Solitary Pulmonary Nodule

Meet the patient

A 62-year-old smoker (40 pack-years) had a CT chest for haematuria workup, and the report lands on your desk: an 8 mm spiculated nodule in the right upper lobe. He feels entirely well. The question you now own is the question that owns every solitary pulmonary nodule: is this a curable early lung cancer, or a benign scar?[1][6]

Two facts decide his next three months: is the nodule over or under 3 cm? (over 3 cm is a mass, presumed malignant — a different pathway), and what does a validated risk model say? (Brock, then Herder). Hold those two facts and the whole algorithm slots into place. The apparatus of nodule management — risk models, surveillance intervals, PET, biopsy — exists for exactly one reason: to catch the curable stage IA cancer without over-investigating the great benign majority.[1]

The 3 cm gate — nodule versus mass

A solitary pulmonary nodule is a single rounded opacity 3 cm or less; over 3 cm is a mass, presumed malignant until proven otherwise. That size threshold is the single most examinable number in the topic, and it is the gate to the whole algorithm — because a mass is investigated and staged as probable lung cancer, not worked up as an SPN.[1][5]

The clinical importance of the SPN is not that it is dangerous in itself — most are benign — but that a minority represent early, surgically curable lung cancer. A resected stage IA cancer detected as a nodule has a 5-year survival of over 80 to 90 per cent, compared with under 20 per cent for symptomatic late-stage disease. That survival gap is the entire justification for the surveillance-and-tissue apparatus.[5][6]

[1]

By density — solid, part-solid, ground-glass

Density is the most examinable axis because it tracks the underlying biology. A solid nodule behaves differently from a pure ground-glass nodule, and a part-solid nodule carries the highest malignant risk of all.[1]

[1]

Everyone forgets: a part-solid nodule's solid component is what drives risk, not the total size. The ground-glass is the lepidic (in-situ) tumour spreading along alveolar walls; the solid component is the invasive adenocarcinoma breaching them. Because invasion is what makes a cancer a cancer, a part-solid nodule with a solid component of 6 mm or more is high-risk even if the overall lesion is modest.[1]

The pre-test probability — what every history computes

Patient-level risk factors are the backbone of every risk model and every examiner's first follow-up question, in roughly descending weight:[2][4]

• Age — risk rises steeply after 50; an SPN in a person under 35 is much more likely benign.
• Smoking — the dominant modifiable risk; current over former, pack-years matter.
• Personal history of extrathoracic cancer — especially melanoma, sarcoma, breast, colon, renal (raises the metastasis differential).
• Family history of lung cancer.
• Chronic lung disease — COPD and pulmonary fibrosis both raise lung cancer risk independently of smoking.
• Occupational exposure — asbestos, uranium, radon, silica, chromium, nickel.[2][4]

The nodule-level features that raise malignant probability are the other half of every model: size (cancer probability tracks diameter roughly linearly — under 10 per cent for a 5 mm nodule, around 50 per cent at 2 cm, over 80 per cent at 3 cm in a smoker); spiculated or lobulated border; upper-lobe location; growth on serial imaging; part-solid morphology; and cavitation with a thick, irregular wall (a wall over 15 mm favours malignancy or infection).[1][4]

[1]

The adenocarcinoma sequence and the granuloma

A malignant SPN is most often a pulmonary adenocarcinoma travelling along a defined sequence. The earliest lesion is atypical adenomatous hyperplasia (AAH) — a small ground-glass focus. AAH progresses to adenocarcinoma in situ (AIS), a pure lepidic (along-alveolar-wall, non-invasive) growth that appears radiologically as a persistent pure ground-glass nodule; then to minimally invasive adenocarcinoma (MIA); and finally to invasive adenocarcinoma, where a solid component appears within the ground-glass as stroma-invasive tumour breaks through the alveolar wall.[1]

The driver mutations are the examinable molecular detail: in non-smokers and women, EGFR mutations (and ALK or ROS1 rearrangements) are more common; in smokers, KRAS dominates. Each defines the biology and the systemic-therapy options.[1]

