In 2020 as per WHO  1.5 million people died due to TB worldwide [1]. To stop the global TB epidemic, in 2014 World Health Organization (WHO) defined the “End TB Strategy”, underlining the need to develop diagnostic methods, to ensure an earlier and correct diagnosis². In resource – limited countries, where the disease has a high prevalence, the diagnosis of pulmonary tuberculosis (TB) is based on clinical findings, chest radiography (CR)  and the demonstration of acid- fast bacilli in sputum. However often anti- tubercular therapy (ATT) has to be started based on a presumptive diagnosis [2].  Very few studies are available on the use of lung ultrasound (LUS) in pulmonary TB (PTB)  in adults [2-7].

Sputum smear microscopy for the identification of acid fast bacilli (AFB), is widely available, quick and inexpensive; but its sensitivity is low (approximately 45 %) [8].Culture methods have a sensitivity of 80% but the results are available ,only after several weeks. The rapid molecular test ( Xpert MTB/ RIF) that can diagnose TB may not be available everywhere. Chest radiography cannot distinguish between early bacterial pneumonia and PTB easily .Thus in many patients a definitive diagnosis of PTB is not possible ;and generally after failure of first or second line antibiotic drugs , ATT is started [3].

Ultrasound has a high sensitivity in identifying lung abnormalities adjacent to the pleura. The most frequent LUS finding(96.6%) in patients with PTB is subpleural nodule (SUN) [3]. SUN are hypoechoic, round or oval nodules with well defined margins ,up to 1.0 cms in diameter, with hyper echogenicity of the posterior wall and acoustic enhancement [3]. The histological basis of the SUN remains unclear , but its very high prevalence indicates that it  may be an important indicator of TB.³ Consolidation is also seen in TB on LUS,   and in miliary TB ,LUS  shows comet tails radiating from echogenic bright ‘granular’ artefacts located in subpleural zone [2,3,5]. Whether pulmonary TB is characterized by other sonographic signs, is unclear at present.  Also the relative frequency of different LUS findings and their specificity is also not clear. For example, SUN has also been found in other diseases e.g. sarcoidosis, Cytomegalovirus , Pneumocystis , Legionella pneumonia and pulmonary embolism [3].

To achieve control of PTB in endemic areas, the diagnosis and treatment of new cases is of paramount importance. A delay in diagnosis may lead to spread of infection, especially in developing countries with crowded living conditions, malnutrition, and limited access to healthcare services [3]. Presently chest radiographs and chest CT are being used for the diagnosis of pulmonary TB, but equipment and skilled radiographers for these imaging modalities are scarce in low- resource settings [9]. Also , both CR and CT use ionizing radiations . So, the purpose of this study was to evaluate the role of LUS in endemic areas for the early diagnosis of pulmonary TB, to avoid the hazards of ionizing radiation.

.1 Study design and patients

This prospective study was approved by the Institutional Ethics Committee [ IEC-HR-2022-55-116] and was conducted according to the principles of the Declaration of Helsinki. Written informed consent was obtained from the patients. From September 2022 to February 2024, LUS was conducted in 100  indoor patients or outpatients in the age group of 12-60 years ,selected by Radiologist 1 (R1) , having 30 years’ experience in radiology. Recruitment of patients with clinical suspicion of pulmonary TB was done based on chest radiograph and/or CT scan of the chest being positive. Available laboratory investigations were recorded. Patients who were hemodynamically unstable; had subcutaneous emphysema or skin diseases; or had a past history of ATT for pulmonary tuberculosis ; or had obvious fibrotic lesions on chest radiograph / CT scan, were excluded. Microbiological tests i.e., sputum for acid fast bacilli (AFB) or sputum CNBAAT positive, were used as the reference standard. In those patients who were microbiologically negative, clinic-radiological response to empirical ATT for 6 months; or response to any alternative treatment, was used as the reference standard.

2.2 Ultrasound Protocol

The study was conducted using Samsung, RS80 EVO, Korea. Convex probe (2-5MHz) was used first followed by the linear probe (3-16MHz), when any parenchymal abnormality was detected. Tissue harmonics and compound imaging modes were not used.  Patients were studied in the supine position for evaluation of the anterior and lateral aspects of the thorax and in sitting /decubitus/ prone position, as per convenience, for evaluation of the posterior thorax. During the examination of the posterior regions, the patients were asked to raise the arms above the head in order to displace the shoulder blade and uncover the area of the lung which is usually masked by the scapula. All intercostal spaces were evaluated along the ribs and the whole surface of the lung was systematically analyzed. The examination of the lung apices was performed by applying the probe between the clavicle and the trapezius muscle anteriorly; and directly on the cranial part of the trapezius muscle on the back.

