ABSTRACT
We present a case in which lung Ultrasound was relevant to reach an
early diagnosis of lung tuberculosis and to manage the patient in the
right setting. Moreover, ultrasound allowed to detect and treat massive
pleural effusion through an ultrasound-guided thoracentesis.
Keywords: lung ultrasound, bedside, tuberculosis, pleural
effusion.
INTRODUCTION
Tuberculosis (TB) is one of the leading causes of death due to
infectious diseases, and it is a global health threat. TB incidence has
undergone a rapid increase in the last decade, attributable to the
increase in risk factors such as HIV co-infections, drug resistance and
emigrations from areas where the disease is endemic. At today, TB is the
major infectious causes of death in the world, with 1.6 million deaths
in 20171. Here, we present the case of a young
policeman referred to our hospital for dyspnoea and respiratory failure.
Ultrasound utilization allowed to reach a fast and specific diagnostic
program, and to perform a diagnostic thoracentesis. Subsequently, we
performed a narrative review on the role of ultrasound in diagnosing
lung tuberculosis.
CASE PRESENTATION
A 36-year-old Caucasian policeman was referred to our Emergency
Department for dyspnoea. He reported persistent cough and fever with
night sweats for about 2 weeks, for which antibiotic therapy with
Levofloxacin 750 mg daily for 5 days was prescribed without clinical
benefit. His past medical history was not suggestive for any disease.
Blood pressure (BP) and cardiac rhythm were normal (BP: 130/75 mmHg;
heart rate: 91 bpm, sinus rhythm), but the arterial blood gas analysis
revealed a respiratory failure (peripheral oxygen saturation: 87% on
room air; pO2: 53 mmHg; pCO2: 41mmHg).
The physical examination showed a hypo-expansion of the left hemithorax,
reduced tactile fremitus, medium-basal hypophonesis with vesicular
murmur abolished.
Blood laboratory tests showed a blood glucose level of 60 mg/dL (normal
value 74-106), a C-reactive protein of 15.10 mg/dl (n.v. 0.00-0.50), and
the following blood count: white blood cells 10.760/uL (n.v.
4.000-11.000), red blood cells 4.850.000/uL (n.v. 4.500.000-5.300.000),
haemoglobin 14.2 g/dl (n.v. 13-16,0), neutrophils 7.810/uL (n.v.
2.100-7.100), lymphocytes 1.800/uL (n.v. 1.100-3.000), monocytes
1.080/uL (n.v. 200-960), eosinophils 50/uL (n.v. 0-500), basophils 20/uL
(n.v. 0 -200).
Urea, creatinine, sodium, potassium, calcium, transaminases,
gamma-glutamyl transpeptidase, amylase, lipase, lactate dehydrogenase,
total and fractionated bilirubin, and procalcitonin were normal.
Chest X-ray was required, and we performed a bedside lung ultrasound
(LUS) to complete physical examination.
LUS showed a massive finely corpuscular anechoic pleural effusion with
thin branches of fibrin on the left, with a consolidated lung parenchyma
collapsing to the hilum as for complete atelectasis; in the context of
the consolidated lung parenchyma, multiple round hypo-anechoic
formations were seen, with finely irregular margins, not assuming
colordoppler signal: cavitations? Abscesses?(Fig.1-2).
Therefore, chest X-ray confirmed the pleural effusion associated with
consensual parenchymal hypoventilation and showed a widespread
non-specific thickening of the bronchial walls. No pleural effusion on
the right and no signs of pneumothorax were appreciable. The cardiac
shape was within limits. Surprisingly, the radiographic images did not
show the anechoic roundish formations identified by US.
A therapy with oxygen supplementation was set and the patient was
admitted to the infectious diseases ward and isolated according to US
findings.
In the ward, an ultrasound-guided diagnostic-evacuative thoracentesis
was performed with drainage of yellow citrine liquid. Moreover, chest
computed tomography (CT) was performed to better characterize the images
observed with US and not confirmed by chest X-ray, thus showing an
excavated formation with fluid-air levels in the apical segment of the
left lower lobe of about 6 x 3.5 cm (Fig.3). This formation was
surrounded by an area of slight increase in the density of the
pulmonary parenchyma, with ground-glass appearance.
Another subpleural area of parenchymal consolidation with
pseudonodular morphology was present in the apicodorsal segment of the
left upper lobe and appeared as excavated and communicating with the
adjacent bronchus. A parenchymal consolidation area with appreciable
subpleural morphology was present in the apical segment of the right
upper lobe, showing a maximum axial diameter of about 2.2 cm and
appeared excavated. Atelectasis striae in the lower segment of the
lingula and in the basal pyramid of the left lower lobe were present.
The findings described were therefore compatible with TB disease with
cavitations. A fibrobronchoscopy was performed showing marked
inflammation of the mucous membrane of the left bronchial hemisystem,
where whitish secretions were detected.
The analysis of pleural fluid, bronchoalveolar lavage fluid and
microbiological examination on sputum were positive for alcohol-acid
resistant elements, confirming the diagnosis of pleuro-pulmonary TB.
First-line specific antibiotic therapy was therefore set up and the
patient was discharged on the seventh day in good clinical condition
with indication for outpatient follow-up. The final diagnosis was
pleuro-pulmonary cavitary tubercolosis.
DISCUSSION & LITERATURE REVIEW
In past years, LUS has shown an increasing utilization in the diagnostic
field alongside traditional diagnostic tests. Although the presence of
an air interface below the pleura prevents the visualization of the lung
in depth, the evaluation of sonographic artifacts allows to evaluate
several pathological conditions and it is a useful tool to address a
more focused and quickly diagnosis.
