Heart-bypass and valve surgeries continue to be a good bet for patients in Pennsylvania, even as the widespread use of stents and statins means that hospitals perform far fewer procedures than they did years ago, a new state report found.
Post-surgery death rates ranged from 1 percent to 4 percent at most facilities during a recent two-year period, the Pennsylvania Health Care Cost Containment Council said Tuesday. The report also found that patients are increasingly unlikely to end up back in the hospital within 30 days of discharge – a quality-of-care indicator sure to please insurers seeking to rein in costs.
The only Philadelphia-area hospital with a better-than-expected mortality rate in any category of heart surgery was Doylestown Hospital, where zero patients died after undergoing valve surgery during the period in question: January 2014 through March 2016.
The lone area hospital with a worse-than-expected mortality rate in any category may come as a surprise: the Hospital of the University of Pennsylvania, an academic medical center that does the most heart procedures in the state.
Of 345 patients who underwent a bypass operation with no accompanying valve surgery, 15 died in the hospital, the flagship facility of Penn's health system. That's a mortality rate of 4.3 percent.
Agency statisticians calculated that the hospital's rate should have been between 1.2 percent and 4.1 percent for the period in question, meaning that with one fewer death, Penn would have performed within the expected range.
The hospital's performance in three other categories – valve surgery with a bypass procedure, valve surgery without bypass, and all valve surgeries lumped together – fell within the expected range.
The state analysis considered how sick each patient was, calculating a different expected mortality rate for each hospital based on factors such whether patients had been suffering from liver or kidney disease in addition to their cardiac issues.
But officials at Penn cautioned that the analysis does not truly reflect the hospital's expertise. Hundreds of times per year, the facility accepts patients that other hospitals deem too difficult to take on. Half of its bypass and valve surgeries were considered so complex that they were excluded from the state analysis.
"We see ourselves as the hospital of last resort," said Patrick J. Brennan, chief medical officer for Penn Medicine, the university's health system.
For all heart surgery at Penn, including transplants, patients have a better-than-expected survival rate, Brennan said. The hospital's mortality index is 0.89 – that is, 11 percent better than the expected baseline of 1, he said.
For the three hospitals in the Main Line Health system — Lankenau, Bryn Mawr, and Paoli — mortality rates fell within the expected range for all types of bypass and valve surgeries, the state analysis found.
That success is due in part to the health system's increasing use of a minimally invasive approach, in which the surgeon makes a two-inch incision rather than the traditional 10-to-12-inch variety, said Konstadinos A. Plestis, system chief for cardiothoracic and vascular surgery.
That makes for fewer complications and shorter recovery times, Plestis said.
"It's usually three weeks vs. three months for the traditional way," he said.
For each of the four surgery categories, the state also evaluated hospitals on the percentage of patients that had to be readmitted within 30 days of discharge. Medicare and other insurers have increasingly pressed hospitals to reduce readmissions; Medicare even assesses a penalty against hospitals that do not fare well.
Hospitals with worse-than-expected readmission rates in at least one of the four categories included Einstein Medical Center Philadelphia, Aria Health, Chester County Hospital, Doylestown Hospital, Hahnemann University Hospital, and HUP.
Just one area facility – Temple University Hospital – had a better-than-expected readmission rate in any category: bypass surgery that did not involve a valve procedure.
Based on Temple's mix of patients, state statisticians calculated that the percentage of patients who would need to come back to the hospital within 30 days should fall somewhere between 11 percent and 20.8 percent. But Temple's actual number was just 10.4 percent for that period.
The readmission analysis was based on discharges from a shorter time period – Jan. 1, 2014 through Aug. 31, 2015 – because of a change in the system used for insurance coding.