We evaluated a total of 16 zones in anterior, posterior and lateral regions of both lungs.

 The anterior regions of both lungs were divided into upper zone ( up to  anterior end of 2^nd rib) : RAUZ,LAUZ;  middle zone ( 2^nd to 4^th ribs): RAMZ,LAMZ; and lower zone  ( below anterior end of  4^th rib) : RALZ , LALZ .The lateral regions of the lungs were divided by 4^th rib into upper and lower zones only; Right side- RLUZ,RLLZ; Left side- LLUZ,LLLZ. The posterior lungs were arbitrarily divided into three parts as upper, middle and lower zones; Right -RPUZ,RPMZ,RPLZ; Left- LPUZ,LPMZ,LPLZ.

Subsequently the convex transducer was placed transversely in the epigastric angle and tilted cranially to evaluate for the presence of a pericardial effusion. Finally the upper abdomen was scanned to evaluate abdominal lymph nodes, spleen, liver and peritoneum.

The scan was performed by Radiologist 2 (R2), a first-year radiology resident trained for one week in LUS, on adult patients who reported for chest US. R2 was specifically trained to identify findings on LUS such as B lines, pleural line irregularity and subpleural consolidation. R2 was blinded to chest radiograph and or CT chest findings; and performed LUS within 2 weeks of the previous imaging. Static and video images of LUS findings were archived and recorded, as per the Case Record form. Radiologist1 (R1) having 10 years experience in LUS, reviewed the ultrasound images about a month later, approximately; to eliminate bias due to memory.

Lung ultrasound findings

 Multiple findings like B lines , converging B lines, consolidation, cavity, subpleural nodules, and confluent B lines are shown in figures (1-4).  Their appearance is described in detail in Table 1.

LUS image analysis

Based on results of previous studies on LUS in PTB and on retrospectively analysing the most frequent LUS observations in those who recovered on ATT  in our study, patients were considered “ LUS positive for PTB” if they showed any two out of the three findings of SUN/ consolidation/ “isolated” Converging B lines [2,5].[Fig. 4a,b]

Statistical Analysis

Data was entered in MS excel software and was analysed by SPSS (v 20.0) ( IBM, Chicago,USA). Descriptive statistics were expressed in form of percentages.  Sensitivity and specificity of LUS and “extended” LUS were calculated. Chi square test was used as test of significance. P value < 0.05 was taken as significant.

3.1 Characteristics of the subjects

The 100 patients recruited included 71% females and 29% males. The mean age was 26.3 years. Most patients had complaints of cough (83%), fever (82%), loss of appetite and loss of weight (32%). Wet cough (66%) was commoner than dry cough (34%) . The duration of complaints ranged from 7 days to 1 year. In this study, 19 % patients had other systemic complaints suggestive of extrapulmonary TB at the time of presentation: pain in the abdomen (3%), vomiting (6%), headache (5%), altered sensorium (5%), neck swelling (2%), cold abscess on the back (3%), seizure (1% )and small bowel obstruction (1% ). In one patient, the chief complaint was dysphagia for 1 year with dry cough off and on; and fever for 5 months. There was a history of contact with TB (15%); diabetes (1%); HIV positivity (1% ) and chronic liver disease (1%),

LUS Findings

LUS showed findings in 98/100 (98%) patients with suspected or proven PTB and 12/12 (100%) patients with proven TB. The most common LUS finding in our study was “Converging” B lines, seen in 77% patients. They were non- homogenously distributed in the lungs, best observed on real time US examination and video recordings, rather than on static images. Other LUS findings in patients with proven or suspected PTB were consolidations (65%) ,B lines (64%) ,subpleural nodule or  SUN (50 %), miliary pattern (43%), confluent B lines (23%) , pleural effusion (15%) , pleural irregularity  (12%) and cavities ( 7 %) . The details of LUS findings are shown in Table 2.

