It’s easy for Filippo Ferrigni, MD, FCCM, to call his hospital’s new insulin protocol a “no-brainer” now that it has resulted in a 32 percent overall reduction in mortality. But that doesn’t mean it was eagerly embraced when he first introduced it two years ago.
“There was a good deal of concern about this when we started,” recalls Ferrigni, director of the intensive care unit and vice president of clinical affairs at St. Joseph Health Center (SJHC) in St. Charles, Missouri, USA, part of SSM Health Care (SSMHC). Sponsored by the Franciscan Sisters of Mary and based in St. Louis, Missouri, SSMHC was the first health care organization to win a Baldrige award, the nation's premier award for performance excellence and quality achievement.
The protocol to which Ferrigni refers was initially introduced in the intensive care unit (ICU), though it is now hospital-wide. It is aimed at aggressively controlling the blood sugar levels of all patients — not just those with diabetes — in an effort to reduce infection. ICU patients typically have elevated glucose levels, brought on by the stress of acute illness and the resulting release of adrenaline and stress hormones. Studies show that higher blood sugar levels are associated with higher rates of infection and mortality.
“This relationship has been understood for at least 25 years,” says Ferrigni. “Elevated blood sugar levels reduce the ability of white cells to fight infection.” What wasn’t known until recently, he explains, is just how low the levels need to be to decrease the risk of infection. “It was commonly thought that white cells were compromised when the blood sugar levels were over 200. Now we know that any abnormal blood sugars are associated with worse outcomes.”
Taking Small Steps Forward
Ferrigni studied the work of Anthony Furnary, MD
, a cardiac surgeon at Providence St. Vincent Medical Center in Portland, Oregon, USA, who used aggressive insulin therapy to help his patients fight post-surgical infections; and the work of a group of Belgian physicians who demonstrated that maintaining blood glucose at a level between 80 and 110 mg per deciliter reduced mortality not only in diabetic patients, but in all surgical ICU patients. (New England Journal of Medicine, November 8, 2001; Volume 345:1359-1367.)
“We discussed the data quite a bit internally,” says Ferrigni, “but there were some who fundamentally didn’t believe it. So we agreed to start slowly. Just measuring glucose routinely in non-diabetic patients was a step forward.”
Through the hospital’s lab system, ICU staff was able to perform bedside blood sugar checks. They found initially the average glucose level in the ICU was 182, about the same as the control group in the Belgium study. Ferrigni knew he needed to introduce the protocol slowly, so he set the first target at 175. “People were afraid of causing low blood sugar and having patients ‘crash,’ so I picked a modest goal that was essentially where we already were.” Even though the goal itself did not represent a challenge, the concept did.
“Under the old model, primary care doctors didn’t want to be called about a patient’s blood sugar unless it was over 300. This required a total shift in thinking,” says Pamela Walsh, RN, MSN, APN, clinical director of the ICU and Cardiovascular Recovery (CVR). “It took a lot of education and prompting, both for doctors and nurses.”
Getting Buy-In, One By One
If controlling blood sugar after surgery was a daunting challenge, controlling it during surgery was even more so. “In the operating room, it is very difficult to get blood sugar down,” says Ferrigni. “In the operating room, anesthesia staff run the insulin, and our anesthesia group didn’t think you could control blood sugar in surgery. The stress to the body is so high during surgery, you have to be really aggressive to control blood sugar. We had to convince them it was worth the extra effort, which is significant.”
Ferrigni says one anesthesiologist agreed to try it, but initial results weren’t impressive. “I put them in touch with the folks at St. Vincent’s, and they were very helpful,” says Ferrigni. “We weren’t being aggressive enough, we weren’t using enough insulin.”
Gradually, as the results began to improve, more anesthesiologists came on board. “Now we start an IV insulin drip the night before surgery. All patients get their blood sugar checked before surgery, and if it is too high, we cancel the surgery.” Getting to this point, says Ferrigni, took about a year.
For post-surgical patients, the average blood sugar goal was reduced at two-month intervals, says Ferrigni. “We used an IHI approach
,” says Ferrigni. “Put a protocol in place, choose a target, achieve the target, make sure there were no bad outcomes, and then reduce the target.”
As staff began to get used to the new way of thinking and caring for patients, they began to teach one another. “If a primary care physician wrote a sliding scale insulin order and it didn’t go with what we had been taught under the new protocol, Dr. Ferrigni would work with the physician to educate him or her,” says Walsh. “He worked with the doctors, and I worked with the nurses.”
When ICU patients’ average blood sugar levels reached about 150, Ferrigni says the data began to show a reduction in ICU mortality. This was due in large part to the dramatic reduction in deep sternal wound infections in patients who had undergone cardiac surgery.
“Deep sternal wound (DSW) infection is a devastating infection involving tissues deep under the sternum,” says Ferrigni. “It is very hard to treat, and increases the heart surgery mortality rate from 2 percent to 20 percent.” With about 300 heart surgeries a year, St. Joseph’s averaged about five DSW infections annually. Presently, says Ferrigni, “it’s been 11 months since we’ve had one.”
Walsh says ICU staff has found it’s easier to control blood sugar in some patients than in others. “We are still tweaking the protocol,” she says. Patients with chronic obstructive pulmonary disease (COPD), for example, are given steroids, which have an impact on blood sugar. “Blood sugar in COPD patients is harder to control than in heart patients,” she says. “But overall, we have very tight control in the ICU. Our goal is less than 130, and our average is currently 128, and that's really good blood sugar control across the whole unit."
Moreover, working to control blood sugar has become second nature to the unit’s nurses, says Walsh. “One nurse said to me today, how long did we moan and groan about the ‘Portland protocol,’ and now we know it like the back of our hands.” Through daily multidisciplinary rounds, all ICU clinicians are reminded of the importance of blood sugar control. “The nurses are so in tune, they always know what the last reading was,” says Walsh. And the intensivists who staff the ICU are all in agreement that the protocol is effective.
Ferrigni says that based on the success of the protocol in the ICU, tighter blood sugar control was also implemented hospital-wide. Today, all patients at SJHC have their blood sugar monitored routinely, and hospital-wide mortality has fallen from 2.5 percent to 1.8 percent, a 32 percent reduction. The cost, says Ferrigni, is remarkably reasonable, given the results. “We used to spend about $2,000 a year on insulin. Now we spend about $6,000.”
Ferrigni is currently in the process of helping to spread the protocol to some or all of the other 20 hospitals in the SSMHC network by facilitating an internal collaborative on improving ICU care. “We learned how to run collaboratives from IHI,” says Ferrigni, who participated in an IHI Breakthrough Series Collaborative on Critical Care several years ago. “Now, just like IHI, we’d like to share the protocols with anyone who wants them.”