Now, a range of new research results show just how dramatic the project's effect has been. And, they provide new information that may help angioplasty patients everywhere.
At the American Heart Association's Scientific Sessions 2003 this week, and in a recent issue of the American Journal of Cardiology, a team led by researchers from the University of Michigan Cardiovascular Center report several significant observational findings from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium.
Today, the researchers will present data showing that in 25,245 artery-clearing procedures performed between July 1997 and September 2002 at five Michigan hospitals, the joint effort improved the delivery of proven medications that can prevent angioplasty complications, reduced the use of potentially toxic dye used during the procedure, and reduced the unnecessary use of a blood thinner after the procedure.
These quality improvements, in turn, were associated with a lower risk-adjusted rate of in-hospital death, unplanned bypass surgery, heart attack, kidney failure caused by the toxic dye and requiring dialysis, stroke, and a composite measure of post-angioplasty adverse cardiac events. The reductions in all these measures were still statistically significant even after the data had been adjusted for patients' individual risk factors.
"These results show just what can be achieved when hospitals cooperate, rather than compete, in a joint effort aimed at improving care," says Mauro Moscucci, M.D., the U-M Health System cardiologist who leads the project in conjunction with BCBSM's David Share, M.D., MPH. "It enables us to collect data on angioplasty and other percutaneous interventions, look for variations in care and deviations from guidelines, determine the risk factors for adverse outcomes, identify opportunities for improvement, and measure the impact of our efforts on patient outcomes."
Adds Share, "The BCBSM-CC project has provided physicians the resources they need to rigorously examine variation in processes and outcomes of care, and to use the learning to optimize quality. It has been exceptionally gratifying to witness the high level of trust evidenced by otherwise competing physicians and hospitals as they work to better-define optimal quality in angioplasty care, and to operationalize that learning. The dramatic decreases achieved in mortality and complications of care are a tremendous reward for the hard work of all involved."
The consortium is funded by BCBSM, which now requires membership in the consortium for any hospital that wants to be listed as one of its Cardiac Centers of Excellence.
Eighteen Michigan hospitals, from Detroit and Ann Arbor to Grand Rapids and Flint, now participate in the consortium, contributing confidential data via a standardized but continually evolving form. Each hospital receives regular audits and reports about its individual performance, and quarterly meetings bring all the lead physicians and administrators together. Meanwhile, Moscucci and his fellow researchers mine the data for trends and useful findings.
Among the other results from the BCBSM Cardiovascular Consortium being presented at AHA or published in the Oct. 15 issue of the American Journal of Cardiology:
Reduction in kidney damage from angioplasty dye: Each patient who has an angioplasty is injected with a dye known as "contrast," which helps physicians see their blood vessels better. But contrast can harm the kidneys of certain patients (a condition called contrast nephropathy), and cause some patients to need dialysis or even die from kidney failure.
Previously, data collected in the early days of BCBSM-CC allowed Moscucci and his colleagues to see which patients were most at risk of developing kidney problems and needing dialysis. From those findings, they developed guidelines aimed at reducing the amount of contrast used and the chance of such problems. These include pre-angioplasty hydration and drug therapy, the use of a more expensive but less toxic dye, better calculations of the amount of contrast needed for individual patients, and other approaches. The team developed a pocket-sized card that clinicians used to calculate maximum contrast doses and determine each patient's risk for developing nephropathy requiring dialysis.
At AHA, U-M cardiologist Stanley Chetcuti will present data from 3,169 patients treated at the U-M between 1998 and 2002, which show a progressive decrease in incidence of contrast nephropathy, nephropathy requiring dialysis and death after the implementation of the guidelines. The guidelines and card are now used at the other hospitals, where preliminary results indicate a similar benefit. "We had known for a while that contrast could cause this effect, and that less was best, but no one had tried to address the issue of how to reduce the amount and the overall risk," says Moscucci. "This project made that possible."
Two quality-improvement projects intersect and boost each other: In addition to the BCBSM Cardiovascular Consortium, many hospitals in Michigan recently participated in the American College of Cardiology's Guidelines Applied in Practice project, which sought to improve the adherence of physicians and patients to a set of guidelines for optimum heart attack care. Eleven of the 18 BCBSM-CC hospitals previously took part in GAP, and many of the quality indicators for heart attack care - such as aspirin, beta blockers, statins and ACE inhibitors - are also good for angioplasty patients. So, Moscucci and his colleagues wanted to see how these two quality-improvement projects intersected.
In data they will present at AHA, the team found that heart attack patients at the 11 GAP/BCBSM-CC hospitals who had angioplasty were far more likely to have been receiving optimum drug therapy than patients at hospitals that did not participate in GAP. This indirectly validates the initial findings of the GAP project, and shows that the GAP effort to optimize guideline adherence was sustained even after the project ended - which may have helped patients do better when they had angioplasties.
