Implementing Surviving Sepsis Campaign Guidelines and Mortality of Adult Patients in Intensive Care Units: An Integrative Review ()
1. Introduction
Sepsis is a life-threatening illness. Every year it affects 30 million people globally [1] and 285 per 100,000 in Taiwan [2]. In Intensive Care Units (ICUs), sepsis prevalence ranged from 10.5% to 37% in different countries, including France, Germany, the Netherlands, China, and Thailand [2] [3] [4] [5] [6]. In addition, sepsis and septic shock (SS) together were responsible for 25.4% of ICU admissions during one pilgrim season in Mecca [7].
Sepsis occurs as a dangerous response of the body to infection accompanied by organ dysfunction; it can cause systemic biologic, biochemical, and physiologic abnormalities [8]. Sepsis is characterized by having tachycardia, hypotension, hyperthermia or hypothermia, tachypnea, and leukocytosis or leukopenia. Signs of acute kidney failure, heart failure, and lung dysfunction may start to appear [9]. In cases where sepsis deterioration into SS may occur, patients will have hypotension and poor perfusion, and they will be unresponsive to intravenous fluids administration, making it necessary to administer vasopressors to manage hypotension [8].
Sepsis has negative consequences on health and health-care cost. It was found to be responsible for six million annual deaths around the world, [10] and 250,000 deaths annually in the USA alone [11]. Without proper and systematic management, sepsis can increase the mortality rate, [12] adding an additional $14 billion dollars to the overall global health expenditure [13]. Moreover, patients with sepsis have been found to have a greater risk of recurrent hospitalization [8] [14] and developing secondary infections later on during hospitalizations, such as pneumonia and bloodstream infections [15]. Fortunately, patients’ bad prognosis can be prevented and the mortality rate can be decreased by early management of sepsis and SS [16]. For this reason, nurses must screen patients for sepsis, and sepsis management should be carried out as early as possible when sepsis is suspected [17].
Surviving Sepsis Campaign (SSC) guidelines are intended to provide guidance for the management of sepsis and SS [18]. These guidelines for the management of sepsis and SS were shown to reduce the mortality rate among patients in ICUs, significantly [12] [18] [19]. In Taiwan, Chou et al. (2014) [20] indicated that applying SSC guidelines also decreased the mortality rate from 34.6% to 24.5% among patients in ICU. Pestana et al. (2010) [21] conducted a retrospective study. The study results revealed that applying SSC guidelines decreased the mortality rate among 184 patients in ICUs (p < 0.001).
Nurses have a vital role in screening and early management of sepsis [22]. In Spain, a national educational program for nurses and physicians led to improvement in the care of patients with sepsis, improved compliance with sepsis management, and reduced mortality rate [23]. In addition, a nurse-driven sepsis management protocol can improve compliance with SSC guidelines and enhance early recognition of sepsis [24]. However, many nurses have poorly adhered to guidelines of sepsis diagnoses and management [25]. In any case, sepsis also occurs with other illnesses concurrently, which can be confusing [26].
Searching the different databases, no Integrative Literature Review (ILR) was conducted with the purpose of examining research literature that studied the effect of implementing SSC guidelines on adult patients’ mortality rate in ICUs. The results of this ILR can help in presenting synthesized evidence, in addition to adding more information to the body of literature about the effect of implementing SSC guidelines on patients’ mortality in ICUs. Moreover, having such information can help nurses and other healthcare providers in enhancing management of patients with sepsis and SS. This review will increase awareness of nursing administrators and educators about sepsis management based on SSC guidelines.
2. Method
2.1. Problem Identification
There are increasing research studies investigating the effect of implementing SSC guidelines for the management of sepsis and SS on adult patients’ outcomes. The authors of this ILR identified the need to reach an in-depth understanding of this relationship. The purpose of this ILR was to examine the literature related to the effect of implementing the Surviving Sepsis Campaign (SSC) guidelines for the management of sepsis and SS on adult patients’ mortality rate in ICUs.
