Innovative Methodology
J Appl Physiol 106: 268 –273, 2009.
First published November 20, 2008; doi:10.1152/japplphysiol.90435.2008.
Measuring partial body potassium in the legs of patients with spinal cord
injury: a new approach
L. Wielopolski,1 L. M. Ramirez,1 A. M. Spungen,2,3 S. Swaby,2 P. Asselin,2 and W. A. Bauman2,3
1
Brookhaven National Laboratory, Environmental Science Department, Upton, New York; 2Department of Veterans Affairs
Rehabilitation Research and Development Service, Merit Review Program and the Center of Excellence for the Medical
Consequences of Spinal Cord Injury; and Medical and Research Services, Veterans Affairs Medical Center, Bronx, New
York; and 3Departments of Medicine and Rehabilitation Medicine, Mount Sinai Medical School, New York, New York
Submitted 21 March 2008; accepted in final form 12 November 2008
of acute spinal cord injury (SCI) includes an
unbridled catabolism, resulting in significant loss of lean tissue
(LT) mass, regardless of nutritional intake. In part, it is attributed to muscle atrophy that is secondary to paralysis below the
level of the lesion. However, muscle loss above the lesion is
due to the stress of the main injury, associated injuries, and
generalized disuse from inherent immobilization after an acute
traumatic event. The extent of loss of viable muscle tissue
following the trauma may, in large measure, determine the length
of time required for rehabilitation—that is, the greater the atrophy of the remaining innervated skeletal muscle (SM) mass,
the longer will be the recovery process.
Because of the difficulties of undertaking sophisticated regional body composition analysis immediately after the catastrophic traumatic event, there is limited knowledge of the
soft-tissue changes. In longitudinal studies of 31 SCI patients
using dual-energy x-ray absorptiometry (DXA), Wilmet et al.
(26) found that there was a rapid 15% decrease in lean mass of
the legs within 1 year; notably, patients with spastic paralysis,
in which motor control is nonvoluntary, lost less lean mass in
their legs than did those with flaccid paralysis. Patients realized
a 30% increase in lean mass in the arms after 6 mo of
rehabilitation (26); however, because initial measurements
were made ⬃10 wk after the acute event, this apparent increase
may have been the result of an equivalent loss (i.e., 30%) in LT
immediately postinjury. Spungen et al. (19) using DXA studied
monozygotic twins discordant for chronic SCI and found no
difference in the lean (or fat) tissue in the arms of the twin
pairs, also supporting the likely possibility that the increase in
the LT of the arms reported by Wilmet et al. (26) was merely
a return to baseline/preinjury status. Rossier et al. (14) studied
17 subjects with SCI within 1 mo after injury and again 2 to 12
mo later demonstrated a significant depletion of potassium and
loss of weight of lean body mass. Individuals with chronic SCI
have a lower energy expenditure that is directly related to the
level of neurological deficit and their LT mass. Spungen et al.
(17, 18) reported a relationship between metabolic rate and
body cell mass in 12 SCI patients. The greater the loss of lean
body tissue, the greater the decline in their resting metabolic
rate (19). Mollinger et al. (9) also described a 12–29% reduction from the predicted values for basal energy expenditure
(BEE) in 48 people with SCI; those with higher levels of injury
and, presumably, less LT mass exhibited the larger reductions
in BEE.
In acute stress, nitrogen wasting may occur despite aggressive nutritional support. Streat et al. (20) reported that eight
patients with sepsis in an intensive care unit who received
ample caloric and protein intake lost 1.5 kg of body protein
over 10 days. This study and others strongly suggest that there
may be substantial losses in lean body tissue during severe
states of stress or injury. In such catabolic states, endogenous
anabolic hormones, growth hormone, and testosterone are
markedly depressed. These anabolic hormone levels are also
low in persons with acute and chronic SCI (1, 3, 4, 10, 21, 22).