Keeping up with the times, to keep benchmarking accurate: Previously, the BCBSM-CC effort yielded data that allowed the researchers to see which patients were most at risk of dying after an angioplasty, to develop a model to predict future patients' risk, and test that predictive tool. The successful results were published in the AHA journal Circulation in 2001, and the 18 hospitals have used that model to help assess individual patients' risk and talk with patients and their families about the potential for complications from the procedure.
But since that time, a lot has changed in angioplasty and other PCI procedures. So the team decided to test how well their risk-predicting model works on today's patients. In data they will present at AHA, they found that indeed, the "Michigan Model" wasn't quite as accurate today as it once had been - though it still beat other models developed in the early 1990s. The finding means that the BCBSM-CC model needs to be updated, which is already under way. But it also sends a warning sign that any model used to measure a hospital's performance on a given procedure - and to "benchmark" that hospital against others for rankings, reporting or reimbursement - needs to change as care changes.
Predicting and reducing angioplasty risks for diabetics: Based on previous BCBSM-CC data showing that diabetics have an increased risk of dying after angioplasty, the researchers decided to make a risk-prediction model that would tell them which factors influence a diabetic patient's chance of death - and therefore, which patients need more aggressive drug therapy. U-M cardiologist Debabrata Mukherjee, M.D., will present the results at AHA.
Using data from 7,223 diabetics treated with PCI at the consortium hospitals between 1997 and 2001, the researchers developed a simple 10-point scoring mechanism that clinicians can use. The more points, the higher the risk. For instance, patients over 70 years of age are assigned one point, and those whose heart-pumping capacity is less than 50 percent get 0.9 points. Those who are having PCI within 24 hours of suffering a heart attack get 1.4 points, but those who have successfully withstood a previous PCI procedure get half a point back. Using this simple bedside scoring mechanism, the team hopes clinicians and patients can make better decisions about angioplasty and drug therapy.
Emergency bypass after failed angioplasty - more common than we think?: In the Oct. 15 issue of the American Journal of Cardiology, a paper based on BCBSM-CC data revealed some surprising trends about emergency bypass surgery in patients who had had a PCI after heart attack. The findings have implications for hospitals that offer angioplasty but don't have a dedicated cardiac surgery service in-house as a backup - an increasingly common practice.
Overall, 2 percent of about 2,300 post-heart attack PCI patients wound up having emergency bypass surgery within 24 hours of their failed angioplasty. Another 1.7 percent of the patients had bypass surgery within the same hospitalization. The death rate for the emergency bypass patients was 20 percent, 8 percent had strokes, 8.3 percent had kidney failure requiring dialysis. Rates for post-heart attack PCI patients who did not require bypass were far lower. "This shows just how important it is for sites without surgical backup to have arrangements for transferring patients immediately if there's a complication requiring surgery, or if the cardiac catheterization identifies problems that warrant urgent or emergency surgery," says Moscucci.
Impact of other vascular disease on angioplasty outcomes: Angioplasty opens up the arteries nearest the heart, but BCBSM-CC data show that the condition of other blood vessels in patients' bodies can indicate how much risk they might face when they have an angioplasty.
In a paper published in the Oct. 15 AJC, Mukherjee and his colleagues showed that patients who had vascular disease in areas outside the heart were at a significantly higher risk of dying in the hospital and suffering other complications from heart attacks to blood transfusions - independent of any other diseases they might have.
Overall, of 25,144 PCI patients whose records were evaluated for signs that they had vascular disease, 17.4 percent had a history of conditions that indicated problems with blood vessels in the heads, abdomens, legs, and other areas. "Overall, we found that in patients who undergo PCI, those who had extra-cardiac vascular disease had significantly worse outcomes compared with those without it, even after adjusting for demographics and co-morbid health conditions," says Moscucci.
References: "Improving Outcomes of Percutaneous Coronary Interventions: The BCBSM-CC Quality Improvement Initiative in PCI", Abstract 3424, Abstract poster session 97.3a, Tues., Nov. 11, 8:30 a.m., Hall A
"Risk Adjustment and In-Hospital Mortality Following PCI: A Moving Target", Abstract 3466, Oral session 97.1, Tues., Nov. 11, 2:15 p.m., Room 231
"Development of a Continuous Quality Improvement Program for the Reduction of Contrast Nephropathy After PCI", Abstract 3471, Oral session 97.1, Tues., Nov. 11, 4 p.m., Room 231
"Quality of Care for AMI Patients Undergoing PCI is Enhanced by Participation in the American College of Cardiology Guidelines Applied in Practice Quality Improvement Initiative", Abstract 3474, Oral session 97.1, Tues., Nov. 11, 4:45 p.m., Room 231
"A Simple Risk Score for Predicting Mortality in Diabetic Patients Undergoing PCI", Abstract 3250, Poster session 99.1, Sunday, Nov. 9, 8:30 a.m., Hall A. American Journal of Cardiology: Oct. 15, 2003, Vol. 92, pp. 967-969 (bypass), pp. 972-974 (vascular)
Contact At AHA meeting: cell 734-358-4910, or
Sally Pobojewski, pobo@umich.edu
At U-M: 734-764-2220