Download Nutrient Intake, Peripheral Edema, and Weight

Journals of Gerontology: MEDICAL SCIENCES
Cite journal as: J Gerontol A Biol Sci Med Sci. 2013 June;68(6):712–718
doi:10.1093/gerona/gls234
Published by Oxford University Press on behalf of The Gerontological Society of America 2012.
Advance Access publication November 26, 2012
Nutrient Intake, Peripheral Edema, and Weight Change
in Elderly Recuperative Care Patients
Dennis H. Sullivan,1,2 Larry E. Johnson,2,3 Richard A. Dennis,1,2 Paula K. Roberson,4 Kimberly K. Garner,1,2,3 Prasad R. Padala,1,2,5 Kalpana P. Padala,1,2 and Melinda M. Bopp1,2
Geriatric Research Education and Clinical Center, Central Arkansas Veterans Healthcare System. 2
Donald W. Reynolds Department of Geriatrics, University of Arkansas for Medical Sciences.
3
Geriatrics and Extended Care Service, Central Arkansas Veterans Healthcare System.
4
Department of Biostatistics, University of Arkansas for Medical Sciences, and
5
Department of Psychiatry, University of Arkansas for Medical Sciences, Little Rock.
1
Address correspondence to Dennis H. Sullivan, MD, Geriatric Research Education and Clinical Center (Building 170, 3J/NLR), Central Arkansas
Veterans Healthcare System, 4300 W 7th Street, Little Rock, AR 72205. Email: [email protected]
Background. It is unclear whether serial measures of body weight are valid indicators of nutritional status in older
patients recovering from illness.
Objectives. Investigate the relative influence of nutrient intake and changes in peripheral edema on weight change.
Methods. A prospective cohort study of 404 older men (mean age = 78.7 ± 7.5 years) admitted to a transitional care
unit of a Department of Veterans Affairs nursing home. Body weight and several indicators of lower extremity edema
were measured at both unit admission and discharge. Complete nutrient intake assessments were performed daily.
Results. Over a median length of stay of 23 days (interquartile range: 15–41 days), 216 (53%) participants gained
or lost more than or equal to 2.5% of their body weight. Two hundred eighty-two (70%) participants had recognizable
lower extremity pitting edema at admission and/or discharge. The amount of weight change was strongly and positively
correlated with multiple indicators of both nutrient intake and the change in the amount of peripheral edema. By multivariable analysis, the strongest predictor of weight change was maximal calf circumference change (partial R2 = .35,
p < .0001), followed by average daily energy intake (partial R2 = .14, p < .0001), and the interaction of energy intake by
time (partial R2 = .02, p < .0001).
Conclusions. Many older patients either gain or lose a significant amount of weight after admission to a transitional
care unit. Because of the apparent high prevalence of co-occurring changes in total body water, the weight changes do not
necessarily represent changes in nutritional status. Although repeat calf circumference measurements may provide some
indication as to how much of the weight change is due to changes in body water, there is currently no viable alternative to
monitoring the nutrient intake of older recuperative care patients in order to ensure that their nutrient needs are being met.
Key Words: Elderly—Malnutrition—Transitional care—Weight loss—Edema—Anthropometrics.
Received April 5, 2012; Accepted October 28, 2012
Decision Editor: Stephen Kritchevsky, PhD
M
any nursing homes have dedicated transitional care
units (TCU) designed to provide recuperative care
and rehabilitation services to older patients recovering from
a recent illness. The goals of admission to these units are
to optimize health status, maximize functional abilities,
and discharge to home when possible. However, many of
the patients on these units are so weak and debilitated that
achievement of such goals is often difficult (1).