The size of SUN observed varied from 1 to 10mm. Size of SUN in patients was 1-2mm (7.1%) [Fig.5a], 2-3mm (18.6%), 3-4mm ( 28.6%), 4-5mm (24.3%), 5-6mm (14.3%), 6-7mm (2.9%). 7-8mm (1.4%), 8-9mm (1.4%) and 9-10mm (1.4%) respectively. More than 50% SUNs measured between 3.0 to 5.0mm.

Most LUS signs were more frequent in anterior aspect and lower zones of lungs than posterior and lateral zones, except for cavities (more in upper and mid zones of lungs) There was a statistically significant difference (p<0.5) in the percentage of patients showing consolidations, SUNs or isolated  “Converging” B lines in anterior , lateral and posterior lung zones on LUS. There was no statistically significant difference (p>0.5) in the percentage of patients showing LUS signs in upper, middle and lower zones on LUS. (Table  3)

In 12 out of 100 patients who were microbiologically positive for  PTB the LUS findings were subpleural nodules (100%) , consolidation with converging B lines ( 83%) [Fig. 5b],   consolidation without converging B lines (58.3%) , isolated converging B lines (41.7%) B lines(66.7%), miliary pattern (41.7%), cavity ( 33.3%), confluent B lines(25%) and pleural irregularity (8.3%)[Fig. 1b] .

 Findings in suspected or proven PTB patients who were false negative on LUS due to absence of 2 out of 3 findings of consolidation/ SUN/ converging B lines, are depicted in Table 4.

EPTB: Pleural effusion was observed in 15% patients (free-10% ; loculated-4%) . The other findings observed on upper abdomen “extended” US in suspected or proven PTB were conglomerated periportal lymph nodes (15%), splenic granulomas (3%), multiple liver granulomas (1%), ascites (1%) and mild pericardial effusion (1%) .

3.3 Sensitivity and specificity of LUS

Among the 12 patients who were microbiologically positive for PTB, sputum was positive for AFB in 6 ; and CB NAAT was positive for TB in 6 patients. Of the remaining 88 patients, 82 were successfully treated with ATT. Thus 94 (12+ 82) patients were diagnosed as pulmonary TB .

Fifty nine patients were true positive [Fig. 1f, 2a] comprising those who were sputum positive microbiologically or improved on ATT , and were  LUS positive for PTB. Thirty five patients who were LUS negative for PTB but improved on ATT, were considered false negative, as they did not show the combination of two of the defined LUS findings ( i.e., consolidation/ SUN/ “isolated” Converging B lines). 

Amongst the remaining 6/100 patients, 5 were true negative - two responded to antibiotics (one had shown sputum positive for Pseudomonas aeruginosa); 01 responded to surgical treatment (diagnosed as achalasia cardia) [Fig 6 a,b];01 was continued on ATT beyond 6 months ( due to poor compliance) and 01 showed no response to ATT ( under investigation for allergic bronchopulmonary aspergillosis ) .

Only one patient out of the six was false positive, because although LUS was positive for PTB, the patient succumbed to his illness after receiving ATT for 2 months; so the diagnosis of PTB could not be confirmed.

3.4 Statistical Analysis

Data was entered in MS excel software and was analyzed by SPSS (v 20.0) software. Sensitivity and specificity of LUS for diagnosis of PTB were 62.8% and 83.3% respectively as shown in table 5. On considering “extended” LUS (i.e. LUS combined with upper abdomen US for EPTB) as positive for PTB, 6 additional patients who were LUS negative, but had periportal nodes, were considered true positive on LUS for PTB. Thus, sensitivity of LUS for PTB increased from 62.8% to 69.1% using “extended” LUS, although specificity remained 83.3%.

B lines was the commonest finding in 57.1% false negative patients, but it is a known fact that B lines are a non-specific lung ultrasound finding and represent interstitial lung disease. Miliary pattern was the second commonest LUS finding (40%) among the 35 patients considered false negative, on basis of LUS (even though they improved on ATT) so it was added to the three defined LUS findings (Consolidation / SUN/ “Isolated” Diverging B lines). This led to an increase in the True positive patients from 59 to 70 (Table 6). Sensitivity and specificity were 74.5% and 66.7% after the inclusion of the miliary pattern on LUS.