Thus an unfavorable metabolic milieu may contribute to the
loss of muscle tissue caused by paralysis and immobilization,
Address for reprint requests and other correspondence: L. Wielopolski,
Brookhaven National Laboratory, Bldg. 490D, Upton, NY 11973 (e-mail:
[email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
in vivo; skeletal muscle; body composition; paraplegia
ONE OF THE SEQUELS
268
http://www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
Wielopolski L, Ramirez LM, Spungen AM, Swaby S, Asselin
P, Bauman WA. Measuring partial body potassium in the legs of
patients with spinal cord injury: a new approach. J Appl Physiol
106: 268 –273, 2009. First published November 20, 2008;
doi:10.1152/japplphysiol.90435.2008.—Patients with acute spinal
cord injury (SCI) with paralysis experience rapid and marked muscle
atrophy below the level of the lesion. Muscle is lost above the lesion
due to enforced bed rest associated with immobilization. Presently,
there is no viable method to quantify muscle loss between the time of
injury to the initiation of rehabilitation and remobilization. Furthermore, to assess the efficacy of any physical or pharmacological
intervention necessitates the ability to accurately determine the impact
of these treatments on muscle mass and function. Our results are
presented from measurements of regional potassium (K) in the legs of
persons with chronic SCI. The intracellular body K, comprising
⬃97% of the total body K, is indicative of the metabolically active
cell mass, of which over 50% is located in the skeletal muscle (SM).
To assess regional variations in SM mass in the legs, a partial body K
(PBK) system designed for this purpose was placed on a potentially
mobile cart. The SM mass measured by PBK in an able-bodied control
cohort (n ⫽ 17) and in patients with chronic SCI (n ⫽ 21) was 17.6 ⫾
0.86 and 11.0 ⫾ 0.65 kg, respectively, a difference of ⬃37.5%.
However, the difference in the lean tissue mass of the legs obtained by
dual-energy absorptiometry (DXA) in the same cohorts was 20.5 ⫾
0.86 and 15.5 ⫾ 0.88 kg, respectively, or a difference of ⬃24.4%.
PBK offers a novel approach to obtain regional K measurements in
the legs, thus allowing the potential for early and serial assessment of
muscle loss in SCI subjects during the acute and subacute periods
following paralysis. The basic characteristics and performance of our
PBK system and our calibration procedure are described in this
preliminary report.
Innovative Methodology
269
PARTIAL BODY POTASSIUM IN THE LEGS IN SCI
METHODS
Subjects. Volunteers were drawn from three groups: 17 healthy
able-bodied subjects, 10 subjects with paraplegia, and 11 subjects
with tetraplegia. These last two cohorts were analyzed separately and
as a single combined SCI group. The groups were of similar age and
weight; time after injury (duration of injury, DOI) and body mass
index (BMI) are also presented (Table 1). There were four obese
subjects in both the control (BMI ⬎ 30 kg/m2) and SCI (BMI ⬎ 26
kg/m2) groups (17, 23). The study was approved by the Institutional
Review Board of the James J. Peters Veterans Affairs Medical Center,
and informed consent was obtained from subjects before initiating the
testing.
Potassium measurement and analysis. Potassium measurement is
based on the long-lived, ⬃109-yr half-life, natural radioisotope 40K
that is in a dynamic equilibrium of 0.0117% with the two stable
isotopes of K, viz., 39K and 41K (2). A complex decay scheme to the
ground state of 40Ar emits isotopically monoenergetic 1.46-MeV
gamma rays at a rate of ⬃200 gamma rays/s per gram of natural K. In
our mobile PBK system, gamma rays are detected by four standard
thallium-doped sodium iodide (NaI[Tl]) detectors that form a detection cavity and stand in a lead shielding block that reduces the
background radiation. The system is mounted on a cart placed adjacent to the foot of a bed (Fig. 1); the system configuration is such that
K contribution from a subject’s trunk is below the detectable change.
The detectors, placed below subject’s legs, were parallelepipeds with
dimensions of 10.1 cm by 10.1 cm by 40.6 cm long; the other two
detectors placed above the legs were cylindrical with dimensions of
15.2 cm in diameter and 7.5 cm in height. Standard nuclear spectroscopy electronics coupled to a multichannel analyzer process the
detectors’ signals, displaying them as a spectral histogram. For the
Fig. 1. A setup for partial body potassium measurement of the legs. The
system is mounted on a cart that is adjacent to the lower end of a bed. The
positioning tray for the subject has been inserted into the counting chamber.
