3.6.1. Gram-negative
The detection rates of extended-spectrum beta-lactamases (ESBL) in Escherichia coli and Klebsiella pneumoniae were 82.7% (43/52) and 65% (26/40), respectively. Escherichia coli and Klebsiella pneumoniae were sensitive to carbapenems, but the resistance rates of Pseudomonas aeruginosa and Acinetobacter baumannii to carbapenems were all over 50%. The main bacterial strains were sensitive to piperacillin/tazobactam, except for Acinetobacter baumannii. The main gram-negative bacteria were sensitive to cefoperazone/sulbactam. A strain of tigecycline-resistant Klebsiella pneumoniae was detected (Table 6).
Table 6
The main gram-negative bacteria resistance rate
Antibiotic type | Escherichia coli | Klebsiella pneumoniae | Pseudomonas aeruginosa | Acinetobacter baumannii |
Ampicillin | 51, (98.1%) | 40, (100.0%) | 36, (100.0%) | 35, (100.0%) |
Ampicillin/sulbactam | 38, (73.1%) | 25, (62.5%) | 36, (100.0%) | 24, (68.6%) |
Piperacillin | 44, (84.6%) | 40, (100.0%) | 12, (33.3%) | 26, (74.3%) |
Piperacillin/tazobactam | 6, (11.5%) | 11, (27.5%) | 12, (33.3%) | 25, (71.4%) |
Amikacin | 0, (0.0%) | 6, (15.0%) | 6, (16.7%) | 7, (20.0%) |
Gentamicin | 29, (55.8%) | 14, (35.0%) | 9, (25.0%) | 21, (60.0%) |
Tobramycin | 8, (15.4%) | 7, (17.5%) | 9, (25.0%) | 18, (51.4%) |
Ciprofloxacin | 39, (75.0%) | 19, (47.5%) | 8, (22.2%) | 26, (74.3%) |
Levofloxacin | 36, (69.2%) | 16, (40.0%) | 8, (22.2%) | 11, (31.4%) |
Imipenem | 1, (1.9%) | 10, (25.0%) | 23, (63.9%) | 25, (71.4%) |
Meropenem | 1, (1.9%) | 10, (25.0%) | 19, (52.8%) | 24, (68.6%) |
Cefazolin | 45, (86.5%) | 27, (67.5%) | 36, (100.0%) | 35, (100.0%) |
Cefuroxime sodium | 43, (82.7%) | 26, (65.0%) | 36, (100.0%) | 35, (100.0%) |
Ceftriaxone | 43, (82.7%) | 26, (65.0%) | 36, (100.0%) | 28, (80.0%) |
Ceftazidime | 37, (71.2%) | 21, (52.5%) | 11, (30.6%) | 26, (74.3%) |
Cefotetan | 12, (23.1%) | 10, (25.0%) | 36, (100.0%) | 35, (100.0%) |
Cefepime | 33, (63.5%) | 20, (50.0%) | 8, (22.2%) | 25, (71.4%) |
Cefoperazone/sulbactam | 2, (3.8%) | 5, (12.5%) | 8, (22.2%) | 13, (37.1%) |
Aztreonam | 37, (71.2%) | 23, (57.5%) | 22, (61.1%) | 35, (100.0%) |
Compound sulfamethoxazole | 32, (61.5%) | 13, (32.5%) | 36, (100.0%) | 8, (22.9%) |
Furantoin | 3, (5.8%) | 21, (52.5%) | 36, (100.0%) | 35, (100.0%) |
Tigecycline | — | 1, (2.5%) | — | 0, (0.0%) |
“—” Not available |
3.6.2. Gram-positive
The detection rate of MRSA was 10.5% (20/190) and that of MRCNS was 2.1% (4/190). Among the enterococci, the resistance rates of Enterococcus faecium to benzyl penicillin and ampicillin were 85.2% and 83.3%, respectively, while the resistance rates of Enterococcus faecalis were both 42.3%. The resistance rates of Enterococcus faecium and Enterococcus faecalis to high concentration gentamicin combined with ampicillin were 64.8% and 57.7%, respectively. There were no strains resistant to vancomycin, temozolomide or linezolid (Table 7).