Using Chi square test p value was calculated. There was significant statistical difference in the percentage of patients showing consolidations(p=0.001), SUNs (p=0.040), isolated “Diverging” B lines (p=0.002) in anterior, lateral and posterior on LUS.

There was no statistically significant difference in the percentage of patients showing consolidations (p=0.43), SUNs (p=0.38), isolated “Diverging” B lines(p=0.33) and cavities (p=0.4) in upper, middle and lower zones of lung on US.

Figures

Figure 1; Characteristics of “Converging” B lines (Linear probe 3-16MHz)

 (a) LUS shows multiple  (eight)  “Isolated” converging B lines (green arrows) directed obliquely(appearing as inverted “V” shape), arising from focal points on the  pleura ( blue arrows) .They are criss-crossing (yellow arrows). (b) “Converging B lines"(blue arrow) arising from a focal area of pleural irregularity( yellow arrow) . The width of these converging B lines varies from 2.9 to 4.3 mm ( between cursors+) , measured at approximately the centre of the converging B lines. (c) Criss crossing of “isolated” converging B lines ( green arrows)arising from a focal area of the pleura(white arrow). (d) a large (size 2.8 x 1.3 cms ) consolidation ( orange arrows) appearing as a hypoechoic area with air bronchogram and irregular margins. It is associated posteriorly with converging B lines( blue arrows),of variable widths 11.7 to 16.7mm (between cursors+) resembling ground glass rockets. (e) A rounded SUN( orange arrow) measuring 6.0mm , associated with converging B lines( blue arrows) showing ill-defined margins, posteriorly in a patient with sputum AFB positive . (f)LUS showed a large consolidation ( orange arrow) appearing as hypoechoic area (size 2.8x1.3cms) showing shred sign. It is associated with converging B lines( blue arrows)  which resemble “Glass rockets”. A solitary “Isolated” converging B line( green arrow) which is not perpendicular to the pleura, is also seen on the right side.

Figure 2 ; LUS  in a True Positive patient (a) shows a  subpleural consolidation (SPC) (between orange arrows)  as a hypoechoic region with confluent pseudo nodules in RPLZ ; associated posteriorly with converging B lines( blue arrows) with ill-defined margins. (b) Subpleural consolidation  (between orange arrows) is seen  as a hypoechoic region with length twice more than the depth (measuring 3.8x0.8cms) with pseudo nodular appearance in RLLZ ;and is  associated with converging B lines( blue arrows) posteriorly.

Figure 3; (a) Miliary pattern on LUS: Multiple comet tail artefacts / incomplete B lines(blue arrows) seen. (b) Cavity on LUS: An anechoic area (yellow arrow) representing clearing in an area of consolidation, shows an air’s sickle (blue arrow) with a slightly convex margin

Figure 4 - True Positive patient (microbiologically) for pulmonary tuberculosis. LUS was considered positive for PTB if it showed the following two LUS features: (a) In RPMZ- A large consolidation (with air bronchogram -green arrows) showing shred sign (blue arrow). (b)In RAMZ- “Isolated” converging B lines (blue arrows) directed obliquely and not vertically , converging at a focal point on the  pleura ( yellow arrow) .

Figure 5; LUS in a microbiologically positive patients of PTB . (a)  “Isolated” converging B lines (blue arrows)  are arising from a focal point on the  pleura ( orange arrow) which shows a miniature Subpleural nodule. (b) In another patient LUS  showed consolidation( between white arrows) ,associated  posteriorly with converging B lines( yellow arrows). Blue arrows represent ribs.

Figure 6; True negative on LUS. CECT chest in a patient with suspected PTB due to complaints of loss of weight / appetite for 1 year; and fever with dry cough for 5 months. (a)Mediastinal window (axial section) (b)Sagittal MPR showed a grossly dilated esophagus (blue arrow) with abrupt narrowing at its lower end (white arrow) suggestive of Achalasia Cardia.

Table 1: DESCRIPTION OF LUS APPEARANCES IN PULMONARY TB IN ADULTS

TABLE 2: LUNG ULTRASOUND FINDINGS DETECTED IN SUSPECTED AND PROVEN PULMONARY TUBERCULOSIS IN ADULTS (n = 100)

B lines* - > 3 B lines in one intercostal space.

