Blood Test Predicts Cardiac Complications After Noncardiac Surgery

Batya Swift Yasgur, MA, LSW

December 27, 2019

Assessment of N-terminal pro-B-type natriuretic peptide (NT-proBNP) level prior to noncardiac surgery may predict whether the patient might experience adverse cardiac outcomes, such as myocardial injury or vascular death after noncardiac surgery (MINS), new research suggests.

Investigators studied more than 10,000 patients, aged 45 years or older, whose NT-proBNP levels were measured before surgery and whose troponin T levels were measured daily for up to 3 days following surgery to evaluate potential cardiac injury. Outcomes were assessed 30 days following surgery.

Higher levels of preoperative NT-proBNP were associated with higher rates of cardiovascular events. Patients whose levels were >1500 pg/mL had a fivefold increased risk for the primary outcome, a composite of vascular death and MINS at 30 days, compared to patients with levels <100 pg/mL.

Risk prediction was further improved when the investigators added NT-proBNP levels to the Revised Cardiac Risk Index, which is currently used to predict perioperative cardiac risk.

"What these findings show is that a simple and relatively cheap blood test can measure in adults undergoing noncardiac surgery and suggest whether they are going to suffer a major cardiovascular event or even death," senior author P. J. Devereaux, MD, PhD, professor of medicine, cardiologist at Hamilton Health Sciences (HHS), and scientific lead for perioperative research at McMaster University and the HHS' Population Health Research Institute, Canada, told | Medscape Cardiology.

The study was published online December 24 in the Annals of Internal Medicine.

Accurate Prediction Needed

MINS, which is "the most common major vascular complication after surgery," is associated with perioperative death, necessitating "accurate preoperative cardiovascular risk prediction," the authors write.

Although the guideline-recommended RCRI is "easy to use, its accuracy in predicting major perioperative cardiovascular complications is limited," they state.

Preliminary evidence suggested that measuring preoperative NT-proBNP can potentially confer "additional predictive value beyond the RCRI," they add.

To investigate the potential for using NT-proBNP in risk assessment, the researchers drew on data from the Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) study, a multinational study of adult patients aged ≥45 years in 16 medical facilities located in nine countries, between August 2007 and October 2013.

MINS included myocardial infarction (MI) and ischemic myocardial injury "that does not satisfy the definition of MI."

The researchers assessed the association between preoperative NT-proBNP level and the primary outcome using a series of "iterative thresholds" up to 4000 pg/mL:

  • ≤100 pg/mL (reference group)

  • 100 to ≤200 pg/mL

  • 200 to ≤1500 pg/mL

  • ≥1500 pg/mL

Of the 10,402 patients included in the study (mean [SD] age = 65 [11] years; 50% men), 40% came from centers in North America, 30% from Asia-Pacific, and 20% from Europe.

The most common surgeries were major orthopedic procedures, followed by major general and major urology or gynecology surgeries (25.3%, 17.9%, and 13.8%, respectively). Of the patients, 33.3% underwent low-risk surgery; 4.4% of the procedures were urgent or emergent.

The primary composite outcome of vascular death and MINS within 30 days post surgery occurred in 12.2% of patients (0.5% vascular death and 11.9% MINS).

Cheap, Convenient

Cox proportional hazards models showed dramatic differences in the composite outcome in those with elevated NT-proBNP measurements compared with the reference group. Composite outcome increased incrementally in association with increasing NT-proBNP level:

  • 100 to ≤200 pg/mL: adjusted hazard ratio [HR], 2.27; 95% confidence interval (CI), 1.90 – 2.70; incidence, 12.3%

  • 200 to ≤1500 pg/mL: adjusted HR, 3.63; 95% CI, 3.13 – 4.21; incidence, 20.8%

  • ≥1500 pg/mL: adjusted HR, 5.82; 95% CI, 4.81 – 7.05; incidence, 37.5%

Preoperative NT-proBNP values were found to be statistically significantly associated with 30-day all-cause mortality:

  • ≤100 pg/mL: 0.3%

  • 100 to ≤200 pg/mL: 0.7%

  • 200 to ≤1500 pg/mL: 1.4%

  • ≥1500 pg/mL: 4.0%

Adding NT-proBNP measurement to the RCRI score improved the ability to predict the primary outcome.

The "optimism-corrected c-statistic," which utilized the RCRI score to predict the primary outcome, was 0.65 (0.64 – 0.67). It increased to 0.73 (0.72 – 0.74) when the NT-proBNP thresholds were included.

"The calibration curve did not show any important miscalibration," the authors report.

The addition of NT-proBNP to the RCRI risk calculator improved risk prediction by reclassifying patients into "more appropriate risk categories."

Including NT-proBNP values for patients with and those without events (21.4% and 26.4%, respectively) resulted in an improved net absolute reclassification improvement of 25.8%.

A risk classification improvement, calculated using a 25% change in predicted probabilities, yielded a net absolute reclassification improvement of 32.1%.

"Our study has demonstrated that NT-proBNP can help identify patients who are at higher risk for postoperative cardiac events and may glean the most benefit from perioperative troponin monitoring," the authors observe.

"The NT-proBNP is commonly used in other areas of cardiology, so most people would be familiar with the test," Devereaux commented.

"It measures both stretch in the heart and also cardiac ischemia that is commonly not recognized clinically, so it allows us to better differentiate beyond what we just see by asking people whether they have a history of prior heart attack or prior stroke," he continued.

"Risk estimates based on that information are helpful, but we know that when we rely on a patient's history, we can underestimate the risk. This test shows we can substantially make predictions based on more than only clinical information," he added.

He noted that many tests are performed prior to surgery, including creatinine and hemoglobin measurements. "To add this test, which can be acquired in a matter of minutes, has the potential to be implemented widely because we commonly perform other preop blood tests."

Moreover, he added, "we spend a lot of money doing other cardiac tests, such as nuclear tests, which cost 20 times more and are not nearly as predictive as this simple blood test. They're also inconvenient, since the tests are long and the patient has to come back to have the results reviewed, while results of this test are available in a matter of minutes."

Ethical Responsibility

An accompanying editorial by Arman Qamar, MD, MPH, and Sripal Bangalore, MD, MHA, both of New York University Grossman School of Medicine, New York City, noted that the "role of biomarkers in routine preoperative cardiac risk stratification is still in its infancy and further study is needed."

They recommend a randomized trial that compares biomarker-guided preoperative risk assessment with use of RCRI alone to potentially "answer whether this approach results in measured improvement in outcomes."

Until then, "data support the use of NT-proBNP to personalize cardiac risk stratification in patients having noncardiac surgery," they write.

Devereaux agreed. "Although more research is needed, right now, we have an ethical requirement to help inform patients not only the benefits of potential surgery but also the risks," Devereaux stated.

"If we are already doing risk assessment, we owe it to patients that if we have something that comes along that's cheaper and more accurate, it doesn't seem rational not to use it," he emphasized.

The primary funding source for the study was the Canadian Institutes of Health. Other sources of funding are listed on the original article. Devereaux is a member of a research group with a policy of not accepting honorariums or other payments from industry for their own personal financial gain. However, members do accept honorariums and payments from industry to support research, as well as costs to participate in meetings. Devereaux reports grants from Abbott Diagnostics, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol Myers-Squibb, Covidien, Octapharma, Philips Healthcare, Roche Diagnostics, Siemens, and Stryker. Qamar and Bangalore report no relevant financial relationships.

Ann Intern Med. Published online December 24, 2019. Abstract, Editorial

Follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.