Abstract:
Background: Pulmonary embolism (PE) is the third leading cause of mortality in
cardiovascular disease, after myocardial Infarction and cerebrovascular stroke.
Diagnosis of PE is based on computed tomographic pulmonary angiography (CTPA),
which is the gold standard. Other tests include clinical, D-dimer test, Chest
Radiograph, Ventilation/perfusion studies, Echocardiography and lower limb Doppler
ultrasound. PE poses a diagnostic challenge due to non-specific clinical presentation
and non-availability of gold standard diagnostic tests especially in resource limited
settings including Kenya, as well as variability in common Chest Radiograph findings
in PE hence the justification for this study.
Objective: To describe and compare chest radiograph and CTPA findings among
adults suspected to have pulmonary embolism at Moi Teaching and Referral Hospital
(MTRH)
Methods: This was a cross-sectional descriptive study conducted at the Radiology
department, MTRH between September 2017 and August 2018. Seventy Five (75)
consecutive patients aged ≥18 years with clinically suspected PE were enrolled after
consenting. Socio-demographic data was recorded using a questionnaire. Study
participants done Chest Radiograph were further subjected to CTPA examination as
per current MTRH Protocol. All chest radiographs and CTPA were reviewed and
findings documented on standard reporting form. Data was analyzed using STATA
software version 13.0. Descriptive statistics was summarized by percentage, whereas
categorical data was analyzed using Chi-square and Fishers exact test. P value <0.05
was considered statistically significant. Cohen Kappa statistics was used to compare
findings between Chest Radiograph and CTPA.
Results: Mean age of the study participants was 46.5 years (SD ±18.7), with females
being the majority at 64 %( 48). The chest radiograph was interpreted as abnormal in
65.52% (19) of the participants with PE. The common chest radiograph abnormalities
associated with PE were cardiomegaly at 45 %( 13) pleural effusion at 38% (11),
atelectasis 24% (7), and Hampton hump at 21% (6). CTPA prevalence of PE was
38.7% (29) and the right pulmonary artery was the most predominant location for PE.
Common parenchymal and pleural abnormalities associated with PE detected on
CTPA were Pleural effusion accounting for 62% (18), atelectasis at 35% (10),
Hampton hump at 31.3% (9), consolidation 17%(5) and Westermark’s sign at
14%.(4).The presence of Hampton hump on Chest Radiograph was significantly
associated with PE (p = 0.012) and there was fair agreement between chest radiograph
and CTPA in the diagnosis of PE at 60% with Cohen Kappa statistics of 0.21.
Conclusions: 1. Majority (65.5%) of the patients with PE had Cardiomegaly, pleural
effusion, atelectasis and Hampton hump as the common abnormal Chest Radiograph
findings 2.There was fair agreement between chest radiograph and CTPA in
identifying pulmonary embolism.
Recommendations: 1. High index of suspicion for PE in patients with Hampton
hump, Cardiomegaly, pleural effusion and atelectasis on chest Radiograph
2. Chest radiograph can be used for investigating patients with suspected pulmonary
embolism in resource constrained settings.