TABLE 3: REGIONAL DISTRIBUTION OF LUNG ULTRASOUND FINDINGS IN SUSPECTED OR PROVEN PULMONARY TUBERCULOSIS PATIENTS ( n= 100 patients)

*There is significant statistical association seen between the presence of these LUS findings namely (consolidation, SUN, Isolated  “Converging” B lines) and regional (anterior , posterior and lateral )zones with p value < 0.5 & No significant statistical association seen between the presence of cavities and regional zones (anterior, posterior and lateral ) with p value being 0.09 (i.e. > 0.05).

#There is no significant association between presence of LUS findings namely (consolidation, SUN, Isolated  “Converging” B lines, cavities) and craniocaudal zones( upper , mid and lower zones) , p value being 0.4 i.e., > 0.05.

Table 4: EXTENDED” LUS FINDINGS OF FALSE NEGATIVE* PTB PATIENTS (PROVEN OR SUSPECTED)

False negative*on LUS- due to absence of 2 out of 3 findings of consolidation/ SUN/ “ isolated” converging B lines

Table 5: Sensitivity and specificity of lung ultrasound in diagnosing pulmonary tuberculosis

Sensitivity: True positive / True positive and false positive 59/94 = 62.8%

Specificity: 5/6=83.3%

Table 6: Sensitivity and specificity of lung ultrasound (including miliary pattern) in diagnosing pulmonary tuberculosis

Pulmonary TB is a global public health problem which can be eradicated. However its diagnosis is still difficult, especially in resource-limited endemic areas; and it is unclear whether LUS can help in the diagnosis of PTB. In the current study, LUS was found to have a sensitivity of 62.8% and specificity was 83.3% in the diagnosis of PTB; and hence can be an important supplementary tool for its diagnosis.

LUS findings were detected in 98/100 (98%) patients of suspected PTB  12/12 (100%) microbiologically positive patients and 94/94 (100%) patients of “proven PTB” in our study. Similarly Agostinis et al found sonographic abnormalities in 100% patients of PTB, while Gianelli et al found pathological LUS findings in 96.6% (56/58) of PTB patients, and in 79/82 (96.3%) patients of suspected TB (only in 3 patients no abnormality was detected) [2,3].

The most common LUS finding in this study was “ Converging” B lines (seen in 77% patients), which has not been described in the available literature on LUS on PTB till date, including a meta-analysis by Bigio et al (2021) [9]. The bronchioles (lobular /terminal/respiratory bronchioles)filled with pus/caseous content along with peribronchiolar inflammation and granulomas in PTB, probably produce  the “ tree” portion  of the “ tree in bud pattern” seen on CT scan in PTB [10]. We postulate that probably these  adjacent centrilobular branching opacities of adjacent “trees” converge towards a focal point on the pleura (supplemental Fig. 1) to form “Converging” B lines . These converging B lines are directed obliquely, instead of perpendicular to the pleura and many of them “criss-cross” each other on real time US. However further studies are needed to confirm this hypothesis.

Converging B lines were “isolated” giving an “ inverted V- shaped appearance” or were seen in association with consolidations / SUN. In some patients “ isolated” Converging B lines arising from a focal point on the pleura seemed to be associated with an ill-defined, miniature SUN.

B lines (as conventionally described in literature), were detected in 64% of the patients with suspected or proven pulmonary tuberculosis in this study, while Fentress et al detected pathological B lines in 89.2% of proven PTB patients [5].

Consolidations were seen in 77%  adult  patients on LUS in PTB in this study . Fentress et al studied LUS in  proven PTB and detected consolidation sonographically in 80.4% patients [5].  Wagih et al studied two groups of pulmonary TB patients based on HIV status. They found that 100% patients in group I (HIV negative group) and 88% patients in group II (HIV positive group) showed consolidations [7].

The lower detection rate of consolidations probably reflects that severity of pulmonary disease in this study was less; and or the number of consolidations reaching the pleural surface was less. For probably the same reason, bilateral consolidations were much less frequent (18 % patients) than consolidations in the left or right lung ( 27 %) ; while Fentress et al had detected consolidation bilaterally in 35.3% cases of proven PTB on LUS [5].

The presence of consolidation alone cannot be useful in making a diagnosis of pulmonary tuberculosis. Malla et al reported consolidation on LUS in children with bacterial, viral and mixed pneumonia [11]. Montuori et al also stated that irregular consolidations of relatively homogeneous texture identified by LUS in PTB were indistinguishable from bacterial pneumonia [4]. Similarly Agostinis et al reported consolidation in only 28/60(46.7%) of PTB patients, which is indistinguishable from bacterial pneumonia [3].

