Initiating antibiotics in sepsis: when late becomes too late

Every few years, new retrospective studies continue to claim that each hour delay in antibiotic administration significantly increases mortality in sepsis and septic shock, implying that any patient with a hint of sepsis should receive antibiotics immediately. However, both sepsis and septic shock can present with widely variable clinical features, making diagnosis challenging. Notably, a post-hoc analysis of 2,579 ICU admissions for sepsis found that infection was considered unlikely in over 40% of cases (1). This blog explores the current evidence regarding the time to antibiotic administration in sepsis and septic shock, addressing the following questions:

1) Why retrospective studies linking each hour’s delay in administration of antibiotics to increased mortality in sepsis shouldn’t be trusted.

2) What the current literature says about the timeframe for antibiotic administration in sepsis and septic shock.

3) What else can be done in the first 24 hours to optimize antimicrobial treatment for sepsis and septic shock?

1) Why retrospective studies linking each hour’s delay in administration of antibiotics to increased mortality in sepsis shouldn’t be trusted.

 

To date, over 30 published studies have investigated the relationship between time to antibiotic administration and mortality in sepsis (2). However, their conclusions are highly heterogeneous due to multiple confounding factors that influence this relationship (2,8). The following table summarizes these factors and the effect on the relationship between time to antibiotics and mortality.

Factor	Increases apparent benefit early antibiotics	Decrease apparent benefit early antibiotics
Time Zero (2)	Often counted from ED triage, which may not reflect true onset of sepsis (2)	Patients with hypotension pre-hospital (e.g.: Half of patients with septic shock might be hypotensive on ED arrival (3)).
Illness Severity (2)	Severely ill patients receive all treatments faster (2), potentially exaggerating the benefit of early administration.	Less severe cases (e.g. 20% of sepsis patients are well enough to be discharged home (4)) may dilute the observed benefit.
Clinical Presentation (2)	Atypical presentations (e.g. geriatric syndromes) increase time to antibiotics (5) whereas young people have classical presentations (2).	—
Other Confounders (8)	E.g.: In 1 study, adjusting for 40 confounders revealed mortality benefit with early antibiotics (8).	In the same study, without adjustment, mortality decreased with each hour delay in antibiotics (8).
Appropriate Antibiotic Therapy (3)	—	Inappropriate antibiotic therapy decreases benefit of early administration.
Time to Other Interventions (6)	Earlier antibiotics often coincide with earlier administration of other therapies.	—
Quality of Care (7)	High-quality hospitals administer antibiotics earlier.	Low-quality hospitals administer antibiotics in general later.
Time of Day (6)	Available doctors and nurses lead to better care (E.g.: In 1 study, independent association with mortality (5)).	Overloaded staff may administer lower-quality care.
Inclusion of Outliers (8)	People treated very late may disproportionately increase observed mortality, skewing results (8).	—

Due to these confounders, the authors of a recent meta-analysis on the relationship between shock management bundles and mortality, for instance, concluded that “due to considerable methodological heterogeneity among the included studies it was not performed as the results could have been misleading” (9).

Furthermore, most retrospective studies report a mortality rate per hour of antibiotic delay, assuming a linear relationship between delay and mortality (2,8). However, large observational studies involving over 50 000 septic patients observed only a slight increase in mortality during the first few hours of delay in antibiotic administration, followed by a rapid rise thereafter (2). While the mortality rate per hour of antibiotic delay is a very comprehensive metric, it may not accurately reflect the observed clinical reality.

2) What the current literature says about the timeframe for antibiotic administration in sepsis and septic shock.

 

Given that antibiotics are essential in the treatment of sepsis, it is reasonable to assume that earlier administration would be beneficial (10). This assumption underpins the current doctrine (11) that antibiotics should be initiated as soon as possible, within 1 hour for septic shock and within 3 hours for sepsis (12). However, clinical practice is more nuanced and antibiotic administration should therefore consider three key factors (13):

1) Certainty of diagnosis

Unlike stroke or myocardial infarction, sepsis is far more challenging to diagnose objectively (14), leading to frequent misdiagnosis of sepsis and septic shock:

• One study, for instance, found that 18% of patients initially treated for sepsis in US emergency departments, had non-infectious conditions (11). Furthermore, the aforementioned, post-hoc analysis of Dutch intensive care patients admitted for sepsis revealed that 43% were unlikely to ever had an infection (1).

