Annals of the American Thoracic Society

To the Editor:

Severe sepsis is a common problem on the general hospital wards, where 32–50% of all cases receive their care (1). Despite an intense focus on improving the recognition and management of patients with severe sepsis and septic shock in emergency departments (EDs) and intensive care units (ICUs), the Surviving Sepsis Campaign (SSC) recently found that 32% of all patients enrolled in the campaign were initially diagnosed on the wards (2). In their most recent guidelines for the management of patients with severe sepsis and septic shock, the SSC stated that these recommendations could be applied to all patients, regardless of their location of care, specifically including those patients who receive all or part of their care on the general hospital wards (3).

Patients with severe sepsis on the wards may differ from those in the ED and ICU in important ways that, in turn, may alter their response to treatment. First, there is an intense selection bias imparted by the specialized care provided in ICUs, with many patients needing circulatory or respiratory support requiring ICU admission. Second, the specific infections leading to sepsis differ on the wards when compared with the ICU (1, 4). Third, patients with sepsis on the wards have different rates and distribution of organ failure (1, 4). The widespread improvement efforts of the SSC have helped to reduce overall sepsis mortality (2), regardless of the initial setting of care. However, patients initially diagnosed on the wards face higher mortality rates (2), while there remains a large fraction of patients with severe sepsis who are unrecognized by treating clinicians (5), many of whom likely receive their care on the wards.

Despite clear recommendations that the SSC Guidelines for Management of Severe Sepsis and Septic Shock be applied to patients on the wards (3), the strength of evidence supporting these recommendations for this heterogeneous population remains unknown. Therefore, we sought to systematically assess the supporting evidence cited in the SSC’s 2012 guidelines for the inclusion of adult wards patients. We defined wards patients as those inpatients who were diagnosed with severe sepsis or septic shock but did not receive care in an ICU during that hospitalization; sepsis care for wards patients may have been initiated in the ED or as inpatients.


Two authors (S.G. and J.S.) reviewed SSC 2012 Recommendations A–E and G, the core wards-relevant recommendations on initial resuscitation (A), screening for sepsis and performance improvement (B), diagnosis (C), antimicrobial therapy (D), source control (E), and fluid therapy for severe sepsis (G), using a structured data abstraction tool. The remaining recommendations were excluded, as they either generally applied only to care delivered in the ED or ICU or were recommendations thought to have less effect on immediate outcomes (see Table E1 in the online supplement). Pediatric recommendations and references were not reviewed.

Every article cited within the text was reviewed for the following data: study design, total number of patients, site of recruitment, and number of wards patients who could be identified. We defined a wards patients as a population specifically identified as such by the study (i.e., cared for outside of the ED or the ICU) or a population subdivided into a non-critically ill cohort by investigators. If a patient ever required ICU-level care, or if we could not determine the clinical status of a study’s population, the patients were categorized as critically ill. All randomized controlled trials (RCTs) not explicitly cited themselves but included in any meta-analysis cited by the guidelines were also specifically assessed for the same information.


A total of 122 studies from 25 specific recommendations were explicitly evaluated (Table 1). No RCT cited in the SSC 2012 Recommendations A–E and G specifically included wards patients; five such trials included in cited meta-analyses did, however, totaling 1,129 wards patients. All five trials were in support of recommendation D-4b (duration of empiric antibiotics). The remaining 24 recommendations relevant for the care of wards patients with severe sepsis or septic shock did not include any wards patients in the RCTs in the supporting citations. Specifically, there were no wards patients in any study, randomized or observational, supporting the initial fluid resuscitation of patients with severe sepsis or septic shock. The administration of intravenous antibiotics within 1 hour for patients with severe sepsis and septic shock for patients on the wards was supported by only one observational study. After including observational studies and adjusting for duplicate citations, 82,514 total patients support the SSC recommendations for wards patients, of whom only 9,290 (11%) actually received their care on the wards, irrespective of study type.

Table 1. Surviving sepsis campaign recommendations stratified by the presence of wards patients in the supportive evidence

No Wards Patients in Any Supporting Study (n = wards patients/total patients in cited studies)Only in Observational Studies (n = wards patients/total patients in cited studies)Randomized Controlled Trials and Observational Studies (n = randomized controlled trial wards patients/observational study wards patients/total patients)
Early goal-directed therapy with normalized lactate (A-2) (n = 0/648)Early goal-directed therapy with goals (A-1) (n = 5,258/35,121)Empiric combination antibiotics duration (D-4b) (n = 1,129/1,627/11,852)
Cultures drawn before antibiotics (C-1) (n = 0/1,017)Screen for severe sepsis (B-1) (n = 5,228/16,916) 
Imaging to identify infectious source (C-3) (n = 0/0)Quality improvement for sepsis (B-2) (n = 30/3,039) 
Reassessing antibiotic regimen (D-2b) (n = 0/0)Invasive candidiasis assays (C-2) (n = 120/599) 
Empiric combination antibiotics initiation (D-4a) (n = 0/551)IV Antibiotics within the first hour (D-1) (n = 5,296/22,218) 
Duration of antibiotic therapy (D-5) (n = 0/0)Activity and penetration of antibiotics (D-2) (n = 1,928/5,726) 
No antibiotics in the systemic inflammatory response syndrome without infection (D-7) (n = 0/0)Biomarkers (D-3) (n = 0/3,297) 
Source control (E-1) (n = 0/201)Antivirals (D-6) (n = 205/371) 
Pancreatic necrosis (E-2) (n = 0/124)  
Intervention with least physiologic insult (E-3) (n = 0/0)  
Intravascular device (E-4) (n = 0/0)  
Crystalloids (G-1) (n = 0/0)  
No hydroxyethyl starch (G-2) (n = 0/8,679)  
Albumin (G-3) (n = 0/6,997)  
Initial fluid challenge (G-4) (n = 0/0)  
Hemodynamic improvement (G-5) (n = 0/0)  

