Dear Editors

We respectfully submit for publication as a Letter to the Editors of Annals of Burns and Fire Disasters our commentary on the following article that appeared in the March 2007 issue (vol. 20, no. 1, pp. 29-34) of the journal: Heparin reduced mortality and sepsis in severely burned children (Zayas G.J., Bonilla A.M., Saliba M.J.).

Sincerely

Mark Oremus, PhD (also on behalf of Drs Hanson and Dal Cin)

Zayas et al. report on two paediatric patient groups, one treated for burns in 1998 and one treated for burns in 1999. The 1999 group received heparin as part of burn therapy, while the 1998 group did not. The authors compared the groups and concluded that the use of heparin relieved burn pain, reduced mortality and sepsis, led to fewer additional treatment procedures, and improved cosmesis. These conclusions are not sufficiently supported by the evidence in the study. We will address the evidence for each conclusion in turn.

Pain relief was subjectively assessed, presumably by the study investigators, in the 1999 group by observing changes in patient behaviour (e.g., crying) before and after receiving heparin, as well as by judging noise levels in the emergency room or burn unit. However, none of these elements were empirically examined using a validated measurement tool, nor were empirical comparisons made with the group of patients who did not receive heparin. It is possible that other factors besides heparin treatment could have influenced patient behaviour (e.g., the calming influence of relatives who were holding the children). We recognize the difficulty of measuring pain in paediatric populations, but validated pain scales do exist for use in children.1 These scales are utilized in paediatric burn care2 and, more specifically, in research regarding pain and discomfort associated with procedural burn care.3,4 An empirical measure of pain would constitute stronger evidence for a benefit of heparin in pain relief, provided comparisons are made between treated and untreated groups.

The improved mortality and sepsis findings in the heparin-treated group may not have been due to heparin. Sepsis itself is severe enough to have a mortality rate of at least 29%.5 In the 1998 group, all nine patients had sepsis and eight died. In the 1999 group, four of 10 patients had sepsis and all four died; the six who did not have sepsis survived. The difference in mortality between groups could have been due to sepsis, not to treatment with heparin. The authors mention that sepsis was a major contributor to death in one patient (in 1999), but were data available and (if so) checked to rule out sepsis as a cause of death in the other 11 patients who died?

The high rate of sepsis in the study sample may in part be explained by the fact that the study hospital appears to be the point of treatment for all major child injuries in El Salvador. However, the authors do not mention how they chose the study sample. Were the patients enrolled consecutively after admission to the hospital, were they a convenience sample, or were they the entire complement of children admitted for burn injuries in 1998 and 1999? Also, the authors do not report whether there were any refusals to participate or losses to follow-up. These selection factors may partially account for the high rate of sepsis in the study sample. The selection factors should also be clarified by the authors so that the internal and external validity of the study can be more adequately assessed.

The authors report that antibacterial creams and antimicrobial baths were discontinued in 1999. Also, less meperidine was used in 1999 than in 1998 and there were also fewer fasciotomies, escharotomies, and grafts performed in 1999 than in 1998. While these changes may have resulted from the implementation of a heparin treatment protocol, the use of a historical comparison group raises the possibility that the changes could have resulted at least in part from independent modifications to the hospital’s standard operating procedures. Additionally, the changes may have reflected (perhaps subconsciously) clinicians’ a priori beliefs about the types of procedures that would be necessary (or not necessary) in heparin-treated patients. Indeed, the authors admit to not knowing the precise reduction in fasciotomies, escharotomies, and grafts, so the impact of heparin on other treatments cannot be adequately judged.

The study authors report improved cosmesis (i.e., less scarring) in heparin-treated patients, but do not provide any quantitative, comparative evidence to support their finding. Their evidence is based on the qualitative observation that the six survivors in the 1999 group had smooth skin after two to three weeks of healing. This essentially constitutes a case series because there was no realistic comparison possible with the 1998 historical controls owing to the 1998 mortality rate. Another issue is the description of the skin as “smooth”, yet this is only one of many features of scar formation. Other features include scar size, colour, texture, and thickness.6 The absence of reporting on these features provides an incomplete assessment of cosmesis. Furthermore, this study is of short duration, precluding the necessary and important assessment of scar formation and maturation over time. Follow-up of longer duration would have provided prospective, developmental data regarding multiple features of scar formation, including colour, thickening, and hypertrophy. The use of a quantitative measurement instrument (e.g., Vancouver Scar Scale6) to compare scar features in treated and untreated groups over a longer follow-up period would have been the ideal standard from which to draw conclusions about heparin’s effect on cosmesis.

In conclusion, Zayas et al. should be congratulated for examining the clinical utility of using heparin to treat bum injury in a paediatric population. Such studies are important and necessary because evidence from the basic scientific literature suggests that heparin could be efficacious in the treatment of burn injury.7 Furthermore, as Zayas et al. noted, investigations regarding sustainable burns treatment in countries with limited technological and financial resources are extremely valuable for improving health care in these settings. However, for the reasons enumerated in the above discussion, the Zayas et al. study does not provide sufficient evidence to conclude that heparin is efficacious in treating paediatric burns.

Matt Oremus, PhD, Assistant Professor (part-time), McMaster Evidence-based Practice Centre, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada

Mark D. Hanson, MD, Med, FRCP(C), Associate Professor, Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ontario, Canada

Arianna Dal Cin, BSc, MD, FRCS(C), Clinical Assistant Professor, Department of Surgery, McMaster University, Hamilton, Ontario, Canada

Bibliography

1. Finley G.A.: How much does it hurt? Paediatric burn pain measurement for doctors, nurses, and parents. Can. J. Anesth., 48: R1-R4, 2001.
2. Stoddard F.J., Sheridan R.L., Saxe G.N., King B.S., King B.H., Chedekel D.S. et al.: Treatment of pain in acutely burned children. J. Burn Care Rehabil., 23: 135-56, 2002.
3. Hernandez-Reif M., Field T., Largie S., Hart S., Redzepi M., Nierenberg B. et al.: Children’s distress during burn treatment is reduced by massage therapy. J. Burn Care Rehabil., 22: 191-5, 2001.
4. Whitehead-Pleaux A., Baryza M.J., Sheridan R.L.: The effects of music therapy on paediatric patients’ pain and anxiety during donor site dressing change. J. Music Ther., 43: 136-53, 2006.
5. Angus D.C., Linde-Zwirble W.T., Lidicker J., Clermont G., Carcillo J., Pinsky M.R.: Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit. Care Med., 29: 1303-10, 2001.
6. van Zuijlen P.P., Angeles A.P., Kreis R.W., Bos K.E., Middelkoop E.: Scar assessment tools: Implications for current research. Plast. Reconstr. Surg., 109: 1108-22, 2002.
7. Oremus M., Hanson M., Whitlock R., Young E., Gupte A., Dal Cin A., Archer C., Raina P.: The uses of heparin to treat burn injury. Evidence Report/Technology Assessment No. 148. AHRQ publication no. 07-E004, Rockville, MD, Agency for Healthcare Research and Quality, 2006.