Aim :
To compare measured REE in people with HIV / AIDS and in controls

• as percentage difference from the HBE estimate
• per kilogram of FFM

Methods :
HIV positive patients and control subjects were recruited from two clinics with diverse patient characteristics. Weight, height and REE, (using Deltrac II, DATEX, Helsinki, Finland), were measured. FFM was measured in a subset on subjects, (N=21), and in controls (N=21) by multifrequency bioelectrical impedance (SEAC SFBIM3, Uniquest, Queensland). All patients were measured at rest after an overnight fast.

Analysis :
Student's t Tests for independent samples were used to compare groups.

Results :
91 male HIV positive subjects and 21 male controls participated. All HIV positive subjects were outpatients and there was no evidence of active opportunistic infections at the time of the study.

Conclusion :

• HIV positive subjects did not have elevated REE when measured REE was compared to the HBE.
• Measured REE was higher in HIV positive subjects than controls both as a percentage of the HBE and per kilogram FFM.

Implications :

• When using the HBE to estimate REE there is no need to use an additional factor for HIV positive subjects.
• Ideally measures of FFM should be incorporated into REE assessments for people with HIV / AIDS.

The decision to commence and cease enteral nutrition in the intensive care environment is often left to the discretion of individual practitioners. Anecdotal eveidence suggests that this practice may lead to patients receiving insufficient nutrition.

In 1997, Monash Medical Centre Intensive Care Unit (ICU) decided to adapt an enteral nutrition algorithm initially developed at Guy's Hospital, London, in 1992. The algorithmis a flow chart that standardises the practice of enteral nutrition delivery. Baseline patient data were gathered from records of 59 patients examining how soon patients were fed, the amount of enteral nutrition received, the frequency of manual gastric aspiration to assess tolerance and whether the feed was ceased and why. Following the introduction of the algorithm, 32 patients were reviewed in the same manner. In addition, a cost comparison of enteral and parenteral nutrition pre- and post- algorithm was measured.

Preliminary analysis indicated patients post-algorithm

• received approximately 34% more enteral nutrition (corrected for energy density and volume).
• commenced enteral feeding earlier :100% had commenced by day 3 compared to 55% by day 3 in the pre-algorithm group.
• experienced fewer interruptions to enteral feeding : 6% (n=2) compared with 58% (n=34) in the pre-algorithm group.
• achieved their target rate of enteral feeding : 94% (n=30) compared to 53% (n=31) in the pre-algorithm group.
• received regular aspiration of gastric contents : 97% (n=31) compared to 47% (n=27) in the pre-algorithm group.

Post algorithm, the average cost per month of enteral nutrition in the unit increased by $369, whislst the cost of parenteral nutrition decreased by $1284. This resulted in a net saving of $915 per month (10,980 per annum) in nutrition support.

This innovative enteral nutrition algorithm is a useful tool in the Monash Medical Centre ICU. Further studies should be directed at measurement of patient outcomes and refining the algorithm.

APPROACHING NUTRITION RESEARCH Ralf G. Heine MD FRACP - Paediatric Gastroenterologist Royal Children's Hospital, Melbourne

Evidence-based medicine

Clinical practice should be based on facts. This is not always as easy as it may seem: most treatment modalities have evolved over many years based on clinical experience and theoretical considerations, and well-designed studies have often not been done. In recent years, more emphasis has been placed on "evidence-based medicine". For those involved in research this requires a thorough understanding of research methods and study design. Those involved in clinical practice need to be able to access studies relevant to their field and critically evaluate the scientific merit of research.

Accessing relevant publications

There is a deluge of medical and allied health publications each year, and it is imperative that we are very selective about what we read. Powerful search engines such as Medline have made it very convenient to find relevant publications based on key words.

The scientific merit of studies depends on their design. Observational studies and case reports have the lowest scientific merit, and randomised controlled trials are the most relevant studies. Well performed epidemiological surveys range somewhere in between.

Critical review of medical publications

When reading a publication it is important not to just skim through the abstract and have a look at figures and the conclusion. In order to be able to critically evaluate the scientific merit of a study one needs to understand the study design. Relevant questions include:

• How were patients selected?
• What were the exclusion criteria?
• Retrospective or prospective study?
• Controlled or uncontrolled intervention?
• Randomised or non-randomised?

