From 2017.01.01 to 2020.12.31, our center treated a total of 6 645 patients with 3 754 cases that underwent active surveillance. 548 (14.6%) patients yielded positive CRGNB surveillance cultures, and 97 cases presented with CRGNB infection on admission. Finally, 451 (12.0%) CRGNB-colonized patients were included, among which 152 (33.7%) developed CRGNB infection. There was no significant difference in gender, age, primary disease, acute physiology and chronic health evaluation (APACHE) II scores, and distribution of surveillance swab sites between the CRGNB infection and control groups. Surveillance cultures with the greatest proportion of positive yields came from throat swabs and sputum, followed by rectal swabs, inguinal swabs, and feces. Interestingly, the proportion of CRKP and CRPA was higher in the infection group, while CRAB was more common in the control group. The relationship between the distribution of swab sites, bacterial type, infection or not, or colonization and outcome is shown in Fig. 1. Compared with the control group, infected patients were associated with prior use of carbapenems (46.7% vs. 14.7%, P < 0.001, OR = 5.08. 95%CI: 3.23–7.97) and glucocorticoids (28.9% vs. 15.7%, P = 0.001, OR = 2.18, 95%CI: 1.36–3.49) use, and invasive procedures (87.5% vs. 76.3%, P = 0.005, OR = 2.18, 95%CI: 1.26–3.78). Moreover, patients in the infection group were more likely to be immunocompromised (47.4% vs. 23.1%, P < 0.001, OR = 3.00, 95%CI: 1.98–4.55), previously hospitalized in another center (59.2% vs. 49.2%, P = 0.043, OR = 1.52, 95%CI: 1.01–2.28) and previously admitted in another department (34.2% vs. 22.4%, P = 0.007, OR = 1.80, 95%CI: 1.17–2.77) than the control group. The incidence of septic shock (55.2% vs. 27.1%, P < 0.001, OR = 3.32, 95%CI: 2.21-5.00) and AKI (48.7% vs. 23.1%, P < 0.001, OR = 3.16, 95%CI: 2.08–4.79) were higher in the infection group, with longer ICU stay, hospital stay and higher mortality(25.6% vs. 14.0%, P = 0.002, OR = 2.11, 95%CI: 1.30–3.44)than the control group.
Furthermore, we identified six risk factors of CRGNB infection after colonization by multiple regression analysis, including CRKP colonization (P < 0.001, OR = 3.27, 95%CI: 1.80–5.95), CRPA colonization (P < 0.001, OR = 2.97, 95%CI: 1.63–5.40), prior admission in other departments (P = 0.029, OR = 1.78, 95%CI: 1.06–2.97), invasive procedure (P < 0.001, OR = 6.22, 95%CI༚2.57–15.03), history of carbapenem use (P < 0.001, OR = 5.48, 95%CI: 3.27–9.19), and immunocompromise (P < 0.001, OR = 7.07, 95%CI: 3.90–12.80). More details are shown in Table 2.
Table 1
Baseline characteristic of the CRGNB colonized patients at active surveillance
| Infection after colonization, N = 152 | No infection, N = 299 | P | OR (95%CI) |
Age, yr, mean | 60.6 | 59.0 | 0.32 | / |
Male sex, no. (%) | 112 (73.6%) | 206 (68.9%) | 0.29 | 0.85 (0.51–1.22) |
Primary disease (%) |
Cerebrovascular accidents, no. (%) | 38 (25.0%) | 79 (26.4%) | 0.74 | 0.93 (0.59–1.45) |
Cardiac insufficiency, no. (%) | 16 (10.5%) | 26 (8.7%) | 0.52 | 1.24 (0.64–2.38) |
Trauma, no. (%) | 61 (40.1%) | 115 (38.5%) | 0.75 | 1.07 (0.72–1.59) |
Infection, no. (%) | 49 (32.2%) | 82 (27.4%) | 0.28 | 1.25 (0.82–1.92) |
Malignant tumor, no. (%) | 11 (7.2%) | 26 (8.7%) | 0.59 | 0.82 (0.39–1.71) |
