Normative Interlimb Impedance Ratios: Implications for Early Diagnosis of Uni- and Bilateral,Upper and Lower Limb Lymphedema

Abstract

Background:

Bioimpedance spectroscopy detects unilateral lymphedema if the ratio of extracellular fluid (ECF) between arms or between legs is outside three standard deviations (SDs) of the normative mean. Detection of bilateral lymphedema, common after bilateral breast or gynecological cancer, is complicated by the unavailability of an unaffected contralateral limb. The objectives of this work were to (1) present normative values for interarm, interleg, and arm-to-leg impedance ratios of ECF and ECF normalized to intracellular fluid (ECF/ICF); (2) evaluate the influence of sex, age, and body mass index on ratios; and (3) describe the normal change in ratios within healthy individuals over time.

Methods:

Data from five studies were combined to generate a normative data set (n = 808) from which mean and SD were calculated for interarm, interleg, and arm-to-leg ratios of ECF and ECF/ICF. The influence of sex, age, and body mass index was evaluated using multiple linear regression, and normative change was calculated for participants with repeated measures by subtracting their lowest ratio from their highest ratio.

Results:

Mean (SD) interarm, interleg, dominant arm-to-leg, and nondominant arm-to-leg ratios were 0.987 (0.067), 1.005 (0.072), 1.129 (0.160), and 1.165 (0.174) for ECF ratios; and 0.957 (0.188), 1.024 (0.183), 1.194 (0.453), and 1.117 (0.367) for ECF/ICF ratios, respectively. Arm-to-leg ratios were significantly affected by sex, age, and body mass index. Mean normative change ranged from 7.2% to 14.7% for ECF ratios and from 14.7% to 67.1% for ECF/ICF ratios.

Conclusion:

These findings provide the necessary platform for extending bioimpedance-based screening beyond unilateral lymphedema.

Introduction

Bioimpedance spectroscopy (BIS) has emerged as a sensitive, specific, and clinic-friendly method for detecting unilateral lymphedema.^1,2 BIS assesses the opposition (as resistance or impedance) to the flow of an electrical current that is primarily conducted by body fluid.³ Current at low frequencies passes through extracellular fluid (ECF) only, while current at high frequencies passes through both intracellular fluid (ICF) and ECF. The measured impedance is inversely proportional to the amount of conductive water and can therefore provide information on the amount of ECF (all body fluids not contained in cells, including lymph, plasma, interstitial, and transcellular fluids), ICF (fluid within the cells), and the combined total body fluid (TBF). However, to date, the application of BIS has relied on the availability of an unaffected, contralateral limb, with which to compare the at-risk limb.

Normative data are available for the ratio of ECF between arms⁴ and the ratio of ECF between legs.⁵ Based on these data, lymphedema is identified if the ratio of ECF between arms or between legs is outside three standard deviations (SDs) of the normative mean (this is equivalent to falling outside approximately 99.7% of a normative sample). The feasibility and clinical utility of this method have been validated and are widely accepted for the early detection of unilateral breast cancer-related lymphedema^6–8 and have shown potential for utility in the unilateral lower limb setting.^5,9 However, this method is unsuitable when both arms or both legs are at-risk or affected, as is the case for bilateral breast and gynecological cancer-related lymphedemas, respectively.^10–15 To address this problem, a proof-of-concept study by Ward et al.¹⁶ investigated the potential for leg ECF/ICF ratios normalized to arm ECF/ICF ratios to detect bilateral lower limb lymphedema. The authors concluded that this approach showed promise but suggested the need for sex, age, and body mass index-related reference ranges, and for a prospective study to determine the minimum change in ratio capable of detecting lymphedema. Other studies that have focused on diagnosing lower limb lymphedema have involved small sample sizes, unilateral lymphedema cases only, or relied on the availability of pretreatment measures.^{5,10,12,17,18} Therefore, the objectives of this work were to (1) present normative values for interarm, interleg, and arm-to-leg ratios of ECF and ECF/ICF; (2) evaluate the influence of sex, age, and body mass index on ratios; and (3) describe the normal change in ratios within healthy individuals over time.

