Since at least the mid-1970s, conventional wisdom has stated that all non-traumatic alterations in mental status (AMS) were presumed to be due to low blood sugars. The preferred treatment was to administer intravenous dextrose to all patients until other reasons for AMS were found.
Recent studies indicating that hyperglycemia may increase morbidity and mortality in patients with cerebral ischemia is causing a reassessment of the appropriateness of this intervention. One investigation, looking at 500 patients admitted for non-traumatic causes of coma, found that more than 50 percent had diagnoses that put them at risk for cerebral ischemia. It is felt that limiting administration of IV dextrose in these patients might minimize adverse outcomes.
“Rapid and reliable determination of blood glucose levels in the pre-hospital setting would allow treatment of hypoglycemia while avoiding the potential ill effects of unnecessary treatments,” noted Jonathan Jones, M.D., of the Department of Emer-gency Medicine at Pitt County Memorial Hospital in Greenville, NC. “We studied the reliability of blood glucose testing by emergency medical technicians in the pre-hospital setting using the reagent strip method.”
There are basically two methods for getting blood sugar readings in the field. One involves placing a drop of blood on a specially designed strip of paper, then visually comparing the color of the strip after a certain period of time with a chart on the side of the bottle. The other way is to use a strip that is then put into a meter device that gives a reading of the patient’s blood sugar at that time.
“With some exceptions, the reagent strips have been shown to be easy to use and reliable under controlled circumstances,” Jones stated. “Since poor technique may adversely effect results, consistency in training is essential.”
To find out if reagent strips are useful in an ambulance setting, Jones and his group studied pre-hospital fingerstick glucose values of 170 patients collected over a 10-week period. Blood was also drawn for laboratory analysis if an IV was started. If not, then one was drawn immediately upon arrival at the hospital emergency department.
“A high sensitivity and negative predictive values are necessary if there is to be confidence in withholding dextrose treatment,” Jones noted. “A high sensitivity identifies all cases of hypoglycemia and a high negative predictive value indicates that if a test is negative, hypoglycemia does not exist.” In their study the sensitivity was 91.7 percent with one false negative. The negative predictive value was 99.3 percent.
Similar findings came from a study done by Robert Lavery and a team from St. Barnabas Medical Center in Livingston, NJ. That series of 180 patients showed a negative predictive value of 98 percent and a sensitivity of 94 percent.
In both studies half of those thought to be hypoglycemic by fingerstick were found not to be on examination of laboratory results. Because untreated hypoglycemia is more devastating to the patient, however, this was deemed to be not as important as the high negative predictive value.
“In the pre-hospital setting, it is clinically more important to identify hypoglycemia than to describe the actual blood glucose value,” Lavery noted. “Use of the reagent strips to diagnose hypoglycemia may decrease the number of patients who receive empiric dextrose.”
Jones and his group noted there are many training issues to be considered when deciding on how to implement a glucose screening program in the pre-hospital setting. It is important that a single large drop of blood covers the pads of the strip and that excess blood is removed with a cotton ball, not a gauze strip. Storage for more than eight weeks has also been shown to decrease effectiveness.
“Other potential adverse factors that are often beyond control include extreme temperatures,” Jones noted. “Color blindness, poor vision and poor lighting, often found in homes and ambulances, could theoretically affect the accuracy in comparing the reagent strips to the printed standards on the side of the bottle.”
The North Carolina team said cited concerns about how well the meters will stand up to the pounding that may occur in an ambulance.
“The use of meters is an alternative to visually reading reagent strips,” Jones said. “While the meters overcome some of the problems associated with accuracy in reading reagent strips, they also have a few technical problems. Since the reagent strip is reliable for ruling out hypoglycemia, the added expense of the meter may not be justified.”
Training paramedics to give and read the test, delays in transport and increased exposure to blood products must be evaluated against the costs of adverse effects of glucose administration on those with cerebral ischemia.
“We have found that EMTs using reagent strops can reliably rule out hypoglycemia in the hospital setting,” Jones said. “However, the reagent strip will not prevent unnecessary dextrose administration in all of the normal or hypoglycemic patients in the pre-hospital setting. … We recommend that all pre-hospital care systems reassess their current protocols to promote rapid glucose testing for all patients with AMS.”
Kurt Ullman, a Firehouse® contributing editor, is a registered nurse and regional chairman of the volunteer Indiana Arson and Crime Association.
