The resistance rate of gram-negative bacteria is increasing. More importantly, it is reported that the multidrug resistance rate is alarmingly high, and steadily increasing(12, 13). But this is not surprising, as our treatment goals previously focused on maximizing clinical and microbiological cure rather than minimizing the emergence of antibiotic resistance(14). According to the mutation selection window (MSW) theory proposed by Zhao(15), when the drug concentration is higher than the MIC value, antibacterial drugs can play an antibacterial role; however, when the concentration is between MIC and MPC value, while drugs play an antibacterial role, but they can also selectively enrich resistant mutants. Finally, when the concentration is higher than MPC value, resistant mutants must have ≥ 2 mutations at the same time for selective enrichment to occur. Therefore, we should not aim to treat patients based on clinical efficacy, but should also consider achieving inhibition of drug resistance on the basis of Zhao’s theory.
In a study by Fink et al(16), monotherapy with 400 mg of ciprofloxacin administered every 8 h resulted in 33% resistance to hospital acquired pseudomonas pneumonia, whilst 500 mg of imipenem given every 6 h or 1000 mg every 8 h resulted in 50% resistance. The resistance rate of levofloxacin was 25% according to a study conducted by Shao Y(17). In this context, we hope to explore the difference between monotherapy and combination therapy for PA.
The traditional combination idea is not to use drug combinations with common mechanisms of resistance because the orthogonality of resistance probabilities will not hold. The emergence of β–lactamase (mainly including AmpC, ESBLs, MBLs, etc.) is the main mechanism of PA’s resistance to β–lactams(18). Fluoroquinolones inhibit DNA gyrase (subunit composition: gyrA and gyrB) and topoisomerase IV (subunit composition: parC and parE) of bacteria(19), but the mutation of any these subunits leads to the resistance of bacteria to fluoroquinolones. As such, we chose to explore the common combination of levofloxacin and ceftazidime, and this study is the first to use a HFIM to investigate the effects of the two drugs on bacterial kill and inhibition of resistance emergence in patients with abnormal renal function. We simulated monotherapy and combination therapy commonly used in clinic, and determined the administration regimens of patients with abnormal renal function according to the instructions. Our study may better simulate patients with low immune function (majority in ICUs) due to the lack of immune system in HFIM, but there may be some differences in the sensitivity of bacteria to drugs in vivo or in HFIM.
The PK of levofloxacin or ceftazidime was similar at the same dose in different groups of monotherapy and combination. When AUC0 − 24h/MIC ≥ 100 and/or Cmax/MIC > 8, the bactericidal effect was good for fluoroquinolones, and when T > MIC is more than 50%, it showed a good bactericidal effect for β–lactams. However, as observed in some studies(20–22), treatment failure and rapid resistance emergence happened even when T > MIC reached or approached 100% for meropenem, or AUC0 − 24h/MIC was 168 for tobramycin which was 4 times of the suggested breakpoint. So it is necessary to find effective ways to inhibit drug resistance. The combination therapy had a better effect on the total and resistant population compared with monotherapy (Fig. 4). The monotherapy produced about a 0.5-log-CFU/mL reduction on the total population before 6 or 8 h (Fig. 2), while the combination therapy for patients with normal renal function achieved a 2-log-CFU/mL bacterial kill on total population, and reduced 3-log-CFU/ml for patients with abnormal renal function within 1–2 days (Fig. 3). Therefore, we can think of this combination as leading to synergistic sterilization. In addition, the effect of bacterial kill on combination therapy in patients with abnormal renal function was superior to that in patients with normal renal function. Our study may be useful for exploring lower doses of levofloxacin combined with ceftazidime to kill bacterial and inhibit resistance in patients with abnormal renal function.
For monotherapy, drug concentration exceeded the MPC value after the first dose when given as 500 or 750 mg levofloxacin once daily, which decreased to less than MPC after 6 h. As such, resistance emergence was seen after 6 h. Drug concentrations of other monotherapy regimens was between MIC and MPC after the first administration, so drugs were able to produce bactericidal effects, but not inhibit resistance, so resistance appeared at 0 h. For combination therapy, the growth of resistant bacteria appeared after 16 h in patients with normal renal function. In the patients with abnormal renal function, resistance emergence was seen at 0 h, but there was a downward trend after 8 h. Therefore, resistance may be reduced if the dose was increased properly to reach the MPC value at the first administration in patients with normal renal function. For patients with abnormal renal function, if their first dose was equal to that of normal patients, they may obtain better effects of bacterial kill and resistance suppression. Of course, the appropriate administration should be implemented according to patient’s Ccr later, and therapeutic drug monitoring should take place. In addition, we suggest priority should be given to combination rather than monotherapy.