Diabetes mellitus, no. (%) | 14 (9.2%) | 41 (13.7%) | 0.17 | 0.64 (0.34–1.21) |
APACHE II scores (median) | 16.8 | 16.4 | 0.29 | / |
Intubation or tracheotomy, no. (%) | 76 (50.0%) | 162 (54.2%) | 0.40 | 0.86 (0.57–1.25) |
Prior cephalosporins history, no. (%) | 116 (76.35) | 198 (66.2%) | 0.028 | 1.64 (1.05–2.56) |
Prior carbapenems history, no. (%) | 71 (46.7%) | 44 (14.7%) | < 0.001 | 5.08 (3.23–7.97) |
Invasive operation, no. (%) | 133 (87.5%) | 228 (76.3%) | 0.005 | 2.18 (1.26–3.78) |
Operation History, no. (%) | 20 (13.2%) | 29 (9.7%) | 0.26 | 1.41 (0.76–2.58) |
Glucocorticoid history, no. (%) | 44 (28.9%) | 47 (15.7%) | 0.001 | 2.18 (1.36–3.49) |
Immunocompromise, no. (%) | 72 (47.4%) | 69 (23.1%) | < 0.001 | 3.00 (1.98–4.55) |
Other hospital history, no. (%) | 90 (59.2%) | 147 (49.2%) | 0.043 | 1.52 (1.01–2.28) |
Another department history, no. (%) | 52 (34.2%) | 67 (22.4%) | 0.007 | 1.80 (1.17–2.77) |
CRGNB type, no. (%) |
CRKP | 60 (39.5%) | 80 (26.7%) | 0.006 | 1.78 (1.18–2.70) |
CRPA | 56 (36.8%) | 90 (30.1%) | 0.15 | 1.35 (0.89–2.04) |
CRAB | 32 (21.1%) | 102 (34.1%) | 0.004 | 0.51 (0.32–0.84) |
Other CRE | 4 (2.6%) | 27 (9.0%) | 0.025 | 0.34 (0.12–0.90) |
Active surveillance sites, no. (%) |
Throat swab and/or sputum culture | 115 (75.6%) | 213 (71.2%) | 0.31 | 1.25 (0.80–1.96) |
Rectal swab and/or fecal culture | 30 (19.7%) | 72 (24.1%) | 0.29 | 0.77 (0.48–1.25) |
Other cultures | 7 (4.6%) | 14 (4.7%) | 0.97 | 0.98 (0.38–2.48) |
Septic shock, no. (%) | 84 (55.2%) | 81 (27.1%) | < 0.001 | 3.32 (2.21-5.00) |
Acute kidney injury, no. (%) | 74 (48.7%) | 69 (23.1%) | < 0.001 | 3.16 (2.08–4.79) |
Death, no. (%) | 39 (25.6%) | 42 (14.0%) | 0.002 | 2.11 (1.30–3.44) |
Length of ICU stay, day | 17.8 | 8.9 | < 0.001 | / |
Length of hospital stay, day | 27.6 | 19.3 | < 0.001 | / |
Table 2
high-risk factors of infection after colonization in multivariate regression analysis
Variable | P | OR (95%CI) |
CRKP colonization | < 0.001 | 3.27 (1.80–5.95) |
CRPA colonization | < 0.001 | 2.97 (1.63–5.40) |
Another department history | 0.029 | 1.78 (1.06–2.97) |
Invasive operation | < 0.001 | 6.22 (2.57–15.03) |
Carbapenems history | < 0.001 | 5.48 (3.27–9.19) |
Immunocompromise | < 0.001 | 7.07 (3.90–12.80) |
Patients infected after colonization were divided into empirically sensitive antibiotic therapy groups (n = 88) and standard antibiotic therapy groups (n = 64), depending on the timing of antibiotic use. No statistical difference in basic characteristics was found between the two groups (Table 3). The incidence of bloodstream infection was higher in empirically therapy groups than standard therapy groups; however, no statistical differences were found for the remaining part of the infection and the bacterial types. The most common antibiotic therapy used was combination therapy based on tigecycline and colistin, followed by ceftazidime-avibactam monotherapy. The average interval between colonization and infection was around 13 days, and ICU hospitalization was about 16 days after CRGNB infection. The mortality was lower in the empirical therapy groups than standard therapy groups (17.0% vs. 37.5%, P = 0.004, OR = 0.32, 95%CI: 0.16–0.73), which was substantiated by survival curve analysis (P = 0.002) (Fig. 2).
