How Lung-RADS Turns a Lung Nodule Into a Management Plan

A Lung-RADS score is the number a radiologist uses to turn a lung nodule seen on a screening scan into two things you can act on: an estimated chance that the spot is cancer, and a specific next step. The American College of Radiology built the system for low-dose CT lung cancer screening, and it sorts every finding into categories from 0 to 4X, where a higher number means higher suspicion and a faster workup. Its whole purpose is to make the plan depend on the nodule rather than on which radiologist happened to be reading films that afternoon.

Key points#

• Lung-RADS assigns each screening finding a category from 0 to 4X; the category carries both a rough cancer probability and a recommended action.
• Categories 1 and 2 are negative screens and cover roughly 90 percent of results; category 3 buys time with short-interval imaging; category 4 triggers closer diagnostic workup.
• Nodule type (solid, part-solid, or non-solid) and size both feed the category, and size is the average of long and short axis.
• The 2022 update did not change the ladder so much as sharpen it, clarifying what counts as growth and adding rules for a few tricky findings.
• A category is a probability and a pathway, not a diagnosis. A high number is not a cancer verdict, and a low one is not a guarantee.

Why screening needs a shared scoring language#

Low-dose CT is sensitive, and sensitivity has a cost: it finds small spots in most people it scans, and the overwhelming majority of those spots are harmless scars, old infections, or benign lymph nodes. The problem this creates is not detection. It is interpretation, and interpretation is where readers used to differ. Without an agreed set of rules, the same 7 mm nodule might send you back for routine imaging and send the next person for a biopsy, depending only on the reader's habits.

Lung-RADS solves that by fixing a common vocabulary. Each category is tied to an approximate probability of malignancy and to one recommended action, so the report that reaches the ordering clinician already contains an implied plan instead of an open question. Categories 1 and 2 are read as negative screens. Categories 3 and 4 are positive and move you into follow-up. Because behavior differs by composition, the system also grades nodule type separately: solid nodules, part-solid nodules that mix a solid piece with a hazy halo, and non-solid (ground-glass) nodules each get their own size thresholds. Size itself is measured as the mean of the long and short axis on a single image, and that number, plus whether your nodule is new or growing, sets the category.

The categories as a risk ladder#

The benign majority: categories 1 and 2#

Category 1 means nothing of concern was seen. Category 2 covers findings that look or behave benign, such as a solid nodule under 6 mm at a first (baseline) scan or a stable ground-glass nodule. Both sit under an estimated 1 percent chance of cancer, and both return you to routine annual screening. Together they account for around 90 percent of screens, which is precisely why the system earns its keep: it lets the large harmless group move through fast so that attention can concentrate on the few results that carry real risk.

Watchful waiting on a clock: category 3#

Category 3 is labeled probably benign, with an estimated 1 to 2 percent chance of cancer. It captures solid nodules from 6 to under 8 mm at baseline, new solid nodules from 4 to under 6 mm, and certain part-solid or larger ground-glass nodules. Rather than jumping to a needle, the recommendation is a repeat low-dose CT in six months. The reasoning is simple economics of risk: a short-interval scan is cheap and non-invasive, and time is often the best test for separating a nodule that will sit still from one that is slowly enlarging.

Suspicion that earns a workup: category 4#

Category 4 is graded by degree. Category 4A, at an estimated 5 to 15 percent chance of cancer, covers solid nodules from 8 to under 15 mm and prompts a three-month CT, and sometimes PET/CT once a solid component reaches 8 mm. Category 4B, above 15 percent, covers solid nodules 15 mm or larger and points toward diagnostic CT, PET/CT, or tissue sampling. Category 4X is the deliberate override: a nodule that the size table alone would place at category 3 or 4 but that carries worrying features, such as spiculated (spiky) margins or an enlarged nearby lymph node, is bumped up to the highest-suspicion tier. That single rule is where a radiologist's pattern recognition is explicitly allowed to outrank the measurements.

What the 2022 update actually changed#

The 2022 version left the ladder standing and patched the rungs that used to wobble. It added explicit handling for atypical pulmonary cysts, the thick-walled or multi-chambered cystic lesions that can hide cancer, placing suspicious ones in category 4A or 4B. It gave a dedicated rule to juxtapleural nodules, so that small, smooth, triangular nodules pressed against the lung lining, usually benign internal lymph nodes, are read as category 2 instead of being funneled into follow-up. It classified airway-centered nodules by location, and it added guidance for findings that look infectious or inflammatory.

Two of the changes matter most for reader-to-reader consistency. First, the update pinned down growth: an increase of more than 1.5 mm in mean diameter now defines a nodule as bigger, so two radiologists judge your enlarging spot the same way. Second, it introduced stepped management for nodules that stay stable or shrink across successive scans, so that stability is rewarded with longer intervals. The RadioGraphics review of v2022 works through these scenarios case by case, while the ACR statement remains the authoritative source for the category definitions themselves.

The trade-off hiding inside every threshold#

Every size cutoff in Lung-RADS is a wager about balancing missed cancers against false alarms, and there is real data on where that line sits. In the National Lung Screening Trial, a large share of people screened by CT had at least one positive result across three annual rounds, and most of those positives were false alarms. When Pinsky and colleagues retrospectively applied Lung-RADS to that trial in Annals of Internal Medicine, the baseline false-positive rate dropped from 26.6 percent to 12.8 percent, and the rate after baseline fell from 21.8 percent to 5.3 percent. Sensitivity also fell, from 93.5 to 84.9 percent at baseline, and that is the honest price: raising thresholds to spare people needless biopsies means a small number of early cancers are watched for a while rather than worked up at once. Each revision, including 2022, tries to shift that balance a little without surrendering the mortality benefit that justified screening. That benefit and its eligibility rules underpin the U.S. Preventive Services Task Force recommendation for annual low-dose CT in adults at high risk because of age and smoking history.

What the score is, and what it is not#

A Lung-RADS category is a snapshot. It grades one nodule, at one moment, against population averages. It does not know your full history, and its percentages are estimates you can fall above or below. A category 4 result is not a verdict, and a category 2 is not a promise. Whether to screen and how to follow a finding are decisions for you and your own clinicians, who weigh the score alongside everything the score cannot see.