Nurse-led sepsis model lifts 1-hour bundle adherence and survival in emergency resuscitation room

Patients with septic shock treated in an emergency resuscitation room after the introduction of a systematic nursing model led by emergency department nurses had significantly higher adherence to the Surviving Sepsis Campaign 1-hour bundle, better short- and medium-term survival, and fewer adverse events than comparable patients treated before the model was implemented, according to a retrospective propensity-matched study published in Frontiers in Medicine. After balancing the two patient groups on key baseline characteristics, the proportion of patients who completed all five elements of the 1-hour bundle within one hour rose from 14.12 percent to 23.66 percent, and 28-day survival improved from 70.23 percent to 81.30 percent.

The study was conducted in the emergency resuscitation room of the First Affiliated Hospital of Wenzhou Medical University, a tertiary, university-affiliated general hospital serving Wenzhou and surrounding areas in China. Researchers consecutively enrolled adult patients who met the Sepsis-3 criteria for septic shock between January 2022 and February 2025. The criteria required confirmed or suspected infection, a need for vasopressors to maintain a mean arterial pressure of at least 65 mmHg after adequate fluid resuscitation, and a serum lactate level above 18 mg/dL. Patients had to receive their initial treatment in the emergency resuscitation room. Pregnant women, patients with end-stage disease and an estimated survival of less than 24 hours, patients with documented do-not-resuscitate orders, patients who had already received complete 1-hour bundle therapy before transfer from other institutions, and patients whose primary cause of death was non-infectious, such as trauma, extensive burns, or cerebral hemorrhage, were excluded.

The nursing model was developed with the Delphi method, a structured process of expert consultation. The research team first drafted core care items through literature review and group discussion, then consulted twenty experts from emergency critical care, nursing management, infection control, and related specialties across two rounds. All experts had at least ten years of working experience and held mid-level or higher professional titles. Both survey rounds achieved a 100 percent response rate, with authority coefficients of 0.87 and 0.89, and all constructed measures met the retention criteria of a mean importance score above 3.5, a coefficient of variation below 0.25, and a statistically significant Kendall’s W coefficient of concordance. After the first round, two items were removed, one group of similar items was merged, and one new item was added. The final model comprised seven core modules, with the nursing team holding a leading or core collaborative role in each.

The first module established a sepsis rapid response team led by emergency nurses, coordinated by the nursing team leader and including senior emergency nurses, emergency physicians, ICU consultants, and laboratory and pharmacy liaisons. All nursing staff completed a two-week systematic training covering the Sepsis-3 criteria, the 2021 Surviving Sepsis Campaign guidelines, 1-hour bundle nursing protocols, electronic alarm response, and multidisciplinary communication, and had to pass both a written test with at least 85 points and a practical assessment. The team operated around the clock: once a nurse received an alert, a preliminary assessment was completed within five minutes, and core team members reached the bedside within fifteen minutes. A second module installed a dedicated alarm terminal at the nursing station, interfaced with the hospital information and laboratory systems to automatically calculate qSOFA and SOFA scores. Alerts were triggered by a qSOFA of 2 or more, a lactate above 18 mg/dL with a mean arterial pressure below 65 mmHg, or infection-related order sets, and nurses were required to record three time points: alarm receipt, assessment completion, and physician notification.

A third module introduced a standardized 1-hour bundle nursing checklist covering the five core interventions: blood sampling for lactate measurement within fifteen minutes, bilateral double-bottle blood cultures collected within thirty minutes before antibiotics, broad-spectrum antibiotic infusion completed within one hour, crystalloid resuscitation at 30 mL/kg through dual-channel infusion for patients with hypotension or lactate of at least 36 mg/dL, and vasopressor administration, primarily norepinephrine, titrated to maintain a mean arterial pressure of at least 65 mmHg with continuous monitoring. Each completed item was timed and verified with dual signatures from the nurse and physician. Further modules added an integrated information platform linking admissions, imaging, laboratory, pharmacy, and physicians, with nurses using handheld terminals to scan wristbands for traceable documentation; dynamic monitoring with bedside ultrasound assessment of inferior vena cava collapsibility, lactate clearance checks every two hours, hourly fluid balance summaries, and real-time vasopressor adjustment; zoned nursing management with isolation resuscitation for patients flagged by the Zhejiang Health Code system; and a hospital-wide nursing quality improvement group applying the Plan-Do-Check-Act cycle to analyze delayed bundle adherence in bi-weekly and monthly meetings.

