Cost-Effectiveness Analysis of New Beta-Lactam Beta-Lactamase Inhibitor Antibiotics Versus Colistin for the Treatment of Carbapenem-Resistant Infections

Abstract

Introduction: Carbapenem-resistant organisms (CROs) present a serious public health problem. Limited treatment options has led to increased use of colistin and polymyxin. Since 2014, the US Food and Drug Administration approved 4 new beta-lactam beta-lactamase inhibitor (BLBLI) combination antibiotics with activity against CROs. These new antibiotics have been shown to be more effective and less toxic than colistin and polymyxin but are considerably more expensive. This study evaluated the cost-effectiveness of the new BLBLIs versus colistin-based therapy for the treatment of CROs. Methods: A decision-tree microsimulation model was used to evaluate the cost effectiveness of the new BLBLIs versus colistin-based therapy for the treatment of CROs. Treatment groups differed in risk of mortality and risk of an acute kidney injury (AKI). The relative risk of mortality was determined by creating a meta-analysis comparing new BLBLIs to colistin. Cost inputs included medication costs and the cost to treat an AKI. The primary outcomes include quality-adjusted life years (QALYs) and incremental cost-effectiveness ratio (ICER). Model inputs included: clinical outcomes and adverse events (30-day mortality and AKI); cost of treatment and adverse drug events; and health utilities. A 3% discount was applied for outcomes. A lifetime horizon was used from the perspective of the US healthcare system with a willingness-to-pay (WTP) threshold of $100 000. A sensitivity analysis was done to incorporate uncertainty. Results: The meta-analysis found the treatment with a new BLBLI was associated with a 50% decrease in the relative risk of 30-day mortality compared to colistin (RR 0.47, 95% CI 0.25-0.88). Treatment with a new BLBLI cost $16 200 and produced 11.5 QALYs, on average. The average colistin based regimen cost $3500 and produced 8.3 QALYs. The new BLBLIs were determined to be cost-effective with an ICER of $3900 per QALY gained. Treatment with a BLBLI remained cost-effective under all uncertainty scenarios tested. Conclusion: New BLBLIs are cost-effective compared to colistin for the treatment of CROs and are associated with improved mortality and fewer AKI events. The use of colistin should be reserved for cases where new BLBLIs are not available or there is documented resistance to these new antibiotics.

Introduction

Carbapenem-resistant organisms (CROs) are an increasing threat to public health. Carbapenem-resistant enterobacterales (CRE) and multidrug-resistant Pseudomonas aeruginosa cause nearly 46 000 infections and 4000 deaths annually in the United States. ¹ Lack of effective treatment options has led to the increased use of the colistin and polymyxin B, which are considered last-line antibiotics. In 2018, the World Health Organization identified CRE and carbapenem-resistant P. aeruginosa as priority pathogens to direct antimicrobial discovery. ² Since 2014, the US Food and Drug Administration has approved 4 new beta-lactam beta-lactamase inhibitor (BLBLI) combination antibiotics for the treatment of CROs: ceftolozane-tazobactam, ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-relebactam. ³ These agents are FDA-approved for the treatment of infections ranging from complicated urinary tract infections to hospital-acquired and ventilator-associated pneumonia.

Beta-lactam antibiotics are typically first line for treatment of many infections due to a favorable safety and efficacy profile. Conversely, colistin is known to cause nephrotoxicity and clinical success rates as low as 25%. ⁴ The incidence of acute kidney injury (AKI) attributed to colistin use ranges from 25 to 75% compared to 10 to 20% among BLBLIs.^5,6 Five studies comparing the efficacy of the new BLBLIs to colistin have suggested a mortality benefit in addition to higher rates of clinical cure and microbiological eradication.^7-11

Despite the preferable safety and efficacy record of BLBLI antibiotics, colistin and colistin-based combination therapies are currently recommended for use by several national and international guidelines.^12-16 Clancy et al ¹⁷ evaluated the treatment of CRE using prescription claims data. From February 2018 through January 2019, colistin and polymyxin treated more CRE infections than the newer antibiotics (ceftazidime-avibactam, meropenem-vaborbactam, and plazomicin). One reason for the slow uptake in use of the new BLBLIs may be acquisition cost. The 4 new BLBLI antibiotics are approximately 10 times more expensive than colistin. However, there are significant non-drug costs associated with colistin treatment including the cost of an AKI and increased mortality that should be considered in treatment and formulary decisions. ¹⁸ The goal of this study was to evaluate the cost-effectiveness of the new BLBLI antibiotics compared to colistin for the treatment of CRO from the perspective of the US healthcare system.

