What an FDG PET Scan Measures and SUV Pitfalls

An FDG PET scan shows where fluorine-18 fluorodeoxyglucose accumulates during the imaging interval. FDG resembles glucose closely enough to enter many cells through glucose transporters. After phosphorylation, much of it remains trapped long enough for a PET scanner to detect the radioactive signal. The paired CT scan shows anatomy and helps correct the PET data.

This is useful because many cancers use glucose avidly, but it is not specific to cancer: the brain, heart, bowel, kidneys, urine, and sometimes skeletal muscle or brown fat can be conspicuous in normal physiology. Infection, immune activation, recent surgery, radiation change, and tissue repair can also be FDG avid, so a PET report interprets a pattern in clinical context rather than treating one number as a diagnosis.

What the tracer signal represents#

The tracer's full name is 2-deoxy-2-[fluorine-18]fluoro-D-glucose. Fluorine-18 decays by positron emission. When a positron meets an electron, the interaction produces two photons traveling in nearly opposite directions. PET detectors register paired photons and reconstruct a three-dimensional distribution of radioactivity.

The biological part comes before detection. FDG moves from blood into tissue, is transported into cells, and is phosphorylated. Unlike ordinary glucose, phosphorylated FDG does not move efficiently through the rest of glycolysis, so it accumulates relative to its rate of delivery and trapping, and uptake depends on blood flow, transporter activity, phosphorylation, competing glucose, time, and loss from tissue.

Calling FDG a “metabolic” tracer is reasonable shorthand, but the scan does not measure all metabolism: it does not directly measure cell division, oxygen use, viability, or a tumor's exact glucose consumption unless a more complex dynamic protocol and model are used. Routine SUV is semiquantitative.

PET and CT answer different parts of the question#

PET has high sensitivity to the tracer but limited anatomical detail. CT shows tissue density, shape, and size. It shows calcification, gas, and structural relationships. Fusing them can show whether a focus sits in a lymph node, muscle, bowel loop, bone, or urinary tract.

The CT component may be a low-dose scan used mainly for localization and attenuation correction, or a diagnostic-quality CT with a protocol suited to the clinical question. Those are not interchangeable. A low-dose, noncontrast CT should not be assumed to provide every detail of a dedicated contrast-enhanced study.

Attenuation correction uses CT information to account for photons absorbed or scattered in the body. Misregistration from breathing, movement, metal, or dense contrast can create apparent PET abnormalities. Reviewing both corrected and uncorrected images can help specialists recognize an artifact.

What SUV means#

The standardized uptake value is commonly calculated as activity concentration in a region divided by injected activity normalized to a body-size measure. A simplified weight-based expression is:

SUV = tissue activity concentration / (injected activity / body weight)

That tidy formula hides several choices. The injected activity must be corrected for radioactive decay and residual activity. Body weight may be replaced by lean body mass or body surface area. The region can be one voxel, a small peak volume, or a larger segmented volume. Reconstruction changes voxel values.

SUVmax reports the highest voxel in a lesion. It is easy to reproduce within one image and avoids drawing a precise boundary, but a single noisy voxel can shift it, while SUVmean averages a defined region and depends heavily on how that region is drawn. SUVpeak averages a small, standardized volume at the hottest area. SUL uses lean body mass normalization and can reduce variation related to adipose mass. Metabolic tumor volume and total lesion glycolysis combine uptake with volume, but segmentation choices add another layer of uncertainty.

No universal SUV cutoff separates benign from malignant tissue across all organs and cancers. A threshold developed for one disease, scanner, protocol, and purpose should not be transferred automatically to another.

Biology can create false-positive patterns#

Activated inflammatory cells use glucose. So pneumonia, granulomatous disease, and vasculitis can resemble tumor. So can arthritis, infection, and immune-related treatment effects. Healing incisions, biopsy tracks, fractures, radiation fields, and recently treated nodes can remain avid. The timing and pattern are therefore essential.

Normal variants also matter. Vocal muscles can light up after talking. Recently used muscles may be symmetric or focal. Brown fat often appears in characteristic neck and upper-chest locations. Bowel uptake varies. Urinary FDG can obscure nearby pelvic structures. Cardiac uptake changes with diet, fasting, insulin, and underlying physiology.

A report may compare uptake with mediastinal blood pool or liver rather than rely only on an absolute number. Disease-specific visual scores, such as the Deauville scale in certain lymphomas, use reference tissues and validated clinical settings. The scale does not generalize to every cancer.