measurements all the subjects were placed in a supine position on a
1.9-cm-thick plastic tray that was slid into the 25-cm-high, 50-cmwide, and 100-cm-long detection cavity until stopped by a cavity
support post, visible in Fig. 1, at the crotch; counts were obtained for
30 min. The background potassium spectrum measured in the room
before taking measurements on a subject is shown in Fig. 2. The
spectrum is divided into three regions: the Compton region is to the left
of the 40K peak, which is our region of interest (ROI), and the highenergy (HE) region is to the right of the ROI. To assess subject’s K,
the radiation in the background (KB) was measured using phantom
legs (thighs and calves) filled with deionized water placed inside the
counter, immediately before and after measuring the subject. The
mean value was subtracted, in the ROI, from the total measured
subject counts (KT), thus yielding the net counts for a subject (KN):
K N ⫽ KT ⫺ KB
(1)
and the error in the net measurement is given by ␴N
␴ N ⫽ 冑共KT ⫹ KB兲
(2)
KN, based on phantom measurement, may also be used to account for
short-term variations in the background or for instabilities (drifts) in the
counter. A slope of the calibration line (mcal; counts per g K per unit
time) is derived from a bottle manikin absorption (BOMAB) phantom
containing a known amount of K, 100 g (Fig. 3), and is given by
m cal ⫽ 共KBOMAB ⫺ KB兲/共K兲
(3)
Equation 3 also defines the system sensitivity s and was determined to
be 175 counts per g K per 30 min. The net counts in the ROI are
Table 1. Demographic characteristics of the control, SCI-paraplegia, SCI-tetraplegia, and SCI-combined groups
Controls (n ⫽ 17)
Age, yr
DOI, yr
Weight, kg
BMI, kg/m2
SCI-Paraplegia (n ⫽ 10)
SCI-Tetraplegia (n ⫽ 11)
SCI-Combined (n ⫽ 21)
Mean
SD
Range
Mean
SD
Range
Mean
SD
Range
Mean
SD
Range
43.7
11.8
20–60
86.6
28.0
16.8
5.1
54.5–115.7
21.3–39.6
52.5
14.6
87.5
27.4
10.6
8.5
23.9
5.6
29–65
4–27
54.7–126.1
20.5–36.9
50.2
10.7
75.8
24.5
13.5
10.6
17.0
5.5
23–71
1–29
48.6–106.5
16.8–36.5
51.3
12.6
81.3
25.9
12.0
9.7
20.4
5.6
23–71
1–29
48.6–126.1
16.8–36.9
SCI, spinal cord injury; DOI, duration of injury; BMI, body mass index. SCI-combined group refers to SCI subjects with paraplegia and SCI subjects
with tetraplegia.
J Appl Physiol • VOL
106 • JANUARY 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
exacerbating the condition. Undoubtedly, accurate knowledge
of the amount and rate of muscle tissue loss following acute
SCI would be beneficial in devising and testing treatments to
preserve muscle mass.
Potassium (K) is an essential element in the human body,
totaling ⬃140 g in adults (16), of which ⬃97% is intracellular.
To maintain the proper potential of the cell membrane, the
amount of K is tightly controlled homeostatically and is distributed uniformly within the metabolically active cells (16).
Consequently, K is a very good indicator of the body cell mass
(BCM) and of intracellular water (11). We propose to use
regional potassium measurements of the legs to directly measure the BCM and the changes in the SM mass affected by the
SCI. Our partial body potassium (PBK) system is mounted on
a cart and in the future has the potential to be wheeled beside
the patient’s bed; Wielopolski et al. (25) earlier constructed a
stationary system for measuring K in the arm. We demonstrate
here the suitability of our leg PBK system and report the
preliminary results of a comparison of two cohorts: healthy
controls and patients with chronic SCI.
Innovative Methodology
270
PARTIAL BODY POTASSIUM IN THE LEGS IN SCI
mounted around each leg of a BOMAB phantom, simulating a layer
of fat in which K is not present.
Finally, we compared PBK results with those obtained from DXA.
For this purpose the DXA images were acquired on a GE Lunar
Prodigy Advance DXA. (GE Lunar enCore2004, version 8.8, Madison, WI). Software algorithms were applied to isolate the legs’ ROI
for quantification of LT mass. The legs were graphically displayed,
and the operator adjusted the final cut lines on each subject. All cuts
were performed by a single investigator to avoid interrater variability.
RESULTS
Fig. 2. Representative background potassium spectrum that was acquired from
a room. ROI, region of interest; HE, high energy.
K Legs ⫽ KN/mcal
(4)
Thus evaluated, the net potassium signal is proportional to the potassium mass in subject’s legs that subsequently is converted into the SM
mass. The linearity of the signal with K content has been published
elsewhere and is not discussed here (5, 12, 15).
