Neuroimaging in psychiatry: Potentials and pitfalls

Author and Disclosure Information



Advances in neuroimaging over the past 25 years have allowed for an increasingly sophisticated understanding of the structural and functional brain abnormalities associated with psychiatric disease.1 It has been postulated that a better understanding of aberrant brain circuitry in psychiatric illness will be critical for transforming the diagnosis and treatment of these illnesses.2 In fact, in 2008, the National Institute of Mental Health launched the Research Domain Criteria project to reformulate psychiatric diagnosis based on biologic underpinnings.3

In the midst of these scientific advances and the increased availability of neuroimaging, some private clinics have begun to offer routine brain scans as part of a comprehensive psychiatric evaluation.4-7 These clinics suggest that single-photon emission computed tomography (SPECT) of the brain can provide objective, reliable psychiatric diagnoses. Unfortunately, using SPECT for psychiatric diagnosis lacks empirical support and carries risks, including exposing patients to radioisotopes and detracting from empirically validated treatments.8 Nonetheless, given the current diagnostic challenges in psychiatry, it is understandable that patients, parents, and clinicians alike have reported high receptivity to the use of neuroimaging for psychiatric diagnosis and treatment planning.9

While neuroimaging is central to the search for improved understanding of the biologic foundations of mental illness, progress in identifying biomarkers has been disappointing. There are currently no neuroimaging biomarkers that can reliably distinguish patients from controls, and no empirical evidence supports the use of neuroimaging in diagnosing psychiatric conditions.10 The current standard of clinical care is to use neuroimaging to diagnose neurologic diseases that are masquerading as psychiatric disorders. However, given the rapid advances and availability of this technology, determining if and when neuroimaging is clinically indicated will likely soon become increasingly complex. Prior to the widespread availability of this technology, it is worth considering the potential advantages and pitfalls to the adoption of neuroimaging in psychiatry. In this article, we:

  • outline arguments that support the use of neuroimaging in psychiatry, and some of the limitations
  • discuss special considerations for patients with first-episode psychosis (FEP) and forensic psychiatry
  • suggest guidelines for best-practice models based on the current evidence.

Advantages of widespread use of neuroimaging in psychiatry

Currently, neuroimaging is used in psychiatry to rule out neurologic disorders such as seizures, tumors, or infectious illness that might be causing psychiatric symptoms. If neuroimaging were routinely used for this purpose, one theoretical advantage would be increased neurologic diagnostic accuracy. Furthermore, increased adoption of neuroimaging may eventually help broaden the phenotype of neurologic disorders. In other words, psychiatric symptoms may be more common in neurologic disorders than we currently recognize. A second advantage might be that early and definitive exclusion of a structural neurologic disorder may help patients and families more readily accept a psychiatric diagnosis and appropriate treatment.

In the future, if biomarkers of psychiatric illness are discerned, using neuroimaging for diagnosis, assessment, and treatment planning may help increase objectivity and reduce the stigma associated with mental illness. Currently, psychiatric diagnoses are based on emotional and behavioral self-report and clinical observations. It is not uncommon for patients to receive different diagnoses and even conflicting recommendations from different clinicians. Tools that aid objective diagnosis will likely improve the reliability of the diagnosis and help in assessing treatment response. Also, concrete biomarkers that respond to treatment may help align psychiatric disorders with other medical illnesses, thereby decreasing stigma.

Cautions against routine neuroimaging

There are several potential pitfalls to the routine use of neuroimaging in psychiatry. First, clinical psychiatry is centered on clinical acumen and the doctor–patient relationship. Many psychiatric clinicians are not accustomed to using lab measures or tests to support the diagnostic process or treatment planning. Psychiatrists may be resistant to technologies that threaten clinical acumen, the power of the therapeutic relationship, and the value of getting to know patients over time.11 Overreliance on neuroimaging for psychiatric diagnosis also carries the risk of becoming overly reductionistic. This approach may overemphasize the biologic aspects of mental illness, while excluding social and psychological factors that may be responsive to treatment.

Second, the widespread use of neuroimaging is likely to result in many incidental findings. This is especially relevant because abnormality does not establish causality. Incidental findings may cause unnecessary anxiety for patients and families, particularly if there are minimal treatment options.

Continue to: Third, it remains unclear...

Next Article: