Table 3

Hypoglycaemia quality indicator subgroups

Indicators
Description
Reference

Hypoglycaemic events
Reported thresholdsa for defining a BGL as hypoglycaemic event:
• <40 [8,10,17-19,30,35,43];
• ≤ 40 [26];
• <45 [44,54];
• <50 [31,49,50];
• <54 [34];
• <57 [32];
• <60 [8,11,45,52,55];
• ≤ 60 [16,21];
• <63 [23];
• <65 [25];
• <70 [9,17,22,27,28,57];
• <72 [20]; or
• threshold was not reported [48]
Represented as:
• percentage and number of measurements and/or patients with hypoglycaemic event during the trial, or normalized for duration of therapy [16];
• mean or median of events per patient-day [57];
• mean of events per patient [27];
• patients with at least one event per day [22]; or
• in three studies, hypoglycaemic events did not occur and therefore were not reported [32,48,54]
31 articles [8-11,16-23,25-28,30-32,34,35,43-45,48-50,52,54,55,57]

Severe or marked hypoglycaemic events
Reported threshold for defining a BGL as severe hypoglycaemic event:
• <40 [15,20,25,29,33,42,46,47,51,56];
• ≤ 40 [12,13,16];
• <48 [24]; or
• in one study clinical finding defined as severe hypoglycaemia [8]
Represented as:
• percentage and number of measurements and/or patients with severe hypoglycaemic event;
• mean or median of events per patient-day [29];
• In two studies severe hypoglycaemic events did not occur and therefore not reported [47,56]
15 articles [8,12,13,15,16,20,24,25,29,33,42,46,47,51,56]

Need for dextrose injection
Reported threshold for dextrose injection:
• <45 [29];
• <54 [44];
• <60 [53];
• <63 [23];
• <65 [25]; or
• threshold was not reported [21,56]
Represented as:
• percentage and number of patient with dextrose injection [21,23,29,44,53,56]; or
• percentage and number of dextrose injections [25]
7 articles [21,23,25,29,44,53,56]

Mild or moderate hypoglycaemic events
Reported threshold for defining an BGL as severe hypoglycaemic event:
• 40–59 [42];
• <60 [46];
• 40–60 [51];
• <63 [15]
Represented as percentage and number of measurements and/or patients with a moderate hypoglycaemic event
4 articles [15,42,46,51]

Hypoglycaemia duration
Represented cumulatively [29], as median [24] and per patient [21]
3 articles [21,24,29]

Time until next in predefined range after hypoglycaemia
Represent as mean [8,49] or median [31] time
3 articles [8,31,49]

Duration of stopping IIT because of hypoglycaemia
Represented as median of percentage of time per patient
2 articles [29,33]

Next BGL after hypoglycaemia
Represented as mean BGL
2 articles [8,49]

Time until reaction to hypoglycaemic event
Represented as maximum time till hypoglycaemia recognition [8] or mean time till IIT adjustment after hypoglycaemic event [28]
2 articles [8,28]

Time from starting IIT until first hypoglycaemia
Represented as mean time
1 article [19]

Time till next BGL after hypoglycaemia
Represented as mean time
1 article [49]

aUnit of all BGL thresholds is mg/dl. BGL, blood glucose level; IIT, intensive insulin therapy.

Eslami et al. Critical Care 2008 12:R139   doi:10.1186/cc7114