Materials and Methods

Study design and data sources

Data from three cross-sectional studies, designated A (n = 486), B (n = 62), and C (n = 230), and two prospective repeated-measures studies, D (n = 33) and E (n = 87), were pooled for this study. All five studies were comparable, including only consenting participants who were healthy by self-report and excluding those with a history of lymphedema, cancer, surgery, or radiotherapy to the regional lymph nodes. Appropriate ethical approvals were obtained from the relevant institutions for the conduct of all initial studies, as well as the present study (ACTRN1500001072).

Data collection

All contributing studies implemented the same, established, validated protocol, outlined by ImpediMed,¹⁹ to measure impedance to TBF (R_inf), ICF (R_i), and ECF (R₀) in arms and legs using a portable, ImpediMed SFB7, SFB3, or U400 BIS device. Differences between the SFB7 and U400 machines are esthetic, with the U400 implementing the same BIS platform utilized in the SFB7, thus generating equivalent raw impedance values.⁹ SFB3 data (used for data collection in studies A and E) were converted to equivalent SFB7 data using the conversion factors provided by Ward.^20–22 BIS measurements were taken with the participant lying supine with legs apart. For any given measure, two measurement electrodes were placed at either end of the arm or leg (i.e., wrist for arm measures and ankle for leg measures) with a current drive electrode placed approximately 10 cm proximal to the base of the middle finger or toe, respectively, and another placed on the ipsilateral or contralateral foot or hand, respectively.²³ Each limb segment was measured in this manner and the resistances corresponding to ECF (R₀) and to TBF (R_inf) were determined. ICF impedance (R_i) was subsequently calculated with the manufacturer's software (Bioimp 4.15.0.0), which uses the formula: R_i = (R₀ x R_inf)/(R₀ - R_inf). Participants were routinely asked to empty their bladder within the previous hour to data collection, and to avoid a higher than normal intake of alcohol or caffeine, vigorous exercise, or more exercise than usual, in the previous 24 hours. Sex, age, height (measured to the nearest mm using a stadiometer), weight (measured to the nearest 100 g using a calibrated scale), and side of dominance were recorded for each participant. Side of dominance was defined by the dominant arm (preferred hand for writing) due to leg dominance being less easily defined and because most people (∼84%) favor the same foot as hand.^24,25 Furthermore, it has previously been shown that unlike in the arms, the effect of dominance in the legs is minimal.⁵

Statistical analysis

The following interlimb ratios were calculated for each participant: dominant arm/nondominant arm (R₀ and R_i:R₀); dominant leg/nondominant leg (R₀ and R_i:R₀); dominant arm/dominant leg (R₀ and R_i:R₀); and nondominant arm/nondominant leg (R₀ and R_i:R₀). For participants with repeated measures, normative change was calculated for each of the four outcomes by subtracting their lowest ratio from their highest ratio. Graphical inspection of the data revealed thirteen BIS ratios from 10 participants as extreme outliers (i.e., arm-to-leg ratios of <0.25 or >4.0; arm-to-arm or leg-to-leg ratios of <0.5 or >2.0). These values were confirmed as outliers by the generalized extreme Studentized deviate test, individually scrutinized, and then excluded due to suspected measurement error (1% of total participants). A random 10% sample of the normative data (n = 90) was retained for future validation of diagnostic criteria, leaving a total of 808 participants. BIS ratios were log-transformed to satisfy the requirements of a normal distribution for regression analyses. To test for a relationship between individuals within data sets the intraclass correlation was estimated using linear mixed-effects models for each of the outcomes with data set as a random factor and sex, age, and body mass index as fixed factors. Intraclass correlation between data sets was found to be very low ranging from 0.006 to 0.042. The mixed models were then compared with corresponding multiple linear regression models without accounting for data set (i.e., fixed models). Comparison of Akaike information criterion values indicated that multiple regression models were superior for all outcomes. Therefore, data sets were pooled for all remaining analyses, and multiple regression models and plots were used to evaluate associations between factors and ratios of R₀ and R_i:R₀ between limbs. Means and SDs were used to describe the normative ratios and normative change in ratios over time. Between-subject coefficient of variation values were calculated by dividing the SD by the mean for each of the outcomes. Mean normative change was presented as a percentage of the mean normative ratio. Analyses were conducted using R version 3.4.0.²⁶ Mixed-effects models were fit using the lme4 package.²⁷

Results

Characteristics of participants

The 808 participants' ages ranged between 18 and 87 years, 65% were female, 93% right-side dominant, and 51% overweight or obese (Table 1). For the 108 participants with repeated measurements, all were taken over 18 months and the number of measurements ranged from 2 to 22 (median = 3; IQR = 2–7). Participants with repeated measures were similar to the total participants in terms of age (median = 50, range = 21–83), right-side dominance (91%), and body mass index (56% overweight or obese), but consisted of a higher proportion of females (90% vs. 65%).

