Association of appropriate empiric antimicrobial therapy with acute kidney injury in gram-negative sepsis

Abstract

Background:

Despite advances in sepsis management, the relationship between appropriate empiric antibiotic therapy and acute kidney injury (AKI) in sepsis remains unclear. This study aimed to examine the association of appropriate empiric antimicrobial therapy with AKI in early onset sepsis caused by gram-negative bloodstream infections.

Methods:

We conducted a retrospective study of gram-negative bloodstream infection episodes in adult patients with early onset sepsis, using the Premier Healthcare Database from 2016 to 2020. The exposure was appropriate empiric antibiotic therapy determined by antibiotic regiments and antimicrobial susceptibilities of pathogens. The primary outcome was development of AKI or death by day 7 after the onset of sepsis. AKI was defined using the Kidney Disease Improving Global Outcome criteria based on serum creatinine levels, as urine output data were not available. The multivariable regression analysis was used to examine the association between appropriate empiric antibiotic therapy and the outcomes.

Results:

We identified 8565 patients with gram negative sepsis. In the total sample, the proportion of appropriate empiric antibiotic therapy was 93.2%, and the prevalence of AKI was 85.3%. Appropriate empiric antibiotic therapy was associated with decreased risk of AKI or death (adjusted odds ratio 0.70, 95% CI 0.52–0.94). For secondary outcomes, appropriate empiric antibiotic therapy was associated with lower AKI, shorter hospital LOS, lower C. difficile infections. However, it was not associated with in-hospital mortality.

Conclusion:

Appropriate empiric antibiotic therapy was associated with lower AKI in gram-negative sepsis. Early administration of appropriate antibiotics may prevent development of AKI.

Keywords

acute kidney injury sepsis gram-negative bloodstream infections empiric antibiotic therapy in-hospital mortality

Introduction

Sepsis is a life-threatening condition characterized by a dysregulated host response to infection, leading to organ dysfunction.¹ Gram-negative bloodstream infections (GNBSIs) are a common cause of sepsis, with pathogens such as Escherichia coli, Klebsiella species, and Pseudomonas aeruginosa being contributors.^2
–5 Acute kidney injury (AKI) is a serious complication of sepsis, including gram-negative sepsis, and is associated with significant morbidity and mortality.⁶ The incidence of sepsis-associated acute kidney injury (SA-AKI) varies widely, ranging from 14% to 87%, with mortality rates between 11% and 77%.⁷ A meta-analysis and systematic review revealed that gram-negative bacterial infections are more likely to result in AKI than gram-positive bacterial infections.⁶AKI in sepsis leads to worse outcomes, including prolonged hospital stays, increased healthcare costs, and higher mortality rates.^8
–10

Despite advances in sepsis management, the relationship between appropriate antibiotic therapy and AKI in sepsis, particularly gram-negative sepsis, remains an area of interest and uncertainty. Appropriate empiric antibiotic therapy is vital in sepsis to promptly and effectively target the causative pathogens and reduce mortality rates.^4,11
–13 However, the specific impact of appropriate antibiotic therapy on the development of AKI in sepsis is not well-defined. Previous studies have not conclusively demonstrated whether the use of appropriate antibiotics modifies the risk of developing AKI in sepsis. AKI in sepsis can result from various factors, including direct renal injury due to systemic inflammation, hemodynamic instability, nephrotoxic medications, and sepsis-induced microvascular dysfunction.¹⁴ Identifying the role of appropriate antibiotic therapy in preventing or mitigating septic AKI could have significant implications for patient care and outcomes. Therefore, this study aims to examine the association of appropriate empiric antimicrobial therapy with the development of AKI in early onset sepsis caused by GNBSIs.

