Table 3.
Analysis of the Risk Factors of Poor Clinical Efficacy After PMB Treatment
| Risk Factors | Effective | Invalid | P value | Binary Logistic Regression | |||
|---|---|---|---|---|---|---|---|
| (N = 98) | (N = 170) | B | OR | 95% CI | P value | ||
| Sex (male) | 71 (72.45%) | 121 (71.18%) | 0.824 | ||||
| Age (years) | 54 (40.75–62.50) | 61 (47.75–75.00) | 0.004 | 0.004 | 1.004 | 0.986–1.023 | 0.666 |
| Mechanical ventilation | 45 (45.92%) | 144 (84.71%) | 0.000 | 1.867 | 6.472 | 2.505–16.719 | 0.000 |
| Vasoactive agents | 28 (28.57%) | 116 (68.24%) | 0.000 | 1.054 | 2.868 | 1.386–5.936 | 0.005 |
| Admission to ICU | 57 (58.16%) | 125 (73.53%) | 0.009 | 1.206 | 3.339 | 1.249–8.927 | 0.016 |
| APACHE II score | 20.0 (12.5–24.0) | 20.0 (13.0–28.0) | 0.215 | ||||
| Comorbidities | |||||||
| Cardiovascular and cerebrovascular diseases | 60 (61.22%) | 109 (64.12%) | 0.636 | ||||
| Respiratory disease | 76 (77.55%) | 150 (88.24%) | 0.020 | 1.024 | 2.783 | 1.165–6.646 | 0.021 |
| Kidney disease | 46 (46.94%) | 74 (43.53%) | 0.589 | ||||
| Liver disease | 16 (16.33%) | 51 (30.00%) | 0.013 | 0.681 | 1.976 | 0.892–4.376 | 0.093 |
| Diseases of digestive system | 24 (24.49%) | 28 (16.47%) | 0.110 | ||||
| Nutritional disease | 15 (15.31%) | 19 (11.18%) | 0.328 | ||||
| Diabetes mellitus | 12 (12.24%) | 32 (18.82%) | 0.161 | ||||
| Infection site | |||||||
| Respiratory tract | 91 (92.86%) | 163 (95.88%) | 0.284 | ||||
| Blood | 21 (21.43%) | 63 (37.06%) | 0.008 | 0.393 | 1.482 | 0.649–3.381 | 0.350 |
| Urinary system | 12 (12.24%) | 17 (10.00%) | 0.569 | ||||
| Intracranial | 11 (11.22%) | 6 (3.53%) | 0.013 | −0.028 | 0.972 | 0.2289–4.131 | 0.970 |
| Digestive tract | 10 (10.20%) | 20 (11.76%) | 0.696 | ||||
| Skin and soft tissue | 6 (6.12%) | 15 (8.82%) | 0.428 | ||||
| Pathogenic bacteria (CR-GNB) | |||||||
| Acinetobacter baumannii | 57 (58.16%) | 121 (71.18%) | 0.030 | 0.130 | 1.139 | 0.368–3.528 | 0.822 |
| Klebsiella pneumoniae | 41 (41.83%) | 51 (30.00%) | 0.049 | −0.613 | 0.541 | 0.174–1.684 | 0.289 |
| Pseudomonas aeruginosa | 19 (19.38%) | 50 (29.41%) | 0.071 | 0.287 | 1.332 | 0.464–3.826 | 0.594 |
| E. coli or E. cloacae | 1 (1.02%) | 9 (5.29%) | 0.149 | ||||
| Multiple site infection | 31 (31.63%) | 88 (51.76%) | 0.001 | 0.446 | 1.562 | 0.742–3.292 | 0.240 |
| Number of CR-GNBs | 1 (1.00–1.00) | 1 (1.00–2.00) | 0.013 | 0.653 | 1.922 | 0.612–6.033 | 0.263 |
| Number of patients with CR-GNB>1 | 16 (16.30%) | 49 (28.80%) | 0.022 | 1.100 | 3.005 | 0.826–10.932 | 0.095 |
| Antimicrobial Susceptibility | |||||||
| Tigecycline MIC >2 (mg/L) | 65 (66.33%) | 99 (58.24%) | 0.497 | ||||
| Polymyxin MIC >2 (mg/L) | 5 (5.10%) | 3 (1.80%) | 0.252 | ||||
| Previous use of carbapenem | 67 (68.40%) | 110 (64.70%) | 0.542 | ||||
| Previous use of tigecycline | 36 (39.90%) | 73 (42.90%) | 0.319 | ||||
| Preemptive therapy | 74 (75.50%) | 133 (78.20%) | 0.608 | ||||
| Delay days of PMB use (days) | 0 (0–2.25) | 0 (−0.25–2) | 0.397 | ||||
| Loading dose≥2.0mg/kg | 27 (27.55%) | 32 (18.82%) | 0.097 | −0.907 | 0.404 | 0.182–0.8895 | 0.026 |
| Maintenance dose≥1.25mg/kg q12h | 14 (14.29%) | 25 (14.71%) | 0.925 | ||||
| Use loading dose | 41 (41.84%) | 69 (40.58%) | 0.841 | ||||
| Maintenance dose (mg/kg q12h) | 0.83 (0.77–1.00) | 0.88 (0.83–1.00) | 0.147 | ||||
| Loading dose (mg/kg) | 1 (0.83–1.80) | 1.01 (0.83–1.67) | 0.802 | ||||
| Treatment duration time (days) | 11 (7.00–16.00) | 9 (5.50–13.00) | 0.002 | −0.075 | 0.927 | 0.832–1.033 | 0.171 |
| Cumulative dose (mg) | 1050 (600.00–1512.50) | 875 (500–1300) | 0.016 | 0.000 | 1.000 | 0.999–1.001 | 0.433 |
| Combination of drugs | |||||||
| Tigecycline | 26 (26.53%) | 58 (34.12%) | 0.197 | ||||
| Carbapenems | 36 (36.73%) | 39 (22.94%) | 0.015 | −0.040 | 0.960 | 0.475–1.943 | 0.911 |
| β-lactams | 36 (36.73%) | 54 (31.76%) | 0.407 | ||||
| Glycopeptides | 13 (13.27%) | 31 (18.24%) | 0.290 | ||||
| Others | 13 (13.27%) | 28 (16.47%) | 0.483 | ||||
Notes: The measurement data are expressed by median and quartile, t-test is used to accord with normal distribution. Non-parametrictric test is used to disaccord with normal distribution, and counting data is expressed by cases and percentage, chi-square test is used. The parameters included in Binary logistic regression were those with P < 0.1 in univariate test. Bold font indicates data with significant differences. B indicates regression coefficient.
Abbreviations: PMB, Polymyxin B; ICU, Intensive Care Unit; APACHE II, Acute Physiology and Chronic Health Evaluation II; CR-GNB, Carbapenem-resistant Gram-negative Bacteria; MIC, minimum inhibitory concentration.