To the Editor:

I read with great interest the two letters about the pros and cons of the “Genotype First Diagnosis” announced in the front page of the May 2009 issue of Genetics in Medicine.1,2 I cannot agree more with Drs. Saul and Moeschler.1 Although the new technologies, mainly microarray-based comparative genomic hybridization, have significantly increased the number of patients for whom we establish a diagnosis,2 there is no question that a good clinical evaluation is still of utmost importance.

The promotion of “testing first” is attractive to many primary care physicians (general pediatricians and some subspecialists) who hope that an abnormal result will provide the diagnosis. The testing first is an example of the “sampling the universe” diagnostic approach described by the late Frank A. Oski.3 This approach is commonly practiced by young physicians. Instead of the use of “hypothesis generation” as the diagnostic method recommended by Oski, and used by most clinical geneticists, a growing number of physicians rely on testing for their diagnoses.

I believe this policy of testing first should not be endorsed by the American College of Medical Genetics (ACMG). It is not an appropriate medical practice, because it increases the cost of medicine, and there are many results that will still need the interpretation of the geneticist and genetic counselor and the evaluation of the family. The attempt to end the “family's diagnostic odyssey” through indiscriminate testing no doubt has a high cost. The number of cases where microarray has been requested by the referring physician before a genetics evaluation is increasing. Furthermore, a significant number of such patients are referred with normal microarray results. By comparison, when a patient has undergone a comprehensive evaluation, including a detailed history and physical examination, microarray seems to yield a higher rate of abnormalities (Lacassie Y, Myrtle V, Sathyamoorthi S, unpublished study).

There is no doubt that our primary goal as physicians is to serve our patients' best interests. In that pursuit, an accurate diagnosis is the cornerstone. In a testing-first approach to genetics, we deemphasize the importance of the family evaluation and phenotype. We may perform unnecessary testing with a consequent high expense. When the diagnostic approach combines clinical skills with the appropriate medical test, the sensitivity of the test is certain to increase. In the case of microarray, it seems clear that the appropriateness of testing is best gauged by a clinical geneticist.

There is also no doubt that that there is a “critical shortage” of medical geneticists, as noted by Dr. Ledbetter. More specifically, however, the shortage may reflect the declining number of clinical geneticists. Such a trend will undoubtedly continue so long as molecular testing advances in its diagnostic ability. However, I would like to point out that the major reason why I am commenting on this controversy is the lack of challenge for the clinical geneticist. The evaluation of patients with an unknown diagnosis is challenging, constitutes clinical research, and is intellectually gratifying when your diagnostic hypothesis is confirmed through a specific test. This is why many geneticists enjoy their job. Otherwise dealing with patients with multiple congenital anomalies, mental retardation, or other developmental issues, for which there are few treatments to offer, other than counseling, is not compelling at all to most physicians and medical students.

I openly admit to the referring physicians that I prefer to evaluate the patient without a diagnosis and before any testing. What makes our specialty interesting is the challenge to try to establish an etiological diagnosis.4,5 Most geneticist will agree that it is quite disappointing when the patient is referred for evaluation and counseling after the diagnosis was already established. Certainly, the ACMG should be the center of discussion for this topic. Before reading these two letters,1,2 I proposed some aspects of this subject to discuss at the next meeting of the ACMG in New Mexico.