Many studies have demonstrated that carbapenem resistance significantly increases mortality compared with carbapenem susceptible groups. In a multicenter study, 30-day mortality rate in bloodstream infections caused by CRKp was 44% (29). In a meta-analysis evaluating CRKp bacteremia, mortality was 42.14% in the carbapenem-resistant group and 21.16% in the carbapenem-sensitive group. Mortality was reported to be 54.30% in bloodstream infections and 13.52% in urinary tract infections (30). The very low mortality rates in the carbapenem susceptible group in the present study may be explained by the fact that carbapenem was used in all of these cases.
Balkan et al reported that colistin resistance increased 28-day mortality in bloodstream infections due to carbapenem-resistant Klebsiella pneumoniae(5). Menekşe et al found that increasing MIC of colistin as a independent mortality factor in Klebsiella pneumoniae bloodstream infections(6).On the other hand, Aslan et al reported that colistin resistance in CRKp had no impact on 30-day mortality(3).
In the present study, we found that the presence of colistin resistance in addition to carbapenem resistance did not affect 30-day mortality. However, we found that significantly increased 14-day mortality. Different colistin susceptibility methods, whether colistin loading dose was given or not, whether meropenem dose and prolonged infusion were used or not, meropenem MIC values, combination or monotherapy and design of studies may have played a role in these different results.
In the present study, meropenem + colistin combination was indifferent (neither synergistic nor antagonistic) by gradient diffusion system.
In vitro synergy studies for carbapenem-resistant Enterobacteriaceae, the combination of polymyxin + carbapenem was generally found to be synergistic (36).
In their study with the time-kill method, Sharma et al. showed that the combination was bactericidal if the MIC of meropenem was ≤ 16 mg/L in colistin-susceptible and meropenem-resistant K. pneumoniae strains (37).
Kulegowski et al. also showed that if the MIC of meropenem is ≤ 32 mg/L, the combination of meropenem + polymyxin B generally has a synergistic effect on polymyxin B-susceptible strains (38).
However, in a study involving 19 patients with bloodstream infection caused by K. pneumoniae, Bono et al. did not reach the PK/PD targets in cases with MIC values ranging from 256 to 1024 mg/L, although the patients were given high dose and prolonged infusion. Furthermore, in vitro synergy with colistin was not detected. However, measured serum meropenem levels could reach the hypothetically determined T > 40% 1xMIC PK/PD targets of meropenem at a rate of 98% for the MIC value of 8 mg/L, 68% for 16 mg/L, and 32% for 32 mg/L (25). The lack of synergistic effect in our study may be attributable to meropenem MIC levels of ≥ 32 mg/L.
There was no therapeutic drug monitoring(TDM) for meropenem to evaluate PK/PD parameters in our study. We consider that TDM could be useful for infections caused by CRKp with MIC ≤ 64 mg/L(39, 40).
In the present study, no significant difference in mortality was noted between high dose - prolonged infusion of meropenem and normal dose - standard infusion of meropenem. In the carbapenem resistant-colistin sensitive and carbapenem resistant-colistin resistant groups, no significant difference in mortality was observed between high dose and prolonged infusion of meropenem and normal dose infusion.
It was shown that meropenem MIC is very important in the success of meropenem plus colistin combination in CRKp infections. If meropenem MIC > 8 mg/L, we can expect higher mortality(8, 9).
Gianella et al. compared carbapenem-containing combinations with carbapenem-free combinations in bloodstream infections caused by CRKp and reported that meropenem-containing combinations were more effective than meropenem-free combinations in terms of 14-day mortality even if meropenem MIC was ≥ 16 mg/L (11).
The fact that the synergy test results were indifferent and that there was no mortality difference between high-dose and prolonged infusion of meropenem and standard dose infusion in both groups suggests that high-dose and prolonged infusion meropenem treatment does not provide any benefit compared to standard dose and infusion if the meropenem MIC is ≥ 32 mg/L.
There are certain limitations of this study. The study was conducted retrospectively. The number of cases was relatively small and the study was single centered. Not all carbapenem-resistant strains were tested for synergy. The results were not compared with a group given colistin monotherapy. Lastly, the strains were not epidemiologically analyzed at the molecular level and carbapenemase enzymes were not analyzed.
In conclusion, 30-day mortality was significantly higher in the carbapenem resistant-colistin sensitive group and in the carbapenem resistant-colistin resistant group compared to the carbapenem susceptible (ESBL positive) group. When combined with colistin, if meropenem MIC is ≥ 32 mg/L, meropenem administration with high dose and prolonged infusion is not superior to standard dose and infusion in both carbapenem resistant-colistin sensitive and carbapenem resistant-colistin resistant K. pneumoniae bacteremia.