Protein–energy undernutrition can be a significant factor
contributing to the older TCU patients’ weakness and
difficulties reaching the targeted outcomes (2–7). At the time
of their admission, many of these patients have established
nutritional deficits. After admission, new or worsening
nutritional problems often develop. Studies indicate strong
712
associations between the severity of such nutritional deficits
and the risk for intervening complications, poor functional
recovery, and other adverse outcomes (2–6). For these
reasons, provision of appropriate nutritional care to TCU
patients is of fundamental importance. At a minimum, it is
essential to ensure that each patient’s nutritional intake is
adequate to maintain energy and nitrogen balance. Optimally,
nutrient intake should sufficiently replete established deficits
and promote muscle accretion and functional improvement.
Unfortunately, many TCU patients have persistent anorexia
and/or impairments of self-feeding abilities (8,9). For these
reasons, it usually requires considerable personnel and
other resources to even approach the goal of achieving an
adequate level of nutrient intake. Because personnel are
Nutrition, Hydration, and Weight Change
almost always a limited resource, efficiency in the provision
of nutritional care is always required.
Although an important component of good nutritional
care in the TCU setting, accurate measures of nutrient intake
are difficult to obtain. Given the limitations of personnel and
other cost constraints and the lack of reliable methods of
estimating nutrient needs, serial weight determinations are
often used as supplementary indicators of patients’ changing
nutritional status and the adequacy of their nutrient intake.
However, little is known about the specificity of weight and
weight change as indicators of nutritional status in the TCU
setting. Many older patients recovering from illness develop
peripheral edema and these fluctuations in total body water
may confound the relationship between weight and nutritional status. Given the importance of nutritional status, this
is an issue needing further investigation. The purpose of this
study is to explore the interaction of edema and nutrient
intake on changes in weight in frail TCU patients.
Participants and Methods
Enrollment
This was a prospective study of patients aged more than
64 years admitted to a TCU for recuperative and rehabilitative care between March 2006 and March 2011. The TCU
was located within a Veterans Affairs hospital nursing home
Community Living Center. Exclusion criteria included having a terminal disease or being admitted for respite care or
long-term custodial care. Of the 1,078 admissions to the
TCU during the study interval, 540 patients were identified
to be eligible for study entry by the attending health care
team. The noneligible participants included 349 admitted
for respite, 77 admitted for long-term custodial care, and
112 repeat admissions that were already enrolled in the
study. After the study’s purpose, procedures, and potential risks were explained to the 540 eligible patients and/
or their surrogates (for those who lacked decision-making
capacity), 94 declined the invitation and were dropped from
further consideration. The remaining 446 patients provided
written informed consent and entered the study. Of these
participants, 42 were dropped from the analyses; 41 due to a
length of stay less than 7 days and 1 because he was a bilateral amputee. This left a final sample of 404 participants.
The study was compliant with the regulations and ethical
standards of the Declaration of Helsinki, Health Insurance
Portability and Accountability Act, Department of Veterans
Affairs, and the Institutional Review Board of the Central
Arkansas Veterans Healthcare System.
At admission, each participant completed a comprehensive standardized evaluation that included a measure of total
body weight, knee height, and several indicators of lower
extremity edema. Weights and knee heights were obtained
by the research staff using validated protocols and calibrated digital scales and a knee height caliper, respectively.
713
Indicators of edema included maximum calf and ankle (1 in
above the malleolus) circumference measures and ratings of
the amount of edema at four sites, the dorsum of the foot,
ankle, mid-calf, and above the knee. Edema was rated as
0–4+ based on the amount the skin could be pitted using
predefined definitions as modified from Brodovicz and colleagues (10). After multiple research team training sessions,
each measure was highly reliable as indicated by interrater
reliability coefficients ranging between 0.99 (weights) and
0.78 (edema ratings). Throughout the remainder of the hospitalization, each participant was monitored on a daily basis.
Components of the admission assessment were repeated
every 10–14 days and, when possible, at discharge. The last
assessment obtained was used as the discharge assessment.
The quality of assessments was maintained throughout the
study by periodic repeat interrater reliability testing.