Table 1.

Characteristics of Participants in Normative Sample

Data set	A (N = 437)	B (N = 57)	C (N = 208)	D (N = 29)	E (N = 77)	Females (N = 526)	Males (N = 282)	Total (N = 808)
Age in years	42 (15)	55 (11)	58 (7)	39 (12)	55 (14)	49 (14)	46 (16)	48 (15)
Mean (SD), range	18, 87	40, 82	44, 69	21, 60	24, 83	18, 85	18, 87	18, 87
BMI (kg/m²)	25.8 (4.5)	27.7 (5.8)	24.4 (3.7)	25.3 (5.3)	26.8 (4.3)	25.5 (4.9)	26 (3.7)	25.6 (4.5)
Mean (SD), range	16.1, 46	16.4, 42.7	17.2, 37.4	17.9, 41.1	18.6, 38.5	16.1, 46	18.5, 40.5	16.1, 46
Female, N (%)	244 (56)	57 (100)	130 (62)	18 (62)	77 (100)	—	—	526 (65)
Right-side dominant, N (%)	414 (95)	49 (86)	194 (93)	27 (93)	70 (91)	490 (93)	264 (94)	754 (93)
BMI categories (kg/m²), N (%)
Under/healthy weight (<25)	207 (47)	19 (33)	127 (61)	14 (48)	31 (40)	280 (53)	118 (42)	398 (49)
Overweight (25.0–29.9)	156 (36)	22 (39)	63 (30)	12 (42)	28 (36)	159 (30)	122 (43)	281 (35)
Obese (30.0+)	74 (17)	16 (28)	18 (9)	3 (10)	18 (24)	87 (17)	42 (15)	129 (16)

N, number of participants; BMI, body mass index; SD, standard deviation; IQR, interquartile range.

Normative interlimb ratios of R₀ and R_i:R₀

Normative means and SDs for interlimb ratios of R₀ and R_i:R₀ are presented for all participants and stratified by sex (Table 2). Comparison of the between-subject coefficient of variation values between outcomes indicates there is approximately twice as much variation among arm-to-leg ratios of R₀ (10.9%–14.9%) compared with arm-to-arm or leg-to-leg ratios of R₀ (5.4%–7.9%). There was greater between-subject variation among R_i:R₀ ratios compared to R₀ values across all limb comparisons (Table 2).

Table 2.

Normative Interlimb Ratios of R₀ (ECF) and R_i:R₀ (ECF/ICF) for All Participants and Stratified by Sex

Sex	N	Mean	SD	% CV	Sex	N	Mean	SD	% CV
Dominant arm R₀/nondominant arm R₀					Dominant arm R_i:R₀/nondominant arm R_i:R₀
All	641	0.987	0.067	6.8	All	639	0.957	0.188	19.6
Female	424	0.989	0.069	7.0	Female	423	0.954	0.227	23.8
Male	217	0.984	0.063	6.4	Male	216	0.963	0.067	7.0
Dominant leg R₀/nondominant leg R₀					Dominant leg R_i:R₀/nondominant leg R_i:R₀
All	642	1.005	0.072	7.2	All	639	1.024	0.183	17.9
Female	426	1.009	0.080	7.9	Female	424	1.027	0.208	20.3
Male	216	0.996	0.054	5.4	Male	215	1.018	0.116	11.4
Dominant arm R₀/dominant leg R₀					Dominant arm R_i:R₀/dominant leg R_i:R₀
All	800	1.129	0.160	14.2	All	796	1.194	0.453	37.9
Female	519	1.178	0.160	13.6	Female	516	1.237	0.491	39.7
Male	281	1.038	0.113	10.9	Male	280	1.114	0.360	32.3
Nondominant arm R₀/nondominant leg R₀					Nondominant arm R_i:R₀/nondominant leg R_i:R₀
All	646	1.165	0.174	14.9	All	643	1.117	0.367	32.9
Female	429	1.214	0.167	13.8	Female	428	1.159	0.388	33.5
Male	217	1.067	0.143	13.4	Male	215	1.032	0.303	29.4

N, number of participants; CV, coefficient of variation.

