VOCAL-Penn: Predicting surgical risk in cirrhosis patients

For individuals with cirrhosis, an advanced stage of chronic liver disease, surgical procedures carry an increased risk of coagulopathy, bleeding, portal hypertension, altered drug metabolism, and other life-threatening complications. Those increased risks are relevant for lifesaving surgeries, but also for elective surgeries such as joint replacements. For surgeons, weighing the risks versus the benefits of a given surgery in cirrhotic patients can be extraordinarily challenging.

Making that calculation requires an accurate understanding of the potential risks, an effort that has led over the years to the development of a variety of risk calculation tools. However, says Nadim Mahmud, MD, MS, MPH, MSCE, these previous surgical risk calculators do not accurately reflect modern data. Dr. Mahmud is an assistant professor of medicine (gastroenterology) at the Hospital of the University of Pennsylvania and the Veterans Administration Medical Center, and an assistant professor of medicine in biostatistics and epidemiology at Penn Medicine.

“There was a sense among the hepatology community, and the surgical community more broadly, that the tools available for risk stratification were inadequate or outdated,” Dr. Mahmud says.

Assessing cirrhosis surgery risks

To bring risk stratification into the modern era, Dr. Mahmud and his colleagues created the VOCAL-Penn Cirrhosis Surgical Risk Score, which predicts postoperative mortality and decompensation risk for patients with cirrhosis. The VOCAL-Penn Risk Score has been widely adopted, allowing hepatologists, gastroenterologists, anesthesiologists, and other physicians to make more accurate predictions about the risk of surgery in patients with liver disease. In the process, the score is helping more patients make informed decisions about procedures that can improve their health and quality of life.

“Our goal was to provide more accurate predictions about surgical risk to inform shared decision-making conversations, so patients and their providers can decide together whether to move forward,” Dr. Mahmud says.

Given the multisystemic risks of cirrhosis, hepatologists are frequently consulted to risk-stratify patients with cirrhosis who might need surgery for conditions unrelated to the liver. To do so, providers use risk calculation tools in patients with cirrhosis that are often used in the context of transplant, such as the Model for End-Stage Liver Disease (MELD) score and the Child-Turcotte-Pugh scoring system to estimate cirrhosis severity, and the Mayo Risk Score to predict postoperative mortality. Although useful, these tools have recognized limitations.

For example, the Mayo Risk Score often gives very high projections of mortality risk, Dr. Mahmud says. “Often, those predictions were very discordant with our own clinical impression about the patient’s ability to tolerate surgery,” he adds. This gap likely reflects advances in medicine since the Mayo Score was developed in the 1990s. Patients having surgery today are more likely to benefit from minimally invasive approaches, for example. Other advances have made surgery safer for patients with comorbidities, including innovations in anesthesiology and perioperative management. Such advances reduce risk and change the risk-benefit calculation, expanding the populations for whom surgery would be of benefit.

“Surgery today looks very different than it did in the late 1990s,” Dr. Mahmud says. As a result, it’s no surprise that risk calculators based on older data would become less accurate over time. “But without another way to assess risk, many patients who would benefit from surgery are excluded from the opportunity,” he adds.

What’s more, the Mayo model didn’t differentiate risk based on the type of surgery, he adds. “You’d get the same projection of risk for a patient getting a knee replacement as you would for an open-heart surgery with major heart valve replacement.”

The VOCAL-Penn Score solves both problems. To create it, Dr. Mahmud and his colleagues—including surgeons, hepatologists, and experts in statistical modeling—used a high-quality dataset of modern surgical cases from the Veterans Administration to develop a new prediction model. Their tool also created different categories of surgeries, including abdominal, vascular, orthopedic, and cardiac surgeries. “Differentiating risk based on the type of surgery was an important innovation,” he says.

The VOCAL-Penn Score—a new tool for risk stratification in surgery

Research has shown that the VOCAL-Penn Score provides more accurate predictions of risk. Dr. Mahmud and his colleagues found that, as they suspected, the Mayo Score was poorly calibrated for assessing the risk for cirrhosis patients. “It was dramatically overestimating the risk of surgery relative to what the actual risk was,” he says. The overall performance of the VOCAL-Penn Score was superior to the Mayo Score and the MELD.

However, Dr. Mahmud cautions, the VOCAL-Penn Score is not meant to replace a provider’s clinical judgment. “It’s an adjunctive tool that’s meant to refine their understanding of the patient’s risk and help them communicate that risk to the patient and to other members of the care team,” he says.

Following its development, the VOCAL-Penn Score was adopted quickly and widely and has now been codified in guidelines from the European Association for the Study of the Liver and the American College of Gastroenterology.

Anecdotal evidence also suggests it is making a difference for patients and their providers, Dr. Mahmud says, citing the example of a colleague treating a patient with compensated cirrhosis and severe obesity. The patient’s physicians thought he would benefit from bariatric surgery but were hesitant to move forward. After a VOCAL-Penn Score determined an acceptable risk profile, the patient underwent laparoscopic bariatric surgery and did very well. “I hear stories like that frequently, where patients were able to get a surgery that might not otherwise have been an option,” Dr. Mahmud says. “And physicians can be a bit more confident in having these conversations about surgical risk.”

In next steps, Dr. Mahmud and his colleagues are exploring whether the score could be adapted to better optimize patients for surgery and reduce risk. For example, the tool might indicate that providers could mitigate risk by using beta blockers or a transjugular intrahepatic portosystemic shunt (TIPS) procedure to address portal hypertension in select patients.

Factors such as nutritional optimization and prehabilitation to improve muscle strength before surgery could also increase the chance of positive outcomes. An updated VOCAL-Penn tool could potentially help estimate how these interventions may influence postoperative mortality risk, providing additional information to guide perioperative optimization. “We’re working on a second iteration that will incorporate such factors so we can better understand how to mitigate risk,” Dr. Mahmud says.

Liver disease care at Penn Medicine

The research that brought about the VOCAL-Penn Score is possible at Penn Medicine because of a strong network of clinical and research-oriented hepatologists who collaborate closely, in a tertiary-care center that cares for a diverse patient population, Dr. Mahmud says.

“At Penn Medicine, we’re always trying to ask questions about how we can improve care for our patients,” he adds. “Our combination of inquisitiveness, methodological expertise, and multidisciplinary collaboration allows us to answer these kinds of questions and move the field forward.”

Referrals and consultations

Learn more about liver disease care and liver transplant at Penn Medicine. To refer a patient or consult with a transplant specialist, please call 877-937-7366, or submit a referral through our secure online referral form.