Table 2.

Clinical studies investigating association between diabetes and susceptibility and outcome of infections

A. Studies investigating susceptibility of diabetic subjects to acquire infections
Author	Year	Infection type	n	Study design	Main outcome measures	Main findings
Zhao (29)	2009	Skin infection	8,655	Longitudinal matched control	Incidence of skin infections	Higher risk for skin infections (adjusted OR 2.8)
Kornum (57)	2008	CAP	34,329	Population-based matched control	Pneumonia-related hospitalization	Increased risk for CAP-related hospitalization (RR 1.26 [95% CI 1.21–1.31])
Benfield (32)	2007	Infectious diseases	10,063	Prospective	Hospitalization, 28-day mortality	Higher risk for infection-related hospitalizations and UTI-related mortality (HR 3.9 [95% CI 1.2–12.7]); no difference in mortality because of sepsis, CAP, skin infection, and other infections
Boyko (30)	2005	UTI	1,017	Longitudinal matched control	Incidence of UTI	Higher risk of UTI (RR 1.8 [95% CI 1.2–2.7]) and antibiotic treatment (RR 2.3 [95% CI 1.3–3.9])
Thomsen (58)	2004	Pneumococcal bacteremia	598	Matched control	Bacteremia	Higher risk for pneumococcal pneumonia (OR 1.9 [95% CI 1.4 –2.6])
Shah (31)	2003	Infectious diseases	513,749	Matched control	Hospitalization, mortality	Higher risk for hospitalization (RR 2.17 [95% CI 2.10 –2.23]) and infection-related mortality (1.92 [1.79 –2.05]); no difference in in-hospital mortality (1.05 [0.89–1.01] and 0.84 [0.87–1.01])
B. Studies showing an adverse association between diabetes and outcome
Author	Year	Infection type	n	Study design	Main outcome measures	Main findings
Kornum (37)	2007	CAP	29,900	Population-based cohort	Complications, bacteremia, mortality	Higher mortality rates (1.2 [95% CI 1.1–1.3]), but similar rates of complications and bacteremia; mortality within patients with diabetes increased when initial glucose levels >14 mmol/L in multivariate analysis (adjusted MMR 1.46 [95% CI 1.01–2.12] compared with patients with glucose <6.1 mmol)
Thomsen (36)	2005	Enterobacteria bacteremia	1,317	National registry	Bacteremia, 30-day mortality	Higher risk for bacteremia (OR 2.9 [95% CI 2.4–3.4]) and a trend toward higher 30-day mortality (1.4 [1.0–2.0])
Fine (35)	1996	CAP	33,148	Meta-analysis	30-day mortality	Higher risk for mortality (OR 1.3 [95% Cl 1.1–1.5])
C. Studies showing no or a protective effect of diabetes on outcome
Author	Year	Infection type	n	Study design	Main outcome measures	Main findings
Stegenga (42)	2010	Septic shock within the ICU	830	Prospective study	28-day mortality	Equal mortality rate (DM 31.4%, non-DM 30.5%)
Vincent (41)	2010	Sepsis within the ICU	3,147	Prospective study	28-day mortality	Similar mortality after adjustment for severity of illness (HR 0.78 [95% CI 0.58–1.07])
Graham (46)	2010	Infectious diseases	1,509,890	Retrospective and prospective cohort	In-hospital mortality	Lower adjusted OR for mortality in both cohorts (0.75 [95% CI 0.74–0.76] and 0.88 [0.79–0.98])
Michalia (59)	2009	Blood stream infection	343	Prospective	In-hospital mortality	Similar mortality rates (25.8 vs. 23.0%, P = 0.751)
Esper (43)	2009	Infectious diseases	12,500,000	National registry	Respiratory failure, in-hospital mortality	Lower risk for respiratory failure (9 vs. 14%, P < 0.05) and mortality (18.5 vs. 20.6%, P < 0.05)
Tsai (40)	2007	Blood stream infection	839	Prospective	30-day mortality	No difference in mortality rates (HR 0.82 [95% CI 0.53–1.26])
McAlister (39)	2005	CAP	2,471	Prospective	Mortality, infection-related complications	No difference in mortality, but hyperglycemia had higher risk for both complications and mortality
Thomsen (45)	2004	Pneumococcal bacteremia	628	Population-based cohort study	30- and 90-day mortality	Lower 30- and 90-day mortality (11.1 vs. 16.5%, P < 0.01 and 16.0 vs. 19.5%, P < 0.01)
Kaplan (38)	2002	CAP	623,718	National registry	In-hospital mortality	No difference in mortality rates, but hyperglycemia carried a higher risk for complications and mortality

Citations are in descending order of publication date. DM, diabetes; HR, hazard ratio.