Similarly, in a preliminary attempt to study the effect of obesity on
the K signal, a 2.5-cm-thick cylinder filled with plain water was
Fig. 3. A whole body bottle manikin absorption (BOMAB) phantom is
displayed. Only the legs were used in the work described in this report.
J Appl Physiol • VOL
Fig. 4. Stability of the partial body potassium (PBK) system over a period of
1 yr. The total number of counts in the ROI of a phantom with (E) and without
100 g K (‚); ƒ represent the net counts, that is, the difference between these
two. The background is offset by 6,000 counts. The solid vertical line
represents the realignment of the detectors (RESULTS, first paragraph).
106 • JANUARY 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
converted into grams potassium in the legs by dividing KN by the
slope of the calibration line
Because the natural background may fluctuate, we monitored the reproducibility of the background in a single location
and the net K counts from a phantom for a year. We noticed
that one of the detectors was misaligned and because the four
detectors are connected to a summing amplifier, small fluctuations affected the counts between two fixed channels. Once
the detector was aligned with the other detectors, there was a
significant reduction in the percent standard deviation (SD) by
a factor of five. The plots with the results are displayed where
the mean and SD in the ROI for the period following alignment
are 19,804 ⫾ 399, 37,257 ⫾ 572, and 17,495 ⫾ 522, respectively, for the background, total, and net counts (Fig. 4). Figure
4 also plots the means and the SDs before the alignment; the
reduction in the SD is apparent by comparing the daily background counts before and after alignment. For a mobile system,
the background may also vary from room to room.
To evaluate for obesity, we wrapped the phantom with a 2.5-cm
layer of insulating water, a fraction of the K gamma rays emitted from
the leg scatter in this extra outer layer losing energy and registering in
the Compton region (see Fig. 2). Thus the losses in the ROI are
partially compensated for by gains registered in the Compton region,
with some being totally lost due to multiple scatterings. To gauge the
scatter, three different set-ups for phantom measurements were obtained: 1) water phantom with an overlay, 2) phantom with K,
and 3) phantom with K and an outer layer; the counts of the
three energy regions (i.e., Compton, ROI, and HE) for these
phantom measurements are presented (Table 2). In this simple
Innovative Methodology
271
PARTIAL BODY POTASSIUM IN THE LEGS IN SCI
Table 2. Thirty-minute counts from background
and phantom measurements from the 3 regions
of the spectrum: Compton, region of interest (40K peak),
and high energy
Background
KCl phantom
KCl phantom with overlay
Table 3. Summary of the PBK and DXA results using 3 g
K/kg SM to determine the SM mass from PBK measurements
SCI
Compton Region
ROI
HE
247,903
305,568
296,497
20,218
36,717
30,807
21,817
21,567
21,188
Values are 30-min counts. ROI, region of interest; HE, high energy.
Fig. 5. Data distribution for each cohort for lean tissue (LT) mass in the legs
assessed by dual-energy X-ray absorptiometry (DXA), and skeletal muscle
(SM) mass in the legs assessed by PBK. Results are expressed as mean ⫾ SD.
Cont, control subjects; Para, paraplegic subjects; Tetra, tetraplegic subjects.
J Appl Physiol • VOL
Paraplegic
subjects
No. of subjects, n
17
10
Mean K (⫾SD) by PBK, g 52.8⫾10.7 30.1⫾10.9
Mean SM mass (⫾SD) by
PBK, kg
17.6⫾3.6 10.0⫾3.6
Mean LT mass (⫾SD) by
DXA, kg
20.5⫾3.6 15.0⫾3.7
Range of K, g
31.3–66.2 17.9–50.2
Range of SM mass, kg
10.3–22.1
6.0–16.7
Range of LT mass, kg
15.3–29.8 10.4–21.0
Tetraplegic
subjects
Combined
11
35.4⫾7.3
21
32.9⫾9.3
11.8⫾2.4
11.0⫾3.1
15.9⫾4.5 15.5⫾4.0
24.8–49.2 17.9–50.2
8.3–16.4 6.0–16.7
10.9–24.8 10.4–24.8
PBK, partial body potassium; DXA, dual-energy X-ray absorptiometry; SM,
skeletal muscle; LT, lean tissue.