The commonest benign SPN worldwide is the healed granuloma — a fibro-calcified scar of an old inhaled infection. Inhaled Mycobacterium tuberculosis (or, in endemic regions, Histoplasma, Coccidioides, or Blastomyces) is walled off by a CD4+ Th1 granuloma; over months to years the centre undergoes caseous necrosis and dystrophic calcification, producing the radiologically dense, central or laminated pattern that ends the work-up.[1]

The hamartoma is the commonest benign lung tumour — a disordered but mature mixture of cartilage, fat, smooth muscle and epithelial clefts. The radiological signature is fat within a nodule (negative Hounsfield units, −40 to −120 HU) with or without "popcorn" chondroid calcification. When fat or popcorn calcification is present, no follow-up is required.[5]

The doubling-time rule and the 26 per cent diameter trap

A tumour grows by exponential cell division, and the examinable metric is the volume doubling time. A malignant nodule doubles in volume every 30 to 400 days; faster than 30 days is too quick for cancer and suggests infection or inflammation; slower than 400 days is too slow for most cancers and suggests benign disease (the lepidic adenocarcinomas are an important exception — they can be very slow).[1][5]

[1]

The differential — name three with three features each

The differential of an SPN is wide, but a structured approach using imaging morphology collapses it. The exam answer requires at least three named differentials, each with at least three distinguishing features.[5]

[1]

The discriminator line: fat or popcorn calcification equals hamartoma (no follow-up); dense central or laminated calcification plus stability equals granuloma; spiculation plus growth in an older smoker equals cancer until tissue says otherwise.[1]

The investigative cascade

The investigative principle is characterise first, then estimate probability, then choose surveillance, further imaging, or tissue. The single most important investigation is a thin-section (1 mm) CT of the nodule, ideally with prior imaging for comparison.[1]

Step 1 — Characterise on thin-section CT. The six features that drive every model: size (greatest diameter), density (solid, part-solid, ground-glass; fat; calcification pattern), border (smooth, lobulated, spiculated), calcification pattern (the four benign patterns are central, laminated, diffuse and popcorn; the malignant-associated are eccentric or stippled), cavitation and wall thickness (over 15 mm favours malignancy or infection; under 5 mm favours benign), and location (upper lobe slightly raises malignant probability).[1]

Step 2 — Compare with prior imaging (the 2-year rule). If a nodule has been stable in size for two years on serial CT, it is benign by the stability rule and needs no further investigation. This single fact — having an old CT — can end the work-up; always hunt for prior imaging before PET or biopsy.[5]

The three risk models — Brock, Mayo, Herder

Step 3 — Quantify pre-test probability with a validated risk model. This is the most examinable step, and the modern BTS 2015 algorithm chains the first two models together.[2][3][4]

[1]

PET-CT and tissue — when and how

Step 4 — Further functional imaging. FDG PET-CT has a sensitivity of about 90 to 95 per cent and specificity 75 to 90 per cent (lower in TB-endemic regions); an SUV max of 2.5 or more is generally suspicious. The classic traps: false positives in TB, histoplasmosis, sarcoid and rheumatoid nodules (all FDG-avid), and false negatives in AIS, MIA and nodules under 8 to 10 mm (low metabolic activity). Contrast-enhanced CT (enhancement over 15 HU suggests malignancy) is largely supplanted by PET but useful when PET is unavailable.[1][3]

Step 5 — Tissue diagnosis, when the post-imaging probability is intermediate-high and the result will change management. The route depends on nodule location and patient fitness:[5]

• Percutaneous transthoracic needle biopsy (TTNB) — best for peripheral nodules. Sensitivity about 90 per cent; principal complication is pneumothorax (about 20 per cent, 5 per cent needing a chest tube).
• Bronchoscopic biopsy — best for central nodules or those reachable by radial EBUS (r-EBUS) or electromagnetic navigation bronchoscopy; lower pneumothorax risk, allows staging of mediastinal nodes in the same sitting.
• Surgical biopsy — when a nodule is high-risk and resectable, diagnosis and treatment combine: a VATS wedge resection with intra-operative frozen section, proceeding to anatomic lobectomy plus mediastinal lymph node dissection if malignant.[5]