Methods

Study population

We conducted a retrospective cohort study using the Premier database (Premier Inc, Charlotte, NC) between January 2016 and March 2020. Adult patients (⩾18 years old) were included if they were diagnosed with sepsis within 2 days of admission. Sepsis was defined using the Centers for Disease Control and Prevention (CDC) adult sepsis event surveillance criteria.^3,15 The onset of sepsis was defined as the timing of blood culture collection. Patients with end-stage renal disease, length of hospital stay < 2 days, and no serum creatinine measurements within 2 days of admission were excluded. The cohort was restricted to patients with monomicrobial gram-negative bloodstream infections from the first blood culture collection within 2 days of admission. We focused on common GNB (Escherichia coli, Klebsiella species, Proteus species, Enterobacter species, Pseudomonas aeruginosa, Serratia species, Acinetobacter species, and Citrobacter species) because the susceptibility for other GNB was not generally reported. This study was approved by the Duke University Institutional Review Board (IRB number: Pro00110302) with a waiver of informed consent.

Exposure

We defined an appropriate antibiotic treatment as initiation of at least one active antibiotic agent against the pathogens based on in vitro susceptibility.^12,16 A microbial result that was either resistant or intermediate was considered as nonsusceptibility. When susceptibilities to antibiotics administered to patients were not reported, imputed interpretation tables for each pathogen were used.¹²

Outcome

The primary outcome was AKI (stage 1, 2, or 3) or death occurring within 7 days from the onset of sepsis, in order to account for the competing risk of death. Using the KDIGO AKI criteria with daily serum creatinine from the onset of sepsis, AKI was categorized into 3 stages: stage 1 (creatinine level that was 1.5–1.9 times the baseline level or increased by ⩾ 0.3 mg/dL), stage 2 (creatinine level that was 2.0–2.9 times the baseline level), and stage 3 (creatinine level that was ⩾ 3.0 times the baseline level, AKI with a creatinine level ⩾ 4.0 mg/dL, or receipt of RRT).¹⁷ The AKI urine output criteria were not used because urine output was unavailable in the dataset. We used the lowest serum creatinine value during the hospital admission as baseline, because serum creatinine records before admission were unavailable.^18
–21

Secondary outcomes included any AKI occurring within 7 days from the onset of sepsis, in-hospital mortality, and C. difficile infections (CDIs). CDIs were identified using the International Classification of Diseases,10th Revision (ICD-10) diagnosis codes.

Statistical analysis

Data were presented as means ± standard deviation, medians and interquartile ranges (IQRs), or percentages, as appropriate. To estimate the association of appropriate empiric antibiotics, we fit multivariable logistic regression for binary outcomes. We fit generalized estimating equations with the negative binomial distribution for continuous outcomes. Random intercepts for individual hospitals were incorporated into the models to account for clustering within hospitals. Covariates included in the model were age, sex, transfer from anther hospital, surgery, intensive care unit admission within 2 days of admission, the van Walraven score (a weighted summary score based on Elixhauser comorbidities using ICD-10 codes), class of empiric antibiotics, infection sites (Supplemental Table 1)^4,12, pathogens, septic shock, vasopressor use, mechanical ventilation, total bilirubin, and platelet count, hospital characteristics (bed size and teaching status). Septic shock was defined as use of at least one vasopressor and lactate >2 mmol/L on the day of sepsis onset.²² A complete case analysis for the regression models was used because laboratory data had small percentages of missing information (0.1% for platelet count, and 2.1% for total bilirubin). Subgroup analysis was performed based on stratifying the cohort according to severity markers and pathogen type. An alpha level of less than 0.05 was considered statistically significant. Analyses were performed using SAS Version 9.4 (SAS Institute, Cary, NC).

Results

Patient inclusion and characteristics

In the cohort of 8565 patients with sepsis caused by GNBSIs, 7998 (93.4%) received appropriate empiric antibiotic therapy (Figure 1). Patient characteristics, including demographic information, comorbidities, and organ dysfunction, are detailed in Table 1.

Figure 1.

Patient flow.

Table 1.

Characteristics of patients with early onset gram-negative sepsis.