Calorie Counts
Complete daily nutrient intake assessments were
generally performed everyday for the first 14 days after
admission and then every other day. They were also
performed daily for 5 days after any significant deterioration
in clinical status. A rigorous protocol for monitoring the
participants and recording nutrient intake from all sources
was followed. As described elsewhere, the protocol utilized
a custom computer program that integrated the hospital
food menus with a nutrition database, direct observation,
and computer entry by a primary assessor and confirmation
of results by a secondary observer using digital photographs
of the food trays obtained before and after meals (11).
Quality assurance for the process was performed regularly
and included validation of estimation skills using weighed
portions and recoding and analysis for random samplings
of the meal photo archive. Energy intake was expressed
as the percent of each participant’s resting metabolic
requirements (Harris–Benedict estimation plus a 25%
adjustment for activity [12]). Protein intake was expressed
as a percentage of calculated requirements defined as 1.5 g/
kg body weight/d (13,14). This value was selected based
on studies that suggest that older adults may require more
dietary protein when ill or hospitalized (15,16).
In addition to the average daily energy intake during
the entire hospitalization, additional variables were created to indicate the adequacy of energy intake over specific
time periods. Admission energy intake was set equal to
the average of the daily totals for the first 5 calorie count
days expressed as a percentage of estimated requirements.
Likewise, discharge energy intake was based on the average
of the last 5 calorie count day totals. Energy intake change
was set equal to the discharge minus the admission energy
intake variables. Similarly, discharge and change variables
were created for protein intake. Several variables were created to indicate the change in amount of edema. All were
based on the average of the measurements taken from the
714
SULLIVAN ET AL.
left and right side of the body at each site. This included
change in maximum calf circumference (discharge −
admission), maximum calf cross-sectional area (area = circumference × pi), and lower calf volume. Volume of lower
calf was estimated using formula for right circular frustum
and calf and ankle circumference measurements (17,18).
Variables were also created to represent clinically assessed
change in edema at each site (possible range: −4 to +4), the
sum of these four variables, and a more simplistic indication of direction of change (−1, 0, +1).
Statistics
The dependent variable of interest was the change in
total body weight between admission and discharge. After
log transformation, the nutrient intake and edema indicator
variables were included in the analyses as potential independent variables.
In order to determine whether the relationships between
the nutrient intake variables and change in total body weight
were influenced by time (ie, TCU length of stay), interaction
terms were created. To do this, each variable of interest was
first transformed to a unitless measure (observation – population mean/standard deviation). A given interaction term
was then created by multiplying the appropriate transformed
variable by the transformed time variable. Along with age,
race, body mass index, use of diuretics, indicators of inflammation, length of stay, and diagnoses, the interaction terms
were included in the analyses as additional covariates.
Variables with skewed distributions were log transformed.
Both univariable and multivariable least squares regression
analyses were utilized to determine the relationship between
a given outcome and the various covariates taken alone or
in combination. For the univariable analyses, the strength
of the correlations was assessed based on Pearson’s correlation coefficients. Only variables that were significantly
correlated with the given outcome by univariable analysis
were included in the multivariable analyses. Analyses were
conducted using SAS v9.1 (SAS Institute Inc., Cary, NC).
A two-sided value of p < .05 was considered significant.
