Report by: Saiqa Hussain, Medical Student

Search checked by: Craig Ferguson, SpR Emergency Medicine

Institution: Manchester Royal Infirmary, Manchester, UK

A short-cut review was carried out to establish whether there is one assessment method that is the most accurate at estimating the surface area of a burn. Over 1500 papers were found using the reported searches, of which five presented the best evidence to answer the clinical question. The author, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of these best papers are shown in table 1. It is concluded that there is no evidence to show benefit of one form of estimation over another in clinical practice.

CLINICAL SCENARIO

A 30-year-old woman is brought to the emergency department with extensive burns after having been trapped in a burning car. After initial resuscitation of the patient you attempt to assess the surface area of skin affected by the burn in order to guide the volume of fluid required for resuscitation. You wonder which method of surface area assessment is the best tool for assessing the surface area of her burns.

THREE-PART QUESTION

In a [patient with a burn] is there one [method of assessment of size] that is most [accurate to assess surface area of the wound]?

SEARCH STRATEGY

Medline using the OVID interface 1966 to May 2009:[burn$.mp OR exp Burns OR heat injur$.mp OR thermal injur$.mp OR scald.mp] AND [assess$.mp OR estimat$.mp OR size.mp OR exp Body Surface Area/or surface area.mp OR SA.mp OR BSA.mp OR TBSA.mp OR Lund Browder.mp OR rule of nines.mp OR serial halving.mp] LIMIT to clinical trial, all and English language and humans.

Embase 1980 to week 22 2009.: [exp Burn/OR burn$.mp. OR scald.mp. OR exp Scald/OR heat injur$.mp. OR thermal injur$.mp.] AND [[assess$.mp. OR estimat$.mp. OR measur$.mp.] AND [size.mp. OR body surface area.mp. OR exp Body Surface/OR tbsa.mp. OR bsa.mp. OR sa.mp.]] OR [exp Diagnostic Accuracy/OR exp Measurement/OR lund browder.mp. OR rule of nines.mp. OR serial halving.mp.]LIMIT to human and english language and “diagnosis (sensitivity)”.

Google scholar searched for first three pages of results using each of the terms, “Lund and Browder”, “rule of nines”, “serial halving” and “burns assessment”.

The Cochrane Library Issue 2 2009: (burns):ti,ab,kw AND (rule of nine):ti,ab,kw 2 records? 1 relevant.

In addition Google Scholar was searched and references of any relevant articles were scanned.

OUTCOME

A total of 1518 papers was found on the Medline search, of which one was relevant; 727 papers were found on the EMBASE search, two were relevant, one was unique. Three papers were found using Google scholar, one of which was already in the EMBASE search result. One further relevant paper was found from the references of papers used. Five papers were therefore available and are summarised in table 1.

COMMENTS

Estimating the size of burns can be difficult to do well. The percentage of total surface area affected is an important figure as it has bearing on whether or not the patient should be referred to a specialist burns centre for management. It is also used to decide if patients require intravenous fluids and in formulae to calculate the requirements for fluid resuscitation. It is relatively difficult to measure the surface area of an irregularly shaped three-dimensional object at any time and pretty much impossible in an acute situation with an unwell patient. It may also be difficult to differentiate simple erythema from partial thickness burns in the immediate presentation. There are four main methods of estimating the percentage of a patient’s body surface area that is involved in a burn: Lund and Browder charts form a fairly detailed diagram with percentage surface areas of different parts of the body, dependent on age, which can be combined for a total figure (6); the “rule of nines” chart splits the body into 11 areas of 9% and the perineum makes up the final 1% (7); serial halving involves dividing the surface area of the patient into areas of 50% and deciding if the burn covers more or less than this area. This process is continued for a further two times if necessary to get an approximate value of the percentage surface area affected; for smaller areas the palmar surface of the patient’s hand is taken to represent 1% of their total body surface area and used to aid estimation. There have also been a couple of studies using computers to calculate the surface area affected from an image of the patient. Although there are several papers comparing different groups of physicians’ estimates for percentage areas of burns, there were only a few papers found in this search that directly compared the different methods used. Due to the impracticality of having several physicians independently assess a real patient with burns, various alternatives were used. Various gold standards were used, the best of which involved measuring the surface area of the burn, although even these studies used formulae to calculate the overall surface area rather than measuring it directly. The literature suggested that there was significantly less variation in the estimates made using the Lund and Browder chart from the schematic burns than using the “rule of nines” technique, although the variation remained substantial. There also appeared to be less variation and greater accuracy in documenting the size of the burns graphically than in converting this to a percentage surface area. The “rule of nines” technique appears to be quicker to use and does not require a special chart as long as the physician can recall the attributed surface areas percentages. Serial halving provides a very approximate result that can be used quickly in the field to establish whether or not intravenous access is required with reasonable consistency. Once a burn has been mapped to a two-dimensional diagram then a computer or planimeter can accurately calculate the percentage surface area affected. Interestingly, when this technique is used to measures the surface areas of the different regions on a blank Lund and Browder chart, they do not correspond with the values that they are purported to represent. This intrinsic error means that there is a limit to how accurately the chart could ever be interpreted. It is not difficult to imagine a computer program that could provide an interactive three-dimensional figure on screen that could be drawn onto and provide an estimate of percentage burn size based on the height, weight and age of the patient. This may improve the accuracy and consistency of estimates but would require access to the computer and software and may still leave problems of documentation in the medical notes and communication with other hospitals. The question that leads on from this discussion is whether more accurate assessment of burn size would actually result in a clinical benefit for patients or whether a rough estimate and then titration of treatment according to response is as good.