Bias

Information in a publication may be biased. Bias is not always easy to recognise and is difficult to avoid in study design. A study that recruits only hospital in-patients may have a bias towards more severe disease (selection bias). A study conducted in one particular suburb may be biased towards a specific socio-economic group. There may be bias in how information is collected (information bias), e.g. differences between observers or laboratory methods in collaborating centres.

p values

Every study tries to answer a fundamental question, the so-called "null hypothesis". An example of a null hypothesis would be "There is no association between malnutrition and death from measles". The null hypothesis is tested against the "alternative hypothesis" which is the opposed statement "There is an association between the two". The study will either confirm or reject the null hypothesis. The probability (p) that the null hypothesis is true is expressed as the p value. A observation is usually considered significant if the probability that it has occurred by chance is less than 5% (p < 0.05).

Type 1 error

A study may falsely suggest a significant difference between two groups. This is of particular concern in studies with multiple comparisons. A difference may have occurred by chance (1 in 20). This phenomenon is called Type 1 error.

Type 2 error

If a study is unable to detect a significant difference between two groups despite a real difference a Type 2 has occurred. This is usually due to inadequate sample size or insensitive diagnostic tools or outcome measures.

Study design

Designing a study begins with an idea for a project. The topic then needs to be researched extensively by reviewing the current literature. The research ("null") hypothesis should be kept simple. It is a common mistake to try answer too many questions in one study. The most suitable study type should then be chosen, e.g. a randomised placebo-controlled study for a treatment intervention. The study population and exclusion criteria need to be carefully defined in an attempt to avoid selection bias. Finally, outcome measures are defined.

Sample size analysis

Before embarking on a study the sample size and statistical power should be assessed. This usually involves the help of a statistician. Sample size depends on the prevalence of a condition in the study population, the expected change in outcome measure and the overall power of the study. Generally, a power of 80% is sufficient. Large numbers of subjects are required if a condition is rare or if the predicted change in outcome measure is small. The following example illustrates this relationship.

Example of a Sample Size Calculation

Number of subjects required for a study (80% power) with an expected reduction in outcome measure by 50% or 25%

Diagnostic tests

A diagnostic test should be both sensitive and specific in detecting a given condition. Specimens should be easily obtained. A test should also be reliable: repeated measurements of the same specimen should give the same result and be independent of the operator (intra- and inter- observer reliability). Ideally, the price of a diagnostic test method should be low.

Sensitivity and specificity

A sensitive test detects all subjects with a certain condition ( "always positive in disease"). Poor sensitivity results in false-negatives (patients with the disease are missed ).

A specific test should exclude a condition in a healthy subject ( "always negative in health"). Poor specificity results in false-positives (healthy subjects are incorrectly identified as having the disease).

Example: Antigliadin antibody (AGA) in 100 subjects with suspected coeliac disease

Placebo

Placebo is an inactive drug that cannot be distinguished from active medication by its appearance or taste. Allocation to either placebo or active medication follows a randomisation code (with assistance from a statistician). The effectiveness of new treatments is often compared against placebo. In the case of an existing effective treatment it is more meaningful to compare the new treatment against the current best treatment rather than against placebo. In order to increase the statistical power, cross-over studies can compare the effect of two interventions within the same subject (i.e. drug A O wash-out period O drug B; order of drugs randomised).

Controls

Double-blind, placebo-controlled randomised studies are the gold standard of interventional studies. The strength of this study design lies in its ability to distinguish between spontaneous improvement and improvement due to the intervention. It also eliminates the placebo effect (improvement due to patient's expectation of a positive effect).

Example:

This graph illustrates the response to formula change in infants with persistent crying. Formula was changed at 6 weeks of age. There was a significant reduction in crying over the following 4 weeks (p< 0.001). However, in control infants there was a similar reduction in crying (median), suggesting that formula change was no better than the spontaneous resolution of infant distress.

