Editorial 

January 21, 2024

Transitioning From SIRS to Phoenix With the Updated Pediatric Sepsis Criteria: The Difficult Task of Simplifying the Complex

Roberto Jabornisky, Nathan Kuppermann, Sebastián González-Dambrauskas

JAMA. Published online January 21, 2024. doi:10.1001/jama.2023.27988

Everyone must acknowledge the difficulty of distinguishing diseases…whoever denies this may as well deny that there is such a thing as medical art.

William Cullen¹

Diagnosis as a Social Construct in Medicine

The art of making a diagnosis is essential to high-quality medical practice and is arguably the most valuable skill of a health care practitioner. Diagnosis serves as the articulation of the language of medical science, guiding medical practice at every clinical encounter and through health policy.¹ Developing medical knowledge entails a social construct rooted in the prevailing medical framework, ethical perspectives, and the socialization of these ideas among physicians. For medical knowledge to evolve, there must be an interaction between a condition’s definition and its scientific identification. This framework is shaped by how these elements interact and how the definition of a condition is configured.²

Medical science has been driven to establish boundaries defining the start of illnesses, although this frequently simplifies complex biological phenomena. The trade-off between sensitivity and specificity unfolds in the face of this complexity and is particularly acute for syndromes such as sepsis. In the face of these challenges, we discuss the new diagnostic paradigm for pediatric sepsis published in this issue of JAMA.³

The Transition From SIRS to the Phoenix Criteria

Pediatricians commonly rely on sepsis definitions established in 2005.⁴ These definitions were based on expert opinions from physicians in high-resource settings and depend on the identification of an infection-induced systemic inflammatory response syndrome (SIRS).⁴ Introduction of adult Sepsis-3 definitions in 2016 marked a shift in the conceptual framework from infection-associated SIRS to infection-associated organ dysfunction.⁵ In response to emerging data indicating that SIRS criteria in children lacked specificity for identifying those at higher risk of mortality,⁶ the pediatric sepsis research community took action. In 2019, a task force of 35 pediatric sepsis experts from 12 countries on 6 continents to develop updated operational definitions was convened.³

Unlike the creation of previous criteria, these investigators developed a 3-pronged data-driven approach, which included a global survey,⁷ a meta-analysis,⁸ and the development of a new organ dysfunction scoring system. Consistent with adult Sepsis-3,⁵ the task force considered that the existing sepsis definitions did not meet clinicians’ needs and suggested that definitions focusing on organ dysfunction would be more pertinent.³ The subsequent research phase led to the creation of the Phoenix Sepsis Score.⁹ This score was developed using robust multivariable regression techniques to derive and validate a composite model using organ dysfunction measures from electronic health data in 10 health systems across 5 countries, including high- and low-resource settings.⁹This involved complex data harmonization across many different large electronic databases encompassing more than 3.6 million pediatric encounters. The primary objective was to identify and validate factors associated with in-hospital mortality in children with sepsis, with the positive predictive value and sensitivity serving as the primary performance measures. The task force then established novel pediatric sepsis definitions following a modified Delphi process.

The Phoenix Pediatric Sepsis Criteria

The updated pediatric sepsis definition is now operationalized by 2 or more points in the Phoenix Sepsis Score (indicating life-threatening organ dysfunction of the respiratory, cardiovascular, coagulation, and/or neurologic systems) in a child with suspected or confirmed infection.³ Septic shock is operationalized by the presence of sepsis in addition to 1 or more points in the cardiovascular component of the Phoenix Sepsis Score (ie, severe hypotension, blood lactate >5 mmol/L, or vasoactive medication infusion).³ These criteria performed better than previous criteria across differently resourced settings. A substantial change in the newly proposed criteria is the removal of SIRS as a diagnostic factor. This adds clarity and minimizes confusion, particularly in the emergency department setting, where many febrile children who do not have sepsis meet SIRS criteria (eg, febrile infants with bronchiolitis who have tachycardia and tachypnea). Additionally, the distinction of severe sepsis as a separate condition has been eliminated, as the term is redundant with sepsis. Considering the worldwide implications of this change, it is important to acknowledge the challenges that will be encountered in adoption by the global pediatric community, especially in low-resource settings.

