|
Author(s)
Andreas M Fette
|
Contents
|
|
Published:
Jan 2006
Last updated: Jan 2006 Revision: 1.0 |
Keywords: wound measurement; clinimetrics; wound tracing; planimetrics; computerised stereophotogrammetry.
Current methods of wound assessment are often subjective so there is a need to standardise assessment to enable accurate data collection and the detection of clinically relevant improvements in patient outcomes.
Wound tracing is a reliable two-dimensional method of recording wound area that is easy to use and requires no special skills. When used in combination with planimetric area calculation it has good validity for shallow, pear-shaped wounds without undermined margins.
Clinimetrics is an effective method of assessing the quality of clinical measurement as it focuses on both the measurement instruments and the performance of measurements, and on ways to improve the quality of both.
The assessment of wound healing is often subjective so there is a need to develop a standard methodology to enable accurate comparisons between treatment outcomes and the accumulation of a reliable body of knowledge in this clinical area. There are many methods of measuring wounds, some simple and practical and others more suited to clinical research. Clinimetrics, which emphasises the quantitative measurement of clinical data through accurate measurement and data collection, was used to assess a number of these methods. The results reveal that wound tracing, combined with an area calculation using a planimetric scale in wounds (with an area greater than 40cm2), is a cheap and reliable method of accurately assessing relative changes in wound area over time.
The way in which wounds are assessed is often subjective, making it difficult to evaluate the effectiveness of treatments and leading to bias and the misinterpretation of data [1][2][3]. The growing demand for evidence-based practice, which aims to achieve optimum treatment outcomes, cost-effectiveness and a reduction in the use of antibiotics, has increased the need to amass a reliable body of knowledge so that wound-healing interventions can be compared and contrasted through the accurate interpretation of results [4]. The adoption of a common and objective approach to wound assessment is, therefore, desirable. Using a clinimetric approach, this paper analyses a number of different methods of measuring wounds to record relative changes in their size over time.
Effective measuring instruments and methods are essential in both clinical practice and research, particularly when evaluating treatments. A framework for the objective measurement and assessment of clinical data was posited by Alvin Feinstein of Yale Medical School in 1983 [5], who called the concept 'clinimetrics'.
Clinimetrics is a methodological discipline concerned with quantitative methods of collecting and analysing comparative data. It focuses on the development of methods to assess the quality of measurement instruments, and the application of these assessment methods to the development and evaluation of instruments of measurement. In particular, it aims to improve the quality of measurement of the distinctively clinical and personal phenomena of patient care, such as diagnostic characteristics [5]. In doing so it relies on a number of established constructs, including:
Validity - does it measure what it is intended to measure?
Accuracy - is the data collected accurate?
Reliability and consistency - do measurements of the same object by two or more individuals differ significantly?
Reproducibility - do repeated measurements yield the same results?
Usability - do users find it convenient, effective and easy to use?
The purpose of any wound measurement is to monitor the progress of healing through changes in the length, width, area or volume of a wound. This can be done using the following techniques:
Simple measurement: The simplest and cheapest method is to calculate the wound surface area by measuring its linear dimensions with a tape measure or ruler. However, this two-dimensional method assumes that the wound has a geometric surface shape, for example a rectangle (length x width), a circle (diameter x diameter) or an oval (maximum diameter x maximum diameter perpendicular to the first measurement). An alternative method of calculating wound surface area is based on the formula for an ellipse (length x width x 0.785).
Wound tracing: Another two-dimensional wound measurement tool is wound tracing, in which a pen is used to trace the outline of the wound directly onto sterile transparent film [6][7][8]. Wound tracing can be performed at the bedside using a minimum of equipment and requires no special skills or training on the part of the clinician. Each tracing in a sequence is easy to compare with the others and tracing is relatively unobtrusive for the patient. Tracings can be immediately stored in the patient's records and could be entered into a data processing system using a simple scanner [6][7][8][9]. The most significant limitation of wound tracing is deciding where the boundary of the wound lies, which affects the reliability and accuracy of the technique.
A Kundin gauge: This is a commercially available three-dimensional ruler used to calculate wound area and volume [10].
Moulds: A three-dimensional mould of the wound can be created by taking a cast of the wound cavity using a saline or alginate filling.
Scaled photographs: This two-dimensional method of assessment uses a photograph that has been processed by a special scanner so that a scaled ruler is incorporated at the edge of the photograph. The ruler is used to calculate length and width, which are expressed in simple measurements. Scaled photographs are useful for comparison but there is the potential for magnification errors [7][11][12].
Planimetrics: This method measures volume by creating a two-dimensional or planar image from a photograph or wound tracing. A transparent sheet of graph paper is laid over the photograph or wound tracing, either manually or using a computer, and the number of complete graph squares within the boundaries of the wound are added up to produce a scale area calculation [13]. This can then be stored in the patient's records or be entered into a data processing system.
