More information about text formats
We read with interest the recent editorial by Siotia and Gunn  which emphasised the need and growing enthusiasm for risk scoring for percutaneous coronary intervention (PCI). This editorial focuses on the recent publication by Wu and colleagues from New York of a risk model to predict in-hospital mortality following PCI . This particular risk model was based on 46,090 patients undergoing PCI between 2...
We read with interest the recent editorial by Siotia and Gunn  which emphasised the need and growing enthusiasm for risk scoring for percutaneous coronary intervention (PCI). This editorial focuses on the recent publication by Wu and colleagues from New York of a risk model to predict in-hospital mortality following PCI . This particular risk model was based on 46,090 patients undergoing PCI between 2002 and 2003, and is clearly the largest study of its kind. However, it only examines in-hospital mortality as an outcome and does not include other important
complications following PCI, such as Q-wave myocardial infarction (MI), emergency coronary artery bypass graft (CABG) surgery, and cerebrovascular accidents.
We have recently published a risk model for PCI which was based on 9,914 patients undergoing PCI between 2002 and 2003 using registry data from the North West Quality Improvement Programme (NWQIP) . Although the NWQIP model was based on a sample size one quarter of that available
in the New York model, it examined outcomes of in-hospital mortality, Q-wave MI, emergency CABG, and cerebrovascular accidents.
We share Siotia and Gunn's enthusiasm to establish risk scoring in PCI and hope that in 2007 this will be possible for all interventional cardiologists. As with all risk models, the NWQIP model will have limitations and more work will be required to examine its applicability in other health care systems and with continuing changes in PCI practice.
However, what the NWQIP model does offer is a contemporary risk model for PCI in a UK setting, which uses variables which are part of the minimum dataset for the British Cardiovascular Interventional Society, which is currently submitted to the Central Cardiac Audit Database (CCAD) .
Currently the NWQIP model is being validated to see how applicable it might be to other hospitals outside of the north west of England; preliminary results look good. Recent analyses using PCI data available on the CCAD database showed promising preliminary results (David Cunningham, personal communication); the NWQIP algorithm appears to be predictive for all MACE occurrences within 30 days, although more detailed analysis will be required to confirm these early findings.
The intervention unit at the Royal Bournemouth Hospital recently assessed their first 1013 cases and concluded that the NWQIP model was a powerful tool to assess interventional performance. Dr Witherow concludes
that the NWQIP model provides an up to date, simple, practical scoring system and should be used to its full potential by as many centres as possible .
Competing interests: ADG, MJ, and RHS are all members of NWQIP and were involved in developing the NWQIP risk model for PCI.
 Siotia A, Gunn J
Risk scoring for percutaneous coronary intervention: let's do it! Heart 2006;92(11):1539-40
 Wu C, Hannan EL, Walford G, Ambrose JA, Holmes DR Jr, King SB 3rd, Clark LT, Katz S, Sharma S, Jones RH
A risk score to predict in-hospital mortality for percutaneous coronary interventions.J Am Coll Cardiol 2006;47(3):654-60
 Grayson AD, Moore RK, Jackson M, Rathore S, Sastry S, Gray TP, Schofield I, Chauhan A, Ordoubadi FF, Prendergast B, Stables RH
North West Quality Improvement Programme in Cardiac Interventions. Multivariate prediction of major adverse cardiac events after 9914 percutaneous coronary interventions in the north west of England.Heart 2006;92(5):658-63
 Central Cardiac Audit Database. Coronary heart disease audit supported by the Central Cardiac Audit Database.www.ccad.org.uk (accessed 19 Oct 2006)
 Witherow FN
What is 'high risk' PCI? Heart Online, 1 Sep 2006 (eLetters: Heart 2006;92(5):658-63)