Effect of sex, age, and body mass index on interlimb ratios of R₀ and R_i:R₀

Visual inspection of the data suggested log-transformed ratios of R₀ and R_i:R₀ between contralateral limbs were not influenced by sex, age, or body mass index (not shown) and multiple regression results supported this observation. Overall, none of the variances in arm-to-arm or leg-to-leg ratios of R₀ or R_i:R₀ was explained by the factors in these models (adjusted R squared values ranged from −0.003 to 0.003).

Figures 1 and 2 show a clear influence of sex on arm-to-leg ratios, with males having consistently lower mean ratios than females. Multiple regression results confirm a statistically significant effect of sex on arm-to-leg ratios of R₀ (Dominant: coefficient = −0.113, p < 0.001; Nondominant: coefficient = −0.118, p < 0.001) and of R_i:R₀ (Dominant: coefficient = −0.121, p < 0.001; Nondominant: coefficient = −0.127, p < 0.001).

FIG. 1.

Log arm-to-leg ratios of R₀ (ECF) and log R_i:R₀ (ECF/ICF) by side of dominance and age. Regression lines were fit separately for males and females. ECF, extracellular fluid; ICF, intracellular fluid.

FIG. 2.

Log arm-to-leg ratios of R₀ (ECF) and log R_i:R₀ (ECF/ICF) by side of dominance and body mass index. Regression lines were fit separately for males and females.

Increasing age was significantly associated with higher arm-to-leg ratios of R₀ (Dominant: coefficient = 0.002, p < 0.001; Nondominant: coefficient = 0.002, p < 0.001) and lower ratios of R_i:R₀ (Dominant: coefficient = −0.008, p < 0.001; Nondominant: coefficient = −0.005, p < 0.001). Figure 2 suggests a stronger effect of aging on arm-to-leg ratios of R₀ in males compared to females. An interaction term between sex and age was included and found to be significant on dominant (coefficient = 0.002, p = 0.002) and nondominant (coefficient = 0.003, p < 0.001) sides of the body. To investigate this difference between the sexes, regression analyses were conducted with raw values of R₀ for individual limbs as the dependent variables. There was no interaction between sex and age for R₀ values in arms, but there was a significant interaction between sex and age for R₀ values in legs (p < 0.01), with R₀ values decreasing more quickly in aging males compared to aging females. There was no sex/age interaction observed for arm-to-leg ratios of R_i:R₀, arm-to-leg ratios of R_i, or for raw R_i values for individual limbs, with age being significantly associated with higher R_i values for all limbs (p < 0.001).

Increasing body mass index was associated with lower arm-to-leg ratios of R₀ (Dominant: coefficient = −0.004, p < 0.001; Nondominant: coefficient = −0.004, p = 0.001) and higher ratios of R_i:R₀ on the dominant side of the body (coefficient = 0.016, p < 0.001), but not on the nondominant side (coefficient = 0.001, p = 0.739).

Regression models with sex, age, and body mass index as independent variables explained 19% of the variance in arm-to-leg ratios of R₀ and 9%–14% of the variance in ratios of arm-to-leg ratios of R_i:R₀. Inclusion of a sex/age interaction term increased the explained variance to 20%–21% for arm-to-leg ratios of R₀. Models, including sex and age, explained 18% for R₀ and 9%–10% for R_i:R₀; while sex alone explained 15% for R₀ and 1%–3% for R_i:R₀. Therefore, arm-to-leg ratios of R₀ and R_i:R₀ are also presented stratified by sex and age (Table 3); and by sex, age, and body mass index (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/lrb).

Table 3.

Normative Arm-to-Leg Ratios of R₀ (ECF) and R_i:R₀ (ECF/ICF) Stratified by Sex and Age

		Arm R₀/leg R₀						Arm R_i:R₀/leg R_i:R₀
		Dominant side			Nondominant side			Dominant side			Nondominant side
Sex	Age (years)	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD
Female	<40	135	1.168	0.112	96	1.202	0.097	135	1.397	0.462	96	1.290	0.408
	40–59	256	1.175	0.162	210	1.213	0.174	253	1.245	0.497	209	1.150	0.383
	60+	128	1.194	0.193	123	1.225	0.196	128	1.055	0.451	123	1.071	0.357
Male	<40	112	1.000	0.094	84	0.999	0.092	112	1.206	0.341	84	1.143	0.374
	40–59	114	1.031	0.090	82	1.069	0.137	114	1.133	0.379	82	1.013	0.226
	60+	55	1.130	0.141	51	1.174	0.157	54	0.884	0.244	49	0.874	0.186

N, number of participants.

