Reliability of wound surface area measurements.

Physial therapists administer a variety of therapeutic procedures to patients with open wounds. The modalities used to facilitate wound healing include hydrotherapy, hyperbaric oxygen electrical stimulation, ultrasound, and topical agents. [1] In order to assess the efficacy of these therapeutic interventions on open wounds, reliable measurements are necessary.

Different techniques have been used to measure wound areas from transparency film tracings. Sterilized transparency film is placed over the wound, and the wound's perimeter is traced with fine-tipped pen. Dyson and Suckling [2] mutliplied length and width measurements taken with a ruler. Fergusson and Logan [3] placed the transparencies over graph paper and counted the 1-mm squares within the perimeter of the wound tracing. A geographer's planimeter was used by Yucel and Basmajian [4] to calculate the area of ulcers traced on transparency paper. Sinacore et al [5] recorded ulcer area from transparency tracings using a digitizer. None of these studies determined the reliability of the techniqued used to record wound area.

Kloth and Feedar [6] reported a technique of tracing a wound perimeter onto plastic wrap and then transcribing the tracings onto metric graph paper to calculate wound area. One examiner performed three repeated tracings of 16 wounds. Tracings were reported to be similar, but the raw data for the repeated tracings were not presented and a statistical analysis of reliability was not performed.

Anthony [7] investigated the intrarater and interrater consistency in calculating pressure sore areas using three different techniques. First, ulcers were photographed with slide film, and a digitizer was used to compute the area of the ulcer from the enlarged slide. Second, transparency tracings were placed over graph paper, and the 1-mm squares were counted. Third, the longest and shortest diameters of the wound were measured with a ruler. Measurement repeatability was assessed by the ratio of standard deviation to the mean expressed as a percentage. An analysis of variance (ANOVA) was used to compare interrater variability.

Repeatability was greatest with the computer method (6.0%), followed by the diameter method (11.4%) and the tracing method (14.0%). The computer method had the highest consistency for intrarater and interrater measurements, whereas the tracing method yielded the lowest repeatability for intrarater and interrater measurements. The diamter method showed high intrarater consistency only.

One aspect of a diabetic foot evaluation presented by Diamond et al [8] involved foot ulcer measurements. The perimeter of a diabetic foot ulcer was traced onto sterilized x-ray film, and a sonic digitizer was used to compute the area. Intraclass correlation coefficients (ICCs) for intrarater and interrater reliability of ulcer area were .99.

In a recent study by Griffin et al, [9] the efficacy of healing pressure ulcers in patients with spinal cord injury was assessed. Serial measurements of wound surface area were taken using a combined photographic slide and tracing technique. Slides taken of the wound were projected at actual size onto paper, and the ulcer perimeter was then traced and digitized. Only intratester reliability of wound surface area measurement was assessed, and an ICC value of .99 was reported.

The literature demonstrates a variety of methods used to measure wound area. Design and statistical differences in these studies prevent adequate comparison of the reliability of measurements obtained with the various would measurement procedures.

The purposes of this study were to determine intratester and intertester reliability of four methods of measuring wound area from transparency film tracings and to compare the reliability estimates. The methods were (1) estimating wound area by multiplying length and width measurements taken with a ruler, (2) placing the transparency on graph paper and counting the squares, (3) using a hand-held planimeter to measure wound area, and (4) using a digitizer to measure wound area. I expected that all methods would be reliable between and within raters, based on results in the literature regarding the similarity between repeated wound tracings.

Method

Subjects

Thirty-one patients attending the Peripheral Vascular Clinic at the Medical College of Virginia Hospitals (Richmond, Va) for management of unhealed venous stasis ulcers were selected for this study. All patients gave verbal consent prior to participation in the study. Each ulcer was located on the lower third of the legal and fit within a 10.2x12.7-cm (4x5-in) transparency film. All tracings were made during the patients' weekly appointment to the clinic. Additional information concerning subject and ulcer characteristics was not recorded.