Ethics Committee in Human or Animal Research

All studies involving human or animal subjects need to be approved by an ethics committee. Most research institutions have their own ethics committee. The project will be assessed for its scientific merit, the risk and benefit of procedures, and whether the project can be achieved. They will ensure that informed consent is obtained and confidentiality of data is maintained. Most peer-reviewed journals request proof of ethics committee approval for publication of a study.

Summary

Success of a research project depends on its careful design and preparation. It requires

• detailed literature review
• formulation of a succinct research hypothesis
• definition of the study population
• selection of suitable diagnostic tests and outcome measures
• statistical advice (incl. sample size analysis)
• ethics committee approval for interventional studies
• development of data collection sheets and documentation of results

Competitive funding and publication in a peer-reviewed journal will depend greatly on proper study design and the overall scientific merit of the project.

SOME INFLAMMATORY AND RADICAL NOTES FROM THE BIOCHEMICAL JOURNALS

C Reactive Protein, (CRP), is a reliable marker of inflammation that is rapidly earning a reputation for itself as a cardiac risk factor. This is due to the combined effects of more reliable and sensitive assays which have now been adapted to near patient testing, NPT, strips systems such as the one manufactured by Nycomed Pharma. CRP is produced in the liver in response to cytokines released by injured tissue and by activated leukocytes. The hypothesis at the moment is that even the presence of small plaques on the walls of the coronary arteries are sufficiently potent stimulators of cytokine release to cause increase in CRP production. The NPT system is able to detect CRP levels between 10 and 50 mg/L to the nearest 5 mg/L, between 50 and 100 mg/L to the nearest 10 mg/L and between 100 mg/L and 225 mg/L to the nearest 25 mg/L. This dynamic range covers what we might expect to see in uncomplicated post surgical patients where the peak CRP concentrations of 25 - 35 mg/L occur between 2 and 3 days post operation. Patients with viral infections have CRP levels of 20 - 50 mg/L and patients with bacterial infections result in levels greater than 50 mg/L. In patients with abdominal sepsis levels in the range 10 - 27 mg/L can be seen. So for the long term TPN and home TPN patients perhaps this NPT CRP test offers the opportunity to investigate those disconcerting 'temperature spikes' immediately and anticipate catheter associated sepsis problems earlier.

Interleukin-6, IL-6, is a potent proimflammatory cytokine which is often mentioned in the nutrition literature. Gudmundsson et al have reported the you will get misleading results if you rely upon measurements taken via an indwelling venous catheter. This is because the catheter itself causes a localised inflammatory response leading to steadily increasing IL-6 levels in the samples taken from that catheter. Their suggestion is that you should sample for IL-6 via discrete venepunctures each taken from a progressively more distal site.

The capacity to resist oxidative stress can be assessed via a number of biochemical tests but some are better than others. Covas et al have performed studies of the biological variation of superoxide dismutase, SOD, in erythrocytes and of glutathione peroxidase, GPX, in whole blood. They have found that although the intraindividual variation of SOD is greater than for GPX, the measurement of SOD is to be preferred because the total biological variability - intraindividual biological variability when combined with interindividual biological variability - is less for SOD than GPX. In plain language this means that because SOD levels are more tightly controlled than GPX levels, an abnormal SOD level can be regarded as more significant than an abnoraml GPX level.

Readers of papers on free radicals and the damage they cause in biological systems will be aware of the popularity of the thiobarbituric acid test, TBA, for the estimation of free radical damaged molecules in plasma. TBA reacts with a variety of compounds in plasma, only some of which can justifiably be described as products of free radical damage. As a consequence purests have poured disdain on this simple and humble test and it was interesting to read that Badcock et al in Adelaide have found that although it might be crude it really does not need to be made more sophisticated by special extra analytical techniques such as high pressure liquid chromatography.

