Discussion:
Chest ultrasound in the diagnosis of Pneumothorax was reported first in 1986 in a veterinary journal. Several researchers have explored its usage since then. The principal method for diagnosing employed an absent sliding lung sign in the air presence between visceral and parietal pleura.26 The presence of sliding lung signs acts as an accurate negative predictor for pneumothorax detection. Dulchavsky et al. 27 described a 100% true negative rate compared to conventional chest radiographs in Pneumothorax diagnosis, with sensitivity ranging from 90% to 100%.
Most of the studies used focused assessment with sonography for trauma (FAST) technique for diagnosing pneumothorax. Literature has shown accurate and fast results. Standard FAST protocols assess four locations (pericardial, perihepatic, perisplenic, and pelvis). Still is more feasible and simpler to widen the scanned regions to assess the chest for haemothorax formally. 28,29
The present systematic review and meta-analysis showed a better performance of chest ultrasound in detecting pneumothorax and can be used as an alternative to CT and MRI.
Pneumothorax diagnosis can be made based on physical examination and symptom presentation. Confirmation is generally via radiography or C.T. scanning. Radiographs taken in the later stages pose difficulty in diagnosing pneumothorax because of the patient’s condition, distance and other considerations. Also, chest radiograph reliability is questionable, and a wrong diagnosis may be expected in 30% of cases.23. The metanalytic study of Ebrahimi et al29 showed ultrasound accuracy in detection of pneumothorax, with sensitivity of 0.87 (95% CI: 0.81–0.92;I 2 = 88.89; P <0. 001) and specificity of 0.99 (95% CI: 0.98–0.99;I 2 = 86.46, P <0. 001). Our study results are also in similar lines with a sensitivity of 0.89 and specificity of 0.96.
The present study included both adult and neonate populations, and the diagnostic accuracy was good. The results of Hamid Dahmarde et al. 30 also support this statement. The sensitivity, specificity, and odds ratio of the chest ultrasound in the diagnosis of pneumothorax in neonates was 96.7% (88.3%–99.6%), 100% (97.7%–100%), and 1343.1% (167.20–10788.9), respectively, depicting superior results.
The study has certain limitations. Firstly, it should be borne in mind that the test characteristics merely a part of evaluation of diagnostic test function, and extent of each test depends on its effect on patient. The second factor which can affect the course of meta-analysis is the chest ultrasound operator’s accuracy. The training quality and nature of training are not considered in several studies. The heterogeneity of studies included was also higher; thus, the random-effects model was employed to obtain better results.