W. C. Ray1, R. W. Rumpf1, A. Rajab1, S. K. Martinez2, C. G. Gerrard2, N. L. Adolphi2, R. Fabia1 1Nationwide Children’s Hospital,Columbus, OH, USA 2University Of New Mexico HSC,Radiology-Pathology Center For Forensic Imaging,Albuquerque, NM, USA
Introduction:
Effectively treating burns relies on rapidly and accurately assessing the total body surface area (TBSA) and the proportion of skin that was burned (TBSA-B). Without accurate measurements of TBSA and TBSA-B, the efficacy of treatment is compromised, increasing the potential for medical complications. As such the burn treatment community requires a rapid and accurate means of assessing the extent and severity of burns. Unfortunately, even burn experts admit that most initial burn assessments, even those performed by experts themselves, frequently have significant errors.
Current methods for determining TBSA-B rely on tables of average body statistics such as the Lund and Browder tables. Today’s pediatric population differs in both size and shape from the population used to derive these charts, and estimates based on historical models are assumed to deviate from today’s reality. In addition to potential errors in the tables themselves, Lund and Browder and similar tables can be procedurally difficult to apply, relying on subjective and relative comparisons that result in significant deviations from the correct value even when the charts themselves are accurate.
Methods:
We initially hypothesized that the preponderance of errors seen in TBSA-B estimates using Lund and Browder were because the chart could not adequately represent a cohort taken from today’s pediatric population. In order to test this hypothesis, we compared the zone percentages as described in the Lund and Browder chart to actual TBSA percentages as measured using CT scans of human cadavers. Previous attempts at evaluating the accuracy of the Lund and Browder chart have been self-referential (e.g., comparing the results that different individuals achieve when applying the chart to the same test subject); to our knowledge this is the first study using an empirical approach to evaluating the chart by comparison to an external “gold standard” measurement of surface areas.
Results:
Unexpectedly, the actually-measured zone percentages were in reasonable agreement with Lund and Browder, even with subjects in the 1% tails of the BMI range. The largest deviation was 5.4%, significantly less than the error seen in real-world TBSA-B estimations. It would appear that while today’s population is morphometrically different from that studied by Lund and Browder, their body-region proportions scale surprisingly well.
Conclusion:
Our data indicates that, contrary to common assumption, the primary source of error in TBSA-B estimation is in the application of the Lund and Browder table rather than in the contents of the table.