Results
The characteristics of the 404 participants are shown in
Tables 1 and 2. Their average age was 78.7 ± 7.5 years, and
the majority was White men, 345 (85%). During their stays
on the TCU, 296 (73%) participants maintained an average daily energy intake of more than 70% of their predicted
needs, whereas 199 (49%) consumed more than 90% of
needs. A majority of the participants (62%) improved their
nutrient intakes during their TCU stays. However, there
was a large variance in the change in energy intakes as indicated in Table 2. The median absolute change in weight for
the study population was 2.1 kg (interquartile range: 0.8–
4.5 kg). A majority of the participants (53%) experienced a
Table 1. Admission Characteristics of Study Participants
Variables
All Participants (n = 404)
Age, y, M ± SD
Body mass index, kg/m2, M ± SD
Education, y, M ± SD
Length of stay, d, median (interquartile range)
Total number of medications, M ± SD
Number of prescription medications, M ± SD
Total number of problems, M ± SD*
Number of active problems, M ± SD*
Charlson’s comorbidity index, M ± SD
Albumin (g/L), M ± SD
Prealbumin (mg/dL), M ± SD
C-reactive protein (mg/L), (interquartile range)
78.7 ± 7.6
26.4 ± 5.8
12.1 ± 3.3
23 (15–42)
14.1 ± 5.0
9.3 ± 3.8
20.6 ± 6.0
3.8 ± 1.9
2.6 ± 1.8
30.7 ± 4.8
19.4 ± 6.9
12.4 (4.8 – 41.4)
Admitted to unit from acute care hospital, n (%)
Married, n (%)
Whites, n (%)
Male, n (%)
Independent in all ADL, n (%)†
Dependent in all ADL, n (%)†
Most prevalent active medical problems, n (%)‡
Hypertension
Dyslipidemia
Arthritis
Anemia
Benign prostatic hypertrophy
Coronary artery disease
Diabetes mellitus, type 2
Chronic obstructive pulmonary disease
Depression
Congestive heart failure
Chronic renal insufficiency§
298 (74)
189 (47)
352 (87)
397 (98)
66 (16)
142 (35)
353 (87)
262 (65)
240 (59)
202 (50)
194 (48)
189 (47)
154 (38)
152 (38)
134 (33)
119 (30)
115 (29)
Notes: ADLs = activities of daily living; M = mean; SD = standard
deviation.
* Total problems = stable + active problems; Active problems defined
as those that are not controlled and require a change of therapy at time of
admission.
† Independent in all of the basic ADLs (bathing, dressing, toileting,
transferring, continence, and feeding) as measured using the Katz index of
ADLs (20). Dependent defined as needing human help or unable to perform.
‡ Includes both primary and secondary diagnoses.
§ Defined as a blood urea nitrogen greater than 30 mg/dL.
weight change of greater than 2.5%, but many of these participants also experienced significant changes in total body
water as indicated by the changes in calf circumference or
the amount of edema present.
Relationships Between Weight Change and
the Covariates
As shown in Table 3, weight change (discharge − admission) was strongly correlated with various indicators of
edema/fluid retention, nutrient intake, and other clinical
parameters such as level of inflammation. The strength of
the relationship between weight change and nutrient intake
increased over time as indicated by the significant nutrient
intake by time interaction. Weight change was not significantly associated with age, race, use of diuretics, or any
715
Nutrition, Hydration, and Weight Change
Table 2. Nutrient Intake and Changes in Weight and Indicators
of Peripheral Edema During Hospitalization
Variables
Average daily energy intake
Kcal/d, M ± SD (range)
As percentage of estimated requirements,
mean ± SD (range)
Average daily protein intake
g/kg body weight, M ± SD (range)
As percentage of estimated requirements,
M ± SD (range)
Weight change from admission to discharge
Lost ≥ 5%, n (%)
Lost ≥ 2.5%, n (%)
Gained ≥ 2.5%, n (%)
Gained ≥ 5%, n (%)
Lower extremity edema at either admission or
discharge, n (%)
Recognizable change in lower extremity edema*
Increased from admission to discharge, n (%)
Decreased from admission to discharge, n (%)
Change in maximum calf circumference
Increased from admission to discharge, n (%)
Decreased from admission to discharge, n (%)
Table 3. Variables Significantly Correlated With Weight Change
by Univariable Analysis*
All Participants (n = 404)
1716 ± 509 (486–3598)
91 ± 32 (22–204)
0.9 ± 0.3 (0.3–2.5)
61 ± 23 (19–167)
61 (15)
116 (28)
100 (25)
49 (12)
282 (70)
104 (26)
80 (20)
130 (32)
105 (26)
Note: *Change in amount of edema defined as a difference between
admission and discharge of greater than or equal to 1 point (on 4-point scale)
in the assessed amount of edema at any of four locations on lower extremities.
given admission diagnostic category such as congestive
heart failure, renal insufficiency, or chronic obstructive lung
disease.