Sensitivity of consolidations was found to be 69% and specificity 57% in this study. Montouri et al found a sensitivity of 78.4% for consolidations and 45.1% for apical consolidations. Sensitivity of multiple consolidations was 46.8% in the present study ,similar to Montouri et al who found 43.1% sensitivity for multiple consolidations [4].

In our study the largest dimension of consolidation seen on LUS in patients with suspected or proven  pulmonary tuberculosis was between 0.4 cm to 7 cm. The mean size of consolidation was 4.2x3.8cms . The maximum dimension of  60% of consolidations detected on lung ultrasound was  between 1 to 4cms. In a study conducted by Fentress et al , on 51 adult subjects with pulmonary TB with a mean age of 33.7 years, the average size of lesion per patient was 2.9 cm [5]. Heuvelings et al mentioned the median size of consolidation in confirmed paediatric tuberculosis patients as 2.78 cm [12].  In a study conducted in children to differentiate bacterial and viral pneumonia by Malla et al , mean size of consolidation was 2.96 cms on LUS  [11]. Hence it appears that the size of a consolidation does not correlate with its etiology.

In our study 50% patients with suspected or proven pulmonary tuberculosis showed presence of SUN, compared to other authors who reported a higher observation of SUN . Probably the higher proportion of young female patients (M:F= 29:71) in this study, made detection of SUN more difficult due to the increased thickness of adipose tissue in the anterior chest wall, resulting in poor penetration of US [supplemental Fig. 2]. Thus SUNs (measuring < 1.0 cm) in the anterior and or lateral zones of lung may have been obscured.

Agostinis et al found SUN in 97% of adult PTB patients ( with M:F = 27:33) , and it was the most frequent chest ultrasound sign in patients with active PTB in sub-Saharan Africa, was often bilateral and randomly distributed [3]. Subpleural nodules are commonly observed in lower respiratory tract infections such as COVID-19 , Pneumocystis pneumonia, and bacterial pneumonia [3,13]. Montuori et al found 73% of adult PTB patients(male : female = 37: 14 ) had small subpleural lesions. [4].  Fentress et al had who studied 51 proven PTB patients with 68.6% male and 31.4% females also found SPC(subpleural consolidation) which is analogous to SUN, in 80.4% adults [5].  Wagih et al who studied 50 patients with sputum positive PTB ,age group 17 to 61 years , with 96% male and 4% female patients stated that 100% patients had SUN [6]. In the present study also, all the 12 patients (100%) who were sputum AFB /CBNAAT positive, showed presence of solitary SUN.

In our study the size of subpleural nodules detected ranged between 1.0 to 10.0mm. Only 4.2% SUN  measured between 7.0 to 10.0mm, while 52.9% SUNs measured between 3.0 to 5.0mm. So the small size of the majority of SUNs, may have prevented their easy identification. Size of SUNs has not been measured by other authors; only Montuori et al reported that maximal pleural extension and depth of SUN measured 8 ± 3mm (range 3–15) x 8 ± 2 (range 3–14) in adult TB patients [4].

The high reported prevalence of SUN indicates that they constitute an important indicator of TB [3].  Fentress et al stated that  patchy or fibronodular consolidations such as those often found in PTB may contact the pleural surface in only a patchy or limited manner leading to SPC detection [5].  Agostinis et al stated that subpleural nodule(SUN) or SPC can be a very specific finding for TB in children, especially in areas endemic for TB [3].

According to Fentress et al, SPCs are also described in pulmonary embolism, early bacterial pneumonia and viral pneumonia e.g. COVID-19 [5].  Montuori et al stated that among 17 adult patients with SUN but no PTB diagnosis, 7 had a previous TB infection and 6 were immunocompromised. Furthermore on CT scan, 7 of them had pulmonary disease such as emphysema with bronchiectasis, pulmonary fibrosis, pulmonary neoplasm and aspergillosis [4].  In previous studies, SUN were also described in cryptococcosis [12].  Agositinis et al stated that a sign similar to SUN, has been found in sarcoidosis, Cytomegalovirus and Pneumocystis pneumonia [3]. Thus reliance on only SUN , as a sensitive marker for the diagnosis of pulmonary TB especially in adult female patients, may be inappropriate.

