Lateral lymph node metastasis in rectal cancer is a decision point with real morbidity attached, because lateral node dissection is not a minor addition to the operation. A meta-analysis of 23 studies and 2,209 patients compared FDG PET/CT against pelvic MRI for detecting it.
Sensitivity was similar: 0.63 for PET/CT against 0.76 for MRI, a difference that did not reach significance. Specificity was not: 0.97 for PET/CT against 0.82 for MRI, and that difference did.
The practical reading follows directly from which number is higher. Neither test is good enough to find every involved node, so neither can be used to rule metastasis out. But PET/CT's specificity of 0.97 means a positive result is rarely wrong, which is exactly what you want before committing a patient to a lateral node dissection they may not need.
So the two are not interchangeable and should not be ordered as though they were. MRI remains the workhorse for local staging. PET/CT earns its place when MRI has raised a suspicion that would change the operation, and a false positive would cost the patient a bigger one.
- Use PET/CT to confirm, not to exclude: sensitivity of 0.63 misses too much to rule out.
- Reserve it for when a suspected lateral node would actually change the operation.
- MRI stays the primary local staging study; this is not an argument to replace it.
- Quote specificity when discussing a positive result with a surgeon, not overall accuracy.
- The evidence is retrospective and mostly patient-level; head-to-head prospective data are still lacking.
The statistics, in plain English
Sensitivity and specificity answer opposite questions and this pair separates cleanly on which one they are good at. Sensitivity of 0.63 means PET/CT missed roughly a third of involved nodes, so a negative scan is weak reassurance. Specificity of 0.97 means only about 3% of uninvolved nodes were falsely called positive, so a positive scan is strong evidence. The comparison on sensitivity gave p = 0.09, which does not establish that the two tests are equivalent — with 23 heterogeneous studies the analysis may simply lack the power to separate them. What is established is the specificity difference, at p < 0.01.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for radiology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free