The proposed method of Whittemore and Knafl (2005) [27] was used to guide this ILR and to improve its rigor. Whittemore and Knafl (2005) [27] modified the integrative literature review method of Cooper (1998) [28], which was composed of five stages: problem formulation; a search of literature; data evaluation; data analysis; and presentation of findings. This updated methodology for ILR enables the rigor to be enhanced. It also enhances data synthesis and combining studies with different methodologies in order to give a wide perspective on phenomena [27]. For the current ILR, this method was beneficial because studies with different methods were included.
2.2. Literature Search
The literature search was undertaken in November 2018 by using MEDLINE, Cab Direct, ProQuest Central, SpringerLink, CINAHL Plus, the Cochrane Database of Systematic Reviews, Scopus and Google Scholar. The following keywords were used in different combinations to guide the search: “sepsis”, “septic shock”, “surviving sepsis campaign”, “guidelines”, “bundle”, “outcomes”, “death”, and “mortality”. The search was conducted by entering the following words: surviving sepsis campaign AND guidelines AND mortality, surviving sepsis campaign AND guidelines AND death, surviving sepsis campaign AND bundle AND mortality, surviving sepsis campaign AND mortality, sepsis AND bundle AND mortality, septic shock AND bundle AND mortality; surviving sepsis campaign AND septic shock AND mortality. The inclusion criteria for research articles were: 1) Research studies published between 2004 and 2018, because the SSC guidelines for sepsis management were first published in 2004 [29] 2) Studies that investigated the effect of implementing SSC guidelines on adult patients’ mortality in ICUs only, with no restriction for study design; 3) Articles written in English.
2.3. Search Results
The initial search resulted in 729 articles. This search was carried out by the principal investigator. The titles and abstracts were reviewed based on the inclusion criteria to assess articles’ eligibility. A total number of 334 research articles were duplicate, they were counted manually by name and frequency, and then duplicates were excluded. An additional 330 articles were excluded based on title and abstract information. The remaining 65 articles were retrieved as full-text and were assessed again for meeting the inclusion criteria; 36 of them were conducted in settings other than ICUs or in mixed settings, and 13 studies were ineligible reviews. Only 16 studies were eligible and the remaining 49 articles were excluded (see Figure 1).
The primary investigator used a research matrix to extract the required data (see Table 1). From each article, the following data were extracted: study purpose, design, settings, sample size, sampling technique, year of implementing the SSC guidelines, and main findings. In addition, the data extracted from the articles were used again to confirm eligibility of the included articles based on the
![]()
Figure 1. Flow diagram of study selection strategy.
Abbreviations: SSC: Surviving Sepsis Campaign; ICU: Intensive Care Unit; P: Power level or α; OR: Odd Ratio; CI: Confidence Interval; USA: United States of America; CVP: Central Venous Pressure; MAP: Mean Arterial Pressure; N: total sample size; n: sample size of a particular group; ScvO2: Central venous oxygen saturation; mm: millimol; ml: millileter; kg: Kiogram; h: hour; mmol: millimol; L: liter; mmHg: millimetre of mercury.
discussed eligibility criteria. Furthermore, the quality of the extracted data was checked by the second investigator by reading all of the 16 eligible articles and confirming the data of the research matrix. All of the 16 eligible articles are quantitative. Five of them were conducted in Spain, four studies were conducted in the USA, and the remaining seven studies were conducted in Portugal, France, Brazil, Saudi Arabia, Taiwan, the Netherlands, and the Czech Republic, with one study for each of the mentioned countries.
2.4. Data Evaluation, Rigor and Data Quality
The quality of methods used in the included articles was assessed by using criteria for assessing the quality of quantitative studies, which was recommended by Kmet et al. (2014) [30]. The criteria have 14 domains, for which the answers and scoring can be as follows: (Yes = 2); (Partially = 1); (No = 0); and not applicable. The quality score of each article was calculated by summing the total score of items and dividing it by the highest possible total score after removing non-applicable items [30]. The calculated summary score for each article can range between zero and two (see Table 2). This assessment was confirmed by the two other researchers.
