Lung CT Screening & Incidental Lung Nodules

The USPSTF recommends annual low-dose CT for adults aged 50 to 80 who have at least a 20 pack-year smoking history and either still smoke or quit within the past 15 years. It carries a Grade B, meaning the Task Force found moderate net benefit and recommends offering it.

A pack-year is one pack a day for one year. Half a pack a day for 40 years counts the same as two packs a day for 10. Work out your own total before the appointment, because eligibility and insurance coverage both turn on that number.

The recommendation also says when to stop: once you have not smoked for 15 years, or once a health problem would prevent you from tolerating curative lung surgery. Screening only helps if a found cancer can be treated.

The National Lung Screening Trial found a 20 percent reduction in lung cancer deaths with annual low-dose CT compared with chest X-ray. The European NELSON trial pointed the same way: after at least 10 years of follow-up, NCI reports a lung cancer mortality rate ratio of 0.76 in screened men compared with usual care.

Overdiagnosis is the harm people most want a number for. NCI does not give one. It says some cancers found by screening do appear to be overdiagnosed, but that published estimates vary a lot, so the true size is not clear.

Small lung nodules are extremely common, especially in people who have smoked. Scars from old infections, healed inflammation, and small lymph nodes all show up as spots.

The Lung-RADS system exists to keep those spots from turning into panic. Most scans land in category 1 or 2, and both mean the same thing: come back in 12 months. Category 3 is "probably benign" and brings a 6-month repeat scan. Category 4A moves to a 3-month scan, and 4B or 4X to further imaging or tissue sampling.

Lung-RADS also changed what gets called positive at all. When researchers applied it retrospectively to the National Lung Screening Trial, the baseline false-positive rate was 12.8 percent, against 26.6 percent under the trial's original definition. The trade-off is real and worth knowing: in the same analysis, baseline sensitivity fell from 93.5 percent to 84.9 percent.

This trips up a lot of people who read about nodules online.

If a nodule is found on a lung cancer screening CT, it is managed under Lung-RADS. The population is high-risk by definition and the thresholds reflect that.

If a nodule is found incidentally, on a CT ordered for chest pain, trauma, an abdominal problem, or before surgery, it is managed under the Fleischner Society criteria instead. Those criteria explicitly do not apply to people under 35, people who are immunosuppressed, people with a known cancer, or people in a screening program. Under Fleischner, a single solid nodule under 6 mm in a low-risk person needs no routine follow-up at all, a threshold that surprises people who expect every spot to be chased.

If you are trying to make sense of a report, the first question is which system was used. A category number means Lung-RADS. A recommendation phrased purely in millimeters and months usually means Fleischner.

Findings outside the lungs. The scan covers the thyroid, heart and coronary arteries, aorta, esophagus, liver, adrenal glands, and spine. They are common: in the Veterans Health Administration's screening demonstration project, 41 percent of screened patients had an incidental finding. Lung-RADS flags them with an S modifier. Coronary artery calcification is among the most common, along with aortic aneurysms, emphysema and thyroid nodules. Others lead to workups that end in nothing.

Radiation. The dose per scan is small. NCI reports a mean effective dose of 1.4 mSv for low-dose CT in the National Lung Screening Trial, and estimates that participants received about 8 mSv over three years of screening once follow-up imaging is counted. It accumulates across years of annual scans, which is part of why eligibility is limited to people whose lung cancer risk is high enough to justify it.

Procedures. A minority of participants go on to biopsy or surgery, and a minority of those turn out not to have cancer.

Screening is a program, not an event. The benefit in the trials came from annual scans over years, with follow-up completed on schedule. The most common failure is a 6-month or 3-month recommendation that quietly never gets booked.

Before you leave, write down the category, the exact next date, and who is responsible for reminding you. If you still smoke, ask about cessation support at the same visit; quitting lowers lung cancer risk more than any scan can.

• ACR Lung-RADS v2022 assessment categories (PDF)
• NCI: Lung Cancer Screening (PDQ) - Patient Version
• NCI: Lung Cancer Screening (PDQ) - Health Professional Version
• Pinsky et al. 2015, Ann Intern Med — Performance of Lung-RADS in the National Lung Screening Trial
• USPSTF evidence report, NCBI Bookshelf — Screening for Lung Cancer With Low-Dose Computed Tomography
• MacMahon et al. 2017, Radiology (RSNA) — Fleischner Society guideline for incidental pulmonary nodules