LYMPHATIC RESEARCH AND BIOLOGY Volume 12, Number 4, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/lrb.2014.0007
Validity and Reliability of Three-Dimensional Imaging for Measuring the Volume of the Arm Maaike Erends,1 Teike van der Aa,1 Andrzej Piatkowski de Grzymala, MD,2 and Rene van der Hulst, MD, PhD 2
Abstract
Background: Lymphedema of the upper limb is a common complication after cancer treatment with axillary lymph node surgery and/or radiation. At this moment, there is no method that can identify lymphedema error free. Currently three-dimensional (3D) imaging is used to measure volumes in aesthetic and maxillofacial surgery. This study aims to assess the validity and reliability of 3D volume measurements of the upper limb compared with the current gold standard (water displacement method). Methods and results: Thirty-three subjects were included which completed both measurements. The accuracy of the aforementioned methods was compared in a within subject design. The analysis showed a mean difference between the two measurements of - 13.8 cc (SD 59.3), this volume difference was not significant ( p = 0.192). Both the intra- and inter-rater reliability of the 3D measurements were high (0.99). Conclusion: The 3D volume measurements of the arm are valid and reliable. Therefore we recommend the 3D method for measuring arm volumes. Since this method is now validated for arms without lymphedema, we plan to validate this new technique for patients with lymphedema.
Introduction
U
pper limb lymphedema is a common complication after cancer treatment with axillary lymph node surgery and/or radiation.1–3 Lymphedema is a swelling of the upper limb due to excess accumulation of interstitial fluid in body tissue caused by obstruction or alteration in lymphatic flow.4,5 At this moment there is no method that can identify lymphedema error free.6 Currently two procedures are commonly used to identify lymphedema. The first is to compare the volume of the affected with the unaffected limb, and second is to compare the volume of the affected limb before and after the onset of lymphedema. Water displacement volume measurement is often considered as the gold standard;7–10 the displaced water equals the volume of the arm. Although this technique is simple, there are many variables that need to be considered. First, the water volume and density are influenced by temperature. This method assumes that 1 cm3 equals 1 mL without taking into account temperature and type of water. Second, the accuracy depends on the manner in which the object sinks to the bottom of the water tank. It is not easy to ensure that every object is submerged to the same level. Third, the water displacement is
not suitable for patients with skin lesions, considering the risk of infection. Additional points for consideration are that it is time-consuming, not portable, and potentially nonhygienic.11 Besides water displacement, it is also possible to measure the upper limb volume indirectly by circumference measurements (CM), using the formula for a truncated cone. This method measures a significant lower volume than water displacement, because it excludes the hand.12 Three-dimensional (3D) imaging is currently used to measure volumes in aesthetic and maxillofacial surgery. The results of these volume measurements are promising.13–15 This study aimed to assess the validity and reliability of 3D volume measurements of the upper limb compared to the water displacement method. Materials and Methods Subjects
Thirty-three subjects completed the study; twenty females and thirteen males. These subjects had no lymphedema or open lesions on their upper extremity at the time of testing. Each subject was measured by two raters, each measurement was performed twice.
1
Faculty of Health Medicine and Life Sciences, Maastricht University, Maastricht. Netherlands. Department of Plastic Surgery. Maastricht University Medical Centre plus (MUMC + ), Maastricht. Netherlands.
275
276 Study design
To assess the validity and reliability of the 3D camera, we used the water displacement method. This method was regarded as the gold standard. To analyze the validity of the 3D volume measurement, we compared the 3D volume measurement with the water displacement. In order to analyze the intra-rater reliability of the 3D volume measurement, the first investigator performed two measurements for each subject. In addition, the inter-rater reliability was analyzed by comparing ten randomly picked 3D volume measurements of two different raters. The right upper limb of the thirty-three subjects was measured. Two lines were drawn at 30% and 80% of the arm’s length, measured from the tip of the third finger to the tip of the acromion. The 30% line correlated with the proximal part of the wrist and the 80% line correlated with the part of the arm that can be submerged into the varitex volumeter (Fig. 1). We used percentages instead of anatomical landmarks because the measurements would be more homogenic.6,12 The hand was excluded from the volume measurement, because 3D volume measurements of the hand are not valid.16 After drawing the two lines, five marks on the top and five marks on the bottom of the arm were drawn. These marks were required as a reference for the 3D volume measurement. When placing the marks, it was important they were in the midline on the top and on the bottom of the arm, so they could
FIG. 1. Anterior and posterior side with the 30% and 80% lines and ten marks.
