What's New in Critical Illness and Injury Science? Antibiotics in critical care: Therapeutic toolbox.

INTRODUCTION

Understanding the nuances of antimicrobial therapy is vital to critical care. Serious infections carry a high morbidity and mortality risk, specifically in septic shock; however, early detection and prompt antimicrobial therapy can dramatically alter the outcome.[1] Intensivists are on the frontline of recognizing impending hemodynamic instability and organ dysfunction, often when findings are too subtle to raise alarm among other providers. Worldwide, antimicrobials are one of the most commonly prescribed therapies for a decompensating patient.[23] As high as, 70% of patients receive empirical or targeted antibiotic treatment on any given day in the intensive care unit (ICU).[4] Although among patients admitted to the ICU for sepsis, only 13%–30% were subsequently adjudicated as having “no” or “probable” infection.[5] Ideally, critical care providers must carefully navigate the fine balance between prioritizing antibiotic stewardship versus ensuring adequate antibiotic coverage, yet ICUs are notorious for inappropriately excessive use of these agents. Such unchecked use is a major driver for the de novo development of antimicrobial resistance, disruption of the microbiome, and the heightened risk of Clostridioides difficile infection or drug-related toxicities. This, in turn, complicates the management of subsequent critical illness by escalating patient complexity, morbidity, and mortality risk. It also inadvertently threatens other patients with nosocomial drug-resistant organisms through colonization pressure.[6] However, stewardship is not only about limiting antibiotics; intensivists must actively engage in efforts to correctly identify patients in need of antibiotics and be vigilant to scrutinize usage daily on every patient they serve.

CHALLENGES IN THE DIAGNOSIS AND MANAGEMENT OF SEPSIS

The diagnosis and management of an infection in a critically ill patient with multiple concurrent disease processes is challenging. Half of the febrile episodes may be noninfectious in origin.[7] Raising the question, “when are antibiotics indicated?” In certain cases, such as a complicated pneumonia or perforated viscus, the answer is clear-cut and the commencement of targeted therapy and source control, if indicated, is pursued. The clinical presentation of infection, however, can be insidious. When the etiology of an ICU syndrome is not as transparent, the decision to begin empiric therapy is dependent on the level of suspicion by the provider. This is rooted in basic clinical judgment, without the availability of a perfect strategy that can be universally applied to the heterogenic critical care population.[8] The tendency to over diagnose is wherein lies the problem, as it leads to resource misuse.

The definition of sepsis has continually evolved over several decades; the most recent 2016 Sepsis-3 definition characterizes sepsis as a “life-threatening organ dysfunction caused by a dysregulated host response to infection.”[9] Severe sepsis was removed and septic shock remains, a subset of sepsis, with circulatory, cellular, and metabolic abnormalities that increase mortality. The Sepsis-3 taskforce introduced the quick Sepsis-Related Organ Failure Assessment Score as a rapid bedside tool (altered mental status, respiratory rate ≥22 breaths/min and systolic blood pressure ≤100 mm Hg) to determine the likelihood of poor outcome, such as a prolonged ICU admission or mortality risk.[9] This score lacks the sensitivity needed to justify antibiotics across the board but should be considered during basic triage.[10] In the future, opportunities for the implementation of screening tools should focus on the risk of deterioration rather than the risk of death, to provide opportunities to intervene.

Some hospitals have adopted this score, Systemic Inflammatory Response (SIRS) and others, into electronic alerts for recognizing signs of sepsis. Alarm fatigue or variations in the presentation of infection make this strategy suboptimal but nonetheless, an innovative attempt to provide early intervention with multidisciplinary support.[11] Similarly, “Code Sepsis” rapid response teams, in the emergency department or hospital ward, are tasked with operationalizing key aspects of the Society of Critical Care Medicine, and the European Society of Intensive Care Medicine's Surviving Sepsis Campaign (SSC) guidelines bundled care.[12]

The SSC has been the authoritative resource to shepherd sepsis management worldwide and has brought momentous attention to its diagnosis and prevention. For example, their collaboration with the Center for Disease Control and Prevention's “Get Ahead of Sepsis Campaign” provides education to the community. In 2017, the World Health Assembly and World Health Organization (WHO), declared sepsis as a global health priority and adopted a resolution which then encouraged United Nation members to set goals to improve the diagnosis, management, and prevention. The WHO has allocated $4.6 million U.S. dollars to help implement their sepsis resolution.[13]

