Table 2.

Recent clinical studies investigating the association between diabetes and mortality for sepsis

Authors	Year	Patients	Study design	Main findings
Studies showing adverse association between diabetes and mortality for sepsis
Zoppini et al. [72]	2018	185,341 diabetic individuals	Retrospective cohort study on a regional electronic archive	Increased risk of death from infection-related causes in diabetic people (especially in female and people aged between 30 and 64 years): Overall SMR 1.83 (95% CI, 1.71–1.94) SMR for septicaemia 1.91 (95% CI, 1.76–2.06)
Shah et al. [12]	2003	513,749 diabetic individuals (matched to an equal number of non-diabetics)	Retrospective cohort study on population-based administrative data	Higher global infection-related mortality in diabetic patients (including home and hospital) (risk ratio up to 1.92 [99% CI 1.79–2.05]) No significant difference in term of infection-related hospital mortality (risk ratio up to 0.94 [99% CI 0.87–1.01])
Bertoni et al. [19]	2001	9,208 individuals (533 with diabetes)	Retrospective cohort study on a national registry	Higher infection-related mortality in diabetic patients with cardiovascular disease (RR 3.0 [95% CI 1.8–5.0])
Studies showing no association between diabetes and mortality for sepsis
Van Vught et al. [74]	2017	41,492 ICU septic patients (8085 with diabetes)	Retrospective large national database review	No association between diabetes and adjusted 90-day mortality In diabetic patients, only severe hypoglycemia in absence of hyperglycemia is associated with increased 90-day mortality (OR 2.95 [95% CI 1.19–7.32]), whereas in non-diabetics several combinations of abnormal glucose level were associated with increased 90-day mortality
Van Vught et al. [34]	2016	1,104 ICU septic patients (241 with diabetes)	Prospective observational study	No association between diabetes and 90-day mortality: HR 0.90 [95% CI 0.69, 1.15] after correction for BMI, age, gender, hypertension, cardiovascular and renal insufficiency HR 1.02 [95% CI 0.81–1.29] after correction for APACHE IV score
Venot et al. [70]	2015	10,911 patients (3,728 with severe sepsis or septic shock; among them, 451 with diabetes)	Case–control study based on a multicenter database	No difference in mortality between diabetic and non-diabetic septic patients (19.8% vs. 15% in the matched case–control analysis; p = 0.08)
Chang et al. [63]	2012	16,497 ICU septic patients (4,573 with diabetes)	Nationwide population-based retrospective cohort study	No association between diabetes and 90-day mortality (OR 0.972 [95% CI 0.890–1.061]) after adjustment for age, gender, comorbidities and number of acute organ dysfunction
Yang et al. [66]	2011	9,221 septic patients (2,943 with diabetes)	Retrospective large database review	No difference in-hospital mortality between diabetic and non-diabetic septic patients (19.2% vs. 20.0%; p = 0.37)
Stegenga et al. [73]	2010	830 ICU patients with septic shock (188 with diabetes)	Retrospective analysis of a previously published RCT	No difference in mortality between diabetic and non-diabetic septic patients: 28-day mortality 31.4% vs. 30.5%, p = 0.8235 90-day mortality 39.1% vs. 39.0%, p = 0.9827
Vincent et al. [33]	2010	3,147 ICU septic patients (226 with insulin-treated diabetes)	Prospective study	No difference in ICU and hospital mortality between diabetic and non-diabetic septic patients
Studies showing a protective effect of diabetes on mortality for sepsis
Esper et al. [21]	2009	12,500,459 septic patients (2,070,459 with diabetes)	Retrospective large national registry review	Lower hospital mortality in diabetic vs. non-diabetic patients (18.5% vs. 20.6%, p < 0.05)

Citations are in descending order of publication date

SMR = Standardized Mortality Ratio – ICU = Intensive Care Unit – RCT = Randomized Controlled Study