Characteristic	Inappropriate	Appropriate
Characteristic	(N = 567)	(N = 7998)	p value
Age, median (IQR), y	72 (60, 80)	70 (60, 79)	0.21
Male, No. (%)	291 (51.3)	3929 (49.1)	0.33
Transfer, No. (%)	53 (9.3)	269 (3.4)	<0.001
Surgery, No. (%)	170 (30.0)	2682 (33.5)	0.09
ICU admission, No. (%)	450 (79.4)	6305 (78.8)	0.81
Comorbidities, No. (%)
Congestive heart failure	154 (27.2)	1912 (23.9)	0.089
Chronic obstructive pulmonary disease	139 (24.5)	1648 (20.6)	0.031
Diabetes	249 (43.9)	3217 (40.2)	0.092
CKD			0.52
No CKD	390 (68.8)	5670 (70.9)
CKD stage 1–2	68 (12.0)	860 (10.8)
CKD stage 3–5	109 (19.2)	1468 (18.4)
Peripheral vascular disease	44 (7.8)	583 (7.3)	0.74
Neurological disorders	119 (21.0)	1375 (17.2)	0.025
Liver disease	61 (10.8)	1074 (13.4)	0.08
Cancer	84 (14.8)	1037 (13.0)	0.23
Weighted comorbidity score, median (IQR)	11 (5, 17)	11 (5, 17)	0.38
Organ dysfunction
Septic shock, No. (%)	205 (36.2)	3060 (38.3)	0.34
Vasopressors, No. (%)	258 (45.5)	3743 (46.8)	0.58
Mechanical ventilation, No. (%)	123 (21.7)	1641 (20.5)	0.54
Creatinine at admission, median (IQR), mg/dL	1.5 (1.0, 2.2)	1.5 (1.0, 2.4)	0.34
Total Bilirubin (max), median (IQR), mg/dL	1.0 (0.7, 1.9)	1.2 (0.7, 2.4)	0.004
Platelet count (min), median (IQR), 10⁹ cells/L	113 (66, 164)	108 (63, 161)	0.23
Lactate (max), median (IQR), mmol/L	3.4 (2.3, 5.4)	3.6 (2.4, 5.5)	0.13
AKI stage			0.011
No AKI	73 (12.9)	1166 (14.6)
Stage 1	231 (40.7)	2756 (34.5)
Stage 2	113 (19.9)	1933 (24.2)
Stage 3	150 (26.5)	2143 (26.8)
Teaching hospital, No. (%)	302 (53.3)	3535 (44.2)	<0.001
Bed number, No. (%)			0.18
<200	121 (21.3)	1960 (24.5)
200–499	245 (43.2)	3428 (42.9)
⩾500	201 (35.4)	2610 (32.6)

IQR: interquartile range; ICU: intensive care unit; CKD: chronic kidney disease; AKI: acute kidney injury.

The median age of patients in the inappropriate therapy group was 72 years (IQR 60–80), compared to 70 years (IQR 60–79) in the appropriate therapy group. The proportion of male patients was 51.3% in the inappropriate group and 49.1% in the appropriate group. COPD (24.5% vs 20.6%) and neurological disorders (21.0% vs 17.2%) were more prevalent in the inappropriate therapy group.

Infection profiles and antibiotic use

The infection profiles of patients with early-onset gram-negative sepsis are shown in Table 2. The most common sites of infection in both groups were genitourinary and pulmonary. Specifically, 48.7% of patients in the inappropriate therapy group and 47.1% in the appropriate therapy group had genitourinary infections, while 22.4% and 21.4%, respectively, had pulmonary infections.

Table 2.

Infection profiles of patients with early onset gram-negative sepsis.