ERENDS ET AL.
be captured on both 180 pictures. The five marks should be equally divided over the length of the arm. First 3D images were taken followed by the water displacement method, both measurements were repeated. Water displacement method
For the water displacement volume measurements, a volumeter was used (Varitex, Haarlem, The Netherlands) (Fig. 2).17 The volumeter was filled with tap water with a temperature between 25C and 32C. Water temperatures across the range of 20–32 Celsius were not found to affect the volume measured.11,18 The volumeter weighed the overflowing water with an integrated calibrated scale with an accuracy of 1 cc. Subjects were instructed to lower the hand straight and slowly into the volumeter. If the 30%-line reached the water surface, subjects were instructed to stop and keep still. Then the arm was measured following the same course until the 80%-line. Both the hand and the arm were measured twice to test the reliability of the water displacement method. To compare the water displacement method with the 3D volume measurement, the volume of the hand was subtracted from the arm volume. The measurements with the volumeter including the marking of the arm took 20 minutes in total. 3D volume measurement
All subjects were photographed using the Vectra XT 3D imaging system (Canfield Imaging Systems, Fairfield, NJ)
FIG. 2.
Volumeter (Varitex, Haarlem, The Netherlands).
THREE-DIMENSIONAL IMAGING OF ARM VOLUME
FIG. 3.
277
Vectra XT 3D camera.
under standard clinical lighting. This system comprises six digital single-lens reflex cameras divided over three stations and four flashes mounted within the Vectra unit. These cameras and flashes fire simultaneously when ‘Cxapture’ is pressed in the computer driven capture software(Fig. 3). The camera is able to capture three-dimensional images in 180. The subject was instructed to abduct the arm up to 90 and keep the hand in horizontal position with the palm facing down. The posterior and anterior sides of the arm were both captured twice in order to investigate the intra-rater reliability of the 3D volume measurement. In order to capture the anterior and posterior side, the subjects were instructed to turn 180. The captured data set was saved and the Vectra software automatically processed the images into a highresolution 3-dimensional model (Fig. 4). The anterior and posterior 3D images were viewed and analyzed using Mirror
FIG. 4.
software (Canfield Imaging Systems). The software allowed merging of the two 180 images using the ten landmarks, resulting in a 360 image. This image was then cropped at the 30% and 80% lines. Both ends of the 360 image were filled. Then the volume was measured using the closed surface (Fig. 5). This procedure, including the marking of the arm and computing the volume, took 15 minutes in total. Statistical analyses
Patient characteristics were presented by mean (SD). The association between water displacement and 3D measurement was assessed by the Pearson correlation coefficient. The mean difference between the values of the water displacement and the 3D measurement, as well as the mean
3D images, (A) Anterior side; (B) Posterior side.
278
ERENDS ET AL.
Table 1. Baseline Characteristics Std. N Minimum Maximum Mean Deviation
FIG. 5.
Stitched 3d image with closed surface.
differences between the single measurements of these methods, were compared using paired-samples t-test or Wilcoxon signed-rank test where appropriate. The normality assumption of the differences was checked using histogram, skewness, and kurtosis. A Bland-Altman plot was used to analyze the agreement between the water displacement method and the 3D volume measurement. The Intraclass Correlation Coefficient (ICC) was used to determine the intra- and inter-rater reliability of the 3D volume measurements. A p value £ 0.05 was considered statistically significant. Data were analyzed in SPSS 20.0 software (SPSS Inc, Chicago, Ill). Results Subject characteristics
Thirty-three participants were recruited to participate in the study. Every subject was able to complete all measurements. The measurements were taken in one session; baseline
FIG. 6.
