Abstract
Background:
The use of bioimpedance spectroscopy (BIS) for the detection of unilateral upper limb lymphedema is growing. Currently, normatively determined diagnostic thresholds using bioimpedance are available for females only. It is unclear if they are suitable for males at risk of unilateral upper limb lymphedema. The aim of the present study was to determine normatively based bioimpedance thresholds for male unilateral upper limb lymphedema.
Methods and Results:
Impedance values (R0) for the upper limbs of 60 healthy adult male participants were assessed using BIS. As has been found in females, dominance was found to significantly impact on R0 values (p = 0.002); therefore, two diagnostic thresholds are required depending on the at-risk limb. Interlimb impedance thresholds were calculated, set at both two standard deviations (2SD) and three standard deviations (3SD) above the mean. The 3SD threshold for unilateral lymphedema in the upper limbs of males is 1.127 for the dominant at-risk limb and 1.094 for the nondominant at-risk limb. The 2SD threshold is 1.090 for the dominant at-risk limb and 1.058 for the nondominant at-risk limb. These differ from what has previously been found for females.
Conclusions:
This study is the first to establish normatively determined bioimpedance thresholds for male unilateral upper limb lymphedema. These thresholds should now be used to improve early detection of male unilateral upper limb lymphedema.
Introduction
Lymphedema is chronic swelling of a body region resulting from either an excess in lymph fluid in the lymphatic system or a deficit in the function of lymphatic pathways. 1 Excess fluid accumulates as a result of an impairment in the drainage capacity of the lymphatic system. 2 For some, abnormalities within the lymph system results in disease progression to adipose and fibrotic tissue hypertrophy.3,4 There is currently no cure for this condition, and management can be particularly difficult once the disease has progressed. 5 Consequently, early diagnosis is advantageous and may enable treatment to commence at a more efficacious stage of the disease. 6
Many tools are available for the measurement of limbs with lymphedema. Bioimpedance spectroscopy (BIS) is a noninvasive device that has been shown to be effective for early diagnosis of breast cancer-related lymphedema in females.6,7 BIS works by passing an imperceptible electrical current through the extracellular fluid compartment of the body, of which lymph is a component.8,9 At a high frequency, the electrical current passes through both the extracellular and intracellular space. 10 However, at low frequencies, the current preferentially passes through the extracellular space. 10 As the current travels through the extracellular compartment, it encounters electrical resistance and reactance due to the presence of bodily fluids, cell membranes, and tissue, which is referred to as impedance. 11 Impedance has an inversely proportional relationship to extracellular fluid volume. 10 Hence, lower impedance values reflect a higher volume of extracellular fluid. BIS has the advantage of being able to directly measure extracellular fluid compared with other methods of lymphedema diagnosis, such as volume and circumference measurements—which may be influenced by unrelated changes such as alterations to muscle, fat, or other fluid types. 12
BIS has been established as both reliable and sensitive tool in lymphedema populations. 13 In the female population, BIS has an inter-rater reliability of 0.987 and an intra-reliability of 0.99312 and has excellent sensitivity (93%–96%) and good specificity (67%–87%) 14 when used for the detection of mild lymphedema. Detection of lymphedema with BIS relies on normatively determined thresholds.8,14 Traditionally, BIS interarm reference thresholds have been set at three standard deviations (3SD) above the mean interarm impedance ratio derived from a control female population. 10 More recently, however, a diagnostic threshold of two standard deviations (2SD) above the mean interarm ratio has also been found to be more sensitive and specific for the detection of mild lymphedema than the higher 3SD threshold. 14 Both the 2SD and 3SD thresholds currently in use have been derived from a sample of healthy adult females. 8 However, males may also develop unilateral upper limb lymphedema as a consequence of cancer treatments, such as surgery and/or radiotherapy for melanoma and breast cancer. 15 Thus far, it is unknown whether established female thresholds are suitable for males. In the lower limbs, the diagnostic thresholds for females and males are different 16 ; therefore, males may require a different threshold from females in the upper limbs as well. Furthermore, it has been established that two different diagnostic thresholds are required in the female population, depending on whether the dominant or nondominant arm is at risk of lymphedema.8,10 It is unknown whether this also holds true for the male population.
The aim of this study, therefore, was to determine the 2SD and 3SD normative BIS diagnostic thresholds for the early detection of unilateral upper limb lymphedema in an adult male population.
Materials and Methods
Participants
Sixty adult male volunteers were recruited from Sydney. The inclusion criteria were healthy male participants who were aged 18 years or older. Participants with metal implants in either of the arms or trunk, such as an indwelling pacemaker or defibrillator, were excluded. Those who had a history of medical conditions or were using medications that may influence limb volumes, for example, congestive heart failure, existing lymphedema, or diuretics, were also excluded from the study.
Informed written consent was acquired from all participants before commencement of the BIS assessment. This study (protocol number: 2016/1009, 12-2009/12400) was approved by the University of Sydney Human Research Ethics Committee.
Measurement of limb impedance
All participants attended a single assessment session. Height was measured to the nearest 0.1 m with a tape measure placed vertically on the wall, and an analogue scale was used to assess weight to the nearest 0.5 kg. A questionnaire regarding medical history, medication, and self-reported hand dominance was also completed.
