Introduction

Severe sepsis and septic shock are associated with a high mortality and cost, affecting approximately 750,000 Americans annually [1, 2]. An estimated 500,000 patients with severe sepsis are treated annually in EDs in the United States [3]. Time is one of the key factors in determining the outcome of patients with sepsis, evidence showing that early identification and management of septic shock can significantly reduce mortality [4–9]. The Surviving Sepsis Campaign recommends accomplishing the resuscitation bundle within 6 h from the onset of septic shock, and the application of adjunctive treatments within 24 h [10]. The early resuscitation bundle, including early goal-directed therapy (EGDT), improves outcomes in patients with septic shock [11–14]. However, few data are available on the impact of late completion of bundle elements, i.e., after 6 h, on outcomes [15, 16]. Furthermore, no studies have been done on the impact of delayed compliance on outcomes when bundle resuscitation is completed in more than 6 but fewer than 12 h from the time of a septic shock diagnosis, especially in ED populations. The aim of this study was to assess whether failure to comply with the resuscitation bundle within 6 h but achieving it by 12 h after the diagnosis of septic shock was related to an improvement in 28-day mortality.

Methods

This was a prospective observational study analyzing all consecutive adult (18 years of age or older) patients with septic shock who were treated with protocol-driven resuscitation bundle therapy including EGDT in the ED of the Asan Medical Center, a 2,800-bed, university-affiliated, tertiary referral hospital center, in Seoul, Korea between January 2010 and July 2012. This study was reviewed and approved by the Ethics Committee of Asan Medical Center.

We included patients in the ED, who met two or more criteria for systemic inflammatory response and who had refractory hypotension or a serum lactate level ≥4 mmol/L. We defined refractory hypotension as a systolic blood pressure <90 mmHg or mean arterial pressure (MAP) <60 mmHg requiring vasopressors even after an intravenous fluid challenge (20 mL/kg over 30 min) [17]. All patients with septic shock were treated with protocol-driven resuscitation bundle therapy, including EGDT, while in the ED [10]. Patients were excluded from the study if they had one of the following: pregnancy, absolute contraindication for a central venous catheter, trauma, ‘do not resuscitate’ status, refusal of invasive therapy, patients who transferred to other hospitals, and central venous oxygenation (ScvO2) not checked at 12 h. The treatment protocol was a sepsis management program called “ED Shock Management,” which was organized and implemented for treatment of severe sepsis and shock at the hospital beginning in 2007. Team members met monthly to review current patient data and to discuss ways to improve patient care. Treatment was carried out according to the standard bundle protocol for sepsis resuscitation: (1) serum lactate measurement made as soon as possible from the time of the onset of severe sepsis; (2) blood culture before antibiotic administration; (3) delivery of an initial minimum volume of 20 mL/kg crystalloid (or colloid); (4) achievement and maintenance of MAP ≥65 mmHg; (5) achievement of central venous pressure (CVP) ≥8 mmHg; and (6) achievement of ScvO2 ≥70 % [17]. Outcomes were evaluated in three patient groups. The early-compliance group consisted of patients who received all six bundle elements within 6 h of diagnosis (Time 0) in the ED. The delayed-compliance group included patients who received all bundle elements within 12 h of Time 0. The non-compliance group consisted of patients who did not achieve all bundle elements within 12 h, including a MAP ≥65 mmHg, CVP ≥8 mmHg, and ScvO2 ≥70 % (Fig. 1). The primary outcome was 28-day mortality following compliance with the sepsis resuscitation bundle.

Fig. 1
figure 1

Flow diagram of study selection process. ED emergency department, ScvO 2 central venous oxygenation, SOFA sequential organ failure assessment. *P value = 0.04

Full size image

Demographic and clinical data, including age, gender, symptoms, previous medical history, initial vital signs, blood results, 28-day course, and diagnosis on admission, were collected. ScvO2 was checked at initial diagnosis, 6, and 12 h. The sequential organ failure assessment (SOFA) score was calculated at ED admission.

Statistical analysis

Continuous variables were expressed as means with standard deviations or medians with range if the assumption of a normal distribution was violated. Categorical variables were expressed as numbers and percentages. Differences between means were assessed for significance using Student’s t test, the Wilcoxon rank sum test, or ANOVA. Chi Squared or Fisher’s exact test was used to compare group differences for categorical variables. Multiple logistic regression models were used to evaluate the association of delayed compliance with 28-day mortality. Data were presented as odds ratios (OR) with 95 % confidence intervals (CI). A two-sided p value <0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS for Windows version 18.0 (SPSS Inc., Chicago, IL, USA).

Results

A total of 426 adult patients were diagnosed with septic shock in the ED during the study period. Of these, we excluded 39 patients who transferred to other hospitals, 32 patients with a consent of do-not-resuscitate, 19 patients who had insufficient data, and 4 patients who refused to have invasive procedures, leaving a total of 332 patients for analysis.

The mean age of the study cohort was 63.9 years; 192 patients (57.8 %) were men. The 6-h compliance with resuscitation bundle elements was 100 % for serum lactate measurement, blood culture before antibiotic administration, and delivery of an initial minimum volume of 20 mL/kg crystalloid (or colloid). Achievement and maintenance of MAP ≥65 mmHg were seen in 319 patients (96.1 %); 287 patients (86.4 %) achieved CVP ≥8 mmHg; and 212 patients (63.9 %) had a ScvO2 ≥70 %.