Computerised stereophotogrammetry: Originally developed for land surveying, computerised stereophotogrammetry uses two pictures of the same area taken from different known positions to produce a three-dimensional image. A computerised matching algorithm searches for corresponding points in the two images and then computes the height of each point, based on the distance between corresponding points in the pair of pictures.
Some of the above techniques can be used in conjunction with others. For example, Yenidunya and Demirseren [9] used a technique in which the contour of a wound was traced on a transparent sheet overlaid on a digital photograph of the wound displayed on a computer screen. The resulting trace, which was relatively unobtrusive for the patient and could be stored digitally, was used for both preoperative planning and wound assessment.
Wound assessments can vary widely and reliable data can be difficult to obtain in clinical settings. When using a clinimetric approach, day-to-day clinical and practical aspects are also important. For these reasons, a literature review was carried out to critically appraise mainly two-dimensional methods of wound area measurement. Key words, such as clinimetrics, tracing and wound measurement, were used to identify relevant papers and the search was extended to sub-topics where necessary. A number of studies of wound measurement tools were then analysed using the following clinimetric constructs.
The validity of any wound assessment tool must be based on its ability to measure accurately what it is intended to measure. This is determined by its performance in relation to all the other constructs.
Calculations of wound parameters such as area, volume and circumference are usually based on mathematical equations [14][15][16]. Even though such calculations can provide valuable information on the progress of healing, unless the wound has a regular mathematical shape they can only be considered an estimate. For example, multiplying length by width to calculate the surface area of an oval wound would result in an overestimate because the formula is based on a rectangle [7].
Because the accuracy of a calculation based on multiplying the width of a wound by its length decreases with increasing wound size, more accurate calculations may be obtained using planimetrics.
Reliability and consistency are determined by the similarities or differences between assessments of the same wound by two or more individuals (also known as observers or raters). Interrater reliability, which is indicated by a high degree of agreement between observers using the same measurement technique on the same wound, is particularly relevant to wound tracing as both the thickness of the line drawn and the subjective assessment of the precise position of the wound boundary are crucial. Intrarater reliability refers to the consistency of individual raters observing different wounds [7][8].
Differences in the manual skill and/or experience of the measuring clinician can result in considerable variations in measurement, such as a precision error (one standard deviation as a percentage of wound size). The standard deviation is a measure of the spread of scores away from the mean (or average), which is the sum of observed values divided by the number of observations. Taking multiple measurements within a short time frame and recording the mean may improve accuracy [17].
Another important parameter, reproducibility is judged by the ability to repeatedly reproduce the same result and is crucial in determining both interrater and intrarater reliability. Reproducibility can only be achieved if measurements are always made under precisely the same conditions (for example the same body posture) using the same procedures. If not, the quality of the measurement is questionable. This highlights the importance of using protocols for each measurement technique used in both research and clinical settings. Unfortunately, authors often fail to report these matters, which can make effective critiquing of their results impossible.
Overall usability is important for any technique to become accepted as users have to be convinced that it offers real benefits in terms of time saved and practicality, and that it results in better patient care and cost savings [18]. Simple and cheap instruments are routinely favoured, which makes manual tracing more attractive in comparison to the often complex and expensive instruments found in research laboratories [18].
According to Plassmann [7], in smaller wounds (less than 40cm2) there is an 'excellent' correlation between planimetrics and calculations based on the formula for an ellipse, although his overview does not provide the statistical data on which this correlation was judged.
In another overview that does not provide statistical data, Plassmann and Peters [18] reported different percentages of precision errors for large and small wounds using three different measuring techniques (rulers, digital photography and wound tracings). They were:
for large wounds (greater than 40cm2) - rulers 8%, digital photography 4% and tracings 3%
for small wounds (in this study, less than 10cm2) - rulers 11%, digital photography 8% and tracings 7%.
In a comparable study, Majeske [19] tested the reliability of measuring venous stasis ulcer wound surface area from transparency film tracings. The results (correlation coefficient 0.97 to 0.99) indicate the reliability of intrarater and interrater measurements using wound tracing in conjunction with both graph paper and a planimeter [19].
Thomas and Wysocki [6] compared wound tracing with assessments made with a Kundin gauge and computerised photographic area quantification in pressure ulcers and venous leg ulcers. A total of 73 observations were made, with the size of the wounds ranging from 15.83mm2 to 3574.0mm2 for pressure ulcers (n=37) and from 84.70mm2 to 11172.5mm2 for venous leg ulcers (n=36). There was no more than a 10 percent error in the measurement of healing venous leg ulcers and pressure ulcers.
In both of the Majeske [19] and Thomas and Wysocki [6] studies, reliability was checked by taking multiple measurements for each observation with each device.