To our knowledge, there are few data about US findings in lung
TB2 focusing on the following fields of interest:
detection of pleural effusion, assessment of residual pleural
thickening, the execution of trans-thoracic needle biopsy, assessment of
mediastinal lymphadenopathies and detection of pulmonary involvement in
miliary TB3.
Sonographic pattern of pulmonary TB can be visualized as consolidations,
subpleural nodules, pleural thickenings, fibrosis, pleural effusion and
pneumothorax together with miliary pattern2,4,5. The
miliary pattern has been described by Hunter et al6 as
characterized by bilateral vertical artifacts (B-lines and/or
“comet-tail artifacts”) in multiple lung areas, and by sub-pleural
granular changes, that may be more characteristic of miliary
TB6. However, these findings are not so specific,
since similar bilateral vertical artifacts have been described in
patients with Pneumocystis jirovecii or cytomegaly virus pneumonia and
sub-pleural US alterations can be found in other pulmonary conditions
such as metastatic thyroid cancer6.
Agostinis et al2 classified the sonographic findings
in patients with lung TB into two categories: lung (subpleural nodule,
pleural effusion, miliary pattern and cavitations) and extra-lung
(pericardial effusion, splenomegaly, abdomen lymph nodes, hepatomegaly,
ascites) US findings2. The hypoecoic subpleural
nodules (SUNs) were the most frequently identified in sixty adults with
diagnosis of lung TB in a rural African setting2.
Consolidations were detected in about half of the patients and are
indistinguishable from bacterial pneumonia, while cavitations were
identified in three patients and described as anechoic or hypoechoic
areas within solid lung consolidation2. SUNs were not
identified in subjects without TB, suggesting a probable important role
in lung TB diagnosis2.
The accuracy of LUS in the diagnosis of pulmonary TB in adults have been
investigated in a recent study4. Montuori et
al4 demonstrated that the combination of apical
consolidations and SUNs found in the same patients has a specificity of
96% and a sensibility of 31%, while when at least one of these signs
is identified, the sensitivity and the specificity are respectively of
86 and 63%4. Moreover, according to Agostinis et
al2, cavitations were found to be the least common
sign and not enough sensitive also in this study, probably due to the
presence of air in cavitations or to the fact that a high number of
lesions do not reach the pleura4.
In our case, the identification of the non-vascularized hypo-anechoic
roundish lesions compatible with cavitations, in relation to the
clinical context, was fundamental in the diagnostic suspicion of
cavitary TB and for the early isolation of our patient in order to avoid
the spreading of the infection in the crowded environment of the
Emergency Room. It must be considered that our patient had not apparent
risk factors for TB (patient medical history and lifestyle, HIV-negative
test), and that the chest X-ray did not show the cavitations identified
by US, so that the patient would have been hospitalized in a medical
setting without the necessary isolation measures. Even if chest CT and
microbiological and laboratory tests were fundamental in the different
diagnosis between cavitations and lung abscesses, the findings provided
by bedside LUS were relevant to confirm our diagnostic suspicion, and
therefore for the instauration of an effective therapy, as already
described in critical care7. In our case, we did not
identify SUNs and apical consolidations because the lung parenchyma was
completely collapsed due to the massive pleural effusion. However, the
chest CT performed after evacuation of pleural fluid documented
parenchymal consolidations at the level of the apical segments of the
left upper and lower lobes, and of the right upper lobe, typical lung
fields where consolidations can be found in post-primary
tuberculosis8.
LUS is increasingly employed to safely perform bedside
thoracentesis7,9–11. In our case, the possibility of
identifying the corpuscular nature of the pleural fluid has contributed
to the diagnostic suspicion5,12, and the US
quantification of the fluid was fundamental to guide the therapeutic
intervention13.
We did not perform a complete US evaluation of the abdomen since, in the
Emergency setting and in the evaluation of the respiratory status of our
patient who presented with a respiratory failure, we focused on the
immediate evaluation of the lung parenchyma to diagnose the possible
life-threatening conditions. However, the abdominal CT performed during
hospitalization did not show lesions compatible with the dissemination
of the disease.
Other possible applications of thoracic US in the diagnosis of TB are
described in literature, especially in patients with suspected
mediastinal TB (Table 1). Diagnosis of mediastinal TB is difficult due
to non-specific clinical features and lack of characteristic
radiographic features14. In order to obtain
histopathological confirmation, a CT guided fine needle aspiration
biopsy (FNAB) or invasive procedures such as mediastinoscopy or open
surgical biopsy are often required14. US-guided FNAB
has proven to be safe, and effective in the diagnosis of mediastinal
TB14, despite its limitations for centrally located
lesions due to the lack of a good acustic window15.
More recently some authors have investigated the possibility of
performing endobronchial ultrasound (EBUS)16,17,
useful in investigating peripheral lung lesions, and also in guiding
transbronchial FNAB and endoscopic aspiration from esophagus in
suspected mediastinal TB18,19.
CONCLUSIONS
Bedside US is a safe, non-invasive, portable, versatile, easily
repeatable and cost-effective imaging modality4,7. Its
use can lead to the identification of specific pleuro-pulmonary
pathological conditions by strengthening the clinical-diagnostic
suspicion. Although nowadays there are few data about LUS evaluation in
tuberculosis, in our case it was fundamental in the formulation of
diagnostic suspect through direct visualization of the characteristic
lung lesions and early isolation of the patient; early suspicion allowed
to promptly request the specific diagnostic investigations and begin the
targeted therapy. Our case has also confirmed that LUS is important for
the identification, quantization and characterization of pleural
effusion and for the execution of diagnostic and therapeutic
ultrasound-guided thoracentesis.
Informed consent was obtained for patient data publication
Declaration of interests
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to
influence the work reported in this paper.
No founding sources.