Accuracy of transthoracic ultrasonography in diagnosis of post procedure pneumothorax


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Universidade do Porto, Ankara Universitesi Tıp Fakültesi, Dahili Tıp Bilimleri, Portekiz

Tezin Onay Tarihi: 2018

Tezin Dili: İngilizce

Öğrenci: Ana Soares Jorge

Asıl Danışman (Eş Danışmanlı Tezler İçin): António Pedro Esteves Da Gama Gonçalves

Eş Danışman: Evren Üstüner

Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu

Özet:

iii
a radiografia do tórax. Possivelmente, pode substituí-la no diagnóstico e exclusão
do pneumotórax iatrogénico. A radiografia pode ser apenas necessária quando a
qualidade do estudo ultrassonográfico não é considerada adequada.

Palavras-chave: Pneumotórax, pós-procedimento, ultrassonografia.

Abstract

Background: Pneumothorax (PNX) is a frequent problem in the emergency and
critically care settings, and can occur spontaneously, due to trauma, or as
complication of interventional procedures as thoracentesis, and lung biopsy. The
gold standard in the diagnosis of PNX is radiographic imaging including
computerized tomography (CT) and chest radiography (CR). Beside the effects of
the irradiation, CT and CR cannot be performed at the bedside of the patient. Thus,
Point-of-care ultrasonography (POCUS) is becoming an integral part of the clinical
practice, being quickly and easily performed.

Purpose: We aimed to analyse the data gained from the available literature
evaluating the accuracy of thoracic ultrasound in the diagnosis of post-procedure
PNX. Our second goal was to propose a methodology for the diagnosis of
pneumothorax, that could be applied in a post-interventional setting, and in the
future included in a study that could support the findings reported by previous
studies.

Methods: In May 2018, we performed a literature review of published research
articles in English, between January 1995 and March 2018 in Medline and
EMBASE, evaluating the test characteristics of US in comparison with CR at a
post-interventional setting. Eligible articles were defined as studies on patients
submitted to thoracic procedures who underwent PNX screening with chest US
and CR. The data extracted were analysed and pooled sensitivity, specificity and
diagnostic odd ratio of US calculated with freeware Meta-DiSc, version 1.4
software.

Results: Eight articles were found suitable for the statistical analysis, comprising a
total of 1,123 patients. Overall, ultrasonography was 89.7% sensitive (95% CI, 82.8
-94.6) and 97.9% specific (95% CI, 96.8-98.7) for the detection of post-procedure
PNX. Chest radiography had a pooled sensitivity of 48.6% (95% CI, 38.8-58.5) and
a specificity of 99.3% (95% CI, 98.2-99.8). The pooled DOR for US was 397.80

iv
(95% CI, 79.204 to 1998.0), whereas for CR, the pooled DOR was 113.52 (95%
CI, 15.378 to 837.96).

Conclusion: Although the gold standard diagnosis modality of pneumothorax is still
computerized tomography, ultrasonography is more sensitive and at least as
specific as CR, and possibly can substitute CR in diagnosis and exclusion of
iatrogenic pneumothorax. CR may only be needed, when the quality of sonography
findings is poor.