneumonia. LUS has good diagnostic accuracy for pneumonia in children, even if the exam is performed by a nonexpert physician [35]. Our algorithm will help us to diagnose bacterial pneumonia accurately, and to prescribe the correct antibiotic treatment. A reduction in patients on antibiotics and in the number of days on antibiotics is expected. Secondarily, a reduction in exposure to ionizing radiation and in costs is expected.

For many years, LUS has been integrated into the management of critically ill paediatric and neonatal patients at our hospital. Some articles have been published by our group regarding the use of LUS in different pathologies, such as prematurity, pulmonary arterial hypertension, during the postoperative period after cardiopulmonary bypass, etc. [36‐38]. For the application of this clinical trial, all the researchers will be intensivists who are experts in LUS, and regular internal training will be essential to guarantee objective results.

In addition, our group has extensive experience in the use of PCT for the diagnosis of bacterial infection in other medical situations, such as after cardiopulmonary bypass in children and new-borns [19, 39]. Furthermore, we have experience in PCT-guided antibiotic policy, with a reduction in the number of days on antibiotics without adverse events in children with nosocomial infections [40] and in children after cardiopulmonary bypass [22].

The use procalcitonin and lung ultrasound algorithm will help us diagnose bacterial pneumonia accurately and prescribe the correct antibiotic treatment. A reduction in patients on antibiotics and in a reduction in exposure to ionizing radiation and in costs is expected. This clinical trial is focused on improving the quality of care for paediatric patients with suspected bacterial pneumonia.