Table 3.
Univariable and multivariable logistic regression analyses of factors associated with a complicated hospital coursea
| N | Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|---|
| Non complicated course | Complicated course | OR [95% CI] | P-value | aOR [95% CI] | P-value | ||
| N = 46 | N = 123 | ||||||
| Demographic data | |||||||
| Age | 169 | 59 [52–69] | 60 [53–68] | 1.00 [0.98; 1.03] | 0.724 | ||
| Sex | 169 | ||||||
| Male | 31 (67.4) | 89 (72.4) | 1.32 [0.63; 2.75] | 0.459 | |||
| Female | 15 (32.6) | 34 (27.6) | Ref | ||||
| Hospital trajectory data | |||||||
| Annual NSTI procedural volume < 3/year | 169 | 5 (10.9) | 33 (26.8) | 3.01 [1.09; 8.26] | 0.033 | ||
| Type of structure after ICU | 167 | 0.139 | |||||
| Medical ward | 19 (41.3) | 66 (54.6) | Ref. | ||||
| Rehabilitation | 5 (10.9) | 5 (4.1) | 0.29 [0.08; 1.10] | 0.069 | |||
| Plastic surgery | 7 (15.2) | 9 (7.4) | 0.37 [0.12; 1.13] | 0.080 | |||
| Other surgical ward | 15 (32.6) | 41 (33.9) | 0.79 [0.36; 1.72] | 0.548 | |||
| Inter-hospital transfer | 167 | ||||||
| No | 18 (39.1) | 61 (50.4) | Ref | ||||
| Yes | 28 (60.9) | 60 (49.6) | 0.63 [0.32; 1.26] | 0.194 | |||
| NSTI characteristics | |||||||
| Hospital-acquired NSTI | 169 | 12 (26.1) | 26 (21.1) | 0.76 [0.35; 1.67] | 0.494 | ||
| Topography | |||||||
| Upper limb | 169 | 7 (15.2) | 17 (13.8) | 0.89 [0.34; 2.32] | 0.817 | ||
| Lower limb | 169 | 32 (69.6) | 88 (71.5) | 1.10 [0.52; 2.31] | 0.801 | ||
| Abdominal/perineal | 169 | 7 (15.2) | 26 (21.1) | 1.49 [0.60; 3.72] | 0.390 | ||
| Multifocal | 169 | 0 | 2 (1.6) | – | |||
| Comorbidities | |||||||
| Immunosuppression | 169 | 9 (19.6) | 34 (27.6) | 1.57 [0.69; 3.60] | 0.286 | ||
| PAOD | 169 | 6 (13.0) | 20 (16.3) | 1.29 [0.48; 3.46] | 0.607 | ||
| Cancer | 169 | 5 (10.9) | 9 (7.3) | 0.65 [0.20; 2.04] | 0.459 | ||
| Neutropenia | 169 | 2 (4.4) | 3 (2.4) | 0.55 [0.09; 3.40] | 0.520 | ||
| HIV infection | 169 | 0 | 4 (3.3) | – | |||
| Diabetes | 169 | 23 (50.0) | 71 (57.7) | 1.37 [0.69; 2.69] | 0.369 | ||
| Chronic heart failure | 169 | 11 (23.9) | 26 (21.1) | 0.85 [0.38; 1.91] | 0.698 | ||
| COPD | 168 | 5 (10.9) | 12 (9.8) | 0.89 [0.30; 2.70] | 0.843 | ||
| Cirrhosis | 168 | 2 (4.4) | 13 (10.7) | 2.62 [0.57; 12.11] | 0.216 | ||
| Chronic hemodialysis | 169 | 0 | 5 (4.1) | – | |||
| Chronic alcohol use | 169 | 8 (17.4) | 41 (33.3) | 2.38 [1.01; 5.55] | 0.046 | ||
| Previous NSAIDs use | 167 | 14 (30.4) | 23 (19.0) | 0.54 [0.25; 1.16] | 0.115 | ||
| Obesity | 169 | 13 (28.3) | 47 (38.2) | 1.57 [0.75; 3.28] | 0.231 | ||
| Microbiological data | |||||||
| Number of isolated bacteria | 156 | 1 [1, 2] | 2 [1–4] | 1.38 [1.07; 1.78] | 0.014 | ||
| NSTI categorization | 154 | ||||||
| Type II (monomicrobial) | 24 (55.8) | 41 (36.9) | Ref. | Ref. | |||
| Type I (polymicrobial) | 19 (44.2) | 70 (63.1) | 2.16 [1.06; 4.41] | 0.035 | 3.18 [1.37; 735] | 0.007 | |
| Positive blood culture | 169 | 6 (13.0) | 44 (35.8) | 3.71 [1.46; 9.45] | 0.006 | 3.29 [1.14; 9.52] | 0.028 |
| Isolation of multi / highly resistant bacteria | 167 | 3 (6.5) | 26 (21.5) | 3.92 [1.13; 13.67] | 0.032 | ||
| Hospital-acquired NSTI | 169 | 12 (26.1) | 26 (21.1) | 0.76 [0.35; 1.67] | 0.494 | ||
| Severity scores | |||||||
| SAPS II | 161 | 33 [22–45] | 53 [33.5–68] | 1.05 [1.02; 1.07] | < 0.001 | 1.05 [1.02; 1.07] | < 0.001 |
| SOFA | 161 | 4 [1–7] | 9.5 [5–12] | 1.09 [1.01; 1.17] | 0.022 | ||
Qualitative variables are shown as number (percentages) and continuous variables as median [quartile 1-quartile 3]
PAOD peripheral artery occlusive disease, HIV human immunodeficiency virus, COPD chronic obstructive pulmonary disease, NSAIDs non-steroidal anti-inflammatory drugs, NSTI necrotizing soft tissue infection, SAPS Simplified Acute Physiology Score, SOFA Sequential Organ Failure Assessment score, OR odds ratio, aOR adjusted odds ratio
aDefined by one of the following: a duration of hospital stay ≥ 6 weeks, a number of debridement performed ≥ 3, or hospital death; Area under the ROC curve = 0.796; model calibration (p = 0.442, Hosmer–Lemeshow test)
Bolded variables are significant at the p < 0.05 level