September 13th, 2010

Three-year SYNTAX Results: Sensible, Not Sensational

In SYNTAX, 1800 patients with multivessel and/or left main disease were randomized to CABG or PCI with DES after a surgeon and an interventional cardiologist reviewed the coronary angiogram and agreed that either procedure was appropriate. (See the CardioExchange News blog for more study information.)

The SYNTAX 3-year results show that patients with a low SYNTAX score have similar outcomes with PCI or CABG, whereas those with an intermediate or high score have a better outcome with CABG.

1) In the cohort as a whole, MACCE rates were higher for PCI than for CABG (28% vs 20%), mainly because of a higher rate of repeat revascularization in the PCI group.

a) Composite safety (death/stroke/MI) was similar in the PCI and CABG groups (14.1% and 12.0%, respectively)

b) MI incidence was nearly twice as high in PCI patients than in CABG patients (7.1% vs. 3.6%)

2) In patients with a low (0-22) SYNTAX score, MACCE rates were similar for PCI and CABG (22.7% and 22.5%, respectively)

3) In patients with an intermediate (23-32) or a high SYNTAX score, MACCE rates were higher with PCI than CABG (27.4% vs 18.9% for an intermediate score; 34.1% vs 19.5% for a high score)

Now be honest….do you calculate a SYNTAX score for your patients?

If you don’t, how do you determine which patients will do as well with PCI as with CABG?

2 Responses to “Three-year SYNTAX Results: Sensible, Not Sensational”

  1. A report in Circulation Jan 9 2010 by Hannan et al noted that 94% of patients for whom PCI was indicated by the ACC/AHA guidelines were recommended for PCI, and that 93% of patients for whom either bypass surgery (CABG) or PCI was indicated by the guidelines were recommended for PCI. Ray Gibbons wrote an interesting editorial worth reading. SYNTAX is more rigorous than data used then.

    Competing interests pertaining specifically to this post, comment, or both:

  2. The overall predictive performance of SYNTAX score in predicting PCI-related outcomes is poor – a PPV of less than 15% and a NPV of about 90%. It is virtually zero for predicting CABG-related outcomes. For example, in the SYNTAX trial, the 2-year CABG-related mortality was 7.9% in those with SYNTAX score of 0-32 compared with a mortality of 4.1% in those with SYNTAX score of greater than 33! Clearly, clinical variables such as age and renal function and LV function are far more useful in predicting outcomes than anatomic variables.

    While I am disappointed in the predictive utility of SYNTAX score, I am reassured that no one will be calculating SYNTAX score for decision-making in the real-world clinical practice. It is only good for writing papers and guidelines!