Hypoglycaemia: history of prior events and lower HbA1c were associated with repeat events.

Clinical bottom line (level 2b)

  1. In patients with type I diabetes on insulin therapy, a history of hypoglycaemic coma and/or seizure was associated with repeat risk.
  2. In patients with IDDM on intensive therapy and compared to conventional treatment, there was an increased risk of coma/seizure (NNF = 2 for 6.5 years) .
Diabetes Control and Complications Trial Research Group : Diabetes 1997; 46: 271-286
Expires June 2003

The study

Prospective cohort study with objective outcomes, adjusted for confounding factors, not validated in an independent set of patients.

Setting: participants in a 29-centre North American RCT

1441 patients (aged mean 27y, 53% male) insulin dependent diabetes without C-peptide secretion

Excluded if
  • aged <13 or >39
  • >2 seizures or coma in previous 2 years
  • significant diabetic complications
  • >1 severe neurological episode without prior warning of hypoglycaemia



    • Factors studied:
  • sex
  • age
  • stimulated C-peptide value
  • IDDM duration
  • HbA1c at entry
  • Insulin dose
  • Prior hypoglycaemic coma/seizure
  • intensive cf conventional therapy
  • history of hypoglycaemia prior to trial
  • prior hypoglycaemic events
  • fall in HbA1c by 10% (at baseline of 9%) varies with baseline HbA1c (falling more with lower baseline results)
  • history of hypoglycaemia prior to trial
  • prior hypoglycaemic events
  • fall of 10% in HbA1c
  • history of hypoglycaemia
  • prior number of hypoglycaemic episodes
  • HbA1c fall by 10% (at baseline 9%) varies by baseline HbA1c - lower risk with lower HbA1c
  • history of hypoglycaemia prior to trial
  • prior number of hypoglycaemic episodes
  • 10% fall in HbA1c


    • intensive = insulin treatment with specific targets, conventional = usual care

    Cox proportional hazards model (and generalised version, multiplicative intensity model) was used to adjust for confounding factors.

    100%. reviewed at monthly (intensive) or quarterly (conventional) examinations and after every episode of hypoglycaemia followed for 6.5y
    Outcomes studied:
  • hypoglycaemic coma/seizure
  • hypoglycaemia requiring assistance in intensive group
  • hypoglycaemic coma/seizure in intensive group
  • hypoglycaemia requiring assistance on conventional therapy
  • hypoglycaemic coma/seizure on conventional therapy

    • The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)    		 NNF
    (95% CI)
    hypoglycaemic coma/seizure 9 years 523/1441 36.3%
    (33.8% to 38.8%)    		 3
    (3 to 3)
    hypoglycaemia requiring assistance 6.5y 518/711 72.9%
    (69.6% to 76.1%)
    hypoglycaemic coma/seizure 6.5y 116/711 16.3%
    (13.6% to 19.0%)
    hypoglycaemia requiring assistance 6.5y 301/730 41.2%
    (37.7% to 44.8%)
    hypoglycaemic coma/seizure 6.5y 186/730 25.5%
    (22.3% to 28.6%)

    prognostic factor for
    hypoglycaemic coma/seizure     		time to outcome adjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    intensive cf conventional therapy 6.5y 3.02
    (2.36 to 3.86)     		2
    (1 to 3)

    prognostic factor for
    hypoglycaemia requiring assistance     		time to outcome adjusted RR
    (95% CI)
    history of hypoglycaemia prior to trial 6.5y 1.33
    (1.17 to 1.52)
    prior hypoglycaemic events 6.5y 1.10
    (1.10 to 1.11)
    fall in HbA1c by 10% (at baseline of 9%) 6.5y 1.53
    (1.43 to 1.64)

    prognostic factor for
    hypoglycaemic coma/seizure     		time to outcome adjusted RR
    (95% CI)
    history of hypoglycaemia prior to trial 6.5y 1.43
    (1.13 to 1.83)
    prior hypoglycaemic events 6.5y 1.29
    (1.26 to 1.32)
    fall of 10% in HbA1c 6.5y 1.33
    ( to )

    prognostic factor for
    hypoglycaemia requiring assistance     		time to outcome adjusted RR
    (95% CI)
    history of hypoglycaemia 6.5y 1.70
    (1.36 to 2.13)
    prior number of hypoglycaemic episodes 6.5y 1.16
    (1.15 to 1.18)
    HbA1c fall by 10% (at baseline 9%) 6.5y 1.62
    (1.53 to 1.71)

    prognostic factor for
    hypoglycaemic coma/seizure     		time to outcome adjusted RR
    (95% CI)
    history of hypoglycaemia prior to trial 6.5y 2.00
    (1.37 to 2.91)
    prior number of hypoglycaemic episodes 6.5y 1.54
    (1.41 to 1.68)
    10% fall in HbA1c 6.5y 1.46
    (1.35 to 1.58)

    Comments

    1. Huge trial, but looking at a group with lower risk of hypoglycaemia; may underestimate number of events in community samples.
    2. Patients on intensive therapy were at much greater risk for hypoglycaemia.
    3. Patients were taken from a randomised trial.

    Citation

    1. Diabetes Control and Complications Trial Research Group , : Hypoglycemia in the Diabetes Control and Complications Trial. Diabetes 1997; 46: 271-286
    Contributor: Bob Phillips and Clare Wotton, Unknown Month 1999
    Reviewer:

    Clinical Question.
    Patient IDDM without residual C-peptide secretion
    Intervention or Exposure taking insulin
    Outcome severe hypoglycaemia