Reduce legal risks of chest pain

Missed MIs are the largest single dollar problem in emergency medicine malpractice nationally, and chest pain liability risks are increasing, warns Eric Knox, MD, medical director of MMI Risk Management Resources, based in Deerfield, IL. "It’s becoming a more expensive category as time goes on because people are presenting at younger ages," he reports. "If you end up with a death following a misdiagnosis, you are liable for more damages if the patient is a 35 year old instead of a 75 year old, because of the way the legal process works."

Here are some ways to reduce risks of chest pain patients:

Adhere to risk management guidelines. A recent 12-year study found that following risk management guidelines led to substantially lower average dollar losses per medical malpractice claim than following no guidelines at all.

More than 200 EDs were asked to collect data on patients complaining of chest pain. Results showed that more than one-half of the chest pain patients older than 30 years who were discharged left with a non-definitive diagnosis. "Those patients have almost twice the rate of unplanned returns to the ED within 48 hours, when compared with all patients discharged from the ED," says Pam Lockowitz, president of MMI, which developed guidelines in response to the study’s findings.

Patients discharged without a definitive diagnosis are more likely to return with an unplanned admission or death, stresses Lockowitz. "It’s important for providers to take a systematic approach to diagnose or rule out MI and reduce negative outcomes," she says.

Implementing chest pain protocols that address risk management issues is key, says Knox. "Timely and accurate diagnosis reduces the number of negative outcomes," he adds. "Cases typically involve not perfectly diagnostic ECGs, or atypical chest pain that gets misdiagnosed as muscular or stomach pain or something else."

A task force of ED physicians, nurses, administrators and risk managers developed guidelines and the results were studied. "If the guidelines are followed in a given lawsuit, the cost is much lower than if one or more had not been followed," says Knox. "In many cases where there was litigation, the case would not have been brought if certain things had been done. If guidelines were applied consistently to every case, there would be fewer lawsuits."

The guidelines state that any patient older than 30 years with chest pain gets an ECG. "A lot of claims involve patients walking out without ECGs," says Knox. "Patients who don’t get ECGs are the ones who end up in court. Be safe and do the ECG."

Guidelines can improve identification of patients who are most at risk, and ensure that appropriate testing is completed. "The goal is to sharpen diagnostic accuracy," says Knox. "Instead of complicated pathways or algorithms, try to hit a median balance. You want to ask people to do important things, but keep it clear and simple."

Consider charting techniques. History is critical in chest pain patients, emphasizes Daniel J. Sullivan, MD, JD, FACEP, chairman of the department of emergency medicine at Ingalls Memorial Hospital in Harvey, IL. "On presentation, you cannot tell the difference between the clinical entities that cause chest pain without a complete history," he says.

"For example, a sudden onset of pain that is maximal at outset suggests a thoracic aortic dissection, whereas a more crescendo type pain suggests cardiac ischemia, a pain in the substernal area suggests ischemic pain, and pain in the substernal area that radiates straight through to the back is more characteristic of a thoracic aortic dissection," notes Sullivan.

Risk factor analysis in the history can also be invaluable, says Sullivan. "For example, there is an increased risk of thoracic aortic dissection in patients with a first-degree relative with the same problem, or in patients with a connective tissue disorder," he notes. "There are few things more important in the chest pain patient than a review of cardiac risk factors."

Documentation is essential in order to establish that the practitioner has met the standard of care in the evaluation of chest pain, Sullivan stresses. "If the practitioner documents these issues, the plaintiff must prove that he did not do the appropriate evaluation," he notes. "If the practitioner does not document these issues, the burden is on the practitioner to prove he did do the appropriate evaluation. It’s easy to see which road the practitioner should choose."

Orientation plays a critical role in risk management of chest pain patients, says Sullivan. "In a number of thrombolytic-related lawsuits, the emergency practitioner claims that he or she did not know that thrombolytics could be administered in the ED," he reports. "During litigation, the hospital presents an ED protocol that obviously includes ED administration of thrombolytics."

Include ECGs in abdominal pain protocols. MIs can be mistaken for abdominal pain, stresses Michael Kohn, MD, MPP, FACEP, attending physician for the department of emergency services and director of quality improvement for the ED at San Francisco General Hospital. "If a 60-year-old patient has a chief complaint of abdominal pain, the chest pain protocol doesn’t get triggered," he notes. "But when the patient is finally examined, it turns out to be very high abdominal pain, without a lot of tenderness, and ends up being an MI."

Abdominal pain protocols need to address this possibility, Kohn urges. "When you have an older patient with abdominal pain, you need to get EKGs on that patient very quickly," he says. "In our protocol, any patient over the age of 50 with abdominal pain gets an EKG soon as they hit the door."