Normative change over time in interlimb ratios of R₀ and R_i:R₀

Table 4 summarizes normative change over time for interlimb ratios of R₀ and R_i:R₀ for all participants and participants stratified by sex. The large SDs relative to means indicate high variation in terms of the amount of change measured for different participants (Table 4). Normative change as a percentage of the normative mean was similar among arm-to-arm, leg-to-leg, and arm-to-leg ratios. Ratios of R₀ varied within an individual by up to 14.7% over time, while ratios of R_i:R₀ varied by up to 67.1% (Table 4). Changes in arm-to-leg ratios of R₀ and R_i:R₀ are presented stratified by sex and age in Table 5; and by sex, age, and body mass index in Supplementary Table S2.

Table 4.

Normative Change Over Time in Interlimb Ratios of R₀ (ECF) and R_i:R₀ (ECF/ICF) for All Participants and Stratified by Sex

Sex	N	Mean	SD	% change ^a	Sex	N	Mean	SD	% change ^a
Dominant arm R₀/nondominant arm R₀					Dominant arm R_i:R₀/nondominant arm R_i:R₀
All	103	0.101	0.083	10.2	All	103	0.592	1.558	61.9
Female	92	0.100	0.085	10.1	Female	92	0.640	1.643	67.1
Male	11	0.115	0.067	11.7	Male	11	0.190	0.107	19.7
Dominant leg R₀/nondominant leg R₀					Dominant leg R_i:R₀/nondominant leg R_i:R₀
All	105	0.115	0.117	11.4	All	105	0.428	0.401	41.8
Female	94	0.120	0.122	11.9	Female	94	0.461	0.411	44.9
Male	11	0.072	0.038	7.2	Male	11	0.150	0.075	14.7
Dominant arm R₀/dominant leg R₀					Dominant arm R_i:R₀/dominant leg R_i:R₀
All	101	0.158	0.116	14.0	All	102	0.503	0.390	42.1
Female	90	0.160	0.119	13.6	Female	91	0.531	0.402	42.9
Male	11	0.135	0.080	13.0	Male	11	0.271	0.125	24.3
Nondominant arm R₀/nondominant leg R₀					Nondominant arm R_i:R₀/nondominant leg R_i:R₀
All	103	0.171	0.107	14.7	All	103	0.642	0.550	57.5
Female	92	0.174	0.107	14.3	Female	92	0.689	0.564	59.4
Male	11	0.141	0.106	13.2	Male	11	0.255	0.098	24.7

Mean change as a percentage of the normative mean extracted from Table 2.

N, number of participants.

Table 5.

Change Over Time in Arm-to-Leg Ratios of R₀ (ECF) and R_i:R₀ (ECF/ICF) Stratified by Sex and Age

		Arm R₀/leg R₀						Arm R_i:R₀/leg R_i:R₀
		Dominant side			Nondominant side			Dominant side			Nondominant side
Sex	Age (years)	N	Mean	SD	N	Mean	SD	N	Mean	SD	N	Mean	SD
Female	<40	23	0.193	0.125	23	0.216	0.125	23	0.380	0.198	23	0.611	0.596
	40–59	44	0.159	0.131	44	0.171	0.098	44	0.506	0.394	44	0.625	0.512
	60+	23	0.131	0.080	25	0.142	0.096	24	0.722	0.493	25	0.871	0.601
Male	<40	4	0.115	0.052	4	0.078	0.036	4	0.265	0.139	4	0.222	0.037
	40–59	6	0.147	0.103	6	0.190	0.124	6	0.297	0.124	6	0.300	0.108
	60+	1	0.140	—	1	0.100	—	1	0.140	—	1	0.120	—

N, number of participants.

