New Risk Score Predicts In-Hospital/30-Day Mortality Following PCI
Using a population-based registry, researchers have developed a new percutaneous coronary intervention (PCI) risk score that quickly and accurately predicts in-hospital/30-day mortality. The risk score study was published June 17 in JACC: Cardiovascular Interventions.
To develop the risk score, researchers used data from 54,233 patients who underwent PCI in 2010 from New York State's Percutaneous Coronary Interventions Reporting System (PCIRS). PCIRS collects data on all PCIs performed in nonfederal hospitals in New York state, and the risk score was validated using 2009 data.
Eleven independent predictors of mortality were identified (age, hemodynamic instability, ejection fraction, pre-procedural MI [with and without ST-segment elevation], peripheral vascular disease, congestive heart failure, malignant ventricular arrhythmia, chronic obstructive pulmonary disease, renal failure, two- or three-vessel disease and left main disease) and assigned a score ranging from 0 to 9 points. The lowest possible risk score was 0, which was associated with a 0.09 percent risk of in-hospital/30-day mortality. The highest score was 43 points, which was associated with a 99.94 percent chance of in-hospital or 30-day death.
The authors note that use of such a database that includes virtually all PCIs in a large region "contributes to the accuracy and generalizability of our findings." Further, they note that "risk score performed reasonably well as a predictor of complication rate and of length of stay. However, any attempts to predict complications and length of stay with strong accuracy should involve developing separate risk scores for these adverse outcomes," they add.
In an accompanying editorial, Lloyd W. Klein, MD, FACC, and Justin Maroney, MD, from the Advocate Illinois Masonic Medical Center and Rush Medical College, Chicago, wrote that the risk model has several strengths. "It is easy to calculate, and the score is easily comprehended. The variables used to construct the model are objectively defined. The population cohort from whom the score is derived is large ... Registry data collection is compulsory rather than voluntary and is routinely audited for accuracy and completeness."
However, they add that there are inherent limitations to such models when applying the findings to specific, individual patients. "The value of the models is determined entirely by the specific variables collected by the sponsoring registry," they wrote. "For example, existing registries do not collect specific data concerning patient frailty, incomplete revascularization, patient preferences, many comorbid conditions or other extenuating circumstances that may be highly relevant to the decisions being made." Overcoming such limitations to develop more comprehensive predictive models is the future challenge for PCI risk scores, they concluded.
< Back to Listings