Despite severe sepsis and septic shock among patients on the wards being common (1, 2), mortal (1, 2, 6, 7), and underrecognized (4, 5), strikingly little of the supporting evidence guiding the care of wards patients included any evidence specific to this population. We found that of the 25 specific recommendations purported to apply to patients on the wards, only the duration of empiric antibiotics included any wards patients in a prospective clinical trial. There were no wards patients in any type of trial supporting the aggressive fluid resuscitation that is often a cornerstone of the care of patients with severe sepsis and septic shock.

The SSC has recently published evidence that high compliance with the SSC Resuscitation Bundle is associated with lower mortality among patients diagnosed with severe sepsis or septic shock on the wards (2). Although these results make an argument that the current recommendations may improve outcomes on the wards, the SSC collaborative did not, until recently, focus their efforts on the recognition and management of this population.

Furthermore, because of the differences in mortality rates, phase of illness, infections, and organ dysfunctions, we cannot assume that the recommendations will work for all patients in this heterogeneous group (8, 9). For example, optimal fluid resuscitation strategies may differ in patients on the wards because of differences in prevalence of comorbidities such as renal failure or diastolic heart failure that may affect response to fluids; the likely differences in incidence of pulmonary capillary leak may also lead to differences in tolerance of large fluid volumes. Likewise, such wards patients are at lower overall risk for death, leading to systematic differences in the risk–benefit trade-offs of more aggressive therapies, potentially including invasive lines (8, 9). Because of this large evidence gap for the management of patients with severe sepsis and septic shock on the wards, prospective, randomized controlled trials are urgently needed to ensure we do not unwittingly do harm to this large, vulnerable population.

1 . Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:13031310.
2 . Levy MM, Rhodes A, Phillips GS, Townsend SR, Schorr CA, Beale R, Osborn T, Lemeshow S, Chiche JD, Artigas A, et al. Surviving sepsis campaign: association between performance metrics and outcomes in a 7.5-year study. Crit Care Med 2015;43:312.
3 . Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, Sevransky JE, Sprung CL, Douglas IS, Jaeschke R, et al.; Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;41:580637.
4 . Rohde JM, Odden AJ, Bonham C, Kuhn L, Malani PN, Chen LM, Flanders SA, Iwashyna TJ. The epidemiology of acute organ system dysfunction from severe sepsis outside of the intensive care unit. J Hosp Med 2013;8:243247.
5 . Liu V, Escobar GJ, Greene JD, Soule J, Whippy A, Angus DC, Iwashyna TJ. Hospital deaths in patients with sepsis from 2 independent cohorts. JAMA 2014;312:9092.
6 . Sundararajan V, Macisaac CM, Presneill JJ, Cade JF, Visvanathan K. Epidemiology of sepsis in Victoria, Australia. Crit Care Med 2005;33:7180.
7 . Esteban A, Frutos-Vivar F, Ferguson ND, Peñuelas O, Lorente JA, Gordo F, Honrubia T, Algora A, Bustos A, García G, et al. Sepsis incidence and outcome: contrasting the intensive care unit with the hospital ward. Crit Care Med 2007;35:12841289.
8 . Hayward RA, Kent DM, Vijan S, Hofer TP. Reporting clinical trial results to inform providers, payers, and consumers. Health Aff (Millwood) 2005;24:15711581.
9 . Kent DM, Rothwell PM, Ioannidis JP, Altman DG, Hayward RA. Assessing and reporting heterogeneity in treatment effects in clinical trials: a proposal. Trials 2010;11:85.

This work was supported by the U.S. Department of Veterans Affairs Health Services Research and Development Services: IIR 11-109 (to T.J.I.).

Author Contributions: S.G. and J.S. performed the primary data extraction, and A.J.O., S.G., J.S., and T.J.I. contributed substantially to the study design, to data analysis and interpretation, and to the writing of the manuscript.

This letter has an online supplement, which is accessible from this issue's table of contents at

Author disclosures are available with the text of this letter at


No related items
Comments Post a Comment

New User Registration

Not Yet Registered?
Benefits of Registration Include:
 •  A Unique User Profile that will allow you to manage your current subscriptions (including online access)
 •  The ability to create favorites lists down to the article level
 •  The ability to customize email alerts to receive specific notifications about the topics you care most about and special offers
Annals of the American Thoracic Society

Click to see any corrections or updates and to confirm this is the authentic version of record