Montouri et al who studied 51 patients with confirmed PTB reported a specificity of 96% when apical consolidations and subpleural nodules were found in the same patient, with sensitivity of 31%. Similarly  in this study presence of both these findings in the same patient had a  specificity of 85.7%  and sensitivity of 35.5%. Montouri et al attained a sensitivity of 86% and specificity of 63% for LUS,  when the LUS was considered positive with the detection of at least one of these LUS signs ;similarly in our study a sensitivity of 83.9% and specificity of 57% was attained [4]. Fentress et al stated that SPCs and consolidation (96%) may be a sensitive marker for diagnosing PTB in adults. In our study presence of both SUN and consolidation combined had sensitivity of 35.5% while presence of either SUN or consolidation had a sensitivity of 83.9% [5].   

Recognition of cavities on LUS was difficult in the initial phase of our study of 100 patients. In our study cavities were the least common finding seen only in 7% patients on LUS, in cases with proven or suspected pulmonary TB.  Similarly Fentress et al detected cavity on LUS in 5.9% cases [5]. In this study cavities were most commonly detected on LUS in the posterior aspects of lung in 5 out of 7 patients , consistent with known predominant involvement of posterior/ apico posterior segment of upper lobes and apical segment of lower lobes, in pulmonary TB. The sensitivity of LUS was rather low (39.5%) , as many cavitated consolidations did not reach the pleural surface [14].

Fentress et al stated that no patients with cavitary disease would have been missed by a composite finding of subpleural nodule or lung consolidation [5]. In our study 14 out of the 28 patients who showed cavities on CR would not be missed, if presence of both SUN and consolidation were considered as composite finding.

In our study, miliary pattern was seen in a few areas of the lungs on LUS, in 43 % of the patients with suspected or proven pulmonary TB .  This was better appreciated on real time US images . Agostinis et al found 6.7% patients with miliary pattern [3].  Fentress et al, Wagih et al and Montouri et al have not reported miliary pattern on LUS [4-6].   In 35 patients who were false negative on LUS in this study, B lines (57.1%) and miliary pattern ( 40%) were the commonest LUS findings. Adding “miliary pattern” as an additional criteria for considering a patient LUS positive for PTB , the sensitivity of LUS increased to 74.5%  while the specificity decreased to  66.7%.

Due to the presence of typical matted periportal abdominal lymph nodes, 6 patients who were LUS negative for PTB out of 15 patients with abdominal lymph nodes , were considered as  LUS positive on “extended” LUS for PTB. On “ extended LUS”  sensitivity of LUS increased slightly (69.1%) and specificity  of LUS  remained the same (83.3%).

Limitations-

A major limitation of this study was that adequate number of patients with non- tubercular, non-resolving chest infections were not evaluated on LUS. So studies are needed to evaluate LUS in cases of other chronic pulmonary infections in the future, to validate our results. The second limitation was that clinic-radiological response to ATT given empirically was considered as “proven” PTB; due to scarcity of laboratory facilities to prove the diagnosis of PTB in low-resource settings. Thirdly, US beam penetration may be affected due to fatty breasts in young females who comprised two thirds of our patients. This may be a limiting factor in the LUS detection of abnormalities such as small subpleural nodule (SUN), which have been relied upon as an important indicator of PTB in adults, till date.

LUS is an appealing alternative to chest radiography or CT scan for the diagnosis and screening of PTB, as it has a steep learning curve, is readily available, is not expensive and does not use ionizing radiation .“Converging” B lines were a unique new LUS feature observed in PTB in adults in the present study,  more frequent than SUN reported in similar studies till date. They may be analogous to the stems or branches of “tree in bud” appearance  seen on CT. However LUS is able to detect only the subpleurally located features; and the interpretation of LUS findings in an adult patient with suspected PTB still remains a challenge.