When all of the covariates listed in Table 3 were
included in a multivariable least square regression analysis employing a stepwise procedure, the strongest correlate of change in weight was the change in maximal calf
circumference (Table 4). Based on the partial R squared
statistics from this model, the change in maximal calf circumference accounted for 39% of the variance in weight
change. Nutrient intake was also an important independent
correlate of weight change. As indicated by the inclusion
of the energy intake by time interaction term in the final
model, the strength of the relationship between nutrient
intake and weight change increased with longer lengths of
stay on the unit.
As a secondary objective, attempts were made to
identify other linear or nonlinear models based on the same
independent variables that better fit the data. None of the
alternate models performed better (ie, explained as much
of the variance in hospital weight change) than the model
presented. Substituting change in either maximal calf
cross-sectional area or calf volume for change in maximal
calf circumference in the final regression model produced
essentially identical results.
Based on the regression model, each 1 cm change in
calf circumference was associated with a 1.2 kg change
in weight. Between admission and discharge from the
unit, 235 (58%) participants experienced a change in
Dependent Variable Change
in Total Body Weight
Independent Variables
Maximum calf circumference change
Calf cross-sectional area change
Calf volume (frustum) change
Calf volume change
Average daily energy intake
Average daily protein intake
Admission protein intake
Discharge energy intake
Discharge protein intake
Mid-calf edema change
Total edema score change
Ankle edema change
Foot edema change
Admission body mass index
Percentage of weight lost prior 6 mo
Interaction (protein intake by time)
Percentage of weight lost prior year
Discharge albumin
Admission C-reactive protein
Interaction (energy intake by time)
Change in blood urea nitrogen to creatinine ratio
Type 2 diabetes mellitus
Admission albumin
Correlation coefficient
0.62
0.55
0.53
0.51
0.48
0.47
0.44
0.38
0.36
0.33
0.32
0.31
0.30
−0.29
0.20
0.20
0.20
0.20
−0.19
0.18
−0.17
−0.13
0.13
Note: *Based on univariable least squares regression analysis and listed
in order of the strength of the correlation. All variables had a p < .0001 except
the last four for which the p value was < .01.
their calf circumference of this magnitude. Their median
absolute weight change was 2.7 kg (interquartile range:
1.3–5.5 kg). However, 58 (25%) of these participants never
had any detectable edema of their legs. More than 73% of
the participants experienced significant changes in either
lower extremity edema or calf circumference during their
TCU stays.
For the overall study population (n = 404), both average daily nutrient intake and the edema assessment scores
correlated with calf circumference change (p < .001 all
variables); nutrient intake accounted for 3% of the variance
(partial R squared = .027) and edema 17% of the variance
in calf circumference change. When only those participants
(n = 205) with a length of stay less than 23 days (median
length of stay for population) were included in the analysis,
calf circumference change was highly correlated with edema
(partial R squared = .14, p < .0001) but not with nutrient
intake (p > .6). When only participants whose calf circumference increased (n = 236) were included in the analysis,
change in calf circumference remained highly correlated
with edema (p < .0001) but not nutrient intake (p > .4).
Conversely, when only those whose calf circumference
decreased (n = 168) were included in the analysis, change
in calf circumference was correlated with edema (partial
R squared: .14, p < .0001) and nutrient intake (partial R
squared: .03, p = .03).
716
SULLIVAN ET AL.
Table 4. Results for Multivariable Analysis
Dependent Variable Change in Total Body Weight
Independent Variables
Maximal calf
circumference change
Average daily energy
intake
Interaction (energy intake
by time)
Mid-calf edema change
Admission C-reactive
protein
Percentage of weight lost
prior 6 mo
β coefficient
Standard error Partial R squared
1.167*
0.073
.39
0.029*
0.006
.12
0.626*
0.155
.02
0.215†
−0.007†
0.088
0.003
.02
.01
0.099†
0.047
.01
Model R squared = .54
Notes: Variables listed in order in which they entered the least squares
regression model when a stepwise procedure was utilized and all variables
listed in Table 3 were included in the analysis.