Plotting the PBK results against DXA, SM mass by PBK
is generally lower than LT mass by DXA, as expected, with
the majority of the subjects falling below the 1:1 line (Fig. 6).
The values obtained by PBK are approximately 60 –90% of the
value of the LT mass value by DXA. However, the correlation
between PBK and DXA in the combined control and SCI
groups is significant (r2 ⫽ 0.325 and P ⫽ 0.0002), albeit weak.
We would expect the regression lines to be at least parallel to
the 1:1 line, suggesting a possible offset by either PBK or DXA
or both. The underestimation of the LT mass by DXA in SCI
subjects compared with SM mass by MRI was reported by
Modlesky et al. (8). This would skew the regression line in Fig.
6 in the expected direction, that is, more parallel to the 1:1 line.
It is appreciated, however, that PBK detectors were over the
upper leg whereas the DXA ROI included the proximal and
distal leg, from the crotch down.
DISCUSSION
The results demonstrated the possibility of using a PBK
system for monitoring SM mass in patients with chronic SCI.
The difference between SM masses in the SCI and control
subjects determined by PBK is ⬃1.5 times larger (37.5 vs.
24.4%) than that recorded for LT masses by DXA. The PBK
Fig. 6. SM mass of legs assessed by PBK vs. LT mass in legs assessed by
DXA for the combined cohorts.
106 • JANUARY 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
model, the signal in the ROI is reduced by 36%. However, by
combining the Compton and ROI regions, the reduction in the
signal is only 20%. Furthermore, the error in the K net counts
in the ROI with the overlay, 2.13%, is reduced to 1.30% when
using both regions. Finally, using both regions improves the K
signal relative to the natural K background, i.e., signal-to-noise
ratio improves when using both regions due to availability of
more data. Linear attenuation of a narrow photon beam through
2.5 cm of water at K energy lowers the signal only by 14%,
clearly underestimating the mean free path length through the
outer layer. During these experiments the third HE region
above the ROI remains unchanged, demonstrating that there
were no changes in the counting system.
The distributions of the SM mass in the legs were determined by PBK, and the LT mass was determined by DXA for
control and SCI subjects (Fig. 5). The mean values, SDs, and
ranges for PBK and DXA are reported (Table 3). A conversion
factor of 3 g K/kg SM, from grams potassium to kilograms
skeletal muscle mass (6), was used (Fig. 5; Table 3). We noted
that LT mass is slightly larger than the SM mass because of the
additional tissues in the former. The SM mass of the legs by
PBK was 37.5% lower in the combined SCI group than of the
control group (P ⬍ 0.0001), whereas the difference in LT mass
by DXA between SCI and control groups was only 24.4%
lower (P ⫽ 0.003). Similarly, PBK measurements provided
greater discrimination between the groups with paraplegia and
tetraplegia than did DXA.
Controls:
Men
Innovative Methodology
272
PARTIAL BODY POTASSIUM IN THE LEGS IN SCI
J Appl Physiol • VOL
DXA measurements in this unique cohort of those with paralysis and striking leg soft tissue changes.
In summary, a PBK system mounted on a cart for monitoring K of the lower extremities was constructed and applied to
persons with chronic SCI and able-bodied controls. The system
has the potential to provide immediate, highly specific, and more
sensitive monitoring of SM atrophy in immobilized individuals. It
is envisioned that this system, unlike DXA and other nonmobile
body compositional techniques, may be used at bedside to test
the efficacy of a physical or pharmacological intervention to
preserve muscle mass immediately after lower extremity paralysis.
GRANTS
Partial support for this work was provided by the Department of Veterans
Affairs Rehabilitation Research and Development Service, Center of Excellence for the Medical Consequences of Spinal Cord Injury (no. B4162C).
Partial support by the U.S. Department of Energy, under Contract No. DEAC02-98CH10886, is also recognized.
REFERENCES
1. Bauman WA, Spungen AM, Flanagan S, Zhong YG, Alexander LR,
Tsitouras PD. Blunted growth hormone response to intravenous arginine in subjects with a spinal cord injury. Horm Metab Res 26:
149 –153, 1994.
2. Bin Samat S, Green S, Beddoe AH. The activity of one gram of
potassium. Phys Med Biol 42: 407– 413, 1997.
3. Bors E, Engle ET, Rosenquist RC, Holinger VH. Fertility in paraplegic
males: a preliminary report of endocrine studies. J Clin Endocrinol Metab
10: 381–398, 1949.