(Refs : Dahler-Eriksen et al. (1997), Clin Chem, v43, #11, 2064-75
Shentag et al. (1984), Arch Surg, v119, 300-4
Gudmundsson et al. (1997), Clin Chem, v43, #11, 2199-2201
Covas et al, (1997), Clin Chem, v43, #10, 1991-3
Badcock et al. (1997), Clin Chem, v43, #9, 1655-7 )

CALENDAR OF NATIONAL AND INTERNATIONAL MEETINGS

March 21st - 26th, 1999 : 4th Winter Research Conferences on Free Radicals , Valloire, France. Contact Arlette Alcaraz, CHU Grenoble, Laboratoire de Biochimie C, BP 217, 38043 Grenoble Cedex 09 France. Ph +33-4-7676-5754 Fax +33-4-7676-5664.
email: Arlette.Alcaraz@ujf-grenoble.fr or Alain.Favier@ujf-grenoble.fr

April 23rd - 26th, 1999 : Clinical Nutrition in Health and Disease Southport, Queensland. Contact : Marilyn Strauss.
Ph : +61-7-5531-3810 Fax : +61-7-5532-6199
email : medical@pain-education.com

May 2nd - 7th, 1999 : Tenth International Symposium on Trace Elements in Man and Animals , Evian, France. Contact Arlette Alcaraz, CHU A Michallon, Laboratoire de Biochimie C, 38043 Grenoble Cedex 9, France. Ph +33-4-7676-5754. Fax +33-4-7676-5664.
email: cerlib@ujf-grenoble.fr

June 17th - 19th, 1999 : Eighth European Nutrition Conference , Lillehammer, Norway. Sponsored by the Norwegian Nutrition Society, Federation of European Nutrition Societies, and the European Academy of Nutritional Sciences. Contact Lillehammer Arrangement AS, PO Box 14, N-2601 Lillehammer, Norway. Ph +47-61-251705. Fax +47-61-256515.
email: lillarra@sn.no Webpage :www.nutrition.uio.no/nse/8thFENS_EANS

September 4th - 8th, 1999 : 21st ESPEN Congress , Stockholm, Sweden. Abstracts are due by 22nd March 1999. Local Organising Committee : Ms Helene Jansson, Dept of Surgery, Huddinge University Hospital, SE-141 86 Huddinge, Sweden. Ph +46-858-582431 Fax + 46-858-582340
email helene.jansson@karo.ki.se
Secretariat : ESPEN 99, c/o MCI Congress, 75 rue de Lyon, CH-1211, GENEVA 13 - SWITZERLAND. Ph +41-22-345-3600 Fax +41-22-340-2363,
email : chantal@mcitravel.com

Programme Outline :

• Tracks
  • Determining outcome in clinical nutrition
  • Maintenance of gut integrity and function
  • Molecular biology in nutritional research
  • Body wasting syndromes and therapeutic strategies
  • Feeding the elderly and the chronically ill
  • Control of food intake - a major challenge for the next millennium

• Symposia
  • Regulation of protein metabolism
  • The role of new fat emulsions
  • Perioperative care of the surgical patient
  • Nutritional immunology- where do we stand?
  • The gut flora and special nutrients
  • Metabolic studies in vivo
  • Nutritional aspects of the short bowel syndrome
  • Nutrition in the critically ill

• Special lectures
  • Opening lecture
  • 20 years of ESPEN
  • Sir David Cuthbertson Lecture
  • Arvid Wretlind Lecture
  • Stockholm Honorary Lecture

October 7th - 9th, 1999 : International Symposium on in vivo Body Composition Studies, Brookhaven National Laboratory, Upton, NY. Contact S Yasumura, Bldg 490, Medical Department, PO Box 5000, Brookhaven National Laboratory, Upton, NY 11973-5000. Ph +1-516-344-3606. Fax +1-516-344-5311.
email: yasumura@bnl.gov.

October 18th - 21st, 1999 : The American Dietetic Association Annual Meeting and Exhibition, Georgia World Congress Center, Atlanta. Contact The American Dietetic Association, 216 West Jackson Boulevard, Chicago, IL 60606-6995. Ph +1-800-877-1600, ext 4866. Fax +1-312-899-0008.
email: mtgsinfo@eatright.org Webpage : www.eatright.org

October 22nd - 23rd, 1999 : ANNUAL SCIENTIFIC MEETING OF THE AUSTRALIAN SOCIETY FOR PARENTERAL AND ENTERAL NUTRITION. GOLD COAST INTERNATIONAL HOTEL, SURFERS PARADISE, QUEENSLAND.