Patients treated before the model’s implementation in August 2023 formed the control group (331 patients) and received conventional nursing, in which assessment and execution depended on the individual duty nurse without standardized sepsis recognition tools, electronic alert response, a dedicated rapid response team, a bundle checklist, or information system support. Patients treated from August 2023 onward formed the observation group (343 patients). Because the two periods could differ in baseline illness severity, the researchers used propensity score matching with a 1:1 ratio and a caliper of 0.2 standard deviations, adjusting for age, sex, SOFA score, Charlson Comorbidity Index, site of infection, initial lactate level, and antibiotic exposure history. Before matching, the groups differed significantly in antibiotic use, lactate level, APACHE score, and SOFA score. After matching, 262 patients in each group were well balanced, with no statistically significant differences on any measured variable.

The primary outcome findings were clear. The 1-hour bundle adherence rate, defined as completion of all five core interventions within one hour of admission and verified through nursing checklist records, was 23.66 percent in the observation group versus 14.12 percent in the control group, a statistically significant difference, although the authors emphasized that the absolute adherence rate remained low in both groups. The 24-hour rescue success rate, defined as survival without cardiac arrest or in-hospital death within 24 hours of diagnosis, reached 93.51 percent in the observation group versus 80.92 percent in the control group. The 28-day survival rate was 81.30 percent versus 70.23 percent, and the 90-day survival rate was 74.43 percent versus 64.89 percent, both significant differences.

Secondary outcomes showed a pattern of gradual rather than immediate benefit. SOFA scores, a measure of organ dysfunction, did not differ between the groups on day 1 or day 3, but by day 7 the observation group had significantly lower scores. The observation group also had shorter emergency department length of stay, shorter ICU stay, shorter total hospital stay, and shorter duration of vasoactive agent use, all statistically significant. Mechanical ventilation duration did not differ between the groups, which the authors attributed to the influence of primary lung infection, underlying lung function, and weaning strategy on ventilator decisions, factors the bundle intervention mainly targeting systemic inflammation and hemodynamics would not be expected to change. Adverse events were less frequent overall in the observation group, at 12.98 percent versus 32.82 percent, driven by significantly lower rates of arrhythmia, 2.67 percent versus 8.78 percent, and nausea and vomiting, 5.34 percent versus 13.74 percent. Differences in fluid overload and allergic reaction were not statistically significant, and neither group experienced healthcare-associated infections.

The authors placed their adherence figures in context, noting that the observation group’s 23.66 percent rate remains below the 57.7 percent reported after a quality improvement initiative in a tertiary emergency department elsewhere, but is comparable to rates of 9.95 percent to 18 percent reported among Chinese emergency and critical care populations, where low nurse-to-patient ratios and limited resuscitation room resources make completion of all five bundle elements within one hour difficult. They also compared their survival findings with prior literature, including a 2022 Japanese study in which in-hospital mortality was 18.0 percent in bundle-adherent patients versus 30.3 percent in non-adherent patients, and a 2026 systematic review associating 1-hour bundle compliance with reduced short-term mortality. They cautioned that a longer 7.5-year quality improvement study reported a larger 90-day mortality reduction, suggesting that more mature programs may achieve greater benefits than this early-stage implementation.

The study’s limitations are substantial and acknowledged by the authors. It is a single-center retrospective matched study with a limited sample size, and despite propensity matching on measured variables, unmeasured residual confounding cannot be excluded; the temporal before-and-after design means other changes over the study period could have contributed to the observed differences. Assessment of 90-day outcomes depended on the completeness of post-discharge follow-up, and loss to follow-up may have introduced information bias. The authors also noted that the study did not include dynamic monitoring of inflammatory mediators or microcirculation indicators, and that multi-center prospective studies are needed to verify the external generalizability of the results. Within those constraints, the findings indicate that an integrated, nurse-led model combining electronic alerts, a rapid response team, a standardized checklist, information-platform collaboration, and continuous quality improvement was associated with better bundle adherence and survival in this emergency resuscitation room setting.

Subject of Research: A nurse-led systematic septic shock nursing model and its effects on 1-hour bundle adherence and mortality in an emergency resuscitation room

Article Title: Emergency department nurse-led systematic nursing model for septic shock: impact on 1-hour bundle adherence and mortality

Article References: Wei, S., Li, X., Wang, X., Ke, T., Lin, C., Zhao, Y., & Sun, H. (2026). Emergency department nurse-led systematic nursing model for septic shock: impact on 1-hour bundle adherence and mortality. Frontiers in Medicine, 13, Article 1967565. https://doi.org/10.3389/fmed.2026.1967565

Image Credits: AI Generated

DOI: 10.3389/fmed.2026.1967565

Keywords: septic shock, sepsis, 1-hour bundle, emergency department, nursing model, propensity score matching, Surviving Sepsis Campaign, mortality, SOFA score, Delphi method, rapid response team, quality improvement