Please use this identifier to cite or link to this item: http://ir.mu.ac.ke:8080/jspui/handle/123456789/5359
Title: Level of agreement between lung ultrasonography and chest radiography findings among children with pneumonia at Moi Teaching and Referral Hospital
Authors: Kwamboka, Loyce William
Keywords: Lung ultrasonography
Chest radiography
Pneumonia
Air bronchogram
Issue Date: 2021
Publisher: Moi University
Abstract: Background: Pneumonia is the leading cause of morbidity and mortality among children. Globally, it accounts for over 1,400 per 100,000 childhood mortalities. Chest radiography is known to have a low dose of ionizing radiation although children are more susceptible to the effects of radiation exposure than adults. Ultrasound (US) has no ionizing radiations therefore its safe in children and is portable. The study seeks to explore whether lung ultrasonography (LUS) can be considered as an alternative diagnostic test to chest radiography (CXR) in the management of pneumonia in children. Objectives: To describe lung ultrasonography and chest radiography findings and determine level of agreement between LUS and CXR findings among children with clinical diagnosis of pneumonia. Methods: This was a descriptive cross-sectional study conducted at Radiology department, MTRH between1st April 2019 to 31st March 2020. Sample size was determined using Cohen’s kappa formula. One hundred and twenty-three consecutive patients aged ≤ 18 years with clinical diagnosis of pneumonia were enrolled. Details of the age, gender, lung ultrasonography and chest radiography findings were recorded in data collection form. Study participants had CXR done and further subjected to LUS examination as per MTRH Protocol. A Mindray M7 ultrasound machine was used. Radiological diagnosis of pneumonia was made as per World Health Organization criteria of 2001 based on either lung consolidation, pleural effusion and/or pulmonary infiltrates. Continuous variables were summarized using median and interquartile range while descriptive statistics were summarized using frequency, percentages, tables and pie charts. Cohen’s kappa coefficient statistic was used to determine the level of agreement between CXR and LUS findings and corresponding p-values were recorded. Significance level was set at 0.05. Results: Median age of study participants was 4 years (IQR 3-8). On CXR, 90 (73.17 %) had lung consolidation, 58 (47.12 %) pleural effusion, 38 (30.89 %) pulmonary infiltrates and 28 (22.76 %) had normal CXR. Among them, 95 (77.24%) were diagnosed with pneumonia on CXR. On LUS, 85 (69.11%) had lung consolidation, 63 (51.22%) pleural effusion and 30 (24.39%) had normal LUS. 93 (75.60%) had diagnosis of pneumonia on LUS. There was a nearly perfect agreement between the CXR and LUS in diagnosis of pneumonia, κ = 0.865 (95% CI, 0.759 to 0.971), p < 0.0001. Conclusion: Lung consolidation was the commonest radiological finding detected on both CXR and LUS. LUS was better in detection of pleural effusion. There was a near perfect agreement between LUS and CXR in diagnosis of pneumonia. Recommendations: There is need to consider LUS as a diagnostic alternative to CXR based on nearly perfect agreement between the CXR and LUS in diagnosis of pneumonia, κ = 0.865 (95% CI, 0.759 to 0.971), p < 0.0001.
URI: http://ir.mu.ac.ke:8080/jspui/handle/123456789/5359
Appears in Collections:School of Medicine

Files in This Item:
File Description SizeFormat 
Dr. Kwamboka Loyce William 2021.pdf789.95 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.