Interventionalists and the Syntax Score—Still Learning

Download this article's Factoid (PDF & PPT for Gold Subscribers)

When it comes to accurately assessing Syntax scores for patients with complex coronary artery disease, interventional cardiologists new to the task fall short of angiographic core laboratory technicians. Even after improving their skills with intensive training, they tend to underscore important angiographic components, according to a small study published online October 25, 2011, ahead of print in Circulation: Cardiovascular Interventions.

To assess the reproducibility of the Syntax score and thus its utility in clinical practice, Gregg W. Stone, MD, of Columbia University Medical Center (New York, NY), and colleagues had 3 interventional cardiologists and 4 angiographic core lab technicians undergo basic training from the Syntax score Web site and then independently review and assign scores to 30 multivessel disease angiograms. The interventional cardiologists then participated in an extensive in-person training session with the core lab team, and all 7 readers each analyzed an additional 50 cases.

After 12 weeks, the interventionalists were asked to read the same 50 cases again in order to assess intraobserver variability. As a basis for comparison, researchers performed standard quantitative coronary angiography (QCA), thereby generating a QCA-derived Syntax score for each of the 80 cases. Interobserver statistic values were determined, with higher levels of achievement being closer to 1 and lower levels closer to 0.

Interventionalists Improve with Training—but Still Underscore 

Overall, the strength of agreement among technicians for both periods was substantial or greater (k = 0.82; 95% CI 0.72-1.00 and 0.84; 0.76-1.00) and not different from QCA. In contrast, the interventionalists’ agreement originally was assessed as fair (k = 0.33; 0.18-0.44) but increased to substantial or greater after more intensive training (k = 0.76; 0.64-1.00).

After 12 weeks, the interventional cardiologists repeated their reading of the 50 cases and showed at least moderate agreement with their previous scores for all components.

Despite their improved performance, however, the interventional cardiologists still underscored the number of lesions, bifurcation, and small vessel disease, which resulted in lower Syntax scores (table 1).

Table 1. Mean Difference in Quantitative Components of the Syntax Score

When the researchers classified patients into Syntax score tertiles (low: ≤ 22, intermediate: 23-32, and high: ≥ 33), they found that slightly more patients were stratified as low risk by the interventionalists than the technicians, and fewer were stratified as high risk.

Increased Reproducibility Still Needed

Dr. Stone told TCTMD in a telephone interview that he was surprised at both how accurate the technicians were and how poorly the interventionalists initially performed. “Obviously, experience does count, and you get very good when you do something over and over,” he said.

Most concerning, he said, was that “one interventional cardiologist would say a patient has a high Syntax score while another might say the patient has a low Syntax score. [That means] the first interventionalist might refer the patient to surgery, while the second interventionalist might say this is an easy case for PCI. And this obviously has a lot of clinical implications.”

Tarun Chakravarty, MD, of Cedars-Sinai Medical Center (Los Angeles, CA), said that overall the study reinforces the value of the Syntax score but suggests that its use should be limited to clinical trials until the reproducibility of the tool can be improved.

“You already know that it’s a very strong predictor of outcomes, so all we need is a reproducible score that can be accurately assessed in the cath lab,” he told TCTMD in a telephone interview. “Then it would be a very strong tool that can be used to stratify patients [to] PCI or CABG.”

Further Training Imperative 

As for what could be done to decrease reader variability, Dr. Chakravarty suggested further research into a Syntax score based on fractional flow reserve (FFR), as did the study authors.

The current Syntax score asks readers to count lesions based on their subjective evaluation of the degree of stenosis, Dr. Chakravarty said, but “with an FFR Syntax score, you can get an accurate assessment of the obstruction—it’s objective.” He added that a computerized Syntax score assessment, similar to computerized QCA, might also make the tool more appropriate for routine clinical use.

Drs. Chakravarty and Stone agreed that interventional cardiologists need more training than merely completing the online Syntax score tutorial. Dr. Chakravarty said more focus should be placed on the bifurcation analysis “because that’s a significant predictor of outcomes when you think of PCI.”

Dr. Stone added that interventionalists should undergo additional training with angiographic lab technicians, similar to that in the study. “There needs to be an advanced tutorial for people to get really comfortable [with the Syntax score],” he said. “It should be a series of cases where people score cases and then get automated feedback to show what they did right and what they did wrong. It probably would take a good 6 or 8 hours before interventionalists get good at this.”

Regardless of how much training interventionalists receive, Dr. Stone emphasized, “no diagnostic test is perfect.” However, the Syntax score has the potential to be a useful clinical tool if implemented properly, he added.

“There’s still a lot of work to do in this field, but I think this study was an important first step to exposing some of the subtleties with any sort of measuring technique,” Dr. Stone said. “Differences in interpretation and measurement affect clinical judgment and decision making.”

Note: Dr. Stone and the study coauthors are faculty members of the Cardiovascular Research Foundation, which owns and operates TCTMD.

 

---

Source:
Généreux P, Palmerini T, Caixeta A, et al. Syntax score reproducibility and variability between interventional cardiologists, core laboratory technicians, and quantitative coronary measurements. Circ Cardiovasc Interv. 2011:Epub ahead of print.

 

 

Related Stories:

• Clinical Variables Added to Syntax Score Increases Predictive Ability
• FFR Plus Syntax Score Equals Better Risk Stratification in Multivessel Patients
• The Syntax Score: Can Learning Curves and Old Habits Be Overcome?