	Inappropriate	Appropriate
	(N = 567)	(N = 7998)	p value
Sites of infection, No. (%)
Pulmonary	127 (22.4)	1714 (21.4)	0.63
Genitourinary	276 (48.7)	3765 (47.1)	0.49
Intra-abdominal	48 (8.5)	959 (12.0)	0.014
Skin or soft tissue	51 (9.0)	531 (6.6)	0.039
Bone or joint	14 (2.5)	151 (1.9)	0.42
Central nervous system	3 (0.5)	48 (0.6)	0.99
Other	182 (32.1)	2010 (25.1)	<0.001
Gram-negative organisms
Escherichia coli	306 (54.0)	4494 (56.2)	0.32
Klebsiella species	76 (13.4)	1646 (20.6)	<0.001
Proteus species	29 (5.1)	733 (9.2)	0.001
Enterobacter species	21 (3.7)	244 (3.1)	0.46
Pseudomonas aeruginosa	74 (13.1)	519 (6.5)	<0.001
Serratia species	8 (1.4)	175 (2.2)	0.28
Acinetobacter species	43 (7.6)	94 (1.2)	<0.001
Citrobacter species	10 (1.8)	93 (1.2)	0.29
CTX-RE/ESBL	301 (53.1)	705 (8.8)	<0.001
CRE	26 (4.6)	53 (0.7)	<0.001
Empiric antimicrobial therapy
Aminoglycoside, No. (%)	64 (11.3)	977 (12.2)	0.56
Cephalosporin, No. (%)	413 (72.8)	6313 (78.9)	0.001
Penicillin, No. (%)	269 (47.4)	4672 (58.4)	<0.001
Carbapenem, No. (%)	341 (60.1)	3033 (37.9)	<0.001
Quinolone, No. (%)	158 (27.9)	2344 (29.3)	0.50
Vancomycin, No. (%)	376 (66.3)	5205 (65.1)	0.58
Antimicrobial duration, day, median (IQR)	9 (6, 12)	8 (6, 11)	0.17

CTX-RE: ceftriaxone-resistant Enterobacterales; ESBL: extended-spectrum β-lactamase producing gram-negative organism; CRE: carbapenem-resistant Enterobacterales; MSSA: methicillin-susceptible Staphylococcus aureus; IQR: interquartile range.

Regarding the specific gram-negative organisms identified, Escherichia coli was the most prevalent in both groups (54.0% in inappropriate vs 56.2% in appropriate). However, significant differences were noted with Klebsiella species (13.4% vs 20.6%), Proteus species (5.1% vs 9.2%), Pseudomonas aeruginosa (13.1% vs 6.5%), and Acinetobacter species (7.6% vs 1.2%). The presence of CTX-RE/ESBL-producing organisms was notably higher in the inappropriate therapy group (53.1% vs 8.8%).

Cephalosporin use was lower in the inappropriate group (72.8% vs 78.9%), while carbapenem use was higher (60.1% vs 37.9%). The median duration of antimicrobial therapy was longer in the inappropriate therapy group (9 days vs 8 days).

Outcomes associated with appropriate empiric antibiotic therapy

The analysis of outcomes associated with appropriate empiric antibiotic therapy is presented in Table 3 and Figure 2. Among the patients, the combined outcome of AKI or death by day 7 occurred in 88.4% in the appropriate therapy group compared to 86.1% in the appropriate therapy group. The unadjusted odds ratio (OR) for this outcome was 0.81 (95% CI: 0.62–1.06), and the adjusted OR was 0.70 (95% CI: 0.52–0.94). For the outcome of AKI by day 7, appropriate empiric antibiotic therapy was associated with lower AKI (adjusted OR 0.75, 95% CI: 0.56–0.99). In the subgroup analysis for the primary outcome, there was no association between the appropriate empiric antibiotic therapy and the primary outcome by severity markers or pathogen type (Figure 2).

Table 3.

Outcomes associated with appropriate empiric antibiotic therapy.

	Inappropriate	Appropriate	OR (95% CI)
	(N = 567)	(N = 7998)	Unadjusted	Adjusted
AKI or death by day 7	501 (88.4)	6886 (86.1)	0.81 (0.62 - 1.06)	0.70 (0.52 - 0.94)
AKI by day 7	494 (87.1)	6832 (85.4)	0.86 (0.67 - 1.12)	0.75 (0.56 - 0.99)
Clostridioides difficile	22 (3.9)	174 (2.2)	0.58 (0.39 - 0.85)	0.54 (0.36 - 0.81)
In-hospital mortality	67 (11.8)	785 (9.8)	0.81 (0.62 - 1.06)	0.99 (0.72 - 1.36)
			Mean ratio (95% CI)
LOS, median (IQR), day	8 (5, 12)	7 (5, 11)	0.95 (0.89 - 1.00)	0.93 (0.88 - 0.98)

Abbreviations: AKI, acute kidney injury; OR, Odds ratio; CI, confidence interval; LOS, length of stay.