Procedure

Subjects were seated with the heel of the involved leg resting on a stool. All wound were cleansed with soap and water by the clinic nurse prior to tracings being obtained. Tracings were made by placing a piece of sterilized transparency film over the wound and tracing the wound's perimeter on the film with a fine-tipped transparency marker. A separate transparency was used for each wound. Both the film and the marker were of the type used to make overhead transparencies.

To assess reliability of measurements, three physical therapists made two tracing of each wound. The two tracings of each wound were designated as a pair of tracings. One therapist traced 34 wounds, and the other two therapists each traced 18 wounds. Therapists were instructed to hold the transparency film against the subject's leg and to avoid movement of the film. Care was also taken not to exert pressure over the wounds to avoid distortion of the wound surface. No training or pratice sessions were conducted prior to data collection.

All tracings of a given wound were done on the same day. Each therapist completed the two tracings on each patient prior to the next therapist entering the room for data collection. All tracings were coded by the clinic nurse so that the therapists could not associate pairs of tracings during calculation of the wound areas.

Each therapist independently determined the area of the tracings by each of the following four methods: (1) the length and width dimensions (in centimeters) of the tracings were measured with a ruler (*) and multiplied; (2) the tracings were placed over metric graph paper, and the number of 1-mm squares within the tracing were counted (only full 1-mm squares inside the perimeter were counted, and the area were converted to square centimeters); (3) the wound tracings were traced with a hand-held Compensating Polar Planimeter, (+) which gave a digital reading of the area in square centimeters; and (4) the wound areas were measured with a digitizer, (++), which also gave a square centimeter digital reading. All measurements were recorded to the nearest 1.0 [cm.sup.2].

Data Analysis

Reliability was assessed using the ICC equation (1,1). The type of ICC selected was based on a one-way ANOVA, as described by Shrout and Fleiss. [10] A minimally acceptable ICC of .70 was arbitrarily chosen. The data of therapist 1 (n=34) were analyzed by ANOVA to determine whether there was a significant difference among the mean wound areas for each of the four methods.

Table 1. Intratester Reliability of Wound Area Measurements

Therapist   Method     ICC
1 (n=34)    Ruler         .99
            Graph paper   .99
            Planimeter    .99
            Digitizer     .99
2 (n=18)    Ruler         .99
            Graph paper   .99
            Planimeter    .99
            Digitizer     .99
3 (n=18)    Ruler         .99
            Graph paper   .99
            Planimeter    .99
            Digitizer     .99

Results

The results showed that intratester reliability for each therapist using all four methods of calculating wound area was high (Tab. 1); all correlation coefficient were .99. Intertester reliability was also high (ICC=.97-.99) for each method of determining wound area (Tab. 2).

The overall range in ulcer area measurements for all three therapists was from 0.1 to 112.8 cm [2] (Tabl. 3). For each therapist, the mean wound areas calculated for the graph paper, planimeter, and digitizer methods were smaller than the mean area obtained with the ruler. The means for the ruler ranged from 7.9 to 10.2 cm [2] as compared with an overall range of 6.1 to 7.2 cm [2] for the other three methods.

Discussion

This investigation demonstrated the good reliability of each tested method of measuring wound area from transparency tracings. The intratester and intertester reliability values of .99 using the digitizer to calculate wound area agree with the results of Diamond et al [8] and Griffin et al [9].

Table 2. Intertester Reliability of Wound Area Measurements for Three Therapists (n=18)

Method         ICC
Ruler          .97
Graph paper    .99
Planimeter     .98
Digitizer      .99

The use of a ruler for measuring wound area in the clinical setting is easy, quick, and inexpensive. As higher wound areas were consistently obtained with the rule, the accuracy of this method is questionable. The overestimation of the real wound area may be attributed to calculating area by fitting a rectangular model (multiply width x length) to an irregular shape.