Age at follow up Body mass index Circumference upper-arm Circumference lower-arm Length arm
33
20
53
31.3
10.0
33
18.69
26.17
22.62
1.96
33
25.00
33.00
29.18
2.20
33
15.00
20.00
16.76
1.23
33
66.00
84.00
75.03
4.98
characteristics from each subject were collected before testing (Table 1). The mean age at time of measurement was 31 years (SD 10), the mean body mass index (BMI) was 22.6 (SD 2.0). The measurements of the water displacement, as well as the volumes obtained by the 3D measurements, were normally distributed. Validity
Figure 6 shows a plot of the volumes obtained by water displacement and 3D measurements. Volumes that were calculated from the 3D measurements were on average higher than obtained from water displacement. Comparison between the two methods showed that there was a high correlation (0.98). In addition, paired samples t-test showed there was no significant difference between the 3D volume measurement and the water displacement ( p = 0.192). The mean volume difference between the two methods was - 13.8 cc (SD 59.3). The Bland-Altman plot (Fig. 7) showed
Plot of the water displacement versus 3D measurements.
THREE-DIMENSIONAL IMAGING OF ARM VOLUME
FIG. 7.
Bland-Altman plot of water displacement and 3D measurement.
there was agreement between the two single measurements; differences ranging from - 134.0 to 88.0 cc. Reliability
The intra-rater reliability of single 3D volume measurements was 0.99. Paired samples t-test showed no significant difference between the two 3D volume measurements ( p = 0.673). The mean volume difference between the two 3D measurements was - 1.39 cc (SD 18.8). A strong agreement between both 3D volume measurements was seen in the Bland-Altman plot (Fig. 8); differences ranging from - 30.0 to 30.0 cc. The water displacement volume measurements had a high intra-rater reliability (0.99), but on the contrary paired sam-
FIG. 8.
279
ples t-test showed a significant difference between the two single volume measurements ( p < 0.001), with a mean volume difference of 51.00 cc (SD 54.2).The Bland-Altman plot of the two water displacement measurements indicated that there was less agreement between the two measurements than the 3D volume measurements; differences ranging from - 89.0 to 163.0 cc. Inter-rater reliability
Inter-rater reliability was determined with 3D volume measurements of two different raters (Table 2). The first rater executed all the 3D measurements and the second rater, who was blinded to the results of the first rater, executed
Bland-Altman plot of the single 3D and water displacement volume measurements.
280
ERENDS ET AL.
Table 2. Descriptive Statistics of Two Separate Raters N Minimum Maximum Rater 1 10 Rater 2 10
1448 1412
2564 2593
Mean
Std. Deviation
1939.70 1945.50
348.37 365.32
measurements of ten randomly picked subjects. The correlation between the two raters of these ten single measurements was 0.99. This means there is a high inter-rater reliability. Discussion
The results from this study demonstrate good validity and reliability for the 3D volume measurement of the upper limb; 3D volume measurements were more reliable as compared to the water displacement method. In addition, it was shown that 3D volume measurements had also high intra- and inter-rater reliability. A disadvantage of the water displacement method is that the second measurement was significantly lower than the first measurement. Second, the water displacement could be less exact, because the eventually calculated volume consisted of two single measurements. Therefore, the risk of measuring errors was higher. Over the years, varying definitions and methods of measurements for lymphedema have been utilized. Currently an increase of ‡ 10 % is a common criterion for lymphedema.19 Looking at the results, there is not one 3D measurement that measures more or less than 10% compared with the gold standard. A shortcoming of the water displacement was the size of the reservoir. Subjects with long arms ( > 80 cm) had difficulty putting their arm into the reservoir to the 80% line. Eventually every subject was able to lower their arm till the 80% line, but if the arm was longer it caused more undulation of the water. Finally, the volumeter is not suitable for patients with open lesions due to the risk of infection. In contrary to the water displacement, 3D volume measurements are suitable for patients with skin lesions. This technique is noninvasive and has no contra-indications. A disadvantage of the 3D volume measurement is the impossibility of measuring the hand volume, because the web spaces are difficult to capture.16 Another difficulty of the 3D volume measurements was the positioning of the marks on the upper limb. In order to capture all ten marks on both 3D images, it was important to draw the marks exactly in the midline of the anterior and posterior side. Therefore the physician has to be trained to perform the measurement. Last, the 3D camera is valuable and not every medical center has one at its disposal. Conclusion
In this article a new method for measuring arm volumes is proposed. This study showed promising evidence for validity and reliability of 3D volume measurements for arms without lymphedema. In further research we plan to validate this new technique for patients with lymphedema. Author Disclosure Statement
The authors indicated no conflict of interest. The authors alone are responsible for the content and writing of the article.
