Volume 14, Number 6—June 2008
Dispatch
Nosocomial Outbreaks Caused by Leuconostoc mesenteroides subsp. mesenteroides
Table 2
Model for predicting infection by L. mesenteroides subsp. mesenteroides (LM), 2003–2004, La Coruña, Spain*
Variable | Cases, n = 42 |
Controls, n = 61 |
Crude OR† (95% CI) | Adjusted OR (95% CI) | p value | |||
---|---|---|---|---|---|---|---|---|
No. (%) | Mean (SD) | No. (%) | Mean (SD) | |||||
Age, y | 34.3 (28.2) | 44.4 (31.7) | 0.99 (0.98–1.0) | |||||
Time between admission and infection | 33.5 (38.4) | 37.5 (100.9) | 0.999 (0.994–1.0) | NS | ||||
Charlson score |
2.94 (2.13) |
4.28 (2.38) |
0.76 (0.61–0.96) |
NS |
||||
Previous surgery | 29 (69) | 23 (37.7) | 3.7 (1.6–8.5) | NS | ||||
Previous infections | 31 (73.8) | 15 (24.6) | 8.6 (3.5–21.3) | 4.2 (1.2–14.7) | 0.023 | |||
Previous antimicrobial drug therapy | 37 (88.1) | 42 (68.9) | 3.3 (1.1–9.9) | NS | ||||
Teicoplanin | 12 (28.6) | 4 (6.6) | 5.7 (1.7–19.2) | NS | ||||
Vancomycin | 5 (11.9) | 3 (4.9) | 2.6 (0.6–11.6) | |||||
Central venous lines |
39 (92.9) |
41 (67.2) |
6.3 (1.7–23.0) |
NS |
||||
Sex | ||||||||
Male |
24 (57.1) |
41 (67.2) |
0.7 (0.3–15) |
|||||
Urinary catheter | 28 (66.7) | 21 (34.4) | 3.7 (1.6–8.6) | NS | ||||
Enteral nutrition | 18 (42.9) | 23 (37.7) | 1.2 (0.6–2.8) | |||||
Parenteral nutrition | 40 (95.2) | 26 (42.6) | 26.9 (6–121.6) | 27.8 (5.5–141.1) | <0.000 | |||
Blood transfusion | 24 (57.1) | 31 (50.8) | 1.3 (0.6–2.8) | |||||
Intubation | 18 (42.9) | 17 (27.9) | 1.9 (0.8–4.4) | |||||
Tracheostomy | 4 (9.5) | 3 (4.9) | 2.0 (0.4–9.6) | |||||
Treatment with steroids | 8 (19) | 12 (19.7) | 1 (0.4–2.6) | |||||
Alteration of gastrointestinal barrier‡ | 29 (69) | 30 (49.2) | 2.3 (1.0–5.3) | NS |
*Values for 42 case patients and 61 control patients. SD, standard deviation; OR, odds ratio; CI, confidence interval; NS, variable did not meet criterion for remaining in the multivariate model. All these variables were considered as potential risk factors for LM infection. The LM infection was considered as an outcome variable.
†Predictor variables with p<0.10 in univariate analysis were included in the multivariate model to enable simultaneous adjustment.
‡Any process that modifies the gastrointestinal barrier (inflammation, atresia, resection, obstruction).