Recently, I spoke with a primary care physician (PCP) about a young runner who had a syncopal episode. Because of the increasing awareness of sudden cardiac death in athletes, she had an electrocardiogram and an echocardiogram to look for structural abnormalities of the heart.

The PCP was inclined to dismiss the syncope as an isolated episode. However, the echocardiogram, otherwise normal, equivocated: “possible hypertrabeculation of the left ventricular apex, consider cardiac MRI to exclude non-compaction of the left ventricle.” The PCP was inquiring how to order a cardiac MRI.

I asked about the circumstances that led to the echocardiogram.

With the absence of a family history, no palpitations and a normal left ventricular function (she ran five minute miles, enough said), I speculated that the finding on the echocardiogram was most likely an overcall. The cardiac MRI would likely overcall as well, as these diseases are defined by numbers that are inevitably shared with normal individuals. I expressed lukewarm enthusiasm for the cardiac MRI. The PCP agreed. The athlete was spared another diagnostic test, a cardiology referral and possibly a lifelong label.

This is not a discussion of overdiagnosis of non-compaction. This is to restate a banal truism: physicians should speak to one another. In particular, PCPs and radiologists should talk to one another, preferably when an expensive test is being contemplated.

One silver lining of the preceding decade of over utilization of imaging is that radiologists have developed a rich mental atlas of imaging findings of clinical irrelevance. Such findings are charitably known as incidentalomas. However, they do not remain incidental for long. Like the miscommunications that led to Romeo’s poisoning, incidentalomas can balloon out of control unless nipped (ignored) in the bud.

Furthermore, most radiologists, whether general or sub-specialized, have encountered uncommon diseases and uncommon manifestations of common entities, because of the sheer volume of cases read.

PCPs and radiologists have complementary knowledge. The combination of strong knowledge of the patient and deep knowledge of the limitations and downstream consequences of imaging can rationally gate keep costly advanced imaging, costlier specialists and even costlier hospitals.

To be sure, there will be challenges. My conversation with the PCP took less than five minutes. Because of the historical inertia this may appear insurmountable.  Then, there is the belief that all problems in healthcare can be solved by a checklist-regurgitating computer program. Some can, many still need recourse to that quaint habit of talking.

There is the prevailing culture etiquettes to deal with. Radiologists might not wish to offer unsolicited opinion that might be ignored or misconstrued as questioning of clinical reasoning, fears though grossly inflated nonetheless prevail.

Finally, there is a feeling that no good deed in medicine can take place without a CPT code. How will physicians bill for consultation about a diagnostic test that was not performed? But value is what we add in between billing for medical services. Billing codes follow value. Value does not follow CPT codes.

The health of a healthcare system, regardless of how it is financed, depends on the strength of its most decentralized units. This is basic epistemology. The alliance of primary care and radiology could be that strengthening force, an improbable alliance, even a marriage of inconvenience. But these are inconvenient times calling for improbable solutions.

Saurabh Jha is a radiologist.