References
1. Monleon S, Murta-Nascimento C, Bascuas I, Macia F, Duarte E, Belmonte R. Lymphedema predictor factors after breast cancer surgery: A survival analysis. Lymphat Res Biol May 16, 2014. Epub ahead of print. 2. Ugur S, Arici C, Yaprak M, Mesci A, Arici GA, Dolay K, Ozmen V. Risk factors of breast cancer-related lymphedema. Lymphat Res Biol. 2013;11:72–75. 3. Petrek JA, Senie RT, Peters M, Rosen PP. Lymphedema in a cohort of breast carcinoma survivors 20 years after diagnosis. Cancer 2001;92:1368–1377. 4. Bates DO, Levick JR, Mortimer PS. Change in macromolecular composition of interstitial fluid from swollen arms after breast cancer treatment, and its implications. Clin Sci 1993;85:737–746. 5. Cheville AL, McGarvey CL, Petrek JA, Russo SA, Thiadens SR, Taylor ME. The grading of lymphedema in oncology clinical trials. Seminars Radiat Oncol 2003;13:214– 225. 6. Lopez Penha TR, Slangen JJ, Heuts EM, Voogd AC, Von Meyenfeldt MF. Prevalence of lymphoedema more than five years after breast cancer treatment. Eur J Surg Oncol 2011;37:1059–1063. 7. Megens AM, Harris SR, Kim-Sing C, McKenzie DC. Measurement of upper extremity volume in women after axillary dissection for breast cancer. Arch Phys Med Rehab 2001;82:1639–1644. 8. Sagen A, Karesen R, Skaane P, Risberg MA. Validity for the simplified water displacement instrument to measure arm lymphedema as a result of breast cancer surgery. Arch Phys Med Rehab 2009;90:803–809. 9. Kaulesar Sukul DM, den Hoed PT, Johannes EJ, van Dolder R, Benda E. Direct and indirect methods for the quantification of leg volume: Comparison between water displacement volumetry, the disk model method and the frustum sign model method, using the correlation coefficient and the limits of agreement. J Biomed Engineer 1993;15:477–480. 10. International Society of Lymphology. The diagnosis and treatment of peripheral lymphedema: 2013 Consensus Document of the International Society of Lymphology. Lymphology 2013;46:1–11. 11. Boland R, Adams R. Development and evaluation of a precision forearm and hand volumeter and measuring cylinder. J Hand Therapy 1996;9:349–358. 12. Brorson H, Hoijer P. Standardised measurements used to order compression garments can be used to calculate arm volumes to evaluate lymphoedema treatment. J Plastic Surg Hand Surg 2012;46:410–415. 13. Kau CH, Richmond S, Incrapera A, English J, Xia JJ. Three-dimensional surface acquisition systems for the study of facial morphology and their application to maxillofacial surgery. Intl J Med Robotics Computer Assisted Surg 2007;3:97–110. 14. Liu C, Luan J, Ji K, Sun J. Measuring volumetric change after augmentation mammaplasty using a three-dimensional scanning technique: An innovative method. Aesthet Plastic Surg 2012;36:1134–1139. 15. Koch MC, Adamietz B, Jud SM, Fasching PA, Haeberle L, Karbacher S, et al. Breast volumetry using a three-dimensional surface assessment technique. Aesthet Plastic Surg 2011;35: 847–855. 16. Koban K, Leitsch S, Holzbach T, Metz P, Giunta R. Einzeitige Messung geschlossener Ko¨rper mittels Vectra 3D
THREE-DIMENSIONAL IMAGING OF ARM VOLUME
Surface Imaging System: Genauigkeit des Mergings zweier Modelle. 2013. 17. Damstra RJ, Glazenburg EJ, Hop WC. Validation of the inverse water volumetry method: A new gold standard for arm volume measurements. Breast Cancer Res Treat 2006;99:267–273. 18. Pellecchia GL. Figure-of-eight method of measuring hand size: Reliability and concurrent validity. J Hand Ther 2003;16:300–304. 19. Specht MC, Miller CL, Russell TA, Horick N, Skolny MN, O’Toole JA. Defining a threshold for intervention in breast cancer-related lymphedema: What level of arm volume
281
increase predicts progression? Breast Cancer ResTreat 2013;140:485–494.
Address correspondence to: Maaike Erends Department of Plastic Surgery Maastricht University Medical Centre plus (MUMC + ) P. Debyelaan 25 6229 HX, Maastricht Netherlands E-mail:
[email protected]
