Medical Imaging is Less Revealing and More Subjective Than Patients Think

Imagine watching two people moving around in dim light. You see one person limping and the other walking smoothly.

Then, imagine looking at retina-sharp still pictures of these two subjects. One person has one leg that is shorter than the other and one person is unable to fully extend his left knee.

Which one has a limp and which one doesn’t? That’s anybody’s guess. Form and function don’t always match. Our bodies can compensate for a lot of things.

Then imagine looking at two lumbar MRIs. One has moderate multilevel disc disease and the other one does not. Which person has more back pain that the other? Again, that’s anybody’s guess. They say 10% of asymptomatic people have a herniated disc. Yet, many patients with modest symptoms insist on getting an MRI to “know what’s going on”.

Now, imagine two different radiologists looking at a CT scan of someone’s liver. One of them sees innocent adenomas and the other sees multiple metastases. That happened in a case I was involved with. The two doctors looked at the same pictures and their respective, opposite interpretations were merely opinions, guesses if you will.

I saw a report of a nuclear stress test the other day. My new patient had this test several months ago. He experiences pain between his shoulder blades and profound shortness of breath with moderate exertion. The report said “small reversible perfusion defect inferiorly and apically that appears artifactual”. That sounds very much like just an opinion to me. The cardiologist he saw in consultation took the report at face value and declared his chest pain non-cardiac. I don’t believe that, especially since the man’s echocardiogram showed a slightly abnormal wall motion at the tip of his heart.

To make matters worse, because of COVID, the hospitals around here don’t do exercise stress tests, but instead stress the heart with injectable drugs. I assume they don’t want people huffing and puffing on a treadmill, spreading aerosols in the room. My patient’s resting heart rate was 74 and it peaked at 90. My understanding is that the accuracy of the test is lower at lower increases in heart rate, similar to an exercise test but on a different scale.

I prescribed isosorbide mononitrate and ordered a new echocardiogram. I told my patient he may need a transesophageal echo and possibly a stress echo.

Even something as simple as using a plain chest x-ray to diagnose pneumonia is not straightforward. Does a patient with cough, green sputum, fever and crackles in part of one lung have pneumonia, even if the x-ray looks normal?