Hypoglycaemia: reagent strips could rule out hypoglycaemia

Clinical bottom line (level 1b)

  1. In patients with suspected hypoglycaemia, a normal or high sugar reading on a reagent strip helped rule out hypoglycaemia (LR-0.07)
  2. In patients with suspected hypoglycaemia who had a hypoglycaemic reading on reagent strip, hypoglycaemia was moderately likely (LR+6.0)
Lavery et al: American Journal of Emergency Medicine 1991; 9: 304-308
Expires September 2003

The study

Setting: five mobile medical intensive care units in USA

181 patients (aged ?, ?% male) diabetes or altered mental status or explicit direction of physician

Excluded if
  • glucose too high for reagent strip


    • visual reading of reagent glucose test strips
    Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
    Reference standard:
    • laboratory glucose values <60 mg/dL

    The evidence

    pre-test probability of hypoglycaemia (<60 mg/dL): 18%, (95% CI: 12% to 24%)

    diagnostic test hypoglycaemia (<60 mg/dL) normal or euglycaemia (>60 mg/dL) LR+
    (95% CI)    		 post-test probability LR-
    (95% CI)    		 post-test probability
    strip reading <60 mg/dL 31 23 6.0
    (4.1 to 8.9)     		57% 0.07
    (0.02 to 0.28)     		2%
    total 33 148

    Comments

    1. Excellent study of reagent strips used in usual practice

    Citation

    1. Lavery RF, Allegra JR, Cody RP, et al: A prospective evaluation of glucose reagent test strips in the prehospital setting. American Journal of Emergency Medicine 1991; 9: 304-308
    Contributor: Bob Phillips and Clare Wotton, September 1999
    Reviewer:

    Clinical Question.
    Patient patients with diabetes or altered mental status
    Intervention or Exposure glucose reagent strips
    Outcome predict laboratory hypoglycaemia