                Abbreviations:

AFB                    -       Acid fast bacilli

ATT                          -        Anti Tubercular Therapy

CBNAAT            -        Cartridge-based nucleic acid amplification test

CMV                    -       Cytomegalovirus

CR                             -        Chest Radiograph

CT                         -           Computed Tomography

EPTB                        -        Extra Pulmonary Tuberculosis

HIV                      -       Human immunodeficiency virus

LALZ                        -        Left Anterior Lower Zone

LAMZ                       -        Left Anterior Mid Zone

LAUZ                        -        Left Anterior Upper Zone

LLLZ                        -        Left Lateral Lower Zone

LLUZ                        -        Left Lateral Upper Zone

LPLZ                         -        Left Posterior Lower Zone

LPMZ                        -        Left Posterior Mid Zone

LPUZ                        -        Left Posterior Upper Zone

LUS                           -        Lung Ultrasound

PTB                           -        Pulmonary Tuberculosis

RALZ                        -        Right Anterior Lower Zone

RAMZ                       -        Right Anterior Mid Zone

RAUZ                       -        Right Anterior Upper Zone

RLLZ                        -        Right Lateral Lower Zone

RLUZ                        -        Right Lateral Upper Zone

RPLZ                        -        Right Posterior Lower Zone

RPMZ                       -        Right Posterior Mid Zone

RPUZ                        -        Right Posterior Upper Zone

RUZ                          -        Right Upper Zone

SUN                          -        Subpleural Nodule

1. World Health Organization. Global tuberculosis report 2016 [Internet]. World Health Organization; 2016 [cited 2022 Dec 6]. 142 p. Available from: https://apps.who.int/iris/handle/10665/250441
2. Giannelli F, Cozzi D, Cavigli E, et al. Lung ultrasound (LUS) in pulmonary tuberculosis: correlation with chest CT and X-ray findings. J Ultrasound. 2022;25(3):625-634.
3. Agostinis P, Copetti R, Lapini L, Badona Monteiro G, N'Deque A, Baritussio A. Chest ultrasound findings in pulmonary tuberculosis. Trop Doct. 2017;47(4):320-328.
4. Montuori M, Casella F, Casazza G, et al. Lung ultrasonography in pulmonary tuberculosis: A pilot study on diagnostic accuracy in a high-risk population. Eur J Intern Med. 2019;66:29-34.
5. Fentress M, Ugarte-Gil C, Cervantes M, et al. Lung Ultrasound Findings Compared with Chest X-Ray Findings in Known Pulmonary Tuberculosis Patients: A Cross-Sectional Study in Lima, Peru. Am J Trop Med Hyg. 2020;103(5):1827-1833.
6. Wagih K, Abdelhalim H, El Y. Chest ultrasound patterns in patients with pulmonary tuberculosis with and without HIV. The Egyptian journal of tuberculosis and chest diseases. 2020;69:421–8.
7. Steingart KR, Henry M, Ng V, et al. Fluorescence versus conventional sputum smear microscopy for tuberculosis: a systematic review [published correction appears in Lancet Infect Dis. 2006 Oct;6(10):628]. Lancet Infect Dis. 2006;6(9):570-581.
8. Hunter L, Bélard S, Janssen S, van Hoving DJ, Heller T. Miliary tuberculosis: sonographic pattern in chest ultrasound. Infection. 2016;44(2):243-246.
9. Bigio J, Kohli M, Klinton JS, et al. Diagnostic accuracy of point-of-care ultrasound for pulmonary tuberculosis: A systematic review. PLoS One. 2021;16(5):e0251236.
10. Im JG, Itoh H. Tree-in-Bud Pattern of Pulmonary Tuberculosis on Thin-Section CT: Pathological Implications. Korean J Radiol. 2018;19(5):859-86
11. Malla D, Rathi V, Gomber S, Upreti L. Can lung ultrasound differentiate between bacterial and viral pneumonia in children?. J Clin Ultrasound. 2021;49(2):91-100.
12. Heuvelings CC, Bélard S, Janssen S, et al. Chest ultrasonography in patients with HIV: a case series and review of the literature. Infection. 2016;44(1):1-10.
13. Muljadi R, Koesbandono, Octavius GS. A systematic review and meta-analysis of diagnostic test accuracy of chest ultrasound in diagnosing pediatric pulmonary tuberculosis. Pediatr Pulmonol. 2024.doi:10.1002/ppul.27044
14. Cozzi D, Bartolucci M, Giannelli F, Cavigli E, Campolmi I, Rinaldi F, Miele V. Parenchymal Cavitations in Pulmonary Tuberculosis: Comparison between Lung Ultrasound, Chest X-ray and Computed Tomography. Diagnostics. 2024; 14(5):522.