Methods

A meta-analysis was conducted to establish an average treatment difference between new BLBLIs and colistin. Articles were identified for inclusion through PubMed using the search term: (“meropenem vaborbactam” OR “ceftolozane tazobactam” OR “ceftazidime avibactam” OR “imipenem relebactam”) AND (“carbapenem resistant”). Peer-reviewed studies were included in the analysis if 1 of the new BLBLIs was compared directly to colistin or polymyxin B for the treatment of CROs. Studies that compared a new BLBLI to multiple other therapies were only included if mortality data was available specific to the colistin-treated patients. The primary outcome for the meta-analysis was 30-day mortality, so all studies had to include mortality as an outcome. The meta-analysis was created using Review Manager (RevMan), Version 5.3. ¹⁹ Given the expected heterogeneity of the studies, a random effects model was used to conservatively estimate the risk of mortality between treatment groups.

Decision Tree Model

A cohort-level decision-tree model was created using Microsoft Excel® to simulate the treatment of CROs. The 2 main clinical inputs that varied between treatment groups were 30-day risk of mortality and risk of an AKI. Figure 1 displays the model which tracks patients from the start of treatment until death. Each box in the decision tree separates into 2 branches that represent the possible outcomes for an individual. The possible outcomes are the same between groups, but the probability of events (AKI and mortality) differ based on existing literature. The first box branches into the treatment groups: new BLBLI or colistin-based combination therapy. The second set of branches represents the possibility of an individual experiencing an AKI. The final set of branches represents the mortality rate in each group.

Figure 1.

Decision tree model for evaluating the cost-effectiveness of the new beta-lactam beta-lactamase inhibitor antibiotics versus colistin-based combinations in the treatment of carbapenem-resistant infections.

CRO = carbapenem-resistant organism; BLBLI = beta-lactam beta-lactamase inhibitor; AKI = acute kidney injury.

The estimated relative risk of mortality and associated confidence interval from the meta-analysis was applied to the BLBLI mortality estimate in the decision tree. Site of infection was not included in the decisions tree as that data in unavailable for infections caused by CROs. The total cohort represents 2000 patients, 1000 in each treatment group. This cohort size was arbitrarily selected to aid in the interpretation of results.

Clinical Inputs

Table 1 provides a complete description of model inputs. Patients entering the model had an average age of 60. The mortality estimates are based on the combined groups in the meta-analysis. Mortality at 30 days was a consistent outcome across all studies in the meta-analysis. Given the heterogeneity in definition of AKI from the studies in the meta-analysis, the incidence rates of AKI for each treatment was not derived from the meta-analysis. Rutter and Burgess ⁶ estimated the rate of AKI for BLBLIs at 9% (95% CI 6-12) based off the incidence of AKI with older BLBLIs (piperacillin-tazobactam and ampicillin-sulbactam. Oliota et al ²⁰ conducted a systematic review and meta-analysis which found the incidence of AKI with colistin was 26.7% (95% CI 22.8-36.7).

Table 1.

Model Inputs.

Clinical outcomes	Estimate	Lower bound	Upper bound	Distribution	Source
Relative risk for BLBLI mortality	0.47	0.25	0.88	Lognormal	Shields et al, 7 van Duin et al, 8 Wunderink et al, 9 Alraddadi et al, 10 Motsch et al 11
BLBLI mortality	17.9%
Colistin mortality	35.6%
BLBLI AKI	9%	6%	12%	Beta	Rutter and Burgess 6
Colistin AKI	26.7%	22.8%	36.7%	Beta	Oliota et al 20
Health utilities	Estimate	Lower bound	Upper bound	Distribution	Source
Survived without AKI	0.85	0.7	1	Beta	Kauf et al 21
Survived with AKI	0.65	0.5	0.8	Beta	Johansen et al 22 Assumption
Treatment	Estimate	Lower bound	Upper bound	Distribution	Source
Duration of therapy	14 days	7 days	21 days	Gamma	Assumption
Discounting health outcomes	3%	3%	3%	Gamma	ICER 24
Drug Costs	Cost per day (2019 USD)		Cost per course a (2019 USD)		Source
Meropenem-vaborbactam	$1188.00		$16 632.00		RED BOOK Online 23
Ceftolozane-tazobactam	$820.37		$11 485.15		RED BOOK Online 23
Ceftazidime-avibactam	$1291.71		$18 083.94		RED BOOK Online 23
Imipenem-relebactam	$1284.00		$17 976.00		RED BOOK Online 23
Colistin	$7.20		$100.80		RED BOOK Online 23
Polymyxin	$37.50		$525.00		RED BOOK Online 23
Meropenem	$195.56		$2737.85		RED BOOK Online 23