Table 7
Main gram-positive bacteria resistance rate
Antibiotic type | Enterococcus faecium | Enterococcus faecalis | Staphylococcus epidermis | Staphylococcus aureus |
Ampicillin | 45, (83.3%) | 11, (42.3%) | — | — |
Oxacillin | — | — | 31, (93.9%) | 17, (63.0%) |
Benzyl penicillin | 46, (85.2%) | 11, (42.3%) | 33, (100.0%) | 25, (92.6%) |
Compound sulfamethoxazole | — | — | 18, (54.5%) | 7, (25.9%) |
Erythromycin | 50, (92.6%) | 16, (61.5%) | 27, (81.8%) | 20, (74.1%) |
Ciprofloxacin | 45, (83.3%) | 10, (38.5%) | 20, (60.6%) | 17, (63.0%) |
Levofloxacin | 44, (81.5%) | 10, (38.5%) | 7, (21.2%) | 15, (55.6%) |
Moxifloxacin | 45, (83.3%) | 10, (38.5%) | 4, (12.1%) | 9, (33.3%) |
Gentamicin | — | — | 9, (27.3%) | 14, (51.9%) |
High level gentamicin | 35, (64.8%) | 15, (57.7%) | — | — |
Clindamycin | 54, (100.0%) | 26, (100.0%) | 22, (66.7%) | 18, (66.7%) |
Rifampicin | — | — | 0, (0.0%) | 6, (22.2%) |
Tetracycline | 28, (51.9%) | 16, (61.5%) | 3, (9.1%) | 13, (48.1%) |
Tigecycline | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) |
Vancomycin | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) |
Linezolid | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) | 0, (0.0%) |
Quetiapine/darfuridine | 0, (0.0%) | 26, (100.0%) | 0, (0.0%) | 0, (0.0%) |
“—” Not available |
Discussion |
Our study found that there was no significant difference in the severity of SAP between the MDR and non-MDR groups. The mortality rate in the MDR group was significantly higher than that in the non-MDR group, which indicated that MDR bacterial infection was an important cause of death in SAP patients. This is because as the disease progresses, compensatory anti-inflammatory response syndrome (CARS) and SIRS compound one other and gradually worsen, resulting in mixed antagonistic response syndrome (MARS). The advantage of a proinflammatory response over an anti-inflammatory response is gradually reversed, and the patient sustains low levels of inflammation with severe immunosuppression development eventually [20]. SAP changes from an early aseptic chemical inflammation to a secondary multisite MDR bacterial infection; uncontrolled pancreatic and severe systemic infections cause sepsis, infectious bleeding, digestive tract spasms and other complications leading to death [21]. However, such findings must be interpreted cautiously because they are probably correlated with the fact that the peak of death occurred in the first and second stage, which was more frequent among patients who were more severely ill, while the MDR bacterial infection occurred later. |
The total hospitalization days did not differ significantly between the two groups, which was related to the abandonment of treatment in some patients. Precautionary antibiotic use, kinds of antibiotics used, receipt of carbapenem, tracheal intubation, hemofiltration and number of hospitalization days in the intensive care unit were significantly higher in the MDR group. This indicates that the above interventions were important causes of MDR bacterial infections. Endoscopic surgery was a safe treatment measure for patients [22]. Unconditional logistic regression showed that ICU hospitalization was a risk factor for MDR bacterial infection [23]. When 4 to 6 different kinds of antibiotics used in patients, the risk of MDR bacterial infection was approximately 3 times that of patients given 1 to 3 antibiotics; therefore, we can draw the conclusion that using a variety of antibiotics increases the risk of MDR bacterial infection [24]. |
SAP infection was caused by pathogens that passed through the blood and bile duct systems or retrograded through the duodenum and ascended into the main pancreatic duct. At the same time, intestinal pathogens crossed the intestinal barrier and then translocated into the lymphatic system and the parenteral system to cause infection [25]. Although gram-negative bacteria were still dominant, the proportion of gram-positive bacteria increased notably compared with 27.9% and 23.9% reported by Ma2 and Su [26]. One of the reasons is that drainage or postoperative infections occur after the appearance of pancreatic or anastomotic fistula, leading to the emergence of multiple infection foci. However, the reason for the increase in the number of enterococci in SAP patients remains unclear and may be related to the prophylactic use of antibiotics [27]. Fernanda S. Soares et al. [28] found that prophylactic use of meropenem in SAP-affected mice induced Enterococcus colonization of the small intestine and gradually became predominant in the gut, which led to an increase in the number of gram-positive bacteria. A multihospital prospective clinical study showed that the intestinal population of Enterococcus was higher and more positively correlated with the serum levels of IL-6 in SAP patients than in MAP patients, suggesting that the increase in enterococci contributes to the severity of this disease [29]. |