Figure 2.

Forest plot showing odds ratios of acute kidney injury or death within 7 days associated with appropriate antibiotic therapy across different subgroups.

Appropriate empiric antibiotic therapy was associated with decreased odds of Clostridioides difficile infection (adjusted OR 0.54, 95% CI: 0.36–0.81) as well as shorter LOS (adjusted mean ratio 0.93, 95% CI: 0.88 – 0.98). There was no association between appropriate empiric antibiotic therapy and in-hospital mortality after adjustment. Similarly, there was no association for in-hospital mortality across the subgroups (Supplemental figure 1).

Discussion

This study examined the impact of appropriate empiric antibiotic therapy on the development of AKI and other outcomes in patients with early-onset gram-negative sepsis. Our findings highlight that appropriate empiric antibiotic therapy is associated with lower AKI or death by day 7, shorter LOS, and lower CDI, while there was no association for in-hospital mortality.

Approximately 9 of 10 patients in our study received appropriate empiric antibiotic therapy. This proportion is slightly higher than the recent studies including patients with sepsis.^4,13,23 A multicenter retrospective study including 17,430 patients with culture-positive sepsis using the Cerner database from 2009 to 2015 demonstrated that 82% of the patients received adequate empiric antimicrobial therapy.⁴ This finding may be explained by the fact that more critically ill patients with sepsis might be more likely to receive broad-spectrum antibiotics, but the prevalence of resistance pathogens are generally not high in the United States.^4,24,25

Early appropriate antimicrobial administration and source control remain a cornerstone of sepsis treatment, which may lead to prevention of sepsis associated with AKI.¹⁴ In our study, appropriate empiric antibiotic therapy was associated with deceased odds of AKI. Although there is little evidence directly looking at the relationship between appropriate antibiotic therapy, a large retrospective study reported that there was no association between appropriate antibiotic therapy and hospital-onset AKI in patients with community-onset sepsis.⁴ The discrepancy might be explained by differences in the definition of AKI, as they defined AKI as an increase in creatinine during the hospitalization. Sepsis-associated AKI can occur due to either direct mechanisms (infection or the host response to infection) or indirect mechanisms (sepsis therapies including antimicrobials).⁷ Therefore, in addition to early administration of appropriate antimicrobial agents to prevent AKI, the future studies will focus on the choice of empiric antimicrobial therapy and the strategy of antimicrobial de-escalation in patients with sepsis.

Although we observed a trend toward lower in-hospital mortality in patients treated with appropriate empiric antibiotic therapy, the result was not statistically significant. In contrast, several studies showed that appropriate empiric antibiotic therapy among severe infections or sepsis was associated with lower mortality.^11,26 There are possible explanations for this inconsistency. One possibility is that our findings might be confounded by indication, because clinicians might be more likely to administer broad-spectrum antibiotics to patients with more severe status, which would make the treatment appear worse.²⁷ Moreover, the definition of appropriate empiric antibiotic therapy in this study was based only on the relationship between antibiotic regimens administered to patients and in-vitro susceptibility tests. However, antibiotic concordance is not necessarily contributing to therapeutic appropriateness, because the overall effectiveness of sepsis treatment includes multiple factors such as the pathogen, site of infection, severity of illness, the timing of antibiotic use, and baseline patient characteristics.²⁸