2.5. Data Analysis
Based on Whittemore and Knafl’s (2005) [27] method of ILR, analysis has four phases: 1) Data reduction: classifying and dividing data into subgroups; 2) Data display: showing data as they appeared in the research matrix (data extraction sheet), in order to enhance comparison; 3) Data comparison: examining data to identify pattern, relationships, and themes, by which variables can be grouped together and a conceptual map can be drawn; 4) Conclusion drawing and verification: by collecting the different parts that make up the whole general picture and verifying them, followed by synthesizing data and data integration. However, in order to use a more rigorous and well-described process of data analysis, inductive content analysis was used in addition to step number three above, as described by Elo and Kyngäs (2008) [31]. This included: coding, categorizing, collecting categories into higher order headings to decrease the number of headings, and finally abstracting by making a general description of the findings.
3. Results
The reviewed articles revealed that implementing SSC guidelines can reduce mortality rate among adult patients in ICUs [20] [21] [32] - [43]. It was shown that implementing the SSC guidelines reduced the mortality rate among adult patients in ICU from 54.0% to 16.2% (N = 564) [34], from 35.0% to 29.7% (p < 0.001) (N = 48110) [42], and from 56.8% to 36.8% [21]. In addition, implementing the SSC guidelines was associated with improved patient survival [odds ratio (OR), 5.8 (95% CI, 2.2 - 15.1; p < 0.001)] [39] with relative mortality rate declined from 21.2% to 8.7% (p < 0.0001) [40].
![]()
![]()
Table 2. Quality and rigor of the eligible studies.
3.1. SSC Guidelines and In-Hospital Mortality
Implementing SSC guidelines were found to reduce in-hospital mortality rates [20] [40] [41] [42] [43]. Specifically, in-hospital mortality was reduced from 57.3% to 37.5% (p < 0.001) (N = 384) [32], from 44.0% to 32.6% (p < 0.001) (N = 1348) [41], from 61.1% to 20% (p < 0.001) (N = 112) [33], 35.0% to 29.7% (p < 0.001) [42], and from 54.3% to 36.7% [36]. Furthermore, in-hospital mortality was reduced in ICUs by 5.8% over 3.5 years, with a relative in-hospital mortality reduction of 16.7% among adult patients with sepsis and SS in ICUs compared to baseline [43].
3.2. SSC Guidelines, 28-Day Mortality, 30-Day Mortality, and ICU Mortality
Implementing SSC guidelines reduced 28-day mortality among adult patients in ICUs [35] [37] [41], from 40% to 27% (p = 0.02) (N = 538) [37], and from 36.5% to 23.0% (p < 0.001) (N = 1384) [41], with odds ratio (OR) of 0.44 [95% confidence interval (CI) = 0.24 - 0.80] in sepsis and 0.49 (95% CI = 0.25 - 0.95) for association of implementing the SSC guidelines with 28-day mortality [35]. Similarly, 30-day mortality was reduced from 31.3% to 21.1% (p = 0.05) [39]. In addition, implementing the SSC guidelines reduced ICU mortality rates [21] [32] [36], from 48.2% to 27.2% (p < 0.01) [36], from 53.1% to 30.5% (p < 0.001) [32], and from 56.8% to 36.8% (p = 0.036) [21].
3.3. Certain Selected Guidelines and Patient Mortality
Two studies investigated the effect of implementing certain practices selected from the SSC guidelines, such as the administration of antibiotics, which was found to decrease the mortality rate [44] [45]. Specifically, the risk of mortality in the case of administering a broad-spectrum antibiotic during the first hour of sepsis compared with no antibiotic in the first 6 hours had an odds ratio [OR] of 0.67; 95% confidence interval [CI], 0.50 - 0.90; p < 0.01) [45]. In addition, collecting blood cultures and giving vasopressors decreased the mortality rate among adult patients with sepsis and septic shock in ICUs [35]. Castellanos-Ortega et al. (2010) [32] pointed out that an inverse relationship exists between mortality rate and the number of implemented SSC guidelines for the management of septic shock. In more detail, the mortality rate was significantly related to the number of accomplished therapeutic guidelines, with an odds ratio [OR] of 1.64; 95% confidence interval [CI], 1.28 - 2.1 (p < 0.001) [21].