Note. BLBLI = beta-lactam beta-lactamase inhibitor; AKI = acute kidney injury.

^aFourteen-day treatment course.

There is insufficient evidence to suggest a difference in the rate of mortality after AKI between treatment with a new BLBLI and colistin. Short-term mortality after AKI was modeled the same between the groups. It is assumed that the antibiotics are used as definitive therapy, thus no de-escalation or changes in antibiotic regimens were incorporated into the model. Lifetime mortality was modeled using Centers for Disease Control and Prevention (CDC) life tables. ²⁵ Among patients who survived, those who experienced an AKI were assumed to have a 1.41 hazard ratio (95% CI 1.39-1.43) of death compared to patients who did not have an AKI. ²⁶ This hazard ratio was based on a 2-year study of veterans who experienced an AKI that did not require renal replacement therapy. Given the absence of evidence of long-term risk of mortality after AKI, this effect was modeled to diminish by 10% every year after the first 2 years, so the difference in mortality from AKI was gone after 6 years.

Cost Inputs

The model included cost estimates for the antibiotics and for the cost of treating an AKI. The International Consensus Guidelines for the Optimal Use of the Polymyxins recommend the use of polymyxin in combination with at least one other antibiotic when treating invasive carbapenem-resistant infections. ¹² Meropenem is the most commonly used antibiotic used in combination with polymyxins. ²⁷ The antibiotic cost model included the average cost of polymyxin B and colistin plus the average cost of meropenem. Antibiotic costs were estimated using average wholesale prices from Red Book Online, Micromedex Healthcare Series. ²³ Average unit prices were converted to average cost per 14-day treatment course for a 60-kg adult with normal renal function. The average cost of an AKI was $1700, which is based on the cost of AKI events that do not require dialysis. ¹⁸ The cost of the hospitalization was not included in this model because it was assumed to be similar between groups.

Health Utility

Health utility scores are measures of quality of life and range from 0 (death) to 1 (perfect health). Health utilities for patients who survived the infection without an AKI was estimated at 0.85 for both groups, which is consistent with a previous US cost-effectiveness analysis between ceftolozane-tazobactam and piperacillin-tazobactam for the treatment of complicated urinary tract infections. ²¹ Johansen et al ²² estimated health utilities after an AKI using a large study from The Veterans Affairs/National Institutes for Health Acute Renal Failure Trial Network Study. Patients who required renal replacement therapy in an intensive care unit had health utility scores of 0.47 (±0.37). The necessity of dialysis for AKI was not included in this cost-effectiveness model, so a more conservative disutility after AKI was chosen to be 0.65 (95% CI 0.5-0.8).

Outcomes

The primary outcomes were quality-adjusted life years (QALY) and total costs of each group. Differences in these estimates were used to calculate an incremental cost-effectiveness ratio (ICER) by dividing the difference in QALYs by the difference in cost. The willingness-to-pay threshold was set at $100 000 per QALY gained. ²⁴ A budget impact analysis was performed assuming a hospital that treats 10 CRO infections per year and uses BLBLI therapy in 35% of the cases. ¹⁷

Sensitivity Analysis

A one-way sensitivity analysis isolates 1 model parameter at a time holding other parameters fixed to evaluate uncertainty in the parameter and the respective impact on the ICER estimate. The one-way sensitivity analysis used the lower and upper bounds which are consistent with the 95% confidence interval where available. For example, the average treatment duration was 14 days with the range of 7 days (lower bound) to 21 days (upper bound). The primary analysis estimated the ICER using 14 days of treatment. The one-way sensitivity analysis estimated the possible variation in ICER due to the treatment duration parameter by using 7 days and separately 21 days of treatment. This process was repeated for each model parameter with a range specified in Table 1.