Discussion

Diagnosis of bilateral lymphedema, which is common after bilateral breast cancer or gynecological cancer, is challenging due to the unavailability of a comparable unaffected, contralateral limb. To advance measurement precision, this study provides novel normative data for arm-to-leg ratios of ECF (R₀) and ECF/ICF (R_i:R₀) for comparison purposes. A key finding of this study was that interlimb ratios of R_i:R₀ showed greater variation between subjects (up to 39.7%) compared with interlimb ratios of R₀ (up to 14.9%). This observation builds on previous findings by Ward et al.,³ who reported greater variation in R_i:R₀ values versus R₀ values for arms (26.2%–27.4% vs. 12.0%–12.9%, respectively). This most likely reflects the larger error associated with estimation of R_i due to inflation of the combined errors associated with determination of R₀ and R_inf.²⁸ Interlimb ratios of R_i:R₀ also showed greater variation within individuals over time. Mean change ranged from 14.7% up to 67.1% for R_i:R₀ ratios, compared with mean change of only 7.2% to 14.7% for R₀ ratios. This suggests that diagnostic methods based on interlimb ratios of R₀ may outperform interlimb ratios of R_i:R₀.

The between-subject variation for arm-to-leg ratios was twice that for ratios between arms or between legs. Variation in arm-to-leg ratios was partially explained by the influences of sex, age, and body mass index, suggesting the need for stratified reference ranges for arm-to-leg ratios. Females had significantly higher mean ratios compared to males, revealing lower ECF in arms relative to legs compared to males. This is in line with previous research showing that females tend to have a lower proportion of their skeletal muscle mass distributed in the upper body.^29,30

Increasing age was significantly associated with arm-to-leg ratios of R₀. This was explained by decreasing values of R₀ in the legs, indicating accumulation of ECF in legs with age. This was more pronounced in males compared to females. It has previously been shown that males experience a larger age-related reduction in arm, leg, and total skeletal muscle compared to females.³¹ Together, these findings suggest that with increasing age, fluid shifts from skeletal muscle mass into the interstitial space and that this shift affects males more than females, and the legs more than the arms, most likely due to age-related vascular insufficiency.³²

Although increased body mass index was significantly associated with decreased arm-to-leg ratios of R₀ on both sides of the body and increased R_i:R₀ ratios on the dominant side, regression analyses suggest that BMI is less important than sex or age (Figs. 1 and 2). Based on these results, we suggest comparing data derived from at-risk populations to reference values stratified by sex and age (e.g., <40 years, 40–59 years, 60+ years).

Interarm ratios were not influenced by sex, age, or body mass index, suggesting that stratification for these factors is unnecessary in the assessment of unilateral upper limb lymphedema. This is in line with current clinical practice. Furthermore, the interarm ratios of R₀ for females presented here (0.989 ± 0.069) are similar to those published previously (0.986 ± 0.040 and 0.964 ± 0.034).³ However, sex was found to be associated with interleg ratios, with being male associated with lower R₀ ratios (0.996 ± 0.054) compared to females (1.009 ± 0.080). A significant association between sex and interleg ratios was also reported by Ward et al.,⁵ who have previously published mean interleg ratios of R₀ for males (0.994 ± 0.050) and females (0.999 ± 0.056). Despite both studies finding a significant effect of sex, the difference in means between females and males was small, equating to 1.3% for the current work and 0.5% for Ward et al.⁵ making it unclear whether stratification by sex is necessary for interleg ratios. Nevertheless, given that comparing ECF between contralateral limbs is unsuitable for assessing bilateral lymphedema, which occurs in two-thirds of lower limb lymphedema cases,^10–13 we predict that the use of arm-to-leg ratios will be more appropriate in the lower limb setting.

Future studies are required to compare the stratified normative data presented here with comparable data from a cohort of lymphedema cases to optimize diagnostic criteria. In addition to testing cutoffs based on the mean and SD calculated from stratified normative data, future research should investigate whether multivariable logistic regression or network-based discrimination methods can improve diagnostic accuracy. Furthermore, if routine bioimpedance assessment is extended to at-risk populations beyond the breast cancer clinic, information on whether a subject's change from baseline (or earliest measurement) exceeds normative change (e.g., 13.0%–14.3% for arm-to-leg ratios dependent on sex and dominance) could be used to further improve diagnostic accuracy.

