What are the differential diagnoses for a miliary (diffuse uniform 1‑5 mm nodular) pattern on chest CT in a child?

Differential Diagnoses for Miliary Pattern on Chest CT in Children

In children presenting with a miliary (1-5 mm diffuse nodular) pattern on chest CT, infectious etiologies—particularly tuberculosis and endemic fungal infections—are the primary considerations, followed by less common causes including metastatic disease and sarcoidosis.

Primary Infectious Etiologies

Tuberculosis

• Tuberculosis remains the classic cause of miliary lung disease and should be the first consideration in any child with this pattern, particularly in endemic areas or with known TB exposure 1, 2, 3.
• CT characteristically shows randomly distributed 1-3 mm nodules throughout both lungs, often with a mixture of sharply and poorly defined margins 2.
• Upper and middle lung zone predominance may be observed 4.
• Additional CT features include coalescing nodular lesions, lymph node enlargement (often with calcification), and potential cavitation 3.
• CT is superior to chest radiographs in detecting early nodular lesions and lymphadenopathy 3.

Endemic Fungal Infections

• Coccidioidomycosis is a leading cause of miliary disease in endemic regions (southwestern United States, northern Mexico), accounting for 53.7% of cases in one endemic center 5.
• Serum eosinophilia (>500 eosinophils/μL) strongly predicts coccidioidomycosis over other etiologies 5.
• Elevated β-D-glucan levels are also strongly predictive of coccidioidomycosis in patients with miliary patterns 5.
• Blastomycosis can present with diffuse miliary nodular patterns in endemic areas (Mississippi and Ohio River valleys, Great Lakes region) and may be fatal without prompt treatment 6.
• Histoplasmosis should be considered in endemic regions (central and eastern United States) 7.
• Candida albicans can cause miliary patterns, particularly in immunocompromised children 4.

Other Infectious Causes

• Atypical mycobacterial infections (e.g., Mycobacterium simiae) should be considered 5.
• These infections commonly produce false-positive PET scan results due to granulomatous inflammation 7.

Non-Infectious Etiologies

Malignancy

• Metastatic disease accounted for 17.1% of miliary patterns in one series, though this is less common in children than adults 5.
• Metastases can occur from virtually any primary malignancy but are uncommon when no known extrapulmonary cancer exists 8.
• Metastatic adenocarcinoma may show more sparse nodules in the lung periphery 4.
• Multiple cyst-like lesions (up to 12 mm) on CT should raise suspicion for metastatic disease 4.
• Lymphoma can rarely present with a miliary pattern 5.

Sarcoidosis

• Sarcoidosis can produce a miliary pattern with characteristic features 4.
• Key distinguishing CT features include: peripheral distribution of nodules, increased thickened interlobular septa, and notable thickening of interlobar fissures 4.
• Upper and middle lung zone predominance may be present 4.
• Produces false-positive PET scan results due to active inflammation 7.

Diagnostic Algorithm

Initial Clinical Assessment

• Obtain detailed exposure history: TB contacts, travel to endemic fungal regions, immunocompromising conditions, known malignancy 5.
• Geographic location is critical—coccidioidomycosis predominates in southwestern US, while TB is more common in areas with close proximity to TB-endemic regions 5.

Laboratory Evaluation

• Serum eosinophil count: >500 eosinophils/μL strongly suggests coccidioidomycosis 5.
• β-D-glucan levels: Elevation supports fungal etiology, particularly coccidioidomycosis 5.
• Tuberculin skin testing or interferon-gamma release assays for TB 3.
• Fungal serologies based on endemic exposure 7.

Imaging Characteristics to Guide Diagnosis

• Random nodule distribution with upper/middle zone predominance: Consider TB or sarcoidosis 4.
• Peripheral nodule distribution with thickened septa and fissures: Strongly suggests sarcoidosis 4.
• Cyst-like lesions (>12 mm) accompanying nodules: Consider TB or metastatic disease 4.
• Coalescing nodules, consolidation, or cavitation: More typical of TB 3.
• Air trapping on CT: Common in TB and may increase during follow-up 3.

Tissue Diagnosis

• Documentation requires organism recovery from body fluids or tissue since skin testing and serology are unreliable for active disease 6.
• Bronchoalveolar lavage is appropriate for accessible diagnosis 3.
• Consider image-guided biopsy when diagnosis remains uncertain after non-invasive testing 9.

Critical Management Considerations

Empiric therapy should be initiated based on clinical context while awaiting definitive diagnosis, as delays can be fatal, particularly in immunocompromised patients 6.

• In TB-endemic areas with compatible clinical presentation, empiric anti-TB therapy is appropriate 5.
• In coccidioidomycosis-endemic regions with eosinophilia or elevated β-D-glucan, empiric antifungal therapy should be strongly considered 5.
• 90.91% of coccidioidomycosis patients and 75% of TB patients received appropriate empiric therapy in one series, highlighting the importance of geographic and clinical risk stratification 5.

Important Caveats

• The Fleischner Society guidelines explicitly state they are not appropriate for children, as lung cancer is rare in patients under 35 years, and infectious causes are far more likely 1.
• Serial CT should be minimized in young patients to reduce radiation exposure 1.
• PET scan has significant limitations with false-positives in infections (TB, endemic mycoses, sarcoidosis) and should not be the primary diagnostic modality 7.
• HIV status significantly affects differential diagnosis, with immunocompromised patients at higher risk for disseminated infections 2.