Limitation

This study has several limitations that need to be acknowledged. First, the generalizability of our findings may be limited due to the specific patient population and healthcare settings included in the study. Second, the definition of sepsis used in our analysis did not adhere to the Sepsis-3 criteria, as the qSOFA score was unavailable due to a lack of Glasgow Coma Scale (GCS) scores in the Premier database. This discrepancy could affect the consistency and comparability of our findings with other studies using the Sepsis-3 criteria. Third, baseline creatinine levels were not available for all patients. In addition, some patients had fewer creatinine measurements available for calculating AKI. These factors may have affected the accuracy of the assessment of AKI development. Fourth, the definition of appropriate antibiotic therapy relied only on in vitro susceptibility data. Dosage and frequency of antibiotics as well as minimum inhibitory concentration were not considered in the analysis, so that antibiotic agents that were susceptible to a pathogen might not achieve effective concentrations at the infection site. Also, antibiotic timing and duration were not considered in the definition of appropriateness, which might lead to the potential for misclassification bias. Fifth, AKI was defined according to the KDIGO criteria based solely on serum creatinine data, as urine output data were unavailable. These limitations underscore the need for cautious interpretation of our results and highlight the importance of further research with more comprehensive data to validate our findings.

Conclusion

In this retrospective cohort study including 8,565 cases of gram-negative bloodstream infections in adults with sepsis, we found that administration of appropriate antibiotics was associated with a reduced risk of AKI or death within seven days of sepsis onset. Notably, while there was no association between appropriate antibiotic therapy and in-hospital mortality, its impact on mitigating AKI underscores its importance in sepsis management. Further prospective studies are warranted to validate our findings across diverse patient populations and healthcare settings, ensuring broader applicability and guiding optimal treatment strategies in clinical practice.

Supplemental Material

sj-docx-1-inc-10.1177_17511437251377990 – Supplemental material for Association of appropriate empiric antimicrobial therapy with acute kidney injury in gram-negative sepsis

Supplemental material, sj-docx-1-inc-10.1177_17511437251377990 for Association of appropriate empiric antimicrobial therapy with acute kidney injury in gram-negative sepsis by Tetsu Ohnuma, Shreya Khandelwal, Shingo Chihara, Miriam Treggiari, Jamie R. Privratsky, Pattrapun Wongsripuemtet, Julia A. Messina, Karthik Raghunathan and Vijay Krishnamoorthy in Journal of the Intensive Care Society

Footnotes

ORCID iDs

Tetsu Ohnuma

Pattrapun Wongsripuemtet

Ethical considerations

This study was approved by the Duke University Institutional Review Board (IRB number: Pro00110302) with a waiver of informed consent.

Author contributions

Study design was performed by TO, VK, and MMT. Data collection and analysis were performed by TO. Literature search and manuscript preparation were performed by SK and TO. The manuscript was reviewed by all authors.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

References

Singer

Deutschman

Seymour

, et al The third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016; 315(8): 801–810.

NHS Improvement. Guidance on the definition of healthcare associated Gram-negative bloodstream infections. 2017.

Ohnuma

Chihara

Costin

, et al Epidemiology, resistance profiles, and outcomes of bloodstream infections in community-onset sepsis in the United States. Crit Care Med 2023; 51(9): 1148–1158.

Rhee

Kadri

Dekker

, et al Prevalence of antibiotic-resistant pathogens in culture-proven sepsis and outcomes associated with inadequate and broad-spectrum empiric antibiotic use. JAMA Netw Open 2020; 3(4): e202899.

Tabah

Buetti

Staiquly

, et al Epidemiology and outcomes of hospital-acquired bloodstream infections in intensive care unit patients: the EUROBACT-2 international cohort study. Intensive Care Med 2023; 49(2): 178–190.

Liu

Xie

, et al Rates, predictors, and mortality of sepsis-associated acute kidney injury: a systematic review and meta-analysis. BMC Nephrol 2020; 21(1): 318.

Zarbock

Nadim

Pickkers

, et al Sepsis-associated acute kidney injury: consensus report of the 28th Acute Disease Quality Initiative workgroup. Nat Rev Nephrol 2023; 19(6): 401–417.

Alobaidi

Basu

Goldstein

, et al Sepsis-associated acute kidney injury. Semin Nephrol 2015; 35(1): 2–11.

