Table 1.
Lead author, Year | Country | Hospital location | Study design | Total patients or admissions | Total exposed to hypoglycaemia | People with diabetes (%) | Tytpe 1 diabetes | Type 2 diabetes | Treated with insulin (%) | Treated with oral agents (%) | Non‐exposed group | Adjustments during analysis | Hypoglycaemia definition (mmol/l) | Outcome |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Gomez‐Huelgas, 2015 11 | Spain | N/S | RCS | 309 020 | 154 510 | 100 | N/Sd | 291 827 | N/S | N/S | MP | Age, sex, type of diabetes, readmission, CCI, primary or secondary hypoglycaemia | ICD code | ↑LOS, ↑ In‐hospital mortality |
McEwan, 2015 12 | UK | N/S | RCS | 2158 | 1079 | 100 | 424 | 1734 | 100 | 0 | MP | Age, sex, type of diabetes, use of medication, BMI, HbA1c, smoking, geographical region and CCI | ICD code | ↑LOS, ↑In‐hospital mortalitya |
Curkendall, 2009 13 | USA | N/S | RCS | 103 813 | 8234 | 100 | 5261d | 87 610d | 25.5 | N/S | NM | Age, sex, type of diabetes, race, specific complications, events and conditions during hospitalizationd, CCI, insulin use, hospital characteristics and clinically plausible interactions | < 3.9 | ↑LOS, ↑In‐hospital mortality |
Geremakis, 2014 14 | USA | N/S | CCS | 8952 | 2510 | 100 | N/S | N/S | N/S | N/S | CC | Ethnicity, admitting hospital, CCI, surgery, fungal infection, antipsychotics, pulmonary hypertension, antidepressants, beta adrenergic, cephalosporin, antibiotics | <2.2 | ↑LOS |
The NICE‐SUGAR Study Investigators, 2012 15 | USA | ICU | SaRCT | 6026 | 643 | 20 | N/S | N/S | N/S | N/S | NM | Age, sex, APACHE II score, BMI, blood glucose, postoperative status, sepsis, trauma, diabetes, prior insulin or steroid treatment, cardiac failure, intensive vs. conventional insulin | ≤ 3.9 | ↑Mortality within 90 days |
Krinsley, 2007, 2011, 2017 8, 9, 10 | USA | ICU | SaRCT | 2786b | 683 | 100 | N/S | N/S | N/S | N/S | NM | Age, sex, APACHE II score, APACHE III score, mechanical ventilation | < 3.9 | ↑In‐hospital mortality |
Sechterberger, 2013 16 | Netherlands | ICU | RCS | 10,320. | 57 | 16 | N/S | N/S | 98 | N/S | NM | Age, sex, APACHE II score, hypoglycaemia severity, cardiothoracic surgery, glucose variability, mean glucose, glucose levels ≤ 4.7 mmol/l | ≤ 2.2 | ↑ICU mortality |
Arabi, 2009 17 | Saudi Arabia | ICU | NC | 523 | 46 | 40 | N/S | N/S | 87 | N/S | NM | Age, sex, BMI, postoperative status, APACHE II score, diabetes, admission glucose, mechanically ventilated, vasopressor, sepsis, creatinine, dialysis/filtration, intensive insulin | ≤ 2.2 | → ICU mortality |
Turchin, 2009 18 | USA | General ward | RCS | 4368 admissions | 338 admissions | 100 | N/S | N/S | 61.8% received insulin, OHA or both | NM for mortality. MP for LOS | Age, sex, ethnicity, health insurance, weighted mean daily glucose, LOS based on DRG and modified CCI | ≤ 2.8 | ↑LOS ↑In‐hospital mortality | |
Ong, 2015 19 | Singapore | General ward | RCS | 288 | 54 | 100 | N/S | N/S | 91% received insulin, OHA or both | NM | Age, sex, ethnicity, BMI, HbA1c, number of comorbidities, systolic BP, admitting specialty, diabetes treatment, steroids and IV dextrose | <4.0 | ↑LOS | |
Borzi, 2016 20 | Italy | General ward | SaCSS | 3167 | 385 | 100 | 0 | 3167 | N/S | N/S | NM | Age, sex, BMI, concomitant disease, insulin treatment, serum creatinine, HbA1c, fasting glucose and number of treatments other than diabetes | < 3.9 | ↑LOS ↑In‐hospital mortality |
Nirantharakuma 2012 21 | UK | General ward | RCS | 6374 admissions | 648 admissions | 100 | N/S | N/S | 25 | N/S | NM | Age, sex, ethnicity, deprivation, admission type, insulin use, modified CCI | < 3.9 | ↑In‐hospital mortality, ↑ LOSc |
Kim, 2014 22 | USA | General ward | RCS | 1276 admissions | 313 | 99 | 63d | 1198d | 0 | 100 | NM | Age, sex, ethnicity, insulin use, CCI | ≤ 3.9 | ↑LOS |
Boucai, 2011 23 | USA | General ward | RCS | 31 970b | 1717 | 34 | N/S | N/S | 19 | 11 | NM | Age, sex, ethnicity, co‐morbidities, number glucose determinators, diabetes treatment | ≤ 3.9 | ↑In‐hospital mortality, ↑LOS |
Mellbin, 2008 24 | Sweden | Cardiac speciality | SaRCT | 1253 | 153 | 100 | 0 | 1253 | N/S | N/S | NM | Age, sex, smoking status, previous myocardial infarction and coronary interventions, cardiac failure, pharmacological treatment, creatinine, diabetes duration, blood glucose before and during admission | < 3.0 | →Total mortality (median follow up 2.1 years) |
Applicable only to people with diabetes (not included in meta‐analysis as data not available).
Data provided by author.
Not in meta‐analyses as only median data reported.
Remainder had unknown or undisclosed type of diabetes.
N/S, not specified; ICU, intensive care unit; RCS, retrospective cohort study; CCS, case controlled study; SaRCT, sub‐analysis of randomised control trial; NC, nested cohort within a RCT; SaCSS, sub‐analysis of two cross‐sectional studies; OHA, oral hypoglycaemic agents; MP, matched patients; NM, non‐matched; CC, case controlled; CCI, Charlson comorbidity score; DRG, diagnostic related group; LOS, length of stay.