Do physician attitudes and practices limit use of EUS in the staging and the treatment of esophageal carcinoma?

Abstract

Background: Although EUS provides superior local staging of esophageal carcinoma when compared with other tests, EUS seems to be underused by physicians. We designed this prospective study to determine whether EUS is ordered in the evaluation of esophageal cancer and whether staging information obtained would change management.

Methods: A total of 114 physicians were mailed a questionnaire that surveyed which tests are used in evaluating patients with esophageal cancer, the order in which they are requested, and their estimated cost. Physicians were asked to estimate prognosis and to indicate which therapy would be used for each specific TNM cancer stage.

Results: Of 114 physicians, 71 (62.3%) physicians from 4 disciplines responded. Only 47.3% of physicians would use EUS in the patient workup for esophageal cancer. Physicians would only order EUS after first obtaining an endoscopy, then a barium swallow, and then a CT scan ( p < 0.0001). A significantly greater number of internists (78.9%, p = 0.055) would not order EUS, and 31.6% of internists would not use any staging data before referral to another physician for definitive management. Physicians were accurate in their assessment of the prognosis for each cancer stage and the cost of each test. There was no difference in the use of surgery between disciplines for stages O, I, IIA, and IV. However, significantly more surgeons than nonsurgeons would use surgery for stage IIB (100.0% vs. 71.3%, p = 0.019), with a trend toward greater use by surgeons for stage III (64.3% vs. 34.1%, p = 0.11). Except for significantly greater use of chemotherapy by surgeons and oncologists for stage IIA than internists and gastroenterologists (36.6% vs. 3.1%, p = 0.0006), there were no differences between subspecialties with use of chemotherapy for all other stages or use of radiation therapy for any stage.

Conclusions: Clinicians have an adequate understanding of patient survival based on cancer stage and a reasonable appreciation of cost for diagnostic tests regarding esophageal carcinoma. Specific data on cancer staging does impact treatment choices and management decisions. EUS is grossly underused by clinicians for staging esophageal cancer. Although internists may serve as gatekeepers, they fail to order EUS, order EUS only after less accurate tests, or fail to use staging data in management (especially referral) decisions. The ultimate modality of treatment may be more related to the type of physician that the patient is referred to, instead of the specific cancer stage. Education of primary care clinicians may be needed before the full impact of EUS on patient care can be appreciated.