4. Discussion
Common themes that emerged from this ILR were: First, implementing the SSC guidelines can decrease the mortality rate among adult patients with sepsis and SS in ICUs; this corresponds to the findings of Lefrant et al. [37], Leisman et al. [38], Patel et al. [33], Pestana et al. [21], Cardoso et al. [35], Memon et al. [39], Sánchez et al. [41], Thompson et al. [42], van Zanten et al. [43], and Levy et al. [12]. The second emerged theme was that implementing certain selected guidelines, such as collecting blood cultures and administering broad-spectrum antibiotics and vasopressors can also reduce the mortality rate among adult ICU patients, which was consistent with the findings of Uvizl et al. [44] and Ferrer et al. [45]. The calculated scores for rigor of the research articles ranged between 0.60 and 0.86, with a mean average of 0.72 ± 0.078. The distribution of scores on histogram was close to normal distribution with skewness level of 0.26. A cut-point score of 0.75 is conservative, while a cut-point score of 0.55 is liberal [30]. However, the minimum score of the included studies was 0.60 indicating adequate rigor of the included studies; therefore, no studies were excluded based on the calculated scores.
There are some limitations of the reviewed studies. Out of the 16 eligible articles, six research articles were descriptive in nature, while the remaining ten research articles used a quasi-experimental design. However, for the discussed clinical problem, quantitative studies are the type of study expected to investigate the research problem in a suitable way. However, the sample size was small in the observational study of Pestana et al. [21], which may limit its generalizability. The studies of van Zanten et al. [43], Thompson et al. [42], Uvizl et al. [44], Sánchez et al. [41], and Leisman et al. [38], recruited large numbers of participants. Moreover, the studies of Ferrer et al. [45], Castellanos-Ortega et al. [32], Patel et al. [33], Shiramizo et al. [34], Thompson et al. [42], Chou et al. [20], and Herrán-Monge et al. [36] used convenient samples of patients, which might have carried a risk of sampling bias [46]. Furthermore, the studies of Pestana et al. [21], Miller III et al. [40], and Uvizl et al. [44] used a retrospective observational research design, which might not be a robust design to be used for answering such research question.
Some possible confounding variables were not controlled in some studies. For example, the variable of baseline severity of patients’ illness was not measured in the study by Thompson et al. [42]. In addition, around half of the included patients in the study by Pestana et al. [21] were patients with cancer, which might limit the generalizability of the study. The reviewed studies came from a wide variety of countries where healthcare and implementation practices likely differ, and the populations were different also. In addition, the implemented guidelines were belonging to the period between 2004 and 2013, which indicate different updates of the guidelines. These differences might explain some of the variability in the results.
5. Conclusions
The reviewed articles revealed that implementing SSC guidelines for adult patients with sepsis and SS in ICUs has decreased mortality rates. The findings of the current ILR imply that nurses and physicians working in multidisciplinary teams at ICUs are required to implement the guidelines of SSC while providing care for adult patients with sepsis and SS. Educational campaigns and continuous learning programs are needed to teach nurses about the importance of implementing the SSC guidelines and how to implement them. Moreover, internal audit teams can be formed to evaluate nurses’ compliance with the SSC guidelines for managing patients with sepsis and SS in ICUs. Furthermore, the SSC guidelines need to be taught to nursing students to increase their awareness and capability of implementing these guidelines in clinical practice.
All of the eligible studies included in the current ILR were conducted between 2004 and 2013, so studying the effect of implementing the newly released guidelines of SSC 2016 and 2018 for the management of sepsis and SS on adult patients’ mortality is recommended. The findings of this ILR need to be considered by administrators and policymakers in order to integrate the SSC guidelines of sepsis management during providing care for patients in ICUs. Future studies that may seek to investigate the same problem need to take into consideration using more rigorous research designs, such as the use of randomized allocation of participants and blinding techniques.