The task force deserves commendation for incorporating the input of professionals from low-resource settings and for validating the Phoenix Sepsis Score by creating an extensive database that included hospitals in these settings, where the greatest burden of sepsis occurs.¹⁰ The disproportionate impact was confirmed in the validation cohort, in which children with sepsis had a 7.1% mortality rate in high-resource settings and a 28.5% rate in low-resource settings.⁹ Those with septic shock in the first 24 hours had a 10.8% mortality rate at the former sites and a 33.5% mortality rate at the latter sites.⁹ The new criteria demonstrated a higher positive predictive value and comparable or higher sensitivity to the previous criteria for sepsis, severe sepsis, and septic shock. However, only 3.1% of the cohort used to validate the score came from low-resource settings, limiting its precision.⁹ Notably, the Phoenix Sepsis Score demonstrated lower sensitivity compared with previous criteria (23% vs 77%) at 1 lower-resource site,⁹ further underscoring the need for additional study before implementing the Phoenix Sepsis Score broadly in lower-resource settings.

Furthermore, the global survey that served as the starting point⁷ for developing this score reflected the perceptions of physicians primarily working in hospital settings, particularly in pediatric intensive care units (PICUs) (57%), with only 15% based in emergency departments. Moreover, all the low-resource validation sites were institutions with electronic health records and most had PICUs, which does not adequately reflect conditions in most low-resource settings. These factors introduce a distinct bias favoring a “PICU-based consensus,” potentially limiting the generalizability and adoption of the new criteria by health care practitioners in non-PICU and nonhospital settings responsible for recognizing and managing children with sepsis. The authors also acknowledge an additional barrier, the requirement for serum lactate measurement and coagulation parameters in the score. These laboratory tests are not readily available in many low-resource settings,⁷ although it is noteworthy that the Phoenix score performed well even in centers where lactate information was not available.

It is important to recognize that the Phoenix Sepsis Score was not designed as a screening tool for sepsis and should not be misconstrued as an early warning tool or a sepsis test. The purpose of the Phoenix Sepsis Score is to assist clinicians in identifying children with both infection and life-threatening organ dysfunction. The score was not designed to predict which children are at risk of developing sepsis. As the authors have stated, “The new pediatric sepsis criteria for sepsis and septic shock are intended to operationalize the concept of life-threatening organ dysfunction due to an infection in children, not for screening or early identification of patients with suspected sepsis.”³ Awareness of this more limited focus is critical for those who practice outside of the PICU setting.

Finally, the investigators did not eliminate the most subjective part of the diagnostic pathway in sepsis which is the “suspected infection” criteria. Clinicians seeking to implement these criteria must take this important aspect into consideration when managing children with suspected sepsis.

Global Challenges of the Phoenix Sepsis Criteria for Research and Clinical Practice

A clinically satisfactory definition of sepsis has been elusive since the term was first coined by the ancient Greeks.¹¹ Reconceptualizing sepsis based on the enhanced comprehension of its nature and complexity and identifying it as the pivotal juncture when an infection escalates into a life-threatening condition was a bold, complex, and imperative undertaking for which the authors should be praised. Identifying the moment when the immune system begins to deregulate is a critical issue, along with whether clinicians should wait for a patient to present with organ dysfunction to diagnose sepsis. Given the higher mortality in lower-resource settings, are other clinical criteria needed in these settings for earlier detection?

This important work introduces novel challenges for implementation, requiring substantial further investigation to ascertain whether these revised definitions result in improved patient care. These criteria must be further validated in a prospective manner, in differently resourced and varied settings, particularly those with the highest disease burden.

Until then, it is essential to refrain from considering these criteria as an inflexible directive governing medical interventions for pediatric sepsis. No definition can fully substitute for the clinical judgment of an experienced, vigilant clinician caring for an unwell child. We commend the Pediatric Sepsis Definition Task Force for their outstanding work, marking a substantial advance in the approach to pediatric sepsis. Nevertheless, recognizing that sepsis is a complex process, the journey ahead remains expansive. Dismissing the notion of “simple” sepsis, we anticipate ongoing dialogues among clinicians and researchers worldwide to further refine this updated conceptual framework. The departure from the widely adopted SIRS paradigm over the past 2 decades poses an important change yet a formidable challenge, as cultural shifts in medicine are inherently arduous.