Data analysis using Pearson's correlation coefficient in Thomas and Wysocki's study [6] revealed that all three measurement methods studied (tracing, Kundin gauge and computerised photographic area quantification) were highly correlated (r = 0.93 or above, p≤0.001).
Pearson's correlation coefficient is a simple linear correlation used for values sampled from 'normal' populations. Correlation (r) tells us how strong the association between quantitative, continuous variables is, such as age and blood pressure, but it does not tell us about cause and effect in that relationship - two variables may be related but that does not necessarily mean that one causes the other. The size of the correlation also needs to be considered, because if a study is sufficiently large even a small correlation will be highly significant (p). Variation (r2) indicates how much of the variation in one value is caused by the other. The closer the r2 value is to 1, the higher the correlation [17]. Repeated measures of analysis of variance (a statistical technique used to compare the means of two or more samples to see whether they come from the same population) among the three methods revealed that they were significantly different (p ≤0.001). The authors concluded that any of the three methods could be used as a valid index of wound area but that acetate tracing should be used to obtain the most accurate measure of actual wound area [6].
Langemo et al [8] used three plaster of Paris wound models - one fairly symmetrical, one L-shaped, and one pear-shaped - to compare three wound measurement techniques: a ruler (length and width), planimetric tracing and computerised stereophotogrammetry (length and width; area). The baseline measurements of the models' outer wound perimeters were obtained by an aerospace testing authority using a coordinate measuring machine with an absolute accuracy of 0.0002 of an inch in both two- and three-dimensional spaces. The raters were 66 nurses with a variety of skill levels in wound measurement. Each rater measured each wound twice in a randomly assigned order of method and selection. This was a well managed, documented and reproducible trial using randomisation and standardised protocols.
The average error in measurement across raters as a percentage of the actual wound size is described as the relative bias. As the exact surface area for each wound was known it was possible to estimate the relative bias in measurement for each wound shape and each associated technique. Unsurprisingly, the least biased technique was stereophotogrammetry area, followed in order by stereophotogrammetry length and width, planimetric tracing, and ruler length and width. The length and width procedures overestimated the wound size, whereas the area procedure underestimated the wound size. In an absolute value sense, the smallest relative bias was associated with the stereophotogrammetry method. The ruler length and width measurements reflected the largest bias, regardless of wound shape [8].
The standard error of measurement (SEM) is defined as the difference between an observed score and a true score, so the SEM is an indication of the imprecision of a measurement technique [17]. The smallest SEM (the most accurately measured) when all the wounds were combined, was achieved by the stereophotogrammetry area technique, followed in order by the ruler length and width, computer length and width, and planimetric tracing.
The stereophotogrammetry area measurement had the smallest SEM for each wound shape. Both length and width methods were best when measuring circular wounds. Planimetric tracing worked best for pear-shaped wounds, whereas the stereophotogrammetry area measurement was most accurate for L-shaped wounds. Interrater reliability of averaged ratings was high for all techniques, but the clinically relevant reliabilities for single ratings were low (0.30-0.53), except for the stereophotogrammetry area technique (0.87).
Intrarater reliability was high for the planimetric tracing method and low for the stereophotogrammetry area method, with the two other methods lying in between. Planimetric tracing was most valid for pear-shaped wounds [8].
Manual tracings, as described by Thomas and Wysocki [6], prevent magnification and positioning errors and therefore makes them more accurate. However, these tracings could be a source of error if a second transfer of wound shape (from the acetate) is required, which could reduce the accuracy of the measurement significantly. Computer software that automatically identifies the boundaries of the tracing could be faster and increase accuracy [7][18] but, on the other hand, might reduce suitability for practical clinical use.
To improve the validity of wound measurements, Plassmann suggested the use of tracing sheets with planimetric imprints for larger wounds, with area calculations for smaller wounds based on the formula for an ellipse [7]. However, this should be limited to shallow wounds without undermined margins as wound volume or the extent of undermined margins cannot be adequately assessed using these methods. In addition, area measurements do not reflect early changes in deep wounds as they start the healing process by building up granulation tissue from the base of the wound [7].
The overall expense of any method is determined by the material used, which in the case of wound tracing with planimetric area calculation is relatively inexpensive. In particular, its validity for pear-shaped wounds is good and proven [8]. Intrarater and interrater reliability is also good for ratings averaged from multiple measurements [8]. This, combined with good reproducibility of results would make comparative wound tracings (in conjunction with planimetric area calculation) a valuable part of a patient's medical record.
Langemo et al [8] found that intrarater reliability was high for methods with low interrater reliability, suggesting that raters are consistent in the direction of their individual error. The low interrater reliability for single tracing ratings highlights the importance of multiple measurements by an individual rater so that valid or more accurate statistical results are obtained. Put simply, this suggests that when using the tracing method, the same observer should ideally carry out all episodes of multiple wound tracings until the assessment has been completed.