Bagshaw

Uchino

Bellomo

, et al Septic acute kidney injury in critically ill patients: clinical characteristics and outcomes. Clin J Am Soc Nephrol 2007; 2(3): 431–439.

10.

Bouchard

Acharya

Cerda

, et al A prospective international multicenter study of AKI in the intensive care unit. Clin J Am Soc Nephrol 2015; 10(8): 1324–1331.

11.

Marquet

Liesenborgs

Bergs

, et al Incidence and outcome of inappropriate in-hospital empiric antibiotics for severe infection: a systematic review and meta-analysis. Crit Care 2015; 19(1): 63.

12.

Ohnuma

Chihara

Costin

, et al Association of appropriate empirical antimicrobial therapy with in-hospital mortality in patients with bloodstream infections in the US. JAMA Netw Open 2023; 6(1): e2249353.

13.

Song

Han

, et al Effect of appropriate empirical antimicrobial therapy on mortality of patients with Gram-negative bloodstream infections: a retrospective cohort study. BMC Infect Dis 2023; 23(1): 344.

14.

Peerapornratana

Manrique-Caballero

Gomez

, et al Acute kidney injury from sepsis: current concepts, epidemiology, pathophysiology, prevention and treatment. Kidney Int 2019; 96(5): 1083–1099.

15.

Centers for Disease Control and Prevention. Hospital toolkit for adult sepsis surveillance. 2018.

16.

Kadri

Lai

Warner

, et al Inappropriate empirical antibiotic therapy for bloodstream infections based on discordant in vitro susceptibilities: a retrospective cohort analysis of prevalence, predictors, and mortality risk in US hospitals. Lancet Infect Dis 2021; 21(2): 241–251.

17.

Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl 2012; 2(1): 1–138.

18.

Chen

Deng

Calvachi-Prieto

, et al A large-scale multicenter retrospective study on nephrotoxicity associated with empiric broad-spectrum antibiotics in critically ill patients. Chest 2023; 164(2): 355–368.

19.

Sparrow

Swan

Moore

, et al Disparate outcomes observed within Kidney Disease: Improving Global Outcomes (KDIGO) acute kidney injury stage 1. Kidney Int 2019; 95(4): 905–913.

20.

Graversen

Jensen

Vestergaard

, et al Defining baseline creatinine for identification of AKI in population-based laboratory databases: a Danish nationwide cohort study. Kidney360 2022; 3(2): 232–241.

21.

Pant

Chihara

Krishnamoorthy

, et al Association of causative pathogens with acute kidney injury in adult patients with community-onset sepsis. Crit Care Explor 2025; 7(2): e1219.

22.

Shankar-Hari

Phillips

Levy

, et al Developing a new definition and assessing new clinical criteria for septic shock: for the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315(8): 775–787.

23.

Van Heuverswyn

Valik

van der Werff

, et al Association between time to appropriate antimicrobial treatment and 30-day mortality in patients with bloodstream infections: a retrospective cohort study. Clin Infect Dis 2023; 76(3): 469–478.

24.

Prescott

Seelye

Wang

, et al Temporal trends in antimicrobial prescribing during hospitalization for potential infection and sepsis. JAMA Intern Med 2022; 182(8): 805–813.

25.

Rhee

Chen

Kadri

, et al Trends in empiric broad-spectrum antibiotic use for suspected community-onset sepsis in US hospitals. JAMA Netw Open 2024; 7(6): e2418923.

26.

Paul

Shani

Muchtar

, et al Systematic review and meta-analysis of the efficacy of appropriate empiric antibiotic therapy for sepsis. Antimicrob Agents Chemother 2010; 54(11): 4851–4863.

27.

Ohnuma

Krishnamoorthy

Raghunathan

, et al Overcoming confounding by indication in nutrition research using electronic healthcare data. Clin Nutr 2020; 39(4): 985–987.

28.

Haeusler

Thursky

KA.

Electronic health record data for antimicrobial prescribing. Lancet Infect Dis 2021; 21(2): 155–157.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB