Hypoglycemia during hyperosmolar hyperglycemic crises is associated with long-term mortality

Table 3 Multivariate analyses of factors associated with mortality (Cox regression)

Hypoglycemia during hospitalization	Crude analyses		Adjusted for age and sex		Adjusted for Charlson Comorbidity Index		Multiadjusted^a
	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value	HR (95% CI)	P-value
Hypoglycemia during the initial intravenous insulin therapy phase	1.94 (1.24–3.01)	0.003	1.83 (1.17–2.88)	0.008	2.61 (1.65–4.14)	< 0.001	2.10 (1.27–3.46)	0.004
Hypoglycemia during the later subcutaneous insulin therapy phase	1.56 (1.00–2.44)	0.048	1.11 (0.70–1.74)	0.649	1.16 (0.74–1.81)	0.517	1.20 (0.75–1.92)	0.428
Hypoglycemia during any phase of hospitalization	1.69 (1.06–2.68)	0.025	1.34 (0.84–2.15)	0.217	1.36 (0.85–2.16)	0.189	1.40 (0.87–2.26)	0.157

^aAdjusted for sex, Charlson Comorbidity Index, pure hyperglycemic hyperosmolar state (versus diabetic ketoacidosis), probable infection during admission, and initial intravenous insulin therapy phase duration. Age was not added to models including the Charlson Comorbidity Index because age is a component of the index. The models included 170 cases, except models with hypoglycemia during the later subcutaneous insulin therapy phase, which included 163 cases
HR hazard ratio, CI confidence interval

ISSN: 1758-5996