| dc.description.abstract | ABSTRACT
Introduction
Globally, prematurity and its complications, including low birth weight are a leading cause of death in children under the age of 5 years. In Uganda, neonatal deaths represent 42% of under 5 deaths; 27.9% of which are attributed to Prematurity and its complications, making it the second leading cause of neonatal mortality. The predominant reason for these preterm deaths is respiratory distress syndrome (RDS). Currently diagnosis of RDS is based upon clinical signs and symptoms and confirmed on Chest radiography which exposes the neonates to ionizing radiation. The use of lung ultrasound, a safer modality of imaging for diagnosis of RDS, has not been studied in our setting and thus its applicability to our population remains unknown.
Aims; The study aimed to determine the accuracy and incremental detection rate of LUS in diagnosing RDS in very low birth weight neonates.
Methodology; This was a cross-sectional study involving 263 neonates of less than 1500grams who were admitted to the neonatal intensive care units at Kawempe national referral hospital and Mulago specialized women and neonatal hospital. Patients with clinical features of RDS at admission had an Antero-posterior chest radiograph done and bedside lung ultrasound scan was done for all participants. Both investigations were done within 72 hours of admission. The chest radiographs were interpreted by two independent radiologists blinded to the lung ultrasound findings, with a tie breaker in cases of non-agreement. The different CXR features included, ground glass opacities, consolidations, atelectasis. The principal investigator performed the bedside ultrasound scans using a high frequency liner probe and was blinded to the chest radiographic findings. The LUS features included pleural line abnormalities, B-lines, consolidations. The final clinical diagnosis based upon clinical history, physical examination and chest radiographic features was used as the reference standard and compared to LUS to determine the accuracy of LUS in diagnosing RDS.
Results; Overall 263 participants were enrolled and had complete data for analysis. Majority of the participants were recruited from Kawempe national referral hospital. A slight female predominance 140 (53.2%) was seen with a median birth weight of 1320grams and age of 24 -48 hours. LUS was positive in 237(90.1%) cases; of these, 237 (100%) had pleural irregularities, 233(98.3%) had B-lines, 214(90.3%) had A-line disappearance, 55(23.2%) had consolidations, and no pleural effusions were observed. There were 199 CXR of diagnostic quality, of these, 98(49.3%) were positive. Of the positive cases, 90(91.8%) had ground glass opacities, 18(18.4%) had consolidations but no pleural effusions were depicted. There were 204(77.9%) cases who had a final clinical diagnosis of RDS.
LUS had a sensitivity of 96.08% (95%CI; 92.42- 98.29%), specificity of 29.82% (95%CI;18.43% to 43.40%), positive predictive value of 83.05% (95% CI; 80.50% to 85.33%), negative predictive value of 68.00% (95%CI; 49.16% to 82.36%), positive likelihood ratio of 1.37 (95%CI; 1.15 to 1.63) and negative likelihood ratio of 0.13 (95%CI; 0.06 to 0.29). The incremental detection rate of LUS over CXR in diagnosing RDS in VLBW neonates was found to be 49%.
Conclusion; LUS was found to have a high sensitivity for the diagnosis of RDS among VLBW neonates and can thus be used as a screening tool to aid in the diagnosis of RDS especially in cases where CXR is negative and the neonate is symptomatic. | en_US |