• Similar issues were observed In the few randomized trials on time to antibiotics in sepsis. For instance, a study across 23 emergency departments in France and Spain found that over 70% of patients ultimately did not meet sepsis criteria, based on a SOFA score < 2 (16). Even the most renowned randomized controlled trials on this topic, the PHANTASi trial, faced criticism because only 57% of participants presented with organ dysfunction (10).

My view: Diagnosing sepsis is inherently difficult. Allowing more time for investigations could improve diagnostic accuracy (although this hypothesis has not been confirmed in the literature as far as I know).

2) Risk of delaying antimicrobial treatment

The literature continues to debate whether delaying antimicrobial therapy significantly increases mortality in sepsis and septic shock. Below is an overview of some of the most interesting studies:

a. Kumar et al

This retrospective study - arguably the most cited on this topic - found that each hour of delay in administering effective antibiotics after the onset of hypotension was associated with a 7.6% increase in mortality (17). However, the 558 excluded patients, who were already receiving appropriate antibiotic therapy before hypotension developed, exhibited similar mortality rates to those who got their first dose five hours after hypotension onset (11). This observation challenges the whole conclusion of Kumar et al.

b. PHANTASi trial

The aforementioned PHANTASi trial, a randomized controlled study, compared usual care with pre-hospital antibiotic administration in patients with sepsis and septic shock (18). Of the 2672 participants, 57% had severe sepsis (organ dysfunction) and 4% had septic shock. Patients in the intervention group received antibiotics an average of 96 minutes earlier than those in the usual care group, yet no survival benefit was observed. Additionally, cultures were less frequently positive in the intervention group (14).

c. Hranjec et al.

This quasi-experimental before/after study contrasted an aggressive antimicrobial approach (immediate initiation after culturing) with a conservative approach (initiation only after microbiological evidence of infection) in surgical intensive care patients (13). The authors concluded that, in hemodynamically stable patients, waiting for microbiological evidence improved survival (19). The data suggest this finding might also apply to patients in shock (20). Nevertheless, the overall fatality rate in this study was high.

d. Multitude of observational studies

As previously noted, several large observational studies identify a breakpoint a few hours after sepsis presentation, beyond which survival declines significantly (2, 8, 21). This breakpoint could be used as an indicator of the reasonable timeframe in which antibiotics should be given in sepsis. However, some experts debate the biological plausibility of such cutoff (11).

3) Risk of unjustified antimicrobial treatment

Inappropriate antimicrobial treatment carries several risks:

I. Inducing antimicrobial resistance (21).

II. Disrupting the microbiome, leading to increasing fungal and C. Difficile infections (11).

III. Missing alternative diagnoses due to the false belief that the patient has sepsis. As mentioned before, sepsis mimics occur in one out of five patients initially diagnosed with sepsis in US emergency departments (11).

My view: I believe that the 2021 Dutch guidelines strike a pragmatic balance between these three factors. They recommend antibiotics as soon as possible for septic shocks and a more nuanced approach for sepsis without shock, weighing the risks of delay against those of overuse (22). No timeframe is mentioned, allowing for clinical judgment.

3) What else can be done in the first 24 hours to optimize antimicrobial treatment for sepsis and septic shock?

 

While early antibiotic administration is critical in sepsis and septic shock, optimizing therapy in the first 24 hours is equally important. In what follows, there is a short discussion of key strategies to do this (although future posts will delve deeper into specific interventions):

3) What else can be done in the first 24 hours to optimize antimicrobial treatment for sepsis and septic shock?