*p < .0001.
†p < .05.
Discussion
This study indicates that clinically significant weight
change is common among older patients during their stays
on a dedicated TCU designed to provide recuperative care
and rehabilitation services to older patients. The data also
suggest that it is changes in hydration status, more so than
adequacy of nutrient intake that accounts for such weight
fluctuations. This conclusion is based on the finding that
there was a direct correlation between weight change
and the change in various indicators of peripheral edema
including maximal calf circumference and a clinical
assessment of the degree to which the skin on the lower
extremities could be easily pitted. There was also a direct
correlation between the adequacy of nutrient intake and
the amount of weight change. As indicated by the nutrient
intake by time interaction, the strength of the relationship
between nutrient intake and weight change increased with
time. However, the changes in the indicators of peripheral
edema explained a much larger portion of the variance
in weight change. This finding is consistent with prior
studies that also found that fluid retention appears to
mask the effects of nutritional depletion on weight during
hospitalization in older patients (19–22). In one study, of
nearly 500 older adults followed throughout their acute
care hospital stays, 102 (21%) participants were identified
as having maintained an average daily nutrient intake
during their hospitalization that was less than 50% of their
calculated maintenance energy requirements. The amount
of weight lost by these participants between admission and
discharge was only modestly greater than that experienced
by the remaining participants (1 kg; interquartile range:
0–3.5 kg) compared with 1 kg (interquartile range: 0–2 kg;
p = .02) but they were more than 6 times more likely to
have become edematous (19).
This study suggests that measurement of maximal calf
circumference is an easy and reliable way of assessing
change in hydration status in older TCU patients. Based on
the regression model, each 1 cm change in maximal calf circumference represents a gain or loss of more than 2.6 lbs
(1.2 kg) of water. Although many of the study participants
probably experienced changes in nonwater body mass during the study, it was probably change in total body water
rather than change in nonwater body mass that accounted
for most of the change in calf circumference. This is suggested by the fact that edema assessment scores explained
much more of the variance in calf circumference change
than did nutrient intake and that there was a direct correlation between nutrient intake and calf circumference change
only for participants whose calf circumference declined.
Apparently, nutrient intake did not contribute significantly
to any increase in calf circumference.
The other methods of assessing tissue hydration status
used in this study were not as easy to perform as calf
circumference measurements and did not have the same
level of accuracy. Despite repeated research team training
in the technique, the assessment of lower extremity edema
did not correlate as closely with weight change as did calf
circumference measurements. In routine clinical practice,
assessment of edema is not standardized and is probably
even less reliable and accurate. Other investigators have
explored the use of more complex anthropometric models to
account for changes in tissue hydration, but the measurement
techniques employed in these studies were laborious
and utilized specialized water displacement volumetry
equipment that is not readily available (10,17,18,23–26). In
contrast, it is relatively easy to obtain calf circumference
measurements. For this reason, a simple formula based on
calf circumference measurements may have greater clinical
utility in helping clinicians assess for changes in total body
water during illness and recovery, which indirectly provides
an indication of nonfluid changes in body mass.
In this study, the amount of weight change during the
hospitalization was not related to any given admission diagnosis or to the use of diuretics. Apparent changes in total
body water appeared to be the most important determinant
of weight change. As assessed by changes in lower extremity edema or calf circumference, more than 73% of the
participants experienced significant changes in total body
water during their TCU stays. None of the participants had
decompensated left-sided heart failure at the time of either
weight measurements. Otherwise, it was not possible to
determine with certainty what mechanisms were responsible for the development of edema in any given participant
or whether the edema was a marker of clinical recovery
(eg, an increase in erect posture or greater nutrient and salt
intake) or persistent clinical instability. There were multiple
factors prevalent within the study population that may have
contributed to such changes in total body water including
venous insufficiency, low albumin, right-sided heart failure, inflammation, medications, and, possibly, the effects of
refeeding.
