An Ounce of Prevention

| |

  

  

Protecting Against Missed Diagnoses

by Dawn Penney, JD, RN, Regional Patient Safety/Risk Manager, and Howard Marcus, MD, FACP, Chair, Texas Alliance for Patient Access, and Chairman, The Doctors Company Texas Physician Advisory Board

^{Robin Diamond, MSN, JD, RN}

As a follow-up to Dr. Troxel’s Director’s Forum article, Dawn Penney and Howard Marcus, MD, FACP, examine the underlying causes that resulted in the patient’s mortality and how it might have been prevented. 


—Robin Diamond, MSN, JD, RN; AHA Fellow–Patient Safety Leadership; Senior Vice President, Department of Patient Safety

Emergency departments care for more than 100 million patients annually and provide accurate and effective diagnoses in well over 99 percent of the cases.¹ Unfortunately, in complex and busy settings such as the ED and the ICU, errors occur. In Dr. Troxel’s article “A Patient with Chest Pain of Sudden Onset,” a 42-year-old man presents to the ED with hypertension and severe chest pain. He is admitted to the ICU with the working diagnoses of PE and poorly controlled hypertension. Two days later, he dies of an undiagnosed aortic dissection.

The Commentary 
According to data by the Physician Insurers Association of America (PIAA), missed and delayed diagnoses, often due to multiple factors, continue to represent many of the most severe and frequent malpractice claims.

The challenging case outlined in Dr. Troxel’s article involves a rare clinical event that illustrates both a failure to develop a comprehensive differential diagnosis and a failure of communication between physicians.

Failure to Consider the Correct Diagnosis 
In the rapid pace of emergency medicine and the ICU, physicians are required to quickly establish a working diagnosis and treatment plan. The use of heuristics—a methodology that uses shortcuts involving testing and therapeutic trials to arrive at a correct diagnosis—is very common in clinical practice. It is important, however, not to ignore alternative diagnoses and clinical findings that are incongruous or inconsistent with the working diagnosis.

Mark Graber, MD, recommends the following techniques to improve diagnostic accuracy:²

• Force yourself to consider alternative diagnoses, and obtain second opinions. Be aware of the odds of being wrong. They are often closer to 10 percent than 1 percent. In this case, without the information about the aortic dilation seen on the CT scan and without a bedside evaluation, the cardiologist anchored on the working diagnosis of pericarditis.
• Remember that once a working diagnosis has been established, there is an inherent bias toward maintaining that diagnosis—which may result in errors.
• Once you have established a working hypothesis, continue to examine it carefully. Consider alternatives, rethink key assumptions, and, in particular, think about diagnoses that you can’t afford to miss.

Failure to Communicate Effectively 
Emergency room health care personnel operate under strict guidelines and directives, both from the hospital and, most importantly, pursuant to the standards of emergency and trauma care medicine. These standards must be maintained in assessing and implementing the mode and timing of emergency treatment. If contraindicated treatment is rendered or an incorrect diagnosis made, the effect on the patient, as in this case, can be tragic.

• Although the radiologist verbally recommended a repeat CT with contrast to the ER physician, that vital recommendation was not clearly communicated to the physicians in the ICU or to the consulting cardiologist. Thus, the opportunity for obtaining a CT scan with contrast was clearly missed.
• The ICU resident spoke with the cardiologist but did not specifically request or order a bedside consultation in this confusing case. (Perhaps this is an example of “house staff mentality,” in which residents try to avoid inconveniencing consulting physicians in the middle of the night.) As a result, the consulting cardiologist was unaware of the CT scan findings suggestive of a possible dissection and did not challenge the working diagnosis of pericarditis.
• There was a delay in dictation and transcription of the TTE that may have impacted the follow-up plan or differential diagnosis.

In Summary

• A CT scan demonstrating modest dilation of the ascending aorta and mediastinal infiltrate was not satisfactorily explained by the diagnosis of pericarditis. In his book How Doctors Think, Dr. Jerome Groopman says, “Hoof beats don’t always mean horses, sometimes there really are zebras. Too often doctors stop testing.”³ When there is conflicting or confusing data, consider alternative causes for a pericardial effusion and chest pain.
• While the CT scan with contrast could not be completed initially in the ER, there was no order to repeat the CT with contrast even though venous access would have been available after admission to the ICU. The best chance for a premortem diagnosis or a thoracic aortic dissection was, therefore, missed.
• The cardiologist never saw the patient, even though the patient was admitted and then expired two days later. Be just a bit skeptical of the working diagnoses in any complex patient—particularly relayed during a 1:00 AM phone call. Also, the house staff may not have made it clear that this patient required an urgent bedside cardiology consultation or, at the very least, a consultation first thing in the morning.
• A TTE lacks the accuracy of the TEE in the diagnosis of aortic dissection—a fact that most non-cardiologists don’t know. Once called, the consultant must see the patient in a timely fashion and make up his or her own mind using all the available data and knowledge.

| |