The resistance rates of Escherichia coli and Klebsiella pneumoniae to quinolones were higher than those of nonfermentative bacteria, but the resistance rates to aminoglycosides were the opposite. Of the nonfermentative bacteria, Pseudomonas aeruginosa and Acinetobacter baumannii gradually exhibited resistance to carbapenems through an active efflux system and decreased permeability of the outer membrane. The drug resistance rates were higher than those of Escherichia coli and Klebsiella pneumoniae; therefore, it was necessary to combine treatment with β-lactamase inhibitors in the clinic [30]. The resistance rates of Escherichia coli and Klebsiella pneumoniae to cephalosporins were high, while the rates of Acinetobacter baumannii to ceftazidime and cefepime were higher than those of Pseudomonas aeruginosa. The rates of the main gram-negative bacteria to aztreonam were also higher, but Klebsiella pneumoniae and Acinetobacter baumannii were more sensitive to the compound sulfamethoxazole.
The detection rate of resistant Enterococcus faecium was higher than that of Enterococcus faecalis, and the resistance rates of these two bacteria to penicillin were quite different; additionally, the resistance rates to high concentration gentamicin were all over 50%. Therefore, the antibacterial effect was poorer for those pathogens when using aminoglycoside-penicillin or benzyl-penicillin for synergistic effects and screening should be performed for clinical use. The detection rate of resistance in Staphylococcus epidermidis was higher than that in Staphylococcus aureus; however, the resistance rates of Staphylococcus aureus to quinolones and gentamicin were higher than those of Staphylococcus epidermidis. Staphylococcus epidermidis was more sensitive to tetracycline, which was similar to the resistance of Staphylococcus aureus to compound sulfamethoxazole.
The prophylactic use of antibiotics for the prevention of secondary pancreatic infection remains controversial [1, 31], and the relevant guidelines are not recommended for patients with SAP and aseptic necrosis [32]. Although prophylactic use of carbapenem antibiotics may lead to bacterial translocation, according to the characteristics of the spectrum, drug sensitivity and antibiotic characteristics, the most appropriate method of empirical antibiotic treatment is meropenem. A short-term, full dose regimen of broad-spectrum antibacterials, especially carbapenems, in the early stage of SAP can eliminate sensitive pathogens quickly, reduce the dual-infection, and reduce the production of drug-resistant strains caused by bacterial flora disturbance [2]. Subsequent use of antimicrobial agents should be based on the results of drug susceptibility testing of pathogens to ensure an effective antibacterial effect, shorten the course of treatment, reduce the production of drug-resistant strains and reduce the probability of fungal infection [33]. Although aminoglycosides have good germicidal efficacy against many major pathogenic bacteria, they cannot be used as routine drugs for the treatment of pancreatic infections because they have difficulty crossing the blood-pancreatic barrier to reach the pancreatic lesions [34]. For gram-negative bacteria and anaerobic infections, compounds such as carbapenems, third, or fourth generation cephalosporins, β-lactamase inhibitors and quinolones are the first choice antimicrobial agents with strong lipid solubility and effective penetration of the blood pancreatic barrier. Piperacillin/tazobactam, metronidazole, imipenem, ciprofloxacin and ofloxacin can produce high blood drug concentrations in pancreatic necrosis tissue. The extensive bactericidal effect of ertapenem on common intra-abdominal pathogens shows that its use in IPN is reasonable [35]. Treatments for gram-positive bacterial infections can be selected according to bacterial culture and drug sensitivity results. If necessary, antibiotics such as tigecycline, vancomycin and linezolid can be used.