Nutrition, Hydration, and Weight Change
The techniques used in this study for assessing change
in lower extremity edema may have underestimated the
amount of change in total body water in the study participants. All of the weight and other anthropometric measures
were obtained early in the morning. Even among the more
ambulatory participants, body water may have redistributed during the night, which may have weakened the association between the amount of leg edema and total body
weight. The study personnel did not assess for edema at
sites other than the lower extremities (eg, the lower back)
and did not monitor how much weights or lower extremity
edema fluctuated throughout the day. Despite these limitations, a simple linear model incorporating calf circumference change explained more than 50% of the variance
in weight change. Although the amount of edema apparent in the participants varied from mild ankle swelling to
three-plus pitting of the skin up to the thigh, more accurate
alternate models for gauging weight change could not be
identified.
The results of this study indicate that serial weight
determinations obtained in the recuperative care and rehabilitation setting lack specificity as indicators of the adequacy of a patient’s nutrient intake or overall nutritional
status. Because weight change in this setting is often a
reflection of changes in the patient’s total body water and
such changes in fluid status are common and often significant, weight measurements may actually provide an
inaccurate assessment of a patient’s nutritional status.
As shown in prior studies, patients with poor nutrient
intake and likely loosing lean body mass often retain fluid
and develop edema (19,22,27). In these individuals, the
retained fluid increases body weight and may mask the
nutritional deficits. Given that clinicians sometimes forget that even small amounts of edema can represent large
volumes of water and thus significant amounts of weight,
they may be lulled by serial weight measurements into a
false sense of how well the patient is doing and fail to recognize serious developing nutritional deficits. Although
repeat assessments of lower extremity edema may provide
some indication as to when there are changes in a patient’s
hydration status, this can only serve to alert the clinician
as to the need to interpret serial weights with caution.
There are no standardized, well-validated, and easily performed clinical assessment methods for accurately measuring changes in total body water that can be used as part
of the nutritional assessment. This suggests that there is
no viable substitute to monitoring the nutrient intake of
older recuperative care patients in order to ensure that
their nutrient needs are being met.
Conclusion
Many older patients either gain or lose a significant
amount of weight within several weeks after admission to a TCU. Because of the apparent high prevalence
717
of co-occurring changes in total body water, the weight
changes do not necessarily represent changes in nutritional
status. This suggests that serial weights are not viable alternatives to monitoring the nutrient intake of older recuperative care patients in order to ensure that their nutrient needs
are being met.
Funding
This work was supported by VA Health Services and Clinical Science
Research and Development programs (HSR&D and CSR&D—IIR
04-298) and a University of Arkansas for Medical Sciences Tobacco
Settlement award.
References
1. Sullivan DH, Walls RC. The risk of life-threatening complications in a
select population of geriatric patients: the impact of nutritional status.
J Am Coll Nutr. 1995;14:29–36.
2. Sullivan DH, Patch GA, Walls RC, et al. Impact of nutrition status on
morbidity and mortality in a select population of geriatric rehabilitation patients. Am J Clin Nutr. 1990;51:749–758.
3. Sullivan DH, Walls RC. Impact of nutritional status on morbidity in
a population of geriatric rehabilitation patients. J Am Geriatr Soc.
1994;42:471–477.
4. Thomas DR, Zdrowski CD, Wilson MM, et al. Malnutrition in subacute care. Am J Clin Nutr. 2002;75:308–313.
5. Gill TM, Gahbauer EA, Han L, Allore HG. Factors associated with
recovery of prehospital function among older persons admitted to a
nursing home with disability after an acute hospitalization. J Gerontol
A Biol Sci Med Sci. 2009;64:1296–1303.