While there is general consensus that SSC has catalyzed tremendous sepsis focused quality improvement initiatives, which are a cornerstone to improve system-based practice and patient outcomes, there is criticism that the stringent application of campaign components may do harm. Specifically, in the face of rising sepsis awareness and stimulus for intervention, advocating indiscriminate 1 h benchmarks to antibiotic delivery could be taken out of context and applied to patients who do not have an indication for antibiotics, but have triggered an “alert.” The protocolization rather than individualization of sepsis management could result in inappropriate prescribing to meet predefined goals. Investigators have presented concern how this, in turn, has the potential to impact government regulatory bodies and their outcome measures and thus influence future care.[1415] As such, neither does sepsis, nor its management fit a one-size-fits-all mold. In that vein, while risk stratification of the individual's clinical presentation may be a practical strategy to dictate acute management, is the diagnosis of sepsis reliable, consistent, and specific enough to be used as a benchmark for performance and quality?

EARLY INTERVENTION: FRIEND OR FOE?

The timing of antibiotic administration is guided by the urgency of the disease. In the ICU, this means identifying high-priority illnesses such as septic shock, necrotizing skin infections, febrile neutropenia, and bacterial meningitis among others. This is in contrast to infections that occur subacutely or chronically (i.e., endocarditis or osteomyelitis). In those circumstances, premature administration of antibiotic therapy, prior to blood or tissue culture and/or specimen obtainment, there is potential for indirect harm by yielding invalid microbiologic results thereby possibly denying a patient who requires a long-duration of therapy an opportunity to receive specific pathogen-targeted management.[16]

There is a paucity of evidence supporting the use of early antibiotics for sepsis without shock. The current SSC guidelines do not differentiate between suspected sepsis and suspected septic shock. The notion to universally and expeditiously administer antibiotics in a 1 h time window has been met with amicable contention, including from the Infectious Disease Society of America (IDSA).[17] For each 1 h delay in therapy a significant increase in mortality has been cited.[1819] However, other investigators have been reluctant to blindly accept this.[2021] A lack of crucial data confounds results; such as confirmation of the presence of a true infection, factoring in the selection of appropriate antibiotics, adequate therapeutic dosing, and the role of source control.[21] In a recent randomized controlled trial, of prehospital intervention with antibiotics, results did not show improvement in patient outcome with earlier administration.[22] Notably, only 10% were critically ill, and their diagnosis of sepsis was not confirmed. Ferrer et al. included approximately 18,000 ICU patients, and demonstrated that a delay in antibiotic administration was associated with an increased mortality with hourly increments, after adjusting for sepsis severity, onset of hypotension and location prior to admission.[23]

It seems intuitive and biologically plausible, that the strength of the relationship between time-to-antibiotic and clinical outcome may be most relevant in patients with evidence of shock and higher severity of illness. As a general rule, antibiotics are indicated and should be given as swiftly as possible in cases of undifferentiated shock, but with judicious consideration of the patient's individual risk-profile, while further investigation of etiology ensues. In patients with suspicion for infection, but without clinical evidence of cardiovascular or biochemical insufficiency, a goal of 3 h may be more realistic and acceptable.[24] Unknown variables still exist, such as variations in biologic phenotypes, where the host-response pattern to intervention is not as predicted.[25]

RISK ASSESSMENT AND EMPIRIC ANTIBIOTIC THERAPY

Deciding on empiric treatment in the ICU for suspected sepsis and septic shock varies by host immune status, comorbidities, and site of suspected infection. Quick synthesis of copious amounts of data occurs, including mining through recent events, following physiologic and biochemical trajectories, and assessing radiologic evidence. Risk-stratification of the severity of illness is a fundamental aspect of this process. Several prognostic scoring systems have been proposed, such as the Acute Physiology and Chronic Health Evaluation, Sequential Organ Failure Assessment, or National Early Warning Score, that display varying predictive performance in determining sepsis-related acuity and mortality risk. While these scoring systems have been useful for risk stratifying patients in research and surveillance, their true utility in bedside management, especially as a tool to capture patients who are more likely to benefit from timely antibiotic therapy remains unclear.[26]

For several frequently encountered presentations of infection in the ICU such as health care/ventilator-associated pneumonia, neutropenic fever, skin and soft-tissue infections, genitourinary and intra-abdominal infection, the IDSA evidence-based practice guidelines provide a useful guide for selecting empiric treatment.[27] It provides microorganism and site-based duration recommendations for both community and health-care infections. When the etiology is not as obvious, the choice of therapy is a composite of many features. The host immune reaction and the local response to an organism's virulence factor is an integral part of infection pathophysiology.[28] For example, endotoxin, which comprises a large portion of the outer membrane of Gram-negative bacteria, precipitates a cascade of circulatory insufficiency, adult respiratory distress syndrome, and disseminated intravascular coagulation resulting in death.[28] These serious sequelae occur even in the presence of antibiotic-susceptible infections.

