New protocol also reduces LOS by nearly 11%
A protocol of intensive monitoring and treatment to maintain proper blood glucose levels in all intensive care unit (ICU) patients at The Stamford (CT) Hospital reduced mortality rates by 29% and length of stay in the ICU by nearly 11%. The initiative, headed by James Krinsley, MD, medical director of the ICU, also:
- decreased mean glucose from 152.3 mg/dL to 130.7 mg/dL, marked by a 56.3% reduction in the percentage of glucose values greater than 200 mg/dL, without an increase in hypoglycemia;
- gained buy-in of the ICU nurses, integrating their input into the initiative and the target glucose values;
- created a data-driven ICU;
- increased motivation, empowerment, and skill development of the nursing staff;
- demonstrated that close monitoring and rapid treatment of patients using this simple and low-cost intervention can profoundly affect morbidity and mortality of critically ill patients.
In addition, the development of new renal insufficiency decreased by 75%, as well as the number of patients receiving transfusions of packed red blood cells (by 18.7%).
Stamford Hospital is a not-for-profit, community teaching hospital that has 305 inpatient beds in medicine, surgery, obstetrics/gynecology, psychiatry, and adult and neonatal critical care units.
This initiative involved 800 consecutive patients admitted after institution of the protocol (between Feb. 1, 2003, and Jan. 10, 2004) and 800 patients admitted immediately preceding institution of the protocol (the baseline group, between Feb. 23, 2002, and Jan. 31, 2003). The setting was the 14-bed medical-surgical ICU in a university-affiliated community teaching hospital.
The glycemic management protocol used intensive monitoring and treatment to maintain blood glucose values less than 140 mg/dL. It incorporated frequent FSG (fasting serum glucose) checks; minimal initial monitoring was every three hours — once an hour during IV insulin infusions.
Continuous intravenous insulin was used if glucose values exceeded 200 mg/dL on two successive occasions; subcutaneous regular insulin was used for milder hyperglycemia. After institution of the protocol, the mean glucose value decreased from 152.3 mg/dL to 130.7 mg/dL.
Examining the database
After reading a paper on intensive insulin therapy in critically ill patients in the New England Journal of Medicine,1 "The first thing I did was look at our own database [of ICU patients] to see if [there was a similar] relationship between glucose levels and mortality," Krinsley recalls.
What he found was dramatic: Survivors had mean glucose values of 137.9 mg/dL, while nonsurvivors had mean glucose values of 172. The lowest hospital mortality, 9.6%, occurred among patients with mean glucose values between 80 mg/dL and 99 mg/dL.
Hospital mortality increased progressively as glucose values increased, reaching 42.5% among patients with mean glucose values exceeding 300 mg/dL. "The nurse director, with whom I share a wall, must have heard me scream," he says.
This led to the first of two papers Krinsley would publish in the Mayo Clinic Proceedings.2 "At the same time, I started discussions with the nurses [in early 2002], and then we finally put the protocol in place," he says. (The second Mayo paper3 describes this initiative.)
In his earlier study, Van den Berghe had set a goal of 110 mg/dL, but the nursing staff were uncomfortable with that goal, and the staff were inadequate, Krinsley notes. "So we had further discussions and compromised with a more traditional 140. This was OK with the nurses, and it fit our data."
Krinsley and the staff wrote the protocol together. "We sat down together in a room, hashed it out, then revised and revised," he recalls. "And we are continually tweaking it. In fact, as of two weeks ago, every patient whose blood glucose is over 125 now gets treated, and [insulin] dripped over 180."
The nurses actually initiated this most recent change, Krinsley notes. "They wrote the protocol, and there was no resistance; it shows you how the culture has changed in two years."
The new protocol is even more flexible than the old one, he says, because it allows a greater degree of clinical flexibility at the bedside. "With drips, for example, you specify an initial infusion rate, but the subsequent drip rates are really at the nurse’s direction. You can’t possibly write a protocol that is as complex and accurate as the integration of data that goes on inside a good nurse’s brain."
The protocol itself has been relatively inexpensive, Krinsley adds. No additional staff have been hired; a few extra bedside glucose monitors have been purchased. And while the use of insulin has increased, the cost of the drug is minimal.
A data-driven ICU
One of the keys to the success of the initiative was the creation of a data-driven ICU, he adds. "The use of data permeates the culture of the ICU. I have the databases, and everything we do has data associated with it. I print out colored copies to the nurses, so they know that everything that can be measured is measured."
Glucose is the most complex issue that has been tackled using the database, Krinsley notes. "There are a whole set of treatment issues, and we would not have been able to do this three or four years ago. It requires experience with protocols and a good data management system."
Krinsley also has noted increased motivation, empowerment, and skill development on the part of the nursing staff.
"A real esprit has developed because they know they are doing something that really matters, and it has given us national recognition [with the Ernest A. Codman Award], so there’s a good deal of pride involved."
Protocol extended
One of the other positive results of the initiative is that the protocol has been extended to the hospital’s intermediate care unit, called the special care unit.
"We’ve not gone to 125, because the nursing ratio is not as tight," he says. "However, while there are no formal data yet, we have had a good impact on lowering glucose levels.
"The whole project has been a win-win,’ Krinsley adds. "My next paper, which I’m working on right now, is about the cost savings associated with glucose management," he reveals.
While he could not share any data before publication, Krinsley did say, "It will be very big number. So, here we have a low-cost intervention that not only has a positive mortality effect, but also a major cost-savings effect."
References
1. Van den Berghe, et al. A low-cost intervention that will become the standard of care in the ICU. N Engl J Med 2001; 345:1,359-1,367.
2. Krinsley JS. Association between hyperglycemia and increased hospital mortality in a heterogeneous population of critically ill patients. Mayo Clin Proc 2003; 78:1,471-1,478.
3. Krinsley JS. Effect of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004; 79(8):992-1,000.
Need More Information?
For more information, contact:
• James Krinsley, MD, Medical Director, Intensive Care, Stamford Hospital, 30 Shelburne Road, Stamford, CT 06904-9317. Phone: (203) 276-5506. E-mail: [email protected].
A protocol of intensive monitoring and treatment to maintain proper blood glucose levels in all intensive care unit (ICU) patients at The Stamford (CT) Hospital reduced mortality rates by 29% and length of stay in the ICU by nearly 11%.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content