Avoid giving medications to patients assumed to have abdominal pain without ruling out chest pain, says Knox. "Don’t rely on a GI cocktail. It may make people with ulcers feel better, but will also make people with heart pain feel better."

Educate staff about Troponin I. In the near future, ED chest pain protocols may include Troponin I testing. "There is a movement to use enzymes to improve our clinical judgment," says Kohn. "The initial cardiac Troponin I is going to possibly prevent us from sending out a patients if results are positive, which will help prevent missed MIs."

At San Francisco General’s ED, Troponin I is drawn upon the patient’s arrival. "Chest pain patients over the age of 35 with abnormal but non-diagnostic ECGS will get an initial cardiac Troponin I and be monitored for six hours, receiving a repeat ECG at three hours and six hours," notes Kohn.

If the initial Troponin I is positive or if the three-hour ECG shows concerning ischemic changes, the patient is admitted. "If not, a repeat Troponin I will be drawn six hours after the initial level," says Kohn. "If this second Troponin I is also negative, the patient will be discharged with precautions and close follow-up."

The strategy is also cost effective. "By sending these patients to the observation unit instead of admitting them to the hospital, we save money on false positives and reduce unnecessary admissions, which are a huge dollar expenditure," Kohn notes.

Still, cardiac enzymes are no substitute for clinical judgment, stresses Kohn. "They can only be used as an adjunct," he says. "For abnormal but nondiagnostic EKGs, clinical judgment is more sensitive, meaning the ED physician is going to send fewer MI patients home than just blind usage of cardiac Troponin I."

Use observation units appropriately. "If patients with normal EKGs are sent to observation units as a kind of hedging maneuver, the rule in rate would then be extremely low," notes Kohn. "What you’ll end up doing is keeping people in the ED unnecessarily who should have [been] sent home."

The solution is to follow a protocol that includes a chest pain observation unit, serial EKGs, and Troponin I, advises Kohn. "But you also need to do very aggressive quality control in that unit, to make sure the patients being admitted do have abnormal EKGs, and you are not admitting patients with such low risk for MI that they should have been sent home," he says.

The rule in rate for those patients should be higher than 10%, says Kohn. "If it’s lower than that, it means you are not using clinical judgment and filtering out very low risk patients who should be sent home without any observation," he explains.

An example of that type of patient is a 47-year-old man with a runny nose and cough who complains of six hours of intermittent chest pain, with no significant past medical history, says Kohn. "The exam reveals BP 150/90, otherwise normal vital signs, nasal congestion, and injected oral pharynx. No chest wall tenderness. CXR and ECG are normal," he notes. "This kind of patient has no business in a chest pain observation unit. In fact, we should be prohibited from enrolling any patient with a truly normal ECG into a chest pain observation protocol.’

On the other hand, take the example of a 47-year-old man with six hours of intermittent chest pain, says Kohn. "He thinks he might have a cold’ but denies runny nose or cough, has no significant past medical history, and the exam reveals a BP of 150/90, but is otherwise unremarkable. CXR is normal, but ECG meets voltage criteria for LVH and there appear to be repolarization abnormalities in the ST segments and T waves," he adds. "Because of his abnormal but non-diagnostic ECG, this patient might be appropriate for a chest pain observation protocol."

Getting a thorough history is key, Kohn stresses. "Obviously, we would get a much more detailed history from these patients, including severity, position, quality, radiation, and timing of the chest pain," he says. "We would also ask about risk factors for venous thromboembolism and cardiac disease."

Look for patients who don’t fit patterns. "You may see patients younger than 50 [years] or female nonsmokers with chest pain and say, this just doesn’t fit the profile of MI," says Lockowitz. "There is a nonspecific diagnosis that gets placed on a number of these patients, who are discharged, and within 48 hours have an unplanned return/death."

Have a high index of suspicion for atypical chest pain. "If a patient describes a pain in their shoulder and was playing touch football, think heart attack anyway," advises Knox.

The main thing to emphasize is the ECG, Kohn stresses. "If a young patient presents with atypical chest pain, is sent home, and ultimately turns out to have an MI, the ED care is much more defensible if it included a truly normal ECG," he says. "It is possible for someone with an MI to have a normal initial ECG, but it is very rare. Of a hundred chest pain patients who ultimately turn out to have MI, no more than one will have a normal ECG (although 50 will have ECGs that are abnormal but not diagnostic of acute MI)."

Editor’s Note: For more information on risk management guidelines and MMI’s report, "Transforming Insights into Clinical Practice Improvements: A 12-Year Data Summary Resource," contact MMI Companies, Inc., 540 Lake Cook Road, Deerfield, IL 60015-5290. Telephone: (847) 374-2400. Fax: (847) 940-2372.