Although the five studies all recruited healthy volunteers, samples were a mixture of population-based and convenience samples. Volunteers who come forward for a lymphedema-related study may differ from the general population. Another limitation of this study was the small sample size available for assessment of normative change, especially for males. It was not possible to investigate the effects of ethnicity or mixed laterality (e.g., right handed and left footed) on ratios in this study due to these data not being collected in all of the included studies.

In conclusion, this study has described the influence of sex, age, and body mass index on interlimb ratios of ECF and ECF/ICF, and presented stratified normative data and normative change data according to those factors. These findings provide the necessary platform for future research seeking to optimize early diagnosis of bilateral lymphedema and to extend bioimpedance-based screening beyond unilateral, cancer-related, upper limb lymphedema.

Footnotes

Acknowledgments

We thank and acknowledge the following organizations for contribution of funds that enabled this work to be completed: ImpediMed Limited (funds supported the data analysis for this body of work but not the original data collection studies); Cancer Council Queensland (SH fellowship and RB grant), National Health and Medical Research Council (MJ fellowship), Department of Health and Aging (RB grant), Fonterra Co-operative Group, and the New Zealand Ministry for Primary Industries (funds supported the New Zealand Primary Growth Partnership Post-Farm Gate Dairy Programme). Funding received had no influence over data analysis and reporting.

Author Disclosure Statement

No competing financial interests exist. LW provides consultancy services to ImpediMed Limited. ImpediMed Limited donated dual-tab electrodes to SG. SDP holds the Fonterra Chair in Human Nutrition at the University of Auckland. AR is an employee of Fonterra Co-operative Group. A.O. is the founder and managing director of SurgicalPerformance Pty Ltd., an Australian, private company that provides surgeons with a platform for collection of surgical outcome audit data; A.O. received travel grants from the O.R. Company (formerly Tyco Healthcare) and is a consultant for Covidien, NSW, Australia.

References

Smoot

, Wong

, Dodd

. Comparison of diagnostic accuracy of clinical measures of breast cancer-related lymphedema: Area under the curve. Arch Phys Med Rehabil, 2011; 92:603–610.

Czerniec

, Ward

, Lee

, Refshauge

, Beith

, Kilbreath

. Segmental measurement of breast cancer-related arm lymphoedema using perometry and bioimpedance spectroscopy. Support Care Cancer, 2011; 19:703–710.

Ward

, Dylke

, Czerniec

, Isenring

, Kilbreath

. Confirmation of the reference impedance ratios used for assessment of breast cancer-related lymphedema by bioelectrical impedance spectroscopy. Lymphat Res Biol, 2011; 9:47–51.

Cornish

, Chapman

, Hirst

, et al. Early diagnosis of lymphedema using multiple frequency bioimpedance. Lymphology, 2001; 34:2–11.

Ward

, Dylke

, Czerniec

, Isenring

, Kilbreath

. Reference ranges for assessment of unilateral lymphedema in legs by bioelectrical impedance spectroscopy. Lymphat Res Biol, 2011; 9:43–46.

Laidley

, Anglin

. The impact of L-Dex((R)) measurements in assessing breast cancer-related lymphedema as part of routine clinical practice. Front Oncol, 2016; 6:192.

Shah

, Arthur

, Wazer

, Khan

, Ridner

, Vicini

. The impact of early detection and intervention of breast cancer-related lymphedema: A systematic review. Cancer Med, 2016; 5:1154–1162.

Soran

, Ozmen

, McGuire

, et al. The importance of detection of subclinical lymphedema for the prevention of breast cancer-related clinical lymphedema after axillary lymph node dissection; a prospective observational study. Lymphat Res Biol, 2014; 12:289–294.

Gaw

, Box

, Cornish

. Bioimpedance in the assessment of unilateral lymphedema of a limb: The optimal frequency. Lymphat Res Biol, 2011; 9:93–99.

10.

Suehiro

, Morikage

, Yamashita

, et al. Distribution of extracellular fluid in legs with venous edema and lymphedema. Lymphat Res Biol, 2016; 14:156–161.

11.

Abu-Rustum

, Alektiar

, Iasonos

, et al. The incidence of symptomatic lower-extremity lymphedema following treatment of uterine corpus malignancies: A 12-year experience at Memorial Sloan-Kettering Cancer Center. Gynecol Oncol, 2006; 103:714–718.

12.

Halaska

, Novackova

, Mala

, et al. A prospective study of postoperative lymphedema after surgery for cervical cancer. Int J Gynecol Cancer, 2010; 20:900–904.