Perspective

The model assumed the perspective of the US healthcare system over a lifetime horizon. Outcomes were discounted at 3%. ²⁴ All costs were related to the index event, so no costs required discounting.

Results

The literature search yielded 270 results and 5 studies met inclusion criteria. The meta-analysis included 5 studies that compared the efficacy of BLBLI antibiotics and colistin for the treatment of CROs (Table 2). Shields et al ⁷ conducted a retrospective study comparing ceftazidime-avibactam to colistin plus a carbapenem. Ceftazidime-avibactam had a lower rate of 30-day mortality (1/13, 8%) compared to colistin plus a carbapenem (9/30, 30%). Van Duin et al ⁸ conducted a multicenter, prospective observational study comparing ceftazidime-avibactam to colistin. Ceftazidime-avibactam had a 30-day mortality rate of 3/38 (8%) compared to 32/99 (32%) mortality rate for patients treated with colistin. Wunderink et al ⁹ compared meropenem-vaborbactam to “best available therapy” in a multicenter randomized-controlled trial. Twenty-eight day all-cause mortality occurred in 5/32 (16%) of the meropenem-vaborbactam group compared to 4/13 (31%) of the colistin group.

Table 2.

Summary of Studies Included in the Meta-Analysis of Mortality Rates Between New Beta-Lactam Beta-Lactamase Inhibitor Antibiotics Versus Colistin in the Treatment of Carbapenem-Resistant Infections.

Citation	Comparison	Design	Population	Mortality outcome
Shields et al 7	Ceftazidime-avibactam (n = 13) vs colistin plus carbapenem (n = 30)	Single-center, retrospective cohort	Patients with CR-Kp bloodstream infections	30-day all-cause mortality a
van Duin et al 8	Ceftazidime-avibactam (n = 38) vs colistin (n = 99)	Prospective, multicenter, observational study	Adults with documented CRE infections	30-day mortality
Wunderink et al 9	Meropenem-vaborbactam (n = 32) vs colistin combinations (n = 13 b )	Phase III, randomized, prospective, multicenter, multinational, open-label, active-controlled trial	Adults with infections due to confirmed or suspected CRE	28-day all-cause mortality
Motsch et al 11	Imipenem-relebactam (n = 21) vs colistin plus imipenem (n = 10)	Phase III, randomized, controlled, multicenter, double-blind trial	Adults with HAP/VAP, cIAI, or cUTI caused by imipenem-nonsusceptible pathogens	28-day all-cause mortality
Alraddadi et al 10	Ceftazidime-avibactam (n = 10) vs colistin (n = 28)	Retrospective cohort study	Adults with clinically established CRE infections	30-day all-cause mortality

Note. CRE = carbapenem-resistant enterobacterales; HAP/VAP = hospital-acquired pneumonia/ventilator-associated pneumonia; cIAI = complicated intraabdominal infection; cUTI = complicated urinary tract infection; CR-Kp = carbapenem-resistant Klebsiella pneumoniae.

^aOutcome was converted from 30-day survival to 30-day mortality.

^bOriginal study comparison arm (best available therapy) included regimens without colistin, so mortality data for the colistin/polymyxin subset of the BAT arm was obtained directly from the manufacturer, Melinta Therapeutics.

Motsch et al ¹¹ conducted a randomized-controlled clinical trial comparing imipenem-relebactam to colistin plus imipenem. Imipenem-relebactam had decreased 28-day all-cause mortality compared to colistin plus imipenem (2/21, 9% vs 3/10, 30%, respectively). Alraddadi et al ¹⁰ compared ceftazidime-avibactam to colistin in a retrospective cohort. The overall mortality rate was higher than previous studies, but similar trends were observed between groups. The mortality rate for patients who received ceftazidime-avibactam was 5/10 (50%) compared to 16/28 (57%) for patients treated with colistin. These 5 studies were combined in the meta-analysis displayed in Figure 2. The relative risk of mortality from the meta-analysis was 53% lower in the BLBLI group compared to colistin (RR 0.47, 95% CI 0.25-0.88).

Figure 2.

Forest plot results of the meta-analysis of mortality rates between new beta-lactam betalactamase inhibitor antibiotics versus colistin in the treatment of carbapenem-resistant infections.

Treatment of 1 CRO with a BLBLI cost $16 200 and produced 11.5 QALYs. Treatment of 1 CRO with a colistin-based regimen cost $3500 and produced 8.3 QALYs. The ICER for treatment with one of the new BLBLI agents was determined to be cost-effective with an ICER of $3900 per QALY gained. Complete results are presented in Table 3.