The assessment of wound healing is, in many cases, subjective and difficult to interpret. It follows that if data is not assessed accurately some necessary interventions might not be applied or performed [5][20]. From the perspective of evidence-based medicine, it is essential to accumulate objective data so that clinically relevant improvements that are due to therapeutic interventions can be detected [4]. The concept of clinimetrics may offer a way forward in the clinical setting by aiding the objective assessment of the effectiveness of interventions.
Several trials have examined wound measurement techniques. It is perhaps inevitable that the most expensive and complex equipment (stereophotogrammetry) produces the most accurate results. In the clinical setting, tracing has emerged as a useful tool, particularly when combined with a planimetric scale area calculation for larger wounds. Tracing is cheap, quick and easy to use and requires no special skills, and when used in combination with planimetric area calculation it has proven validity for shallow, pear-shaped wounds without undermined margins. It is also largely independent of the patient's position.
Further challenges lie in improving the quality of both the measurement instruments and performance of the actual measurements, especially in relation to formal research. More research is required to assess the benefits of adopting clinimetrics in other clinical areas, such as infection control and tissue viability.
1. Lazarus GS, Cooper DM, Knighton DR, Margolis DJ, Pecoraro RE, Rodeheaver G, et al. Definitions and guidelines for assessment of wounds and evaluation of healing. Arch Dermatol 1994; 130(4): 489-93.
2. Falanga V, Brem H. Wound bed preparation for optimal use of advanced therapeutic products. In: Falanga V, editor. Cutaneous Wound Healing. London: Taylor & Francis, 2001.
3. Kane D. Chronic wound healing. In: Krasner DL, Rodeheaver GT, Sibbald RG, editors. Chronic Wound Care: A Clinical Source Book for Healthcare Professionals. Wayne, PA: HMP Communications, 2001.
4. de Vet HC, Terwee CB, Bouter LM. Current challenges in clinimetrics. J Clin Epidemiol 2003; 56(12): 1137-41.
5. Feinstein AR. An additional basic science for clinical medicine: IV. The development of clinimetrics. Ann Intern Med 1983; 99(6): 843-8.
6. Thomas AC, Wysocki AB. The healing wound: a comparison of three clinically useful methods of measurement. Decubitus 1990; 3(1): 18-20, 24-5.
7. Plassmann P. Measuring wounds. J Wound Care 1995; 4(6): 269-72.
8. Langemo DK, Melland H, Hanson D, Olson B, Hunter S, Henly SJ. Two-dimensional wound measurement: comparison of 4 techniques. Adv Wound Care 1998; 11(7): 337-43.
9. Yenidunya MO, Demirseren ME. A useful method for preoperative planning with a transparent plastic sheet. Plast Reconstr Surg 2004; 114(1): 271-2.
10. Kundin JI. Designing and developing a new measuring instrument. Perioper Nurs Q 1985; 1(4): 40-5.
11. Resch CS, Kerner E, Robson MC, Heggers JP, Scherer M, Boertman JA, et al. Pressure sore volume measurement. A technique to document and record wound healing. J Am Geriatr Soc 1988; 36(5): 444-6.
12. Berg W, Traneroth C, Gunnarsson A, Lossing C. A method for measuring pressure sores. Lancet 1990; 335(8703): 1445-6.
13. Cooper DM. Assessment, measurement and evaluation: their pivotal roles in wound healing. In: Bryant RA, editor. Acute and Chronic Wounds: Nursing Management. St Louis, MO: Mosby, 2000.
14. Eriksson G, Eklund AE, Torlegard K, Dauphin E. Evaluation of leg ulcer treatment with stereophotogrammetry. A pilot study. Br J Dermatol 1979; 101(2): 123-31.
15. Bulstrode CJ, Goode AW, Scott PJ. Measurement and prediction of progress in delayed wound healing. J R Soc Med 1987; 80(4): 210-2.
16. Melhuish JM, Plassmann P, Harding KG. Volume and circumference of the healing wound. In: Proceedings of the 3rd European Conference on Advances in Wound Management. London: Macmillian Magazines, 1994; 41-44.
17. Harris M, Taylor G, editors. Medical Statistics Made Easy. New York: Taylor & Francis, 2003.
18. Plassmann P, Peters JM. Recording wound care effectiveness. J Tissue Viability 2002; 12(1): 24-8.
19. Majeske C. Reliability of wound surface area measurements. Phys Ther 1992; 72(2): 138-41.
20. van Rijswijk L. Wound assessment and documentation. In: Krasner DL, Rodeheaver GT, Sibbald RG, editors. Chronic Wound Care: A clinical source book for healthcare professionals. Wayne: HMP Communications, 2001; 369-383.