Programme Overview

• Oxidative Stress and Gastrointestinal Disease
• Nutritional Assessment
• Nutrition and Gastrointestinal Disease
• Mucosal Nutrition
• Pharmaceutical Aspects of Nutritional Support
• Diet and Nutrition in Gastrointestinal Cancer
• Workshop on Practical Issues of Nutritional Support

International Guests

• Dr Khursheed Jeejeebhouy MD, Professor of Medicine, Nutrition and Physiology, University of Toronto and Director of the Nutrition Support Service, St Michaels's Hospital, Toronto, who has a distinguished career in gastroenterology and nutrition.
• Dr John McFie MD, Consultant Surgeon at Scarborough Hospital, UK. His areas of expertise are in nutrition and mucosal immunity and the role of glutamine in nutrional support.

Conference Secretariat :
Consult Fleetwood, PO Box 79, ARANA HILLS, QUEENSLAND 4054, Australia.
Ph: +61-7-3264-5970 Fax: +61-7-3264-3520
email : charlesf@gil.com.au

November 25th - 28th, 1999 : Australian Society of Hospital Pharmacists Perth W.A. Contact Jenny Dyer, Motive Conventions, Perth, Western Australia. Phone: + 61-8-9322-2666 Fax: 61-8-9322-1417
email: motive@vianet.net.au
Topics : Allergic disorders, Anaesthesia, Cardiovascular disorders, Clinical pharmacology, Endocrine disorders, Gastrointestinal disorders, General Medicine, Health care management, Infectious and parasitic diseases, Musculoskeletal and connective tissue disorders, Neurologic disorders, Nursing, Nutritional and metabolic disorders, Paediatrics, Pulmonary disorders, Radiology, Sexually related disorders, Surgery, Transplant Surgery

---

INTERNET NEWS

I am pleased to be able to report that I have located the ESPEN Website :

www.espen.org

I am not too sure when this came into existence but the 21st ESPEN Congress page was last updated in August '98. This is the site to visit if you want to register your name for further information on the meeting in Stockholm from the 4th to the 8th September 1999.

While attending the '98 Annual Scientific Meeting I met Dr Mark Bassett and Dr Gabrielle Andrew who run an excellent gastroenterology website :

www.gastro.net.au The content is very much patient orientated and includes some excellent graphics.

At the AGM Julie Bines reported on the plans that ASPEN have for providing educational material and courses over the internet. Their first unit will be for physicians and that will be followed by nursing and pharamacy units. The South American countries are particularly active in web based communication in TPN and EN and they will be collaborating with ASPEN on this project.

If any members are planning trips to Spain and would like to make contact with TPN / EN people over there then I am sure that Dr Carlos Seron (an intensivist at Hospital San Jorge. HUESCA. SPAIN. email : seron@infomail.lacaixa.es would be delighted to hear from you. He contacted me recently with some information on a research project he is carrying out in the area on calorimetry. He has a web site with ten clinical cases for your comment

http://phoebe.cps.unizar.es/ccn/ The pages are in English and Spanish and any responses you give will be kept confidential. His plan is to use the data to develop and validate a computer program that calculates formulae for artificial nutrition. This program has been designed to be used in hospitals and intensive care units that do not have access to calorimetry

Whilst still on the topic of travel and international contacts in TPN/EN you can contact Krishnan Sriram, the President of the Indian Society for Parenteral Nutrition at

ksriram@md2.vsnl.net.in

---

DISCLAIMER

The views and opinions expressed in this Newsletter are not necessarily the views and opinions of the Australian Society of Parenteral and Enteral Nutrition. Reports and articles on techniques, procedures and products are provided for the information of the Members of the Society and their inclusion does not imply any endorsements from the Australian Society of Parenteral and Enteral Nutrition. No liability can or will be accepted by AuSPEN or its agents for the third party use of information in this Newsletter.

Dr Tom Hartley, Editor, 1st February 1999.

---

Newsletter of the Australian Society for Parenteral and Enteral Nutrition Editor : Dr Tom Hartley, 36, Pregnells Road, Sandfly, Tasmania 7150, Australia Ph 03-6239-6475 (AH) 03-6222-8780 (BH) Fax 03-6231-3145 email hartley@southcom.com.au AuSPEN WWW Homepage : www.southcom.com.au/~hartley/home.htm