Patients hospitalized for treatment of cancer or other medical conditions who are diabetic or become diabetic need aggressive glycemic control, MiKaela Olsen, RN, MSN, OCN, said at the Oncology Nursing Society 31st Annual Congress
BOSTONPatients hospitalized for treatment of cancer or other medical conditions who are diabetic or become diabetic need aggressive glycemic control, MiKaela Olsen, RN, MSN, OCN, said at the Oncology Nursing Society 31st Annual Congress (abstract 69).
"Often, doctors do not really listen when you tell them the patient's blood glucose levels are 200 or 250 mg/dL, but the consequences can be severe. One blood glucose level greater than 220 mg/dL results in a 5.8 times increase in nosocomial infection rates. Two hours of hyperglycemia results in impaired white cell function for weeks. And a blood glucose greater than 200 mg/dL in a hospitalized patient causes increased morbidity and mortality," said Ms. Olsen, an oncology nurse specialist at Johns Hopkins Hospital.
Ms. Olsen noted that rapid growth in the incidence of diabetes is forcing practice changes. "If you haven't had them happen in your institution, you will," she said. "They will be mandated at the bedside. And we often are not prepared to deal with diabetics, since the management of diabetes in the hospital is generally considered secondary in importance to the admitting problem, for example, cancer." She also pointed out that oncology providers often lack knowledge about the most up-to-date management of diabetes, such as current research showing major benefits of tight glycemic control in the ICU setting.
Ms. Olsen said that at least one-third of patients admitted to the hospital either have a diagnosis of diabetes at entry or will be found to have high glucose levels during their stay. She reviewed the new American Diabetes Association criteria for the diagnosis of diabetes: a casual plasma glucose of 200 mg/dL or higher, fasting plasma glucose of 126 mg/dL or higher (lowered from 140 mg/dL in the previous guidelines), or a 2-hour glucose of 200 mg/dL or higher during an oral glucose tolerance test.
"It is amazing how many patients are being harmed or even killed because people do not know that a type I diabetic has to have exogenous insulin. Without exogenous insulin, the patient will develop diabetic ketoacidosis within hours, which is potentially lethal," Ms. Olsen said. She also warned that the longer patients have type II diabetes, the more they will begin to look like type I patients and need exogenous insulin.
Tight Control in the ICU
Intervention studies have shown improved outcomes with intensive glycemic control. Ms. Olsen described work by Van den Berghe et al documenting a significant reduction in mortality and morbidity, including sepsis and blood transfusion use, with tight glycemic control (80 to 110 mg/dL) vs conventional control (180 to 200 mg/dL) (N Engl J Med 345:1359-1367, 2001).
She also noted American Association of Clinical Endocrinologists guidelines that recommend maintaining ICU patients' blood glucose levels at less than 110 mg/dL. "Outside the critical care setting, the pre-meal glucose should be not more than 110 mg/dL; and the maximum glucose 2 hours postprandial should not exceed 180 mg/dL," she said.
Patients today expect good glycemic control as a part of their hospital care. They have been taught by their endocrinologist and diabetes educator to keep their blood sugar under control when they are at home. "They have difficulty understanding our casual approach to their hyperglycemia when they come into the hospital," Ms. Olsen said.
In caring for hospitalized diabetic patients, she warned that sliding scale insulin regimens (based on blood glucose values) are inadequate and should be abandoned in favor of newer insulin therapy approaches. "Traditional sliding scales are dangerous when not used with basal insulin, are unlikely to be modified throughout the hospitalization, and are not standardized. Some of you may have preprinted orders, but a lot of physicians just handwrite these. This approach treats hyperglycemia after the fact instead of proactively taking care of these patients," Ms. Olsen said.
Proactive treatment combines basal insulin, which is needed regardless of whether the patient is sleeping or awake, nutritional insulin to cover food intake, and correctional insulin to correct any highs, she said.
"Treatment is moving toward the rapid and the long-acting, because they act more like a pancreas," Ms. Olsen said. She noted that her hospital is switching from regular insulin, which is not quick acting, to the rapid-acting Humalog (insulin lispro) and NovoLog (insulin aspart). "The key with those is to give them with food, not 2 hours before the tray arrives," she said. "They mimic endogenous insulin by working quickly."
At Johns Hopkins, they also use long-acting insulins, such as Lantus (insulin glargine), in some cases. "These have the disadvantage that you cannot mix them with other insulins, and nurses may forget that," she warned.
The number of patients using insulin pumps is also increasing. "You need to develop a plan for caring for these patients, because taking that pump off when they come into the hospital may not necessarily be the right thing to do, especially for a type I diabetic who is well controlled," Ms. Olsen said. "Get procedures in place, and make sure that you are aware of and able to take care of those patients."
Case Study: What Not to Do
In her presentation at the Oncology Nursing Society Congress, Mikaela Olsen, RN, MSN, OCN, reviewed a case study of a 64-year-old man with type I diabetes who had a renal transplant 2 years previously and presented with non-Hodgkin's lymphoma. The patient had been on Lantus 42 units SC every day and NovoLog 6 units SC with meals, and steroids.
He was transferred from a local hospital for continued care of fever and presumed infection, and had several episodes of hypoglycemia at that referring hospital.
His blood glucose in the hospital to which he was referred ranged from 50 to 300 mg/dL with no apparent pattern, and he was not eating. He was being managed with sliding scale insulin, and no basal insulin.
On the fourth day of hospitalization, his midnight blood glucose was 248 mg/dL. On the morning of the fifth day, he suffered a cardiac arrest, at which time his blood glucose was 616 mg/dL and his serum acetone was positive. Despite aggressive ICU care, he died the following day.
"So what happened? The patient had type I diabetes, received no exogenous insulin, and went into diabetic ketoacidosis very quickly. He was managed on a sliding scale, which was inappropriate. Staff were reacting to his blood sugars, not proactively treating him. This death could have been prevented," Ms. Olsen said.