4. Claus-Walker J, Scurry M, Carter RE, Campos RJ. Steady state
hormonal secretion in traumatic quadriplegia. J Clin Endocrinol Metab 44:
530 –535, 1977.
5. Cohn SH, Dombrowski CS, Pate HR, Robertson JS. A whole-body
counter with an invariant response to radionuclide distribution and body
size. Phys Med Biol 14: 645– 658, 1969.
6. Elia M. Organ and tissue contribution to metabolic rate. In: Energy
Metabolism: Tissue Determinants Cellular Corollaries, edited by Kinney
JM and Tucker HN. New York: Raven,, 1992.
7. Hansen RD, Raja C, Aslani A, Smith RC, Allen BJ. Determination of
skeletal muscle and fat-free mass by nuclear and dual-energy X-ray
absorptiometry methods in men and women aged 51– 84 y. Am J Clin Nutr
70: 228 –233, 1999.
8. Modlesky CM, Bickel CS, Slade JM, Meyer RM, Cureton KJ, Dudley
GA. Assessment of skeletal muscle mass in men with spinal cord injury
using dual-energy X-ray absorptiometry and magnetic resonance imaging.
J Appl Physiol 96: 561–565, 2004.
9. Mollinger LA, Spurr GB, El Ghatit AZ, Barboriak JJ, Rooney CB,
Davidoff DD, Bongard RD. Daily energy expenditure and basal metabolic rates of patients with spinal cord injury. Arch Phys Med Rehabil 66:
420 – 426, 1985.
10. Nance PW, Shears AH, Givner ML, Nance DM. Gonadal regulation
in men with flaccid paraplegia. Arch Phys Med Rehabil 66: 757–759,
1985.
11. Pierson RN Jr. Quality of the Body Cell Mass, Body Composition in the
Third Millennium. New York: Springer, 2002.
12. Ramirez LM. Feasibility Study of In Vivo Partial Body Potassium
Determination in the Human Body Using Gamma-Ray Spectroscopy (PhD
Dissertation). Austin, TX: Univ. of Texas at Austin, 2005.
13. Ramirez LM, Wielopolski L. Analysis of potassium spectra with low
counting statistics using trapezoidal and library least-squares methods.
Appl Radiat Isot 61: 1367–1373, 2004.
14. Rossier AB, Favre H, Valloton MB. Body composition and endocrine
profile in spinal cord injured patients. In: The Spinal Cord Injured Patient:
Comprehensive Management, edited by Lee BY, Ostrander E, George J,
Cochran B, Shaw WW. Philadelphia, PA: Saunders, 1991, chapt. 13, p.
163–170.
15. Schneider B, Wang J, Thornton JC, Arbo J, Horlick M, Heymsfield
SB, Pierson RN Jr. Accuracy, reproducibility and normal total body
potassium (TBK) ranges measured using the renovated whole body 40K
106 • JANUARY 2009 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
system is passive involving no external radiation, and it capitalizes on radioactive 40K in equilibrium with natural K as an
indicator of the SM mass. By contrast, the DXA system
involves external radiation, albeit small. Our present system,
which was mounted on a cart, was heavily shielded. However,
with further shielding redesign and optimization, it should be
possible to make the entire system more flexible to be used at
the bedside during the acute and subacute phase of SCI. Data
processing for quantitative analysis can be improved using the
Compton and ROI regions or, alternatively, using library leastsquares (LLS), resulting in a better monitoring system (13).
