European Journal of Cardiovascular Medicine

Miliary shadows on chest radiographs, although not uncommon, represent a significant diagnostic challenge as they are associated with a wide spectrum of pulmonary and systemic diseases. These shadows consist of numerous small, discrete nodules typically ranging from 1 to 3 mm in diameter, diffusely distributed throughout both lungs with well-defined margins (1). The term "miliary" is derived from the resemblance of these nodules to millet seeds, and they are usually indicative of hematogenous or lymphatic dissemination of disease within the pulmonary interstitium (1,2).

Over 80 conditions are known to present with miliary nodular patterns on imaging. Common causes include miliary tuberculosis, disseminated metastases (particularly from adenocarcinomas), various diffuse parenchymal lung diseases (DPLDs) such as sarcoidosis and hypersensitivity pneumonitis, fungal infections like histoplasmosis and disseminated candidiasis, and occupational diseases such as silicosis (3-6). Among these, miliary tuberculosis remains the most frequently reported etiology in endemic regions (7,8). However, the overlap in radiographic appearances often complicates differentiation based solely on imaging.

Traditionally, sputum smear microscopy for Acid Fast Bacilli (AFB) has been the first-line diagnostic tool for tuberculosis. Yet in many patients with miliary patterns, especially those who are sputum smear-negative, alternative diagnostic modalities become essential. High-resolution computed tomography (HRCT) provides better characterization of nodular patterns—whether random, centrilobular, or perilymphatic—and helps narrow the differential diagnoses (9,10).

Fiberoptic bronchoscopy (FOB), particularly when combined with procedures such as bronchoalveolar lavage (BAL), bronchial washings (BW), transbronchial needle aspiration (TBNA), and endobronchial biopsy (EBB), plays a pivotal role in the diagnosis of miliary lung diseases. These minimally invasive procedures allow for the collection of cellular and microbiological samples directly from the affected lung regions, thereby increasing diagnostic yield, especially in cases where conventional sputum tests are inconclusive (11-14). Previous studies have emphasized the utility of bronchoscopy in the diagnosis of miliary tuberculosis and malignancy when sputum smears fail to yield results (15).

In this context, the present study was conducted to investigate the etiological profile of patients presenting with miliary patterns on chest radiograph and to evaluate the diagnostic utility of fiberoptic bronchoscopy and its associated techniques in establishing definitive diagnoses

Study Design and Setting

This was a hospital-based, cross-sectional observational study conducted in the Department of Pulmonary Medicine, S.C.B. Medical College and Hospital, Cuttack, Odisha. The study duration was one year, from August 2017 to August 2018. Ethical clearance and written informed consent were obtained prior to the inclusion of participants.

Study Population

A total of 50 consecutive adult patients aged between 18 and 60 years, presenting with a chest radiograph suggestive of miliary pattern and admitted to the Department of Pulmonary Medicine, were enrolled for this study. All patients were sputum smear-negative for Acid Fast Bacilli (AFB) on direct microscopy.

Inclusion Criteria

1. Patients aged between 18 and 60 years.
2. Chest X-ray showing bilateral miliary pattern involving more than two-thirds of the lung volume.
3. Negative sputum smear for AFB on direct microscopy.
4. Patients who provided informed written consent to undergo fiber-optic bronchoscopy (FOB).

Exclusion Criteria

1. Patients younger than 18 years or older than 60 years.
2. Sputum smear-positive cases for AFB on direct microscopy.
3. Patients who refused to consent for FOB.
4. Patients who were medically unfit to undergo FOB, including:
• Unstable cardiovascular status
• Critically ill condition
• Severe hypoxemia

Clinical Evaluation

All patients underwent a detailed clinical assessment including history-taking and physical examination. Relevant baseline investigations such as complete blood count, liver and renal function tests, serum electrolytes, viral markers (HIV, HBsAg, HCV), and blood glucose levels were performed. Chest X-ray (posterior-anterior view), ultrasonography, and High-Resolution Computed Tomography (HRCT) thorax were done in all patients.

Bronchoscopy Procedure

Flexible fiber-optic bronchoscopy was performed under local anesthesia using 4% lignocaine nebulization and oropharyngeal spray. Nasal passages were anesthetized with lignocaine jelly, and supplemental oxygen was administered throughout the procedure. The procedure was carried out transnasally or transorally using a mouth gag, depending on anatomical feasibility.

The bronchoscope was advanced systematically into the segmental bronchi to inspect for mucosal abnormalities, granulations, ulcerations, extrinsic compression, or endobronchial growths.

Specimen Collection and Processing

• Bronchoalveolar Lavage (BAL): BAL was performed by instilling 100 mL of sterile normal saline in 20 mL aliquots into the subsegmental bronchus, followed by suction and collection into a sterile trap.
• Bronchial Washings (BW): In addition to BAL, 10–20 mL of saline was instilled and suctioned for bronchial washings.
• Transbronchial Needle Aspiration (TBNA): Performed when mediastinal lymphadenopathy was detected, using standard aspiration techniques under bronchoscopic visualization.
• Endobronchial Biopsy (EBB): Obtained from visually suspicious lesions (ulcerated or irregular mucosa).