6.Mudge AM, O’Rourke P, Denaro CP. Timing and risk factors for
functional changes associated with medical hospitalization in older
patients. J Gerontol A Biol Sci Med Sci. 2010;65:866–872.
7. Chevalier S, Saoud F, Gray-Donald K, et al. The physical functional
capacity of frail elderly persons undergoing ambulatory rehabilitation
is related to their nutritional status. J Nutr Health Aging. 2008;
12:721–726.
8. Sullivan DH. The utility of an admission assessment to predict inhospital nutrient intake. J Am Geriatr Soc. 1994;42:478–480.
9. Simmons SF, Schnelle JF. Feeding assistance needs of long-stay nursing home residents and staff time to provide care. J Am Geriatr Soc.
2006;54:919–924.
10.Brodovicz KG, McNaughton K, Uemura N, et al. Reliability and
feasibility of methods to quantitatively assess peripheral edema. Clin
Med Res. 2009; 7:21–31.
11. Baden AL, Cain CM, Sullivan DH. Impact of computer based nutrient
analysis system on reducing time for obtaining calorie counts. J Am
Diet Assoc. 1992; 92.
12. Harris JA, Benedict FG. Biometric Studies of Basal Metabolism in
Man. publication, no. 279 Ed. Washington, DC: Carnegie Institute of
Washington; 1919.
13. Campbell WW, Crim MC, Dallal GE, et al. Increased protein requirements in elderly people: new data and retrospective reassessments. Am
J Clin Nutr. 1994; 60:501–509.
14. Wolfe RR, Miller SL. The recommended dietary allowance of protein:
a misunderstood concept. JAMA. 2008;299:2891–2893.
15. Gaillard C, Alix E, Boirie Y, et al. Are elderly hospitalized patients
getting enough protein? J Am Geriatr Soc. 2008;56:1045–1049.
16. Gaffney-Stomberg E, Insogna KL, Rodriguez NR, et al. Increasing
dietary protein requirements in elderly people for optimal muscle and
bone health. J Am Geriatr Soc. 2009;57:1073–1079.
17. Mayrovitz HN, Macdonald J, Davey S, et al. Measurement decisions
for clinical assessment of limb volume changes in patients with bilateral and unilateral limb edema. Phys Ther. 2007;87:1362–1368.
18. Mayrovitz HN. Limb volume estimates based on limb elliptical vs.
circular cross section models. Lymphology. 2003;36:140–143.
718
SULLIVAN ET AL.
19.Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition
among elderly hospitalized patients: a prospective study. JAMA.
1999;281:2013–2019.
20. Sullivan DH, Bopp MM, Roberson PK. Protein-energy undernutrition
and life-threatening complications among the hospitalized elderly. J
Gen Intern Med. 2002;17:923–932.
21.Campbell IT, Watt T, Withers D, et al. Muscle thickness, meas
ured with ultrasound, may be an indicator of lean tissue wasting in
multiple organ failure in the presence of edema. Am J Clin Nutr.
1995;62:533–539.
22. Asai K. A controlled study on edema in elderly inpatients. Nippon
Ronen Igakkai Zasshi—Japanese J Geriatr. 1990; 27:123–128.
23. Cheah LH, Parkin A, Gilson P, et al. Limb volume measurements in
peripheral arterial disease. Clin Phys Physiol Meas. 1989;10:75–79.
24.Kaulesar Sukul DMKS, den Hoed PT, Johannes EJ, et al. 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 Eng.
1993;15:477–480.
25. Lund-Johansen P, Stranden E, Helberg S, et al. Quantification of leg
oedema in postmenopausal hypertensive patients treated with lercanidipine or amlodipine. J Hypertens. 2003;21:1003–1010.
26. Manning EM, Shenkin A. Nutritional assessment in the critically ill.
Crit Care Clin. 1995;11:603–634.
27. Jensen GL, Wheeler D. A new approach to defining and diagnosing
malnutrition in adult critical illness. [Review]. Curr Opin Crit Care.
2012;18:206–211.