Where: Sunnybrook Health Sciences Centre, a 1,375-bed tertiary care hospital in Ontario, Canada.
The issue: Reducing the incidence of heparin-induced thrombocytopenia (HIT).
For most of the past 30 years at Sunnybrook, HIT was an occasional but dangerous complication that seemed to be part of the cost of caring for patients who needed anticoagulation. But after a near-fatal case of HIT in a cardiac surgery patient, the hospital's cardiac surgeons asked William H. Geerts, MD, director of the thromboembolism program, if there was a way to prevent the condition. He searched the literature for prior HIT-prevention interventions and came up empty. “So we thought, ‘Well, maybe we should try to reduce the burden of HIT,’” said Dr. Geerts, also a professor of medicine at the University of Toronto.
Knowing that low-molecular-weight heparin (LMWH) poses a substantially lower risk of HIT than unfractionated heparin (UFH), he and a multidisciplinary committee came up with an “avoid-heparin” policy in 2006, with clinicians replacing UFH with LMWH wherever possible.
How it works
Originally, the effort was focused on preventing HIT in cardiac surgery patients, who routinely receive UFH and incur the most HIT-related complications. But the Sunnybrook clinicians soon realized that it would be very difficult to restrict the intervention to cardiac surgery patients alone. They also wondered if other surgical and medical patients in the hospital might also benefit from an avoid-heparin policy.
Their avoid-heparin initiative had six main components: 1) replacing heparinized saline with regular saline in venous and arterial line flushes, 2) replacing UFH used as venous thromboembolism (VTE) prophylaxis with LMWH, 3) replacing intravenous UFH with LMWH for acute VTE treatment, 4) replacing UFH with LMWH for postoperative mechanical heart valves and other therapeutic indications after cardiovascular surgery, 5) replacing UFH with LMWH on nearly all order sets, and 6) removing UFH from most patient care areas.
The project was well timed, according to Dr. Geerts, as the hospital had already decided to give LMWH to most patients receiving thromboprophylaxis because of its once-daily dosing regimen (rather than two or three times a day for UFH). “We were actively involved in all the anticoagulant components of order sets,” he said. “We simply didn't include [unfractionated] heparin in any of the order sets.”
This all occurred “fairly painlessly and seamlessly,” with most physicians and nurses failing to even notice that LMWH had been slipped into the order sets, said Dr. Geerts. “We did have input from various doctors, nurses, and pharmacists, but we didn't broadcast the initiative through the whole hospital to say, ‘As of this date, we're doing this policy.’” There were some key exceptions, such as in cardiac and vascular surgery, where UFH continued to be used intraoperatively.
Comparing the three years prior to the intervention (2003 to 2005) with the six years that followed it (2007 to 2012), the team's researchers found that HIT rates diminished dramatically. They saw a 79% reduction in adjudicated HIT, a 91% reduction in HIT with thrombosis, and an 83% reduction in HIT-related costs of care (by $266,938 per year in 2007 Canadian dollars), according to results published in April 2016 in Blood. They also reported a 42% decrease in suspected HIT and a 63% decrease in patients with positive HIT enzyme-linked immunosorbent assays.
Prior to the initiative, the hospital saw about 20 cases of HIT annually, with about 60% leading to HIT-related thrombosis. Now, cases are rare. “We almost never see HIT now. It's not zero, but the rates have gone to almost zero. There are now residents who never see patients with HIT because it's so uncommon here,” Dr. Geerts said.
At first, two groups of clinicians were skeptical of the project: anesthetists and nephrologists. The anesthetists were concerned that line failures would increase without use of heparinized saline, and the nephrologists were concerned that patients with renal failure would bleed with LMWH, said Dr. Geerts. Neither concern manifested in practice.
The one limitation of LMWH is that, unlike UFH, it is partially eliminated from the body through the kidneys, Dr. Geerts noted. “So if you have patients with severe kidney dysfunction, then you have to choose your LMWH dose more carefully because you don't want the drug to accumulate,” he said.
The cost of LMWH may also pose a challenge for some hospitals trying to follow Sunnybrook's model. In Canada, LMWH and UFH are similarly priced. “In a lot of American hospitals, the cost of LMWH is still quite a bit higher than for UFH. If the cost of LMWH in our hospital was many times greater than the cost of UFH, we would not have been able to justify doing this,” Dr. Geerts said. Still, other institutions, such as the University of North Carolina, have begun rolling out their own avoid-heparin policies, he noted.
Sunnybrook now only sees a few patients a year with HIT, which has led to a change in diagnostic practices, Dr. Geerts said. “We're not suggesting that clinicians not consider the diagnosis of HIT, but because we now see it so rarely and the false positive rates are much more common than the true positive rates among those tested, it's appropriate for us now to raise our threshold for considering this disease,” he said.