risk reward

Tuesday, March 20, 2018

Before UMBC beat Virginia, no other 16th ranked team had won a game in the NCAA’s March Madness. Not in 135 previous games. Before that win, #1 never lost and #16 always won. But those words changed from always and never to rarely. The odds of the upset rose to 1 out of 136 or less than 1%. Fans of college basketball fans learned an important lesson; always and never do not exist in the real world and especially in medicine.

Risk-reward is always at play when a patient received advice for medical care. Surgeons who are technically gifted have patients who develop complications during and after surgery, from infection and bleeding to non-healing wounds. A gifted clinical doctor may have a misstep or two in proving a diagnosis caring for a patient who presents with confusing symptoms. Even when the diagnosis is firm, treatment options may have a potential for causing harm. Often there is time for discussion and contemplation before choosing a treatment plan but in emergency situations, decisions need to be made in minutes or less.

Heart Cath

Heart catheterizations have become almost routine. A cardiologist threads a thin catheter through an artery in the groin or arm and directs it into the coronary arteries that supply blood to heart muscle, looking for a narrowing or blockage. If found, blood flow can be improved or restored using a balloon to restore flow and a stent to keep the artery propped open. The benefit is preventing a heart attack and keeping the heart muscle pumping as normally as possible. There are minor complications, like bleeding or infection where the catheter is inserted through the skin. But here are also major complications like heart attack, stroke and death. Depending upon the health of the patient, statistics suggest the risk of complication is between 0.06% and 0.17% (between 1 in 600 and 1in 1600).

In an emergency situation, in the midst of a heart attack, where a coronary artery is completely blocked and heart muscle is dying, most people would accept a 1 in a thousand chance of harm when they are 100% having a heart attack.


The numbers aren’t always so easy. Strokes occur when an artery in the brain is blocked preventing blood supply. That part of the brain turns off and the part of the body it controls stops working. The window of time to intervene with clot busting drugs (thrombolytics: thrombo=clot + lytic= dissolve) is very narrow, about 3-4 ½ hours after the onset of symptoms. If the patient is a thrombolytic candidate, the treatment can help return blood supply to the brain in more than one third of patients, but the complication of causing irreparable bleeding into the brain can be as high as 6%. The sooner the patient gets to the ER and the sooner the clot busting drug is given, the less likely the complication of bleeding. Other therapies, including removing the clot using catheters are also a possible, but not necessarily available at most hospitals. Treatment decisions by patient and family about risk-reward have to be made in minutes, and sometimes that decision is not clear cut.

Atrial Fibrillation

Preventing a stroke in the first place would be the way to go. Atrial fibrillation an irregular heart rhythm, is one of the major risk factors for causing stroke. Blood clots can form inside the heart and then potentially break off and travel (embolize) to the brain, blocking an artery and causing a stroke. Anticoagulation with a blood thinner may be an appropriate treatment, but its complication is bleeding. The question becomes how much benefit versus how much risk. Calculators (like CHADS-VASC) can be used to help decide whether anticoagulation is beneficial and other calculators (HasBled) to decide their risk of bleeding. Together, the doctor and patient can sit down and decide. At the end of the day, regardless of the decision that might be addressed and revised over the years, the patient will either 100% have prevented a stroke or not, and 100% would have bled or not.

Always and never do not exist in medicine. Minimizing risk is the goal of any treatment that has potential benefit, but that risk is never zero and treatment does not always work. Virginia learned a valuable lesson; a #1 team had never lost in the NCAA first round, but never no longer exists for them. They 100% lost and UMBC 100% won, no matter what the statistics might have predicted. It’s that lesson that also applies to doctors, patients and families. Medicine isn’t a black and white science, but people aren’t always happy with shades of gray.




This entry was tagged , , , , , ,

bleeding risk calculator

HASBLED Calculator


 H  Hypertension: (uncontrolled, >160 mmHg systolic) 1
A  Abnormal kidney function: Dialysis, transplant, Cr >2.26 mg/dL or >200 µmol/L

Abnormal liver function:Cirrhosis or Bilirubin >2x Normal or AST/ALT/AP >3x Normal




S  Stroke: Prior history of stroke 1
B Bleeding: Prior Major Bleeding or Predisposition to Bleeding 1
L  Labile INR: Unstable/high INR), Time in Therapeutic Range < 60% 1
E  Elderly: Age > 65 years 1
D  Prior Alcohol or Drug Usage History (≥ 8 drinks/week)

Medication Usage Predisposing to Bleeding: (Antiplatelet agents, NSAIDs)





  • Score greater than or equal to 3: increased risk of bleeding
  • Use caution when using anticoagulation medications
  • Patients should be assessed and risk reviewed routinely
  • HASBLED is only moderately effective in predicting bleeding risk


Shoeb, Marwa; Fang, Margaret C. Assessing bleeding risk in patients taking anticoagulants. Journal of Thrombosis and Thrombolysis. 2013: 35 (3):312–9





This entry was tagged ,