Medical Center

Skip to content. | Skip to navigation

Personal tools

Home > Clinical Nutrition Services > Inpatient Clinical Nutrition Services > Digestive Health > E-journal Club > September08.html


University of Virginia Health System

Nutrition Support E-Journal Club

September 2008




We enjoyed a brief hiatus from our e-journal club in July and August due to teaching opportunities with new residents, new GI fellows and even some vacation fun.  We are back on schedule and had a great September traineeship with trainees from Wilmington, De and Salinas, CA. 


April Citation 1: 

Elke G, Schädler D, Engel C, et al.  German Competence Network Sepsis (SepNet).  Current practice in nutritional support and its association with mortality in septic patients--results from a national, prospective, multicenter study.  Crit Care Med. 2008;36(6):1762-1767.


This was a prospective, observational study carried out as part of the German Competence Network Sepsis (SepNet).  Data was collected from 454 ICUs in 310 hospitals that were visited in a random manner throughout a year.  A cross-section of patient data was obtained by the collection of 24-hours of data from each hospital.

In total, 3,877 patients were screened with 415 patients found to have severe sepsis or septic shock, and data was evaluated on 399 patients with sepsis or septic shock that had complete nutrition information.  Hospitals were stratified by size:

  • ¨ University hospitals (n = 10, 3.2%),
  • ¨ University-affiliated hospitals (n = 106, 34.2%),
  • ¨ General hospitals (n = 173, 55.8%)
  • ¨ Others (n = 21, 6.8%).

Information collected included demographic data (age, gender, weight, height), main diagnoses, co-morbidities, and severity of illness (Acute Physiology and Chronic Health Evaluation [APACHE] II, Sepsis-related Organ Failure Assessment [SOFA] score) as well as diagnostic and therapeutic measures, including nutrition. Total length of ICU and hospital stay (LOS) as well as hospital mortality was assessed after 3 months. 

The investigators conducted multivariate analysis to identify factors associated with receiving exclusive enteral nutrition (EN) and to control for factors such as sepsis, severity of illness, hospital size, length of ICU stay, and clinical diagnosis expected to influence the route of nutrition support.  In addition, multivariate analysis was used to analyze the association of the type of nutrition (enteral, parenteral or combined) with mortality.

Inclusion and Exclusion Criteria were:

Inclusion criteria were adult patients with sepsis or septic shock that had complete nutrition information for the 24 hour data period for that hospital.

Exclusion criteria were patients < 18 years, those without sepsis and those who did not have complete illness or nutrition data available.

Major Results reported by authors:

Overall, 20.1% of patients received only EN, 35.1% received only parenteral nutrition (PN), 34.6% received mixed PN + EN (MIX) and 10.3% of the patients received no nutritional support. 

Patients with mechanical ventilation (OR, 0.48; 95% CI, 0.26-0.91), a GI/intra-abdominal disease (OR, 0.24; 95% CI, 0.11-0.53), and the presence of septic shock (OR, 0.31; 95% CI, 0.16-0.61) were less likely to receive exclusively EN.  In those patients without any GI/intra-abdominal disease, exclusive EN was used only 25.9% of the time.  Mean glucose concentration was not significantly different between the groups, although the MIX group required significantly increased insulin dose (p = 0.0002).

Hospital mortality was significantly higher in patients receiving exclusively PN (62.3%) or MIX (57.1%) vs those exclusively on EN (38.9%) (p = 0.005).  Multivariate analysis revealed that the presence of PN was a significant independent predictor for mortality (OR, 2.09; 95% CI, 1.29-3.37).  The APACHE II score (OR, 1.05; 95% CI, 1.02-1.09) and the presence of renal dysfunction (OR, 2.07; 95% CI, 1.30-3.31) were also found to be significant predictors for hospital mortality.

Author's Conclusions:

The authors concluded that their survey reflected a high rate of PN in patients with severe sepsis or septic shock in German ICUs, and that PN was associated with increased mortality. They also recommended randomized controlled trials on the benefits and risks of nutrition support in the septic population, and suggested that EN could utilized in a greater percentage of this population.


This was an ambitious study, which collected data from a large number of hospitals and patients.  Although 24 hours of data collection may not be representative of practice at a single hospital, random visits to a large number of centers over a year does should provide a reasonable cross-section of practice in a broader sense.  The primary limitation of this study is that, as an observational study, it can only described associations.  Regardless of statistical controls these associations should not be used to imply cause and effect, but should only allow formation of theories that need to be tested in randomized studies.  Other limitations of this study, which the authors outline in the discussion section were: 1) a lack of information on timing of nutrition, 2) the amounts of nutrition provided, and 3) the prior nutrition status of patients.

One factor that distinguishes this survey from most done previously, was the relatively tight glucose control in all groups.  The fact that PN is associated with increased mortality, even in the setting of tight glucose control suggests that PN could play a role in increased mortality.  However, as we have already said, there is no way to completely account for the selection bias that is inevitable in an observational study of this nature, and randomized studies are needed.  Until these randomized studies are conducted, it is unlikely that this is the final chapter in the PN versus EN controversy.

The group felt that the primary benefit of this survey was to draw attention to the current practice that many patients that could, or should, receive EN actually still receive PN.  At the very least, the use of PN in patients that could be nourished with EN is not cost effective, and justifies the need for education of clinical staff, and implementation of protocols that improve the use and delivery of EN.

Our Take Home message:

PN is associated with increased mortality in German ICU's, even in the setting of good glucose control.  Randomized studies are required before any cause-and effect conclusions can be formed regarding this association. 

Other News:

Our next Weekend Warrior 2 day mini-traineeship program is scheduled for Saturday and Sunday, March 7th & 8th, 2009.   If you know anyone who might not be able to get away for our full week traineeship, please let them know about our weekend program-there is one scholarship available to cover tuition also; contact Stacey McCray for details: sf8n@hscmail.mcc.virginia.edu.

Also, coming soon...Nutrition Support webinars!  Keep checking the website for more information or contact Stacey McCray (sf8n@hscmail.mcc.virginia.edu) if you would like to be contacted when more information is available.

Check out the latest Practical Gastroenterology articles/info at:


•1)      Chan LN. Opioid Analgesics and the Gastrointestinal Tract.  Practical Gastroenterology 2008; XXXII(8):37.

•2)      Noakes TD. Water Intoxication-Considerations for Patients, Athletes and Physicians.  Practical Gastroenterology 2008; XXXII(9):46.


Joe Krenitsky MS, RD

Carol Parrish RD, MS

PS - Please feel free to forward this on to friends and colleagues.