Table 3.

Results of Cost-Effectiveness Analysis of New Beta-Lactam Beta-Lactamase Inhibitor Antibiotics Versus Colistin in the Treatment of Carbapenem-Resistant Infections.

	BLBLI	Colistin	Incremental (BLBLI vs Colistin)	ICER per QALY (BLBLI vs Colistin)
Drug costs	$16 000	$3000	$13 000
Cost of AKI	$200	$500	$300
Total cost	$16 200	$3500	$12 700
QALYs	11.5	8.3	3.2	$3900

Note. BLBLI = beta-lactam beta-lactamase inhibitor; AKI = acute kidney injury.

Results from the one-way sensitivity analysis are displayed in a tornado diagram in Figure 3. The parameters with the largest impact on the ICER are the relative risk of mortality and duration of treatment. The uncertainty in the estimate of the relative risk of mortality (0.25-0.88) results in an ICER range of $2936 to $10 351 per QALY gained. The duration of treatment ranges from 7-21 days and results in an ICER range of $2008 to $6218 per QALY gained. The cost of an AKI had little impact on the ICER, despite the large range of values included to represent different AKI severities. Treatment with a BLBLI remained cost-effective under all uncertainty scenarios.

Figure 3.

Tornado diagram displaying one-way sensitivity analysis of the impact of uncertainty in model inputs on the estimated incremental cost-effectiveness ratio.

QALY = quality-adjusted life years; BLBLI = beta-lactam beta-lactamase inhibitor; AKI = acute kidney injury.

The budget impact of increasing the use of BLBLIs from 35% to 100% of CRO infections would increase by $85 000 for every 10 patients treated. Additionally, the increase in new BLBLI use would prevent 1 AKI and save 1 life per 10 patients treated.

Discussion

The results of this study suggest treatment of CROs with a new BLBLI is cost-effective compared to colistin-based combinations. The total ICER for treatment with a new BLBLI was $3900 per QALY gained. Interventions are considered cost-effective in the US if the total ICER is less than $100 000-$150 000 per QALY gained. ²⁴ Treatment of CROs with the new BLBLIs falls significantly below this threshold, so the clinical advantages are worth the increase in cost.

The results from this cost-effectiveness model and meta-analysis are consistent with previous literature. Two studies have evaluated the cost-effectiveness of new BLBLIs for the empirical treatment of complicated urinary tract infections. Kauf et al ²¹ evaluated the cost-effectiveness of ceftolozane-tazobactam compared to piperacillin-tazobactam. Ceftolozane-tazobactam was cost-effective with an ICER of $6128 per QALY gained. A European study by Kongnakorn et al ²⁸ found the ICER for ceftazidime-avibactam compared to imipenem was also cost-effective at €8039 per QALY gained. An additional strength in our study was the evaluation of definitive treatment of carbapenem-resistant infections, which can be a more complex patient population. Simon et al ²⁹ evaluated the cost-effectiveness of ceftazidime-avibactam versus colistin-based treatment of CRE bacteremia and pneumonia. Even after including the cost of discharging to a long term care facility, treatment with ceftazidime-avibactam was cost-effective with an ICER of $95 000 per QALY gained. Our results build on the existing literature by additionally evaluating meropenem-vaborbactam and imipenem-relebactam. Together, the evidence now suggests the entire novel BLBLI class is cost-effective for the treatment of CROs.

Previous studies have suggested a decrease in mortality when treating CROs with a new BLBLI compared to colistin, but these are limited by small samples and heterogeneous populations.^7-11 Van Duin et al ⁸ was the only study that independently showed a statistically significant difference in 30-day mortality between treatment groups. The relative risk of mortality from the meta-analysis was statistically significantly lower in the combined BLBLI group compared to colistin (RR 0.47, 95% CI 0.25-0.88). Our results increased confidence by pooling these studies in a meta-analysis to estimate the relative risk of motality between treatment groups.

