A multicenter study in Spain developed a new predictive model for surgical risk in patients with cirrhosis that may improve the prediction of postoperative mortality.

The surgical risk prediction model, called the Surgical Risk Score (S-RISC) for patients with cirrhosis, was developed by researchers at Hospital del Mar in collaboration with Hospital Clínic de Barcelona, Bellvitge University Hospital, and Parc Taulí University Hospital. The research received the Best Clinical Oral Presentation award at the Annual Congress of the Spanish Association for the Study of the Liver.

To date, the most widely used and standard international tool for estimating surgical risk in cirrhosis has been the VOCAL-Penn score, which was developed in the US.

Study Design

The study analyzed 1818 European patients with cirrhosis who underwent major extrahepatic surgery.

Two independent cohorts were used: one to develop the model and one to evaluate in the validation cohort.

Speaking with Univadis Spain, part of the Medscape Professional Network, José Antonio Carrión, MD, PhD, hepatologist at the Secció d’Hepatologia, Servei de Digestiu, Hospital del Mar, and researcher at Institut Hospital del Mar d’Investigacions Mèdiques in Barcelona, Spain, said, “The development cohort included 1296 patients (71%) from two hospitals, while 522 (29%) from two additional centers formed the validation cohort.”

Carrión noted that the model was derived using European patient data and showed improved performance compared with the US model across a wide range of extrahepatic surgeries performed in patients with cirrhosis.

Prediction Challenges

Cirrhosis affects multiple physiologic processes relevant to surgical outcomes, including coagulation, immune function, wound healing, and drug metabolism.

As a result, procedures considered routine in the general population may carry a substantial risk in patients with cirrhosis.

Carrión said, “The existing VOCAL-Penn model has several limitations. The original cohort consisted primarily of men who were US military veterans, which may limit its generalizability. The model also incorporates albumin levels, which are often unavailable in emergency surgery settings.”

In addition, it relies on the American Society of Anesthesiologists (ASA) physical status classification system, which can be difficult to consistently apply to cirrhosis, and includes BMI, which may be unreliable in patients with ascites.

Finally, surgical procedures in the model are categorized largely by anatomic location without accounting for differences in procedural complexity.

Key Variables

The S-RISC model incorporates several clinical variables that show independent associations with postoperative mortality across nearly 100 different types of major extrahepatic surgical procedures.

These variables included age, ASA classification, bilirubin level, international normalized ratio, creatinine level, sodium level, platelet count, presence of ascites, surgical urgency, and surgical complexity.

Model Performance

It works very well. There are two types of analysis necessary for a model to function: two statistical measures — discrimination and calibration.

“Discrimination reflects the model’s ability to distinguish between patients who will die within 90 days of surgery and those who will survive. Our model’s discriminative ability was 0.89,” said Carrión.

He added that calibration is even more important, “which complements discrimination and is the relationship between the anticipated mortality probability given by the model and the actual probability.”

Clinical Impact

The new model could become a standard tool in Spain for assessing surgical risk in patients with cirrhosis.

It may be particularly useful for hepatologists, surgeons, and anesthesiologists when evaluating patients undergoing urgent or complex surgeries.

It improves surgical safety by allowing for more precise identification of high-risk patients and facilitating more informed decisions.

Because the model was developed using European patient populations, it may also be more applicable in European healthcare settings, where the US model has shown limitations.

“We have the possibility of implementing it in clinical practice, but we’re going to wait until we publish it in a high-impact journal. The fact that it received an award at the National Congress is an important reference point. Once published, we want to create a freely accessible website so that any doctor can use it,” concluded Carrión.

Carrión reported having no relevant conflicts of interest.

This story was translated from Univadis Spain.