13.

Cambria

, Gloviczki

, Naessens

, Wahner

. Noninvasive evaluation of the lymphatic system with lymphoscintigraphy: A prospective, semiquantitative analysis in 386 extremities. J Vasc Surg, 1993; 18:773–782.

14.

Warren

, Janz

, Slavin

, Borud

. The use of bioimpedance analysis to evaluate lymphedema. Ann Plast Surg, 2007; 58:541–543.

15.

Miller

, Specht

, Horick

, et al. A novel, validated method to quantify breast cancer-related lymphedema (BCRL) following bilateral breast surgery. Lymphology, 2013; 46:64–74.

16.

Ward

, Winall

, Isenring

, et al. Assessment of bilateral limb lymphedema by bioelectrical impedance spectroscopy. Int J Gynecol Cancer, 2011; 21:409–418.

17.

Suehiro

, Morikage

, Harada

, et al. Application of the L-Dex score for the assessment of bilateral leg edema. Lymphat Res Biol, 2018; 16:65–68.

18.

Novackova

, Halaska

, Robova

, et al. A prospective study in detection of lower-limb lymphedema and evaluation of quality of life after vulvar cancer surgery. Int J Gynecol Cancer, 2012; 22:1081–1088.

19.

Impedimed Ltd. Impedimed. 2017. www.impedimed.com/. Accessed October 17, 2017.

20.

Hayes

, Battistutta

, Newman

. Objective and subjective upper body function six months following diagnosis of breast cancer. Breast Cancer Res Treat, 2005; 94:1–10.

21.

Hayes

, Cornish

, Newman

. Comparison of methods to diagnose lymphoedema among breast cancer survivors: 6-month follow-up. Breast Cancer Res Treat, 2005; 89:221–226.

22.

Ward

. Inter-instrument comparison of bioimpedance spectroscopic analysers. Open Med Devices J, 2009; 1:3–10.

23.

Cornish

, Jacobs

, Thomas

, Ward

. Optimizing electrode sites for segmental bioimpedance measurements. Physiol Meas, 1999; 20:241–250.

24.

Kumar

, Misra

, Suman

, Suar

, Mandal

. Interrelationship of limb dominance and sensory function across age. Int J Neurosci, 2010; 120:110–114.

25.

Porac

, Coren

. Lateral Preferences and Human Behavior. New York: Springer-Verlag; 1981.

26.

R Core Team

. A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2017.

27.

Bates

, Maechler

, Bolker

, Walker

. Fitting linear mixed-effects models using lme4. J Stat Software, 2015; 67:1–48.

28.

Ward

, Byrne

, Rutter

, et al. Reliability of multiple frequency bioelectrical impedance analysis: An intermachine comparison. Am J Hum Biol, 1997; 9:63–72.

29.

Miller

, MacDougall

, Tarnopolsky

, Sale

. Gender differences in strength and muscle fiber characteristics. Eur J Appl Physiol Occup Physiol, 1993; 66:254–262.

30.

Janssen

, Heymsfield

, Wang

, Ross

. Skeletal muscle mass and distribution in 468 men and women aged 18–88 yr. J Appl Physiol (1985), 2000; 89:81–88.

31.

Gallagher

, Visser

, De Meersman

, et al. Appendicular skeletal muscle mass: Effects of age, gender, and ethnicity. J Appl Physiol (1985), 1997; 83:229–239.

32.

Beebe-Dimmer

, Pfeifer

, Engle

, Schottenfeld

. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol, 2005; 15:175–184.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

Normative Interlimb Impedance Ratios: Implications for Early Diagnosis of Uni- and Bilateral,Upper and Lower Limb Lymphedema

Abstract

Abstract

Background:

Methods:

Results:

Conclusion:

Introduction

Materials and Methods

Study design and data sources

Data collection

Statistical analysis

Results

Characteristics of participants

Normative interlimb ratios of R0 and Ri:R0

Effect of sex, age, and body mass index on interlimb ratios of R0 and Ri:R0

Normative change over time in interlimb ratios of R0 and Ri:R0

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

Supplementary Material

Normative interlimb ratios of R₀ and R_i:R₀

Effect of sex, age, and body mass index on interlimb ratios of R₀ and R_i:R₀

Normative change over time in interlimb ratios of R₀ and R_i:R₀