Table 3
The timing and clinical efficacy of antibiotics for infected patients after colonization.
| Empirically sensitive antibiotic therapy groups, N = 88 | Standard antibiotic therapy groups, N = 64 | P | OR (95%CI) |
Age-yr | 61.2 | 59.8 | 0.60 | / |
Male sex, no. (%) | 66 (75.0%) | 46 (71.8%) | 0.66 | 0.85 (0.41–1.76) |
APACHE II scores (median) | 17.0 | 16.4 | 0.20 | / |
Cerebrovascular accidents, no. (%) | 22 (25.0%) | 16 (25.0%) | 1.00 | 1.00 (0.47–2.10) |
Cardiac insufficiency, no. (%) | 9 (10.2%) | 7 (10.9%) | 0.88 | 0.93 (0.32–2.63) |
Trauma, no. (%) | 38 (43.2%) | 23 (35.9%) | 0.36 | 1.35 (0.69–2.62) |
Infection, no. (%) | 24 (27.3%) | 25 (39.1%) | 0.12 | 0.58 (0.29–1.16) |
Malignant tumor, no. (%) | 9 (10.2%) | 2 (3.1%) | 0.09 | 3.52 (0.73–16.94) |
Diabetes mellitus, no. (%) | 9 (10.2%) | 5 (7.8%) | 0.61 | 1.34 (0.42–4.22) |
Intubation or tracheotomy, no. (%) | 45 (51.1%) | 31 (48.4%) | 0.74 | 1.11 (0.58–2.12) |
Invasive operation, no. (%) | 72 (81.8%) | 51 (79.6%) | 0.74 | 1.14 (0.51–2.59) |
Operation History, no. (%) | 10 (11.3%) | 10 (15.6%) | 0.44 | 0.69 (0.27–1.77) |
Immunocompromise, no. (%) | 46 (52.2%) | 26 (40.6%) | 0.15 | 1.60 (0.83–3.06) |
Septic shock, no. (%) | 44 (50.0%) | 40 (62.5%) | 0.12 | 0.60 (0.31–1.15) |
Acute kidney injury, no. (%) | 40 (45.5%) | 34 (53.1%) | 0.35 | 0.73 (0.38–1.40) |
Infection sites, no. (%) |
Pulmonary infection | 64 (72.7%) | 50 (78.1%) | 0.45 | 0.74 (0.35–1.59) |
Bloodstream infection | 35 (39.8%) | 10 (15.6%) | 0.001 | 3.56 (1.60–7.92) |
Urinary tract infection | 26 (29.5%) | 16 (25.0%) | 0.53 | 1.25 (0.61–2.60) |
Gastrointestinal infection | 21 (23.9%) | 16 (25.0%) | 0.87 | 0.94 (0.45–1.99) |
Other sites infections | 37 (42.0%) | 25 (39.1%) | 0.71 | 1.13 (0.58–2.18) |
CRGNB types, no. (%) |
CRKP | 38 (43.2%) | 22 (34.4%) | 0.27 | 1.45 (0.74–2.82) |
CRPA | 31 (35.2%) | 25 (39.1%) | 0.63 | 0.84 (0.43–1.65) |
CRAB | 17 (19.3%) | 15 (23.4%) | 0.54 | 0.78 (0.35–1.71) |
Other CRE | 2 (2.3%) | 2 (3.1%) | 1.00 | 0.72 (0.10–5.25) |
Antibiotic therapy, no. (%) |
Combinations with Tigecycline | 25 (28.4%) | 14 (21.8%) | 0.36 | 1.41 (0.67–3.01) |
Combinations with Polymyxin B | 14 (15.9%) | 10 (15.6%) | 0.96 | 1.02 (0.42–2.47) |
Ceftazidime–avibactam monotherapy | 13 (14.7%) | 4 (6.2%) | 0.10 | 2.60 (0.81–8.39) |
Other therapies | 36 (40.9%) | 36 (56.3%) | 0.06 | 0.53 (0.28–1.03) |
Duration from colonization to infection, day | 13.5 | 12.8 | 0.54 | / |
Length of ICU stay after infection, day | 16.7 | 16.8 | 0.61 | / |
Length of hospital stay after infection, day | 32.8 | 24.2 | 0.22 | / |
Death, no. (%) | 15 (17.0%) | 24 (37.5%) | 0.004 | 0.34 (0.16–0.73) |