While early antibiotic administration is critical in sepsis and septic shock, optimizing therapy in the first 24 hours is equally important. In what follows, there is a short discussion of key strategies to do this (although future posts will delve deeper into specific interventions):

• Loading dose and prolonged infusion of beta-lactams

Prolonged infusion of beta-lactam antibiotics is associated with reduced 90-day mortality in sepsis and septic shock, according to a recent meta-analysis (24). However, in these cases, loading doses are essential for rapid attainment of target concentrations (25).

• Accounting for pathophysiologic changes in sepsis

Sepsis induces significant pharmacokinetic alterations, such as hypoalbuminemia, increased volume of distribution and changes in the renal function (25). Especially augmented renal clearance, a common phenomenon in sepsis, may lead to underdosing of hydrophilic antibiotics like beta-lactams (25).

• Order of antibiotic administration

Some observational studies suggest that administering beta-lactams before vancomycin may improve survival, though these results may be confounded by other factors (28).

• Minimizing drug loss in IV lines

Up to 25% of the antibiotic dose can remain in the infusion line after administration. Therefore, flushing the line post administration is of utmost importance (26).

• Timely administration of the second dose

One study found that delays in the second doses of antibiotics occur in about 25% of patients, often after transfer out of the emergency department (27).Loading dose and prolonged infusion of beta-lactams

Prolonged infusion of beta-lactam antibiotics is associated with reduced 90-day mortality in sepsis and septic shock, according to a recent meta-analysis (24). However, in these cases, loading doses are essential for rapid attainment of target concentrations (25).

• Accounting for pathophysiologic changes in sepsis

Sepsis induces significant pharmacokinetic alterations, such as hypoalbuminemia, increased volume of distribution and changes in the renal function (25). Especially augmented renal clearance, a common phenomenon in sepsis, may lead to underdosing of hydrophilic antibiotics like beta-lactams (25).

• Order of antibiotic administration

Some observational studies suggest that administering beta-lactams before vancomycin may improve survival, though these results may be confounded by other factors (28).

• Minimizing drug loss in IV lines

Up to 25% of the antibiotic dose can remain in the infusion line after administration. Therefore, flushing the line post administration is of utmost importance (26).

• Timely administration of the second dose

One study found that delays in the second doses of antibiotics occur in about 25% of patients, often after transfer out of the emergency department (27).

References:

 1. Klein Klouwenberg PM, Cremer OL, van Vught LA, et al. Likelihood of infection in patients with presumed sepsis at the time of intensive care unit admission: a cohort study. Crit Care. 2015 Sep 7;19(1):319.

 2. Weinberger J, Rhee C, Klompas M. A Critical Analysis of the Literature on Time-to-Antibiotics in Suspected Sepsis. J Infect Dis. 2020 Jul 21;222(Suppl 2):S110-S118.

 3. Kumar A. Systematic Bias in Meta-Analyses of Time to Antimicrobial in Sepsis Studies. Crit Care Med. 2016 Apr;44(4):e234-5.

 4. Wang HE, Jones AR, Donnelly JP. Revised National Estimates of Emergency Department Visits for Sepsis in the United States. Crit Care Med. 2017 Sep;45(9):1443-1449.

 5. Tantarattanapong S, Hemwej T. Door-to-antibiotic Time and In-hospital Mortality of Elder Patients Presenting to Emergency Department with Sepsis; a Cross-Sectional Study. Arch Acad Emerg Med. 2021 Jun 11;9(1):e44.

 6. Farkas J. The fallacy of time-to-intervention studies. https://emcrit.org/pulmcrit/the-fallacy-of-time-to-intervention-studies/ (Accessed on 25th of October 2025).

7. Pines JM. Timing of antibiotics for acute, severe infections. Emerg Med Clin North Am. 2008 May;26(2):245-57, vii.

 8. Pak TR, Young J, McKenna CS, et al. Risk of Misleading Conclusions in Observational Studies of Time-to-Antibiotics and Mortality in Suspected Sepsis. Clin Infect Dis. 2023 Nov 30;77(11):1534-1543.