The graph paper technique can be easily incorporated into clinical practice, because it is easy to use and low in cost. The clinical application of the graph paper method to assess wound size in response to treatment intervention has been documented. [6] I agree with Bohannon and Pfaller [11] that counting the 1-mm squares can be tedious. With the metric graph paper, 1-cm blocks centrally located within the wound tracing could be counted quickly. Counting the 1-mm blocks at the periphery of the tracing was time consuming. The area of a 70-cm [2] ulcer was calculated in 10 minutes. The areas of the smaller ulcers could be calculated within 1 to 2 minutes.

The hand-held planimeter can be used to calculate wound areas quickly and requires minimal training to use. The planimeter is portable, and wound areas can be determined immediately in any setting. Many clinics or home health agencies, however, may not have this equipment. Purchasing a planimeter ($300-$700) may be cost effective, because of the staff time saved in calculations. The digitizer also gives a rapid calculation of would areas and is similar in price to a planimeter, but requires a personal computer.

Although intertester variability in determining wound area with the graph paper, planimeter, and digitizer techniques was low, research has not demonstrated that these methods may be interchanged when performing

[TABULAR DATA OMITTED]

serial wound measurement for assessing wound healing. Consistent use of the same examiner and technique should be emphasized.

Error in approximating the true wound surface area has been suggested to arise from several sources. Anthony [7] hypothesized that variability in determining sore area was due to individual differences in assessing the wound's border. Placing a transparency over a wound may alter the ulcer size or undermine the wound edges. [9] Bohannon and Pfaller [11] also suggest that error is introduced from the tracing itself. The counting, weighing, and planimetry methods used by Bohannon and Pfaller were more accurate with known areas than with ulcer areas.

Clinicians must also consider that the four measurement techniques presented are limited in that only the wound boundary is measured. This may be sufficient with shallow lesions. Evaluating wound depth in deep tracking wounds, however, may also be necessary. A variety of methods of assessing depth of tissue involvement have been described. [1] Additional research on the reliability of depth measurements is indicated.

Conclusions

The documentation of wound area is one aspect of assessing wound healing. This study has shown that intratester and intertester reliability for calculating wound area with the ruler, graph paper, planimeter, and digitizer methods is high. I recommend the graph paper technique to clinicians who do not have access to a planimeter or digitizer. The graph paper technique gives therapists a low-cost and simple tool to record wound size.

References

[1] Kloth LC, McCulloch JM, Eddar JA. Wound Healing: Alternatives in Management. Philadelphia, Pa: FA Davis Co; 1990.

[2] Dyson M, Suckling J. Stimulation of tissue repair by ultrasound: a survey of the mechanisms involved. Physiotherapy. 1978;64:105-108.

[3] Fergusson AG, Logan JC. Leg ulcers: assessment of response to certain topical medications. Br Med J. 1961;1:871-874.

[4] Yucel VE, Basmajian JV. Decubitus ulcers: healing effect of an enzymatic spray. Arch Phys Med Rehabil. 1974;55:517-519.

[5] Sinacore DR, Mueller MJ, Diamond JE, et al. Diabetic plantar ulcers treated by total contract casting: a clinical report. Phys Ther. 1978;67:1543-1549.

[6] Kloth LC, Feedar JA. Acceleration of wound healing with high voltage, monophasic, pulsed current. Phys Ther. 1988;68:503-508.

[7] Anthony D. Measuring pressure sores. Nursing Times. 1985;81:57-61.

[8] Diamond JE, Mueller MJ, Delitto A, Sinacore DR. Reliability of a diabetic foot evaluation. Phys Ther. 1989;69:797-802.

[9] Griffin JW, Tooms RE, Mendius RA, et al. Efficacy of high voltage pulsed current for healing of pressure ulcers in patients with spinal cord injury. Phys Ther. 1991;71:433-442.

[10] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428.

[11] Bohannon RW, Pfaller BA. Documentation of wound surface area from tracings of wound pertimeters: clinical report of three techniques. Phys Ther. 1983;63:1622-1624.