All collected samples were sent for:

• Microbiological Testing: AFB staining (Ziehl-Neelsen), culture for Mycobacterium tuberculosis, and CBNAAT (Cartridge-Based Nucleic Acid Amplification Test).
• Cytological Evaluation: For malignant cells and differential cell count.
• Histopathological Analysis: On biopsy samples for definitive diagnosis.

Post-Procedure Monitoring

Patients were monitored for 24–48 hours post-bronchoscopy for any procedure-related complications including bleeding, hypoxia, hypotension, chest pain, fever, or pneumothorax. Appropriate interventions were initiated if complications arose.

Data Analysis

Data were recorded, tabulated, and analyzed using descriptive statistics. Diagnostic yield of FOB and its components (BAL, TBNA, EBB) were calculated to evaluate the effectiveness in identifying the underlying etiology of miliary shadows on chest radiograph.

A total of 50 patients were enrolled in the study. The demographic, clinical, and occupational data were analyzed and summarized as follows:

Age and Sex Distribution

Out of 50 patients, 29 (58%) were males and 21 (42%) were females. The most common age group affected was 35–54 years, comprising 54% of the population (15 males and 12 females), followed by the 55–70 age group (28%, including 9 males and 5 females). Only 18% were between 15–34 years, and there were no patients above 70 years of age. The mean age was 46.16 ± 12.47 years (Table 1).

Table 1: Age and Sex Distribution

Distribution According to BMI

Analysis of Body Mass Index revealed that 82% (41/50) of the patients had a BMI greater than 18.5 kg/m². A total of 3 males (6%) and 6 females (12%) had a BMI below 18.5 kg/m². The mean BMI among the participants was 20.73 kg/m² (Table 2).

Table 2: Distribution According to BMI

Co-Morbidities

The most frequently observed co-morbidity was Diabetes Mellitus, seen in 7 patients (14%), followed by Hypertension in 5 patients (10%) and Chronic Kidney Disease in 3 patients (6%). Only 1 patient (2%) had Chronic Liver Disease (Table 3).

Table 3: Co-Morbidities

Occupation Distribution

Occupational data showed that the majority of the participants were housewives (40%), followed by farmers (20%). Other occupations included teachers and stone/tile cutters (8% each), labourers (6%), and electricians and fish sellers (4% each) (Table 4).

Table 4: Occupation Distribution

Past History

Regarding past medical history, 10% of patients had a prior diagnosis of tuberculosis (PTB or EPTB), and 8% had a history of anti-tubercular therapy (ATT) intake. Notably, 16% had a contact history with TB patients. Malignancies were observed in 10% of patients, including thyroid (2%) and breast cancer (4%) (Table 5).

Table 5: Past History

The present study was conducted to evaluate the diagnostic efficacy of fiber-optic bronchoscopy (FOB) in patients with chest radiographs suggestive of a miliary pattern. Miliary shadows on chest imaging can result from a wide variety of conditions, including infections like tuberculosis and fungal diseases, malignancies, and diffuse parenchymal lung diseases (1,2).

In our study, the most frequent age group affected was 35–54 years, with a mean age of 46.16 years, and males were more commonly affected (Table 1). This observation is consistent with studies by Voloudaki et al. and Jin et al., who also reported similar age distributions (3,4). Males had a higher incidence, possibly due to increased exposure to environmental and occupational risk factors.

A significant proportion of patients (82%) had a BMI above 18.5 kg/m², with a mean of 20.73 kg/m² (Table 2), comparable to the findings of Sang-Man Jin et al., who reported a mean BMI of 21.8 kg/m² in their cohort (4). Diabetes mellitus emerged as the most frequent comorbidity (14%), which is noteworthy as immunosuppression can predispose patients to disseminated infections like miliary tuberculosis (5).

Radiologically, all patients had micronodular opacities on chest radiographs, with a predominance of symmetrical distribution across all lung zones in 70% of cases (Tables 9 and 10). High-resolution computed tomography (HRCT) revealed random distribution of micronodules in 84% of cases, with mediastinal lymphadenopathy and “tree-in-bud” appearance seen in 48% and 46%, respectively (Table 11 and 12). These findings are characteristic of hematogenous spread, as seen in miliary tuberculosis and metastatic disease (6,7).

FOB findings were normal in 50% of the patients, while inflammatory changes, mucosal irregularities, and ulcerations were observed in the remaining (Table 14). This supports the idea that visual inspection alone during bronchoscopy may not always be sufficient, and adjunctive sampling techniques are essential (8). Bronchoalveolar lavage (BAL) fluid differential counts revealed lymphocytic predominance in 70% of cases (Table 15), consistent with granulomatous diseases such as sarcoidosis and hypersensitivity pneumonitis (9).