Low uptake can also mislead#

Some malignancies are less FDG avid because of histology, differentiation, or cellularity. The reason can also be mucin content, size, or treatment effect. A negative scan is therefore not a universal cancer exclusion test.

Spatial resolution produces the partial-volume effect. When a lesion is near or below the scanner's effective resolution, its signal spreads across neighboring voxels and the measured concentration is underestimated. A tiny active lesion can have a modest SUV. Respiratory motion further blurs lung, liver-dome, and upper-abdominal lesions.

High blood glucose competes with FDG and may lower tumor-to-background contrast. Insulin can drive tracer into skeletal muscle and alter distribution. Preparation protocols address fasting, glucose measurement, and activity. They address temperature, medicines, and timing, because these variables change what the scan can show. A deviation does not always invalidate a study, but it belongs in interpretation.

Technical choices move the number#

Uptake continues to change after injection. Comparing a scan acquired at 55 minutes with one acquired much later can produce a difference unrelated to disease. Injection infiltration leaves activity near the vein and reduces the amount delivered systemically. The recorded injected activity, assay clock, scanner clock, and decay correction must agree.

Scanner calibration links detected counts to activity concentration. Reconstruction settings govern corrections, resolution recovery, smoothing, and noise. Newer scanners can display small foci differently from older equipment. EARL and QIBA programs exist because quantitative consistency requires phantom testing, quality control, and protocol harmonization.

The QIBA FDG PET/CT profile reports that, under controlled conditions, tumor SUVmax can be measured with useful repeatability, but the expected variation is not zero. A small numerical change can fall within measurement variability. Response frameworks therefore combine relative and absolute change rules with lesion pattern and new findings.

How specialists read a study#

Interpretation starts with the reason for imaging: diagnosis, initial staging, or radiation planning. The reason could be treatment response, suspected recurrence, or infection. It could be inflammation or a neurologic question. Each has different evidence and pitfalls. The interpreter reviews preparation, glucose, and uptake interval. The review covers injected activity, technical quality, and relevant treatment dates.

Next comes distribution. Is uptake focal or diffuse, symmetric or asymmetric, expected or unexpected for the organ? Does CT show a matching mass, consolidation, fracture, surgical change, or no structural correlate? Are there older scans, pathology, laboratory findings, or symptoms that change the probability?

The final report should separate observations from interpretation. “FDG-avid node” describes tracer uptake. “Suspicious for metastasis,” “favored inflammatory,” and “indeterminate” express different probability judgments. Pathology may still be needed when a result would change major treatment and imaging cannot resolve the alternatives.

Comparing serial scans responsibly#

For treatment response, the strongest comparison uses similar preparation, tracer, uptake time, scanner performance, reconstruction, and measurement method, though the same target lesion is not always the most informative lesion if disease distribution changes. New sites, CT anatomy, and overall pattern matter.

Treatment can complicate timing. Chemotherapy, immune therapy, and radiation can change marrow, spleen, muscle, and lesion uptake. So can growth factors, surgery, and infection. An inflammatory flare can transiently look worse even when a therapy is helping. Conversely, lower uptake may reflect treatment effect without proving eradication of every viable cell.

Validated criteria should be named when used. PERCIST is a research-oriented framework for metabolic response in many solid-tumor settings. Deauville-based criteria are used in defined lymphoma contexts. RECIST measures anatomy rather than FDG uptake. A clinical trial, specialty guideline, and routine impression may apply different rules, so their conclusions should not be blended without explanation.

Questions worth asking about a report#

Useful clarifications include the clinical purpose, whether the CT was low-dose or diagnostic, whether the finding has an anatomical correlate, how it compares with reference tissues and prior scans, and whether a disease-specific response framework was applied. For a numerical change, it helps you to know whether acquisition conditions were comparable and whether the change exceeds expected measurement variation.

An indeterminate result is not a failed result. It can narrow the possibilities and guide the next step, such as correlation with a dedicated scan, interval imaging, clinical evaluation, or tissue sampling. Which next step is appropriate depends on the probability of disease, urgency, procedural risk, and how the answer would change your management.

Sources#

1. EANM FDG PET/CT tumor-imaging procedure guideline, version 2.0
2. ACR-SPR practice parameter for FDG PET/CT in oncology
3. National Cancer Institute fact sheet on CT and combined PET/CT
4. EANM-SNMMI FDG guideline for infection and inflammation, version 2.0
5. QIBA profile for FDG PET/CT as a cancer-response imaging biomarker
6. EARL accreditation results for FDG PET/CT quantification

For your own health, talk with your clinician.*