[1]

The four-step algorithm

[1] [5]

Fleischner 2017 — incidental solid nodules

[1]

A consultant confession: Fleischner 2017 applies to adults 35 and older with incidental nodules — it does not apply to lung cancer screening (use Lung-RADS), the immunocompromised, patients with a known primary cancer (metastasis in play), or children. The intervals are shorter in heavy-smoking patients and no follow-up is needed for definite benign patterns.[1]

BTS 2015 — the model-driven algorithm

The BTS algorithm is explicitly chained on the risk models, and is the cleanest single examinable pathway:[2][3]

[1]

Lung cancer screening — NLST and Lung-RADS

When the nodule is found on organised low-dose CT screening in a high-risk smoker, neither Fleischner nor BTS applies — instead use the Lung-RADS scheme. Screening is offered to adults aged 55 to 80 (USPSTF) or 55 to 74 (NCCN) who have 30 pack-years or more of smoking and quit within 15 years, with annual low-dose CT.[6]

The National Lung Screening Trial (NLST, NEJM 2011) showed a 20 per cent reduction in lung-cancer mortality and a 6.7 per cent all-cause mortality reduction with three rounds of annual LDCT versus chest X-ray — the single trial that legitimised organised lung-cancer screening. The burden is that about 96 per cent of positive screens are NOT cancer, which is the over-detection cost screening imposes.[6]

Surgical treatment — lobectomy, sublobar, SBRT

Once a malignant SPN is confirmed, the standard operation is anatomic lobectomy with mediastinal lymph node sampling or dissection — both cure and accurate staging.[5]

Sublobar resection (segmentectomy or wedge) is acceptable for small (under 2 cm), peripheral stage IA tumours, especially with limited pulmonary reserve — it must achieve a 2 cm or greater margin or a margin-to-tumour ratio of 1 or more with hilar and mediastinal node sampling. VATS or robotic lobectomy is now standard for early-stage disease, with lower morbidity and equivalent oncologic outcome to thoracotomy. SBRT (stereotactic body radiotherapy) is for medically inoperable early-stage (T1-2aN0) lung cancer — local control over 90 per cent and 3-year survival about 50 to 60 per cent in unfit patients.[1][5]

The TB-endemic twist

In TB- or histoplasmosis-endemic regions (large parts of India, sub-Saharan Africa, South-East Asia), the pre-test probability of a benign granuloma is much higher, which depresses the specificity of PET-CT (granulomas are FDG-avid). The approach is: correlate with IGRA or tuberculin test, send sputum for AFB, mycobacterial culture and Xpert MTB-RIF, and have a lower threshold for tissue biopsy with stains for AFB and fungi before assuming cancer and embarking on staging.[1]

When it goes wrong — over- and under-investigation

[1]

Prognosis, disposition, and the safety-net

Most SPNs are benign and need no treatment — the goal of management is to catch the curable early lung cancer among them. Outcome is driven by malignant versus benign (the dominant determinant), histological subtype and stage, fitness for surgery, and the timeliness of detection.[5][6]

Disposition follows the risk band: low-risk nodules return to primary care with a structured CT surveillance plan and reassurance; intermediate-risk nodules are referred to the respiratory or lung-cancer MDT for PET plus or minus biopsy; high-risk nodules are referred to thoracic surgery for tissue diagnosis and resection (or SBRT if inoperable). The safety-net is a clear, written surveillance plan with named CT dates and a named responsible clinician — patients lost to follow-up during surveillance are the classic preventable failure.[1]

Special populations

[1]

The evidence — NLST, Brock, Mayo, Herder

The SPN literature is built on a small number of landmark prediction models and trials.[2][3][4][6]