Cytological examination of BAL revealed malignant cells in 18% of cases, indicating the role of bronchoscopy in identifying lung malignancies that mimic infectious pathologies on radiographs (Table 16). Microbiological confirmation of Mycobacterium tuberculosis using CBNAAT was achieved in 44% of cases, and culture was positive in 40% (Table 17). These findings echo previous reports that highlight the importance of BAL in sputum-negative cases of miliary tuberculosis (10,11).

Endobronchial biopsy (EBB) revealed adenocarcinoma in 20% and tuberculosis in 12% of cases (Table 18), reinforcing the notion that histopathological diagnosis is essential in differentiating miliary TB from malignancy. Transbronchial needle aspiration (TBNA) cytology showed caseating granulomas in 14%, suspicious cells in 10%, and anthracotic pigments in 4% of patients (Table 19), reflecting a wide differential and supporting FOB's diagnostic utility (12).

FOB-related complications occurred in 30% of cases, with bleeding being the most common (10%) (Table 20). This rate is within acceptable limits and consistent with other large bronchoscopy series (13).

Overall, the final diagnosis revealed miliary tuberculosis in 50% of cases, followed by adenocarcinoma (14%) and miliary metastasis (10%) (Table 21). These findings align with previous literature, which consistently identifies miliary TB as the leading cause of diffuse micronodular disease, although neoplastic etiologies must not be overlooked (14,15).

Miliary patterns on chest X-ray present a diagnostic challenge due to their association with diverse pathologies. This study highlights that fiber optic bronchoscopy, when combined with BAL, TBNA, and biopsy, significantly enhances diagnostic accuracy. Miliary tuberculosis emerged as the most common etiology, followed by adenocarcinoma and metastatic disease. With an overall diagnostic yield of 86%, FOB proves to be a safe and valuable tool in establishing the underlying cause of miliary lung shadows, especially in sputum smear-negative cases.

1. Sharma SK, Mohan A, Sharma A. Challenges in the diagnosis & treatment of miliary tuberculosis. Indian J Med Res. 2012 May;135(5):703–30.
2. Ray S, Talukdar A, Kundu S, Khanra D, Sonthalia N. Diagnosis and management of miliary tuberculosis: current state and future perspectives. Ther Clin Risk Manag. 2013;9:9–26.
3. Sharma SK, Mohan A. Miliary Tuberculosis. Microbiol Spectr. 2017 Mar;5(2):10.1128/microbiolspec.TNMI7-0013-2016.
4. Taleuzzaman M, Kumar V. Recent Complications and Issues in Tuberculosis Treatment. Recent Pat Antiinfect Drug Discov. 2017;12(2):138–46.
5. Sharma SK, Mohan A, Sharma A, Mitra DK. Miliary tuberculosis: new insights into an old disease. Lancet Infect Dis. 2005 Jul;5(7):415–30.
6. Sharma SK, Mohan A, Sharma A. Miliary tuberculosis: A new look at an old foe. J Clin Tuberc Other Mycobact Dis. 2016 Mar 18;3:13–27.
7. Kwon HH, Baek SH, Park SH. Miliary tuberculosis and necrotizing tuberculous fasciitis--an unusual coexistence in a rheumatoid arthritis patient. Int J Rheum Dis. 2010 May;13(2):171–4.
8. Irfan M, Hussain SF, Jabeen K, Islam M. Drug susceptibility pattern of Mycobacterium tuberculosis in adult patients with miliary tuberculosis. Trop Doct. 2007 Jul;37(3):182–4.
9. Swaminathan S, Padmapriyadarsini C, Ponnuraja C, Sumathi CH, Rajasekaran S, Amerandran VA, et al. Miliary tuberculosis in human immunodeficiency virus infected patients not on antiretroviral therapy: clinical profile and response to short-course chemotherapy. J Postgrad Med. 2007 Oct-Dec;53(4):228–31.
10. Adzic T, Pesu D, Stojsic J, Nagorni-Obradovi L, Stevi R. Specific synovitis of a knee as the first manifestation of miliary tuberculosis. Pneumologia. 2008 Jul-Sep;57(3):156–7.
11. Taniguchi H, Izumi S. [Case of miliary tuberculosis during treatment with infliximab for rheumatoid arthritis]. Kekkaku. 2008 May;83(5):431–4. Japanese.
12. Shah A, Panjabi C, Maurya V, Khanna P. Multidrug resistant miliary tuberculosis and Pott's disease in an immunocompetent patient. Saudi Med J. 2004 Oct;25(10):1468–70.
13. Dagli CE, Guler E, Bakan V, Atilla N, Koksal N. Miliary tuberculosis accompanying paravertebral tuberculosis abscess in an adolescent. J Infect Dev Ctries. 2009 Jun 1;3(5):402–4.
14. Dones P, Di Gangi M, Failla MC, Genova S, Giannitto C, Corsello G, et al. Intestinal tuberculosis in a child living in a country with a low incidence of tuberculosis: a case report. BMC Res Notes. 2014 Oct 27;7:762.
15. API Consensus Expert Committee. API TB Consensus Guidelines 2006: Management of pulmonary tuberculosis, extra-pulmonary tuberculosis and tuberculosis in special situations. J Assoc Physicians India. 2006 Mar;54:219–34.