Patients with prolonged hospital length of stay, use of invasive devices, and prior antibiotic exposure are at increased risk of difficult-to-treat organisms. Yet, no single algorithm exists to capture this risk entirely. Knowledge of prior infection or colonization, the hospital antibiogram and the local epidemiology of multidrug-resistant organisms in the ICU, home institution, or the specific community from where they originated can be a good start.[6]

The selection pressure and the potential for cross-colonization make the ICU an important determinant in the spread of the multidrug-resistant pathogens. Inadequate therapy for an infection is associated with poor outcomes, including greater morbidity and mortality risk and can increase hospital length of stay.[29] Therefore, in the case of presumed septic shock, broad-spectrum antimicrobial agents are traditionally used. This provides coverage of multiple potential pathogens but should be reassessed at a minimum of 72 h. It is exceedingly important to execute an individualized approach when choosing empiric therapy, and recognizing the collateral damage of potentially selecting for resistant organisms.

ASSESSING FOR DISCONTINUATION OF ANTIMICROBIALS

As many as, 33% of pneumonia cases in the ICU are viral in origin, and yet at clinical presentation, they receive antibiotics.[30] In one study, over a quarter (28%) of patients received broad-spectrum antibiotics, despite having nonbacteriologically proven disease.[2] Even with appropriate laboratory data, culture-negative sepsis can account for up to 20%–40% of cases.[31] This is especially true with a primary pulmonary source. Intensivists routinely encounter this dilemma when seeking to discontinue antibiotics at the backend of a critical illness.

With a positive culture data, the SSC guidelines recommend transitioning to definitive monotherapy for 7–10 days. A longer course can be considered in patients with inadequate source control, Staphylococcus aureus bacteremia, or immunodeficiency.[12] However, this approach is overly straightforward.[17] The treatment duration should depend on the host-pathogen dynamic, clinical response to therapy and source of infection. For example, in intra-abdominal infections, with adequate source control, 4 days of appropriate therapy may be sufficient. In contrast, a patient that developed native vertebral osteomyelitis, after a nosocomial bacteremia, may require up to 6 weeks of treatment.

Empirical or definitive antibiotics should be reassessed daily for both necessity and secondary events. Excessive antibiotic load, in the ICU environment, is a driver of selective pressure for resistant bacteria and has untoward adverse effects on the vulnerable population. In as little as 24 h on therapy, adverse drug effects including C. difficile infection or incident multidrug-resistant organism infection can occur.[32] Importantly, Tamma et al. also reported 20% of adverse drug events were attributable to antibiotics prescribed for conditions for which antibiotics were not indicated. Critical care has become a recent focus for antimicrobial stewardship programs, and more work is to be done.

Serial procalcitonin levels are increasingly used by clinicians to guide antibiotic discontinuation after clinical stabilization. The ICU population is subject to other causes of systemic inflammation (i.e., trauma, surgery, and malignancies) which may render confusing results. It is also difficult to draw conclusions on the potential survival benefit due to the low quality of evidence available and potential risk of bias. Importantly, however, five recent meta-analyses consistently demonstrated that procalcitonin testing reduced antibiotic use in patients with sepsis or lower respiratory infections.[3334353637] Thus, there may be a role for testing as an adjunct to clinical judgment, other data points, and stewardship programs.

Since routine culture data for antibiotic susceptibility testing and pathogen identification may sometimes take over 2–3 days, there has been an increased interest in automated rapid molecular-based systems to identify pathogens and their resistance genes. These fast turnaround platforms can provide easier insight into the potential causative agent and its susceptibility profile while awaiting further data. This may allow clinicians to update antibiotics accordingly. However, their associated high cost, lack of universal availability, and the absence of guidelines to direct results, currently limit their role in stewardship efforts. Still, this is an exciting area with promising prospects, and in the future, this platform may offer additional insight to improve patient outcomes.[38]

SUMMARY

Continually, investigating and improving the diagnosis and management of sepsis and septic shock are a fundamental component to our commitment to high-quality and safe patient care. Both inappropriate and suboptimal utilization of antibiotics can inadvertently lead to an increase in adverse events, morbidity, and mortality risk. Intensivists carry antibiotics in their therapeutic toolbox, and thus, they have a unique role in the worldwide fight against antimicrobial resistance by incorporating thoughtful practices daily. This is essential to ensuring positive clinical outcomes for vulnerable critical care patients now and in the future.