[6] [2] [3]

The guideline landscape: Fleischner Society 2017 (MacMahon) is the global default for incidental solid and subsolid nodules; BTS 2015 (Callister) is the UK algorithm, model-driven (Brock then Herder); ACCP (Gould 2013) uses the Mayo or VA models in North America; Lung-RADS is for screening only. In the Indian context, no single national SPN guideline exists — Fleischner and BTS are applied with regional calibration because the high prevalence of TB and granulomatous disease lowers PET specificity.[1][5]

The mantra, and the mnemonic

[1] [1]

The mantra: 3 cm is the gate; 2 years stable is benign; quantify risk with Brock then Herder; do not biopsy a nodule you would resect anyway; in TB-endemic regions, tissue before you stage.[1]

Ward-round test — three stems, thirty seconds each

Stem 1 — the patient from the top of the topic (answer)

The 62-year-old smoker (40 pack-years) with an 8 mm spiculated right upper-lobe nodule found incidentally. No prior imaging. What is your pathway? Model: This is an intermediate-to-high risk incidental SPN. Characterise on thin-section CT (size, density, border, calcification, location), hunt for prior imaging (none here), and apply the Brock model — at 8 mm and spiculated in a smoker, Brock is likely over 10 per cent, so run the Herder model (which adds PET). PET-CT to reclassify: if Herder is over 70 per cent and he is fit, proceed to VATS wedge with frozen section, on to lobectomy if malignant — do not pre-operatively biopsy a nodule you would resect anyway. If intermediate (10 to 70 per cent), biopsy and MDT review. Do not apply Fleischner intervals reflexively — this nodule is over 8 mm and high-risk.[1][2]

Stem 2 — the small change that is not trivial (answer)

A nodule measured 8 mm on a CT six months ago is now 10 mm on the current scan. The registrar says "only 2 mm, probably nothing". What do you say? Model: This nodule has almost doubled its volume. The 26 per cent diameter rule — the cube root of 2 is approximately 1.26 — means a 100 per cent increase in volume shows as only a 26 per cent rise in diameter. A growth from 8 mm to 10 mm is a 25 per cent diameter rise, consistent with one volume doubling, and places the nodule firmly in the malignant until proven otherwise category. The doubling time (under 400 days) fits malignancy. Escalate to PET-CT and tissue diagnosis, or to surgical resection if the probability is high and the patient is fit. Never dismiss a small diameter change as trivial.[1]

Stem 3 — the nodule that is fat and popcorn (answer)

A 45-year-old non-smoker has a 12 mm smooth, well-defined nodule with fat density (−80 HU) and central popcorn calcification on CT. What is the diagnosis, and what follow-up is needed? Model: This is a pulmonary hamartoma — the commonest benign lung tumour. The radiological signature (fat within the nodule plus or minus popcorn chondroid calcification, smooth border, in a younger non-smoker) is diagnostic, and no follow-up is required. The work-up ends here. The trap would be to biopsy or surveil a lesion whose fat and popcorn calcification have already declared it benign — the patient can be reassured and discharged.[5]

References

1. [1]MacMahon H, Naidich DP, Goo JM, et al. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017 Radiology, 2017.PMID 28240562
2. [2]McWilliams A, Tammemagi MC, Mayo JR, et al. Probability of cancer in pulmonary nodules detected on first screening CT N Engl J Med, 2013.PMID 24004118
3. [3]Herder GJ, van Tinteren H, Golding RP, et al. Clinical prediction model to characterize pulmonary nodules: validation and added value of 18F-fluorodeoxyglucose positron emission tomography Chest, 2005.PMID 16236914
4. [4]Swensen SJ, Silverstein MD, Ilstrup DM, et al. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules Arch Intern Med, 1997.PMID 9129544
5. [5]Gould MK, Donington J, Lynch WR, et al. Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines Chest, 2013.PMID 23649456
6. [6]Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening N Engl J Med, 2011.PMID 21714641