In this study, we established that imipenem/relebactam was one of the most effective antimicrobial agents for the treatment of infections caused by P. aeruginosa,including carbapenem-non-susceptible and MDR strains.First, our study echoes that carbapenem-resistantP.aeruginosa, the pathogen listed as a critically prioritized bacteria for drug development by the WHO, is also a serious threat to the public health in China. In the current study, the susceptibility rate of imipenem against P. aeruginosa across China was 55.7%, with the lowest rate of 33.0% in the east Jiangzhe area. These imipenem susceptibility rates were numerically lower than those in other countries as per published SMART reports, with 73% in US/Canada and 66.7% in Europe(18), which reflected a more serious resistant status of P. aeruginosa that China faces. In addition, the China Antimicrobial Surveillance Network (CHINET) Program reported that 63.7% of collected P. aeruginosaisolates was susceptible to imipenem in China in 2017, higher than our finding. The difference between the two studies is likely derived from the difference of the collected isolates. CHINET strains were from both inpatients and outpatients whereas in our study isolates only collected from hospitalized patients.
The percentages of imipenem non-susceptible P. aeruginosa and MDR P. aeruginosa isolates were numerically higher in ICU than in non-ICU samples, which is in agreement with previously published reports(19, 20).
Imipenem/relebactam demonstrated an overall 84.2% susceptibility rate in P. aeruginosa in this study. By adding relebactam, the susceptibility to imipenem in imipenem non-susceptible P. aeruginosa strains rosefrom 0.0–58.6% and 66.7% in ICU and non-ICU derived P. aeruginosa isolates,respectively, and that of MDR P. aeruginosa isolates rose from 21.0–58.5% in ICU and from 26.7–68.5% in non-ICU isolates. Itindicated that imipenem/relebactam is similarly activein ICU and non-ICU derived P. aeruginosa isolates, including carbapenem non-susceptibleones.However, relebactam did less improvement to the imipenem susceptibilityin MDR P. aeruginosa isolates especially in the east Jiangzhe region where the MDR incidence rate was the highest. The reason for regional susceptibility differences of imipenem/relebactamneeds further studies focusing on resistance mechanisms in MDR isolates especially in areas with high resistance rates.
The effect of relebactam is based on the inhibition of β-lactamase (Class A and Class C) activities (13, 21). In addition to oxacillinase (OXA)relebactam cannot restore imipenem activity in strains producing metallo-β-lactamases (12, 22), which have also been found in found in Chinese P. aeruginosa isolates (23, 24). When compared with other antibiotics, we found thatP. aeruginosaexhibited susceptibilities of 84.2% to imipenem/relebactam and about 89.7%, and 94.9% to amikacin and colistin, whereas susceptibilities for other antibiotics were less than 70%, indicating that imipenem/relebactam, amikacin and colistin were the most effective antibiotics.
The majority ofP. aeruginosa isolates were obtained from RTIs (65.6 %) with 26.5% collected in ICUs and 73.5% in non-ICUs, followed by IAIs (21.5%), 20.7% in ICUs and 79.3% in non-ICUs, and UTIs (12.9%) 11.9% in ICUs and 88.1% in non-ICUs. Imipenem susceptibilities of ICU (60.7%) and non-ICU (59.5%) isolates were similar for IAIs, but isolates collected in ICUs were less susceptibleto imipenem compared with non-ICU isolates for RTIs (43.6% vs 53.0%) and UTIs (55.2% vs 77.1%). However, restoration of imipenem susceptibilityby relebactam in imipenem non-susceptible P. aeruginosa isolates was similar in RTI isolates from ICUs (62.2%) and non-ICUs (64.0%), but essentially lower in ICU compared to non-ICU isolates collected from IAIs (50.0% vs 77.8%) and UTIs (30.8% vs 61.2%), a trend which was also exhibited for MDR P. aeruginosa isolates with similar imipenem restoration rates (Supplementary Table 4). These data indicated that P. aeruginosa isolates from RTIs, though with a higher imipenem resistance rate in ICUs, could be restored by relebactam to similar levels as for isolates from non-ICUs, which was not the case for UTI and IAI P. aeruginosa isolates. These findings may indicate that different carbapenem resistance mechanism distributions occurred in various organs and in ICU vs non-ICU infections. In addition, the similar rates of imipenem susceptibility restoration in imipenem non-susceptible and MDR P. aeruginosa isolates strongly suggests a high efficacy of relebactam also against MDR P. aeruginosa strains.
The results from the 2015 SMART surveillance program, involving 17 European countries, showed that the overall susceptibility of A. baumannii to imipenem was only 10.1% (25) and somewhat lower than found in the present study (21.0%). However, in both studies the imipenem non-susceptibilities could not be restored by relebactam in imipenem non-susceptible A. baumanniiisolates, which has also been reported in the US (10) and has been attributed to certain resistance mechanisms of A. baumannii strains (26, 27), since a main mechanism of carbapenemresistance in A. baumanniiis expression of oxacillinases (OXAs) (28). A. baumanniipercent susceptibilities to imipenem and imipenem/relebactamwere virtually identical (both about 20%), whereas for amikacin they were only 16.5–32.7%, indicating amikacin resistance, which has been attributed to aminoglycoside-modifying enzyme expressions in A. baumannii(29). Except for amikacin (32.7%) and colistin (96.2%), the percent susceptibility of A. baumannii to other antibiotics was less than 21%.With the exception of amikacin (17.8%), < 6% of MDR A. baumanniiisolates were susceptibleto the other antibacterial agents tested
Susceptibility rates were high to colistin in all P. aeruginosa (94.6–94.9%) and A. baumannii(96.1–96.5%) isolates, including imipenem-non-susceptible isolates and MDRisolates, which might be explained by the fact that in China colistin was used in veterinary medicine only until 2014, probably due to its previously reported serious nephrotoxicity and neurotoxicity, which limited its clinical use (30).
One major strength of the present study was the large sample size but one weakness is the inherent limitation that no molecular analyses data are conducted.