The meta-analysis used random effects rather than fixed effects to mitigate the heterogenity between studies, caused in part by the inclusion of all infection sites. Infection sites were not separately evaluated in our model because there was not enough data to support subgroups. Most studies in the meta-analysis included infections from various sources and mortality rates were not always delineated by infection site. Random effects allowed more flexibility in the meta-analysis by not assuming identical treatment effects between studies. Weighting of studies in random effects models depends on the variance within studies as well as variance between studies. Because of this, the studies by Alraddadi et al ¹⁰ and Wunderink et al ⁹ carried the most weight in estimating the average relative risk of mortality between groups. The overall mortality rate from the study by Alraddadi et al ¹⁰ was higher than the other studies (55% vs 16-25%, respectively). The authors hypothesized that the higher overall mortality was due to the inclusion of more severe infections and more complex patients as noted by high comorbidity indexes. The Alraddadi study had the smallest mortality risk reduction for the BLBLIs, so including this study in the meta-analysis made the combined relative risk of mortality less favorable for the BLBLIs.

Despite the clinical and economic evidence supporting the safety and efficacy of new antibiotics, clinicians continue to utilize colistin and polymyxin instead of new BLBLIs. Schulz et al ³⁰ found from 2014 to 2018, the median time from FDA approval to use of a new Qualified Infectious Disease Product was 398 days (range 13 to >1478 days). Clancy et al ¹⁷ used prescription claims data to show the underutilization of several new antibiotics for the treatment of CRE. Ceftazidime-avibactam, meropenem-vaborbactam, and plazomicin were only used in 35% of cases where these new agents were considered first-line treatment by clinicians. A plausible explanation for this is the increase in cost of the newer agents. However, the results of this study show that the cost of the new BLBLIs should not be prohibitive to their use and are in fact cost effective from the perspective of the healthcare system.

Most CRO treatment will occur in acute care hospitals due to disease severity and need for parenteral antimicrobial therapy. The additional cost of new antibiotics could have a significant impact on the hospitals’ formulary budget. Our budget impact analysis estimates switching CRO treatment from colistin to new BLBLIs could prevent 1 death and 1 AKI for every 10 patients treated. This scenario increases costs by $85 000, but this is an overestimate for hospitals already using new BLBLIs in more than 35% of CRO cases. It is worth noting that treating CRE infections are expensive regardless of the antibiotic used. Bartsch et al ³¹ recently found the average cost for a hospital to treat any type of CRE infection was $29 157 (95% CI $22 993-$35 503), although antimicrobial regimen had little impact on the estimate. Newer antibiotics can add thousands of dollars for a treatment course, but are not significant drivers of total hospital expenditures. ³²

There are several limitations of the cost-effectiveness model due to parameter selection and uncertainty in those estimates. Antibiotic costs were estimated using average wholesale price, which is not the true price paid by hospitals. Antibiotic costs will differ between hospitals and health systems based on specific contract pricing and discounts. Additonal inputs that could be strengthen the model include the rate and cost of other non-AKI adverse drug events and the total cost of hospitalization. Nephrotoxicity was the only adverse drug event included in our model for 2 reasons. There is strong evidence supporting the increase in nephrotoxicity in colistin compared to other agents, and there is evidence to suggest an AKI is associated with poorer clinical outcomes and increased costs. ¹⁸ The model did not incorporate differences in length of hospitalization between patients who had an AKI because there were too many unmeasured confounders to accurately describe this difference. However, including other adverse events and length of stay after AKI would likely make the BLBLIs even more cost-effective that the result of this study show. Finally, this model used a health utility estimate after a CRO infection that was based off an assumption from a previous cost-effectiveness study. ²¹ More severe infections and longer treatment durations will likely decrease health utility, which this model is unable to capture. More research is needed to better understand health preferences of patients who survive CRO infections to inform future infectious diseases cost-effectiveness studies.

CRE infections present a multitude of difficulties including: few treatment options, costly treatment course, poor patient outcomes, and negative impact on the patients’ quality of life. The use of cost-effectiveness analyses can help illustrate these interactions should be used to help aid in formulary decisions. The results of this study demonstrated that the new BLBLI antibiotics are cost-effective compared to colistin for the treatment of CRE infections and support the preferential use of BLBLIs over colistin-based therapy when appropriate.

Conclusion

When used for the treatment of CRE, the new BLBLIs are cost-effective with an ICER of $3900 per QALY gained compared to colistin. Using the new BLBLIs instead of colistin-based combinations could prevent 1 death and 1 AKI for every ten patients treated. The new BLBLIs should be preferred over colistin-based therapies due to improved 30-day mortality and decreased AKI risk.