Impedance measurements were acquired from both upper limbs using the BIS device (SFB-7; Impedimed Ltd., Brisbane, Australia). Participants were measured according to the measurement protocol detailed by the manufacturer. Participants were positioned in supine on a flat nonconductive surface with slight abduction of all limbs. Sites of electrode placement were cleaned with alcohol wipes, and any accessories or conductive items located close to hands and feet were removed before attachment of Ag-AgCl electrodes. Four electrodes were utilized for each measurement, which consisted of two current generating drive electrodes and two measurement electrodes placed in a tetrapolar arrangement. One drive electrode was placed in between the second and third metatarsal of the right foot, and the second drive electrode at the third metacarpal of the dorsum of the hand. The measurement electrodes were placed at the ulnar styloid processes of each upper limb. The right arm was measured first, followed by the left arm. To ensure ongoing quality of assessments, the BIS device and connecting electrodes were regularly tested between data collection sessions with the manufacturer's test cell.
Data processing
Each BIS file was uploaded, then processed and analyzed according to the Cole theory 17 to extrapolate the impedance value at R0 through the use of Bioimp (v.5.2.4.0; Impedimed Ltd.) software.
Statistical analysis
The mean, standard deviation, and range for participants' characteristics were calculated from collected data (Table 1). Group mean and standard deviations of impedance values at R0 were calculated for both dominant and nondominant limbs using the Statistical Package for the Social Sciences (SPSS, version 24 for Windows by IBM). Paired t-tests were utilized to test for any significant difference of the R0 values between the dominant and nondominant upper limbs. Interarm impedance ratios were then calculated using SPSS for each participant according to hand dominance. When the dominant limb is at risk, the impedance ratio was determined by
Participant Characteristics
Mean ± SD (range).
BMI, body mass index.
For the nondominant at-risk limb, the ratio was determined by
Normality of data for the distribution of interarm ratios was assessed using the Shapiro–Wilk test, and frequency distribution graphs and normality plots were generated for visual inspection of normality. The mean and standard deviation for the interlimb ratios for both the dominant and nondominant at-risk limbs were calculated. Reference thresholds for the diagnosis of unilateral upper limb lymphedema were calculated for both 2SD and 3SD above the mean. Finally, using SPSS, correlation analysis was performed to determine whether final interarm ratios were affected by participants' characteristics of age and body mass index (BMI).
Results
The impedance values found for R0 differed significantly between the dominant and nondominant upper limbs (p = 0.002; Table 2). Two thresholds were therefore required, depending on whether the at-risk limb is the dominant or nondominant arm. The mean interlimb impedance ratios for both the dominant and nondominant arm were calculated (Table 2).
Raw Impedance Values and Calculated Interlimb Mean Impedance Ratios
Mean ± SD.
Paired t-test, p = 0.002.
Interarm ratios were normally distributed as indicated by the Shapiro–Wilk test for both the dominant (p = 0.438) and nondominant (p = 0.605) at-risk limbs.
The 3SD normatively determined diagnostic threshold for the determination of male unilateral upper limb lymphedema was calculated to be 1.127 if the dominant limb was at risk compared with 1.094 for the nondominant limb. The 2SD threshold was 1.090 and 1.058 for the dominant and nondominant limb, respectively.
Neither age (p = 0.319, r = −0.131) nor BMI (p = 0.506, r = 0.088) was found to have a significant correlation with the impedance ratios in this population.
Discussion
This study has, for the first time, established the BIS diagnostic thresholds for the assessment of unilateral upper limb lymphedema in the male population. Currently, there are no known diagnostic criteria published for this population. This study therefore represents a significant step forward in the improvement of diagnosis for early unilateral upper limb lymphedema in males.
As has been found for the female and male lower limb impedance ratios, 16 male impedance ratios were not influenced by participant characteristics such as age or BMI. However, limb dominance did influence the measured impedance. Two different thresholds are required depending on whether the individual's at-risk limb is dominant or nondominant. The determined male thresholds from the current study are different from those found for the female population. The 3SD female diagnostic thresholds are 1.136 for the dominant at-risk arm and 1.066 for the nondominant at-risk arm, 8 in comparison with the male thresholds of 1.127 and 1.094, respectively. These findings have clinical implications. If female thresholds are inappropriately applied to a male, the existing female thresholds may yield false-positive results with nondominant at-risk limbs and overdiagnose the male population. Similarly, for the dominant arm, males may be underdiagnosed using the female normative threshold. This supports the need for the adoption of male-specific diagnostic thresholds. With diagnostic criteria established for both genders, this discrepancy can be avoided.
As has been undertaken after the determination of the female normatively determined thresholds,10,14 further research is required to confirm current determined thresholds. While this study analyzed the influence of BMI and age on male diagnostic thresholds, other participant characteristics, such as height and weight were not analyzed due to sample size. Future studies in male unilateral upper limb lymphedema would benefit from a larger sample size, to achieve sufficient power for further correlation analyses between participant characteristics and impedance ratios. To confirm the utility of the thresholds established in this study, further research is also required to verify these ratios against males who have a confirmed diagnosis of unilateral upper limb lymphedema and males at risk of lymphedema. Furthermore, normative thresholds for segmental bioimpedance should also be determined as for females with early localized lymphedema where whole-arm BIS thresholds may not yet be exceeded, 18 which would delay diagnosis and intervention.
Conclusions
In conclusion, normatively determined BIS diagnostic thresholds for unilateral upper limb lymphedema in the male population have been established. These should now be utilized for early detection of lymphedema in at-risk males.
Footnotes
Author Disclosure Statement
No competing financial interests exist.