 9. Ford JS, Morrison JC, Kyaw M, et al. The Effect of Severe Sepsis and Septic Shock Management Bundle (SEP-1) Compliance and Implementation on Mortality Among Patients With Sepsis : A Systematic Review. Ann Intern Med. 2025 Apr;178(4):543-557.

 10. Vincent JL. Antibiotic administration in the ambulance? Lancet Respir Med. 2018 Jan;6(1):5-6.

 11. Singer M. Antibiotics for Sepsis: Does Each Hour Really Count, or Is It Incestuous Amplification? Am J Respir Crit Care Med. 2017 Oct 1;196(7):800-802.

 12. Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021 Nov;47(11):1181-1247.

 13. Hranjec T, Rosenberger LH, Swenson B, et al. Aggressive versus conservative initiation of antimicrobial treatment in critically ill surgical patients with suspected intensive-care-unit-acquired infection: a quasi-experimental, before and after observational cohort study. Lancet Infect Dis. 2012 Oct;12(10):774-80.

 14. Rello J, Rubulotta F. Best practice for sepsis. J Thorac Dis. 2018 Mar;10(3):1280-1282.

 15. Schinkel M, Nannan Panday RS, Wiersinga WJ, et al. Timeliness of antibiotics for patients with sepsis and septic shock. J Thorac Dis. 2020 Feb;12(Suppl 1):S66-S71.

 16. Helms J, Póvoa P, Jaber S. Refining trial design in sepsis management: balancing realism with ideal outcomes. Intensive Care Med. 2024 Jul;50(7):1126-1128.

 17. Kumar A, Roberts D, Wood KE, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006 Jun;34(6):1589-96.

 18. Alam N, Oskam E, Stassen PM, et al. Prehospital antibiotics in the ambulance for sepsis: a multicentre, open label, randomised trial. Lancet Respir Med. 2018 Jan;6(1):40-50.

 19. Hranjec T, Sawyer RG. Aggressive versus conservative initiation of antibiotics--authors' reply. Lancet Infect Dis. 2013 May;13(5):390.

 20. Reade MC. Aggressive versus conservative initiation of antibiotics. Lancet Infect Dis. 2013 May;13(5):388-9.

 21. Simpson SQ. Prehospital Antibiotics for Sepsis: An Open but Not Shut Case. Chest. 2018 Mar;153(3):588-589.

 22. Llor C, Bjerrum L. Antimicrobial resistance: risk associated with antibiotic overuse and initiatives to reduce the problem. Ther Adv Drug Saf. 2014 Dec;5(6):229-41.

 23. Sieswerda E, Bax HI, Hoogerwerf JJ, et al. The 2021 Dutch Working Party on Antibiotic Policy (SWAB) guidelines for empirical antibacterial therapy of sepsis in adults. BMC Infect Dis. 2022 Aug 11;22(1):687.

 24. Abdul-Aziz MH, Hammond NE, Brett SJ, et al. Prolonged vs Intermittent Infusions of β-Lactam Antibiotics in Adults With Sepsis or Septic Shock: A Systematic Review and Meta-Analysis. JAMA. 2024 Aug 27;332(8):638-648.

 25. Pereira JG, Fernandes J, Duarte AR, et al. β-Lactam Dosing in Critical Patients: A Narrative Review of Optimal Efficacy and the Prevention of Resistance and Toxicity. Antibiotics (Basel). 2022 Dec 18;11(12):1839.

 26. Vanneste D, Gijsen M, Debaveye Y, et al. A good start is half the battle: Antibiotic IV line residuals in the intensive care unit. Intensive Crit Care Nurs. 2024 Jun;82:103663.

 27. Cook ME, Schuler BR, Schontz MJ, et al. Prevalence, risk factors, and outcomes associated with delayed second doses of antibiotics in sepsis at a large academic medical center. Antimicrob Steward Healthc Epidemiol. 2023 Nov 10;3(1):e207.

 28. Cravero JC, Telchik C, Yakubik T, et al. A single-institution retrospective analysis on the administration of β-lactam antibiotics prior to vancomycin in bacteremic patients. Proc (Bayl Univ Med Cent). 2025 Apr 28;38(4):421-426.