When the study cohort was evaluated for compliance with all six bundle therapy elements, early compliance (≤6 h) was achieved in 195 patients (58.7 %) and delayed compliance (≤12 h) in 59 patients (19.8 %). Non-compliance (>12 h) was observed in 78 patients (23.5 %). Overall, 275 patients survived and 57 patients died, for a 28-day mortality rate of 17.2 %. There were no statistically significant differences in gender, previous medical history, baseline vital signs, and laboratory test results between these groups (Table 1). The non-compliance group had a significantly higher median age (69.5 years) than the delayed-compliance (64.0 years) and early-compliance (65.0 years) groups (both p = 0.01). The between-group differences in variables of illness severity mostly had no significant differences (Table 2). However, the non-compliance group received a higher dose of norepinephrine treatment, more days of mechanical ventilator therapy, and renal replacement therapy than the delayed-compliance and early-compliance groups. Furthermore, the non-compliance group had a significantly higher 28-day mortality of 29.5 % than the delayed-compliance or early-compliance group (13.6 and 13.3 %, respectively, p = 0.04). Considering that there were no differences in baseline illness severity, delayed compliance decreased the 28-day mortality compared to non-compliance (OR 0.32, 95 % CI 0.13–0.82, p = 0.02).

Table 1 Demographic and baseline characteristics of patients with septic shock
Full size table
Table 2 Illness severity and outcomes of patients with septic shock
Full size table

Discussion

This study evaluated 28-day mortality in ED patients who achieved resuscitation bundle compliance within 6 h of septic shock diagnosis, more than 6 but fewer than 12 h, and patients who did not achieve compliance within 12 h. Our findings show that delayed compliance with the resuscitation bundle in patients diagnosed with septic shock in the ED has a significantly lower 28-day mortality (13.6 %) than non-compliance (29.5 %), and a better overall mortality (OR 0.32, 95 % CI 0.13–0.82, p = 0.02).

In 2001, Rivers et al. [4] reported that EGDT for severe sepsis and septic shock reduced mortality when begun in the ED before intensive care unit admission. Previous studies that assessed performance of the resuscitation bundle [9, 11–13] primarily addressed outcomes when compliance was achieved within the first 6 h of septic shock presentation. Consequently, the impact on mortality of those patients who completed the protocol beyond that time is unknown. Our results provide additional data on the resuscitation bundle for preventing septic shock-associated mortality. The data indicate that the period to complete the protocol may be extended beyond the time limit proposed by the Surviving Sepsis Campaign. However, treatment should start at the time of shock presentation, with a goal of achieving compliance as early as possible.

In our results, compliance was achieved in 58.7 % of patients within the initial 6-h target. This is much higher than the 31.3 % reported by the international guideline-based performance improvement results of the Surviving Sepsis Campaign [8], and could help account for the low 17.2 % mortality in this study. Furthermore, we studied a cohort including only ED patients, rather than patients presenting from a variety of settings, and likely receiving various methods of initial resuscitation. Our overall mortality rate was consistent with other studies of ED patients receiving bundle resuscitation [11–13]. In this study, delayed compliance until 12 h was seen in 23.5 % of patients. As such, failure to comply might be the failure to reach the target despite the clinician’s attempt because all of the patients in the current study received the same early recognition and aggressive treatment protocol. The failure to achieve a target may be indicative of greater severity, so compliance with the attempt alone may produce the false impression that compliance is associated with reduced mortality. Therefore, attention to adjustment for severity of patient illness at the time of enrollment should be taken into account. The difference in baseline SOFA illness severity between delayed compliance and non-compliant patients was not statistically significant (19.8 vs. 23.5 %, p = 0.28), also in the comorbidities, lactic acid level, PaO2/FiO2 ratio, decreased urine output, and duration of vasopressor use. However, age could have affected the greater mortality in the non-compliance group. Taking this into consideration, delayed compliance still has the benefit of decreased 28-day mortality compared to non-compliance as defined here.

To the best of our knowledge, results regarding the impact on mortality of resuscitation bundle completion within 12 h in an ED population have not been reported previously. In a recent study, Coba et al. [16] evaluated the impact of quality improvement of sepsis bundle therapy on mortality, and report the effectiveness of therapy when completed within 18 h. Compliance at 18 h has a hospital mortality of 37.1 %, which is 10.2 % lower than non-compliance at 18 h (p < 0.03). They conclude that when bundle completion is extended from 6 to 18 h, the mortality reduction remains significant. Although their study included all intensive care unit patients, and early compliance with the 6 h bundle target was only 12.9 %, their results are consistent with our findings that late is better than never. However, importantly, we do not suggest extending the time limit of achievement of bundle target to 12 h, but that delayed compliance might be an extra benefit of the improvement in the process of sepsis care compared to what is achieved when sepsis recognition and application of the resuscitation bundle are early and aggressive.

This study has several limitations that should be considered. We did not assess the timing of antibiotic administration, which may affect mortality and the achievement of management bundle, or other treatments such as low-dose steroid administration, or glucose control. This study is from a single institution, which limits the generalization of the findings to other institutions or patient populations. In addition, the sample size is relatively small.

In conclusion, if sepsis bundle therapy is started at the time of presentation, the outcome of delayed resuscitation bundle compliance within 12 h is the same as that of early resuscitation bundle compliance within 6 h, and these are better than that of the patients in whom bundle compliance goals cannot be met.