While our experiments with the overlaying water layer in the
phantom overestimate the signal’s self-attenuation from that
expected in obese subjects, including the Compton region in
the analysis lowered the effect of obesity by reduction in signal
loss from 36% to 20%. In addition, improvements in spectral
analysis using LLS analysis may ease the requirements for
heavy shielding by better handling the variations in the background (24). Furthermore, for a mobile system, changes in the
background in different rooms will be easily accounted for
using LLS and local room background measurements. Thus we
recommend monitoring variations in the room background
each day, rather than using a standard value. Adopting this
procedure yielded a satisfactory estimate of the phantom net
counts that were contained within 1 SD (Fig. 4), especially
after the energy calibration of one of the detectors has been
aligned with the remaining ones. DXA is a well-established
technique for assessing the body’s soft-tissue lean and fat
components in healthy normal populations. However, its validity for aged, sick, or disabled people has been questioned (7,
8). Thus the higher sensitivity of PBK in revealing body
compositional differences and/or changes might be attributable
to the high specificity with which it measures 40K. The significant, but weak, correlation between measurements of PBK and
DXA is intriguing and requires further study, although the
weak regression coefficients may partially be attributable to the
small cohort sizes. A methodological effect may have been
operative because the PBK detectors were placed over the
upper leg, whereas the DXA measurement included the entire
leg. However, this is probably not the full explanation of the
difference between these methodologies because most of leg
muscle is present proximally and the proportion of muscle
retained (or proportion of muscle lost due to paralysis) in the
upper leg would be expected to be similar to that in the lower
leg. Thus additional work, including more closely approximating the leg regions measured by both methods, needs to be
performed to answer this question. The most likely explanation, in our opinion, is that the soft tissue equations for DXA
were derived from studies in able-bodied individuals, not
persons with marked soft tissue changes that are characteristic
of those with SCI. Although dependent edema is measured as
LT by DXA, this is not a likely explanation for our findings
because most subjects were not grossly edematous. In future
studies, however, leg edema could be more carefully excluded
before performing DXA measurements. It must be emphasized
here that because DXA is of questionable validity if used in
any subjects but in the healthy, able bodied, it is not being used
as a criterion method. PBK was compared with DXA in our
study because it is currently widely used in clinical and
research studies of bone and soft tissue in the SCI population.
However, we did not render judgment as to the validity of the
Innovative Methodology
PARTIAL BODY POTASSIUM IN THE LEGS IN SCI
16.
17.
18.
19.
20.
counter of St. Luke’s-Roosevelt Hospital. Int J Body Comp Res 2: 51– 60,
2004.
Snyder WS, Cook MJ, Nasset ES, Karhausen LR, Parry Howells G,
Tipton IH. Report of the Task Group on Reference Man. ICRP 23.
Oxford, UK: Pergamon, 1975 [CE 6].
Spungen AM, Adkins RH, Stewart CA, Wang J, Pierson RN Jr,
Waters RL, Kemp BJ, Bauman WA. Factors influencing body composition in persons with spinal cord injury: a cross-sectional study. J Appl
Physiol 95: 2398 –2407, 2003.
Spungen AM, Bauman WA, Wang J, Pierson RN Jr. The relationship
between total body potassium and resting energy expenditure in individuals with paraplegia. Arch Phys Med Rehabil 73: 965–968, 1993.
Spungen AM, Wang J, Pierson RN Jr, Bauman WA. Soft tissue body
composition differences in monozygotic twins discordant for spinal cord
injury. J Appl Physiol 88: 1310 –1315, 2000.
Streat SJ, Beddow AH, Hill GL. Aggressive nutritional support does not
prevent protein loss despite fat gain in septic intensive care patients.
J Trauma 27: 262–266, 1987.
273
21. Tsitouras PD, Zhong YG, Spungen AM, Bauman WA. Serum testosterone and insulin-like growth factor-I/growth hormone in adults with
spinal cord injury. Horm Metab Res 27: 287–292, 1995.
22. Wang YH, Huang TS, Lien IN. Hormone changes in men with spinal
cord injuries. Am J Phys Med Rehabil 71: 328 –332, 1992.
23. Weaver FM, Collins EG, Kurichi J, Miskevics S, Smith B, Rajan S,
Gater D. Prevalence of obesity and high blood pressure in veterans with
spinal cord injuries and disorders: a retrospective review. Am J Phys Med
Rehabil 86: 22–29, 2007.
24. Wielopolski L, Cohn SH. Application of the library least squares analysis
to whole body counter spectra derived from an array of detectors. Med
Phys 11: 528 –533, 1984.
25. Wielopolski L, Ramirez LM, Gallagher D, Heymsfield SB, Wang ZM.
Measuring partial body potassium in the arm versus total body potassium.
J Appl Physiol 101: 945–949, 2006.
26. Wilmet E, Ismail AA, Heilporn A, Welraeds D, Bergmann P. Longitudinal study of the bone mineral content and of soft tissue composition
after spinal cord injury. Paraplegia 33: 674 –577, 1995.
Downloaded from http://jap.physiology.org/ by 10.220.33.4 on June 17, 2017
J Appl Physiol • VOL
106 • JANUARY 2009 •
www.jap.org