Abdominal pain is the most common chief complaint among emergency department patients, yet firm diagnosis and etiology elude us in up to 40% of cases. While young patients do not escape the risk of misdiagnosis and bad outcome, diagnosis is particularly difficult in the elderly, and in that special patient population, up to a third will require surgical intervention, with its significant associated mortality. Still, many patients will be discharged with a diagnosis of nonspecific abdominal pain. In some of those patients, the disease process will progress, and they will not fare well; and in others, the diagnosis will be missed because of unusual presentations. This article addresses how properly to manage and discharge those patients, while at the same time limiting one’s liability exposure.

There is a common myth that emergency physicians are risk-takers and “adrenaline junkies.” That may be true for some in terms of outside pursuits, but it is certainly not true when it comes to the disposition of patients. Call it defensive medicine or simply wanting to avoid bad outcomes for patients placed in our care. The question is this: What amount of risk are we willing to accept in the evaluation of a patient with undifferentiated abdominal pain? Zero risk is unrealistic, and would be prohibitively expensive and often impossible to achieve. It is far easier to rule something in than to rule it out. Therefore, we must accept some risk for both our patients and ourselves and manage it appropriately.

The Work-Up

Abdominal pain is a high-risk complaint and requires a thorough and focused physical examination and history. And, while there are some things that all patients should get, such as a pregnancy test in all women of childbearing age, not every test is warranted. One may decide to forgo a pelvic exam, complete blood count (CBC), or CT scan, for example, depending upon the complaint. If the result of a test is not going to change your management, do not order the test. The reason is that if there is an abnormal result, it is much more difficult to explain (or defend) why it was not pursued than why an inappropriate test was not ordered in the first instance.

So order what you think is needed, forget what will not change management, and have good solid reasons for what you do and do not do. For example, an 11-year-old girl presents with acute onset RLQ pain that is 10/10 and sharp in nature. It is associated with nausea. All else in the review of systems is negative. The patient has not had this before. She recently started menstruating and is just finishing her cycle. She is very tender in the RLQ, but all other quadrants are soft, nontender, and nondistended. What do you order? The patient swears that she is not sexually active. Do you order a urinary chorionic gonadotropin (UCG) test? Do a pelvic exam and screen for sexually transmitted diseases? You are concerned about appendicitis; do you order a CT scan or ultrasound? What about a CBC?