Perspective in Nutrition

478

Measurement of muscle mass in humans:validity of the 24-hour urinary creatininemethod1 � 2

Steven B Heymsfield,3 MD, Carlos Arteaga, MD, C4fford McManus, BS,

Janet Smith, MMSc, RD. and Steven Moffltt, PhD

ABSTRACT Measuring muscle mass is an important component of the nutritional assessment

examination and a suggested index of this body space is the 24-h urinary excretion of creatininc.

The method originated from studies in a variety of animal species in whom early workers found a

parallelism between total body creatine and urinary excretion of creatinine. Assuming that nearly

all creatine was within muscle tissue, that muscle creatine content remained constant and that

creatinine was excreted at a uniform rate, an obvious “corollary” was that urinary creatinine was

proportional to muscle mass. The so-called “creatinine equivalence” (kg muscle mass/g urinary

creatinine) ranged experimentally from 17 to 22. One of the limiting factors in firmly establishing

this constant and its associated variability was (and is) the lack of another totally acceptable

noninvasive technique of measuring muscle mass to which the creatinine method could (or would)

be compared. An improved understanding of creatine metabolism and a variety of clinical studies

in recent years has tended to support the general validity of this approach. However, specific

conditions have also been established in which the method becomes either inaccurate or invalid.

While creatinine excretion may serve as a useful approximation of muscle mass in carefully selected

subjects, there remains a need for accurate and practical indices of muscle mass for use in the

individuals in whom the method cannot be reliably applied. Am J C/in Nutr 1983;37:478-494.

KEY WORDS Muscle mass, urinary creatinine

Introduction

Muscle is roughly 80% water and 20% pro-tein, and a 70-kg man has about 28 kg of thistissue (1). Muscle thus represents 4 to 6 kg ofthe body’s total 10 to 12 kg of protein (2).Expressed in terms of metabolizable energy,

this amounts to 20,000 to 30,000 kcal, ofwhich 70 to 80% can be utilized during pe-nods of negative energy balance (3). It is notsurprising therefore, that quantifying musclemass has been an important focus of nutri-

tionists studying or caring for patients suffer-ing from protein-energy malnutrition (4, 5).Unfortunately, this quantification of muscletissue has lacked the accuracy enjoyed by

other body compartments, because there is asyet no simple, accurate, and inexpensivemethod of measuring muscle mass in living

human subjects. Several methods have beensuggested (Table 1), but none of these has aclearly demonstrated validity. This review fo-cuses on one of these methods: the urinarycreatinine method of measuring muscle mass.

Overview of the method

Myers and Fine (22) were the first to show

that urinary creatinine output was directly

I From the Department of Medicine and Clinical

Research Facility, Emory University School of Medicine,

Atlanta, GA.2 Supported by NIH Contract 1-8565 1 and NIH

Grant RR-0039.3 Teaching and Research Scholar of The American

College of Physicians.

Received February 5, 1982.

Accepted for publication September 8, 1982.

The American Journal ofClinical Nutrition 37: MARCH 1983, pp 478-494. Printed in USA

© 1983 American Society for Clinical Nutrition

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

TABLE 1

Thliui�f�bEIll YIYtJM�U[�II1�HI Offfiu�1� I1��

Anthropometry

.‘-� � Method � -�- Comment l&e(erence

3, 6

Ultrasound

Radiographic

Isotopic methods

‘5N-creatinc and‘4Ncreatine

8, 9

10

CREATININE MUSCLE MASS METHOD 479

Limited accuracy; measures individual muscle groups,

but not total muscle mass. Assumes muscle size pro-

portional to muscle composition.

Replaces skinfold caliper for measuring thickness of fat 7

layer.

Computerized tomography allows measurement of limb 6

muscle cross sectional area. Cost, radiation dose,

availability of scanners limit widespread application.

Cannot measure total body muscle mass.

Based on isotope dilution principle; requires muscle bi-

opsy. Some workers argue that biopsy of one muscle

group inadequate for accurate results.

Total body water and extra-

cellular volume

Total body water and total

body potassium

Neutron activation analysis

Nuclear magnetic resonance

Chemical markers

24-h urinary

creatininc

Urinary 3-methyl histidinc

Combined methods

Creatininc andtotal body

potassium

Skeletal measurements plus

total body potassium

* Modified from Reference 21.

Method based on experimental observations in the rat.

Method provides adipose, muscle and muscle-free leantissue.

Limited number of instruments in operation. Values for

total body muscle mass somewhat lower than anatom-

ical dissections.

As for radiography above, can visualize limb musclecross sectional area. May also provide muscle compo-

sition data. Method largely untested in man.

Subject of current review.

Appropriate method in healthy subjects ingesting meat-

free diet. Not applicable in most disease states because

of variable 3-methyl histidine turnover rate.

Method based on prediction equations developed in ratstudies. Calculated variables include muscle protein,

nonmuscle protein and total body protein.

Three component analysis includes adipose, muscle, andmuscle free lean tissues.

11

12

13-15

16

17, 18

19

20

proportional to the total body creatine con-

tent in three species: human, dog and rabbit.Palladin and Wallenburger (23) later cx-tended this observation to rat, sheep, andguinea pig. BUrger (24), assuming that nearlyall body creatine was within muscle, and thatmuscle possessed a constant creatine content,

proposed that urinary creatinine was propor-tional to muscle mass. Burger’s calculation ofthe creatinine equivalence (kg muscle mass/g urinary creatinine) is presented in the cx-perimental studies sections below.

Four assumptions are entailed in the cx-trapolation of urinary creatinine to musclemass: 1) creatine is almost totally within skel-

etal and smooth muscle; 2) on a creatine-freediet, the total creatine pool and the averageconcentration of creatine per kg of muscleremains constant; 3) creatine is converted

nonenzymatically and irreversibly to creati-nine at a constant daily rate; and 4) creati-nine, once formed, undergoes renal excretionat a constant rate. Mathematical expressionof these assumptions is outlined in Table 2,

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

A) Assuming daily Cm excretion is the result of breakdown of a fraction K of Cr each day:

Cm (moles/day) = K (day� ) x Cr (moles/kg) x M (kg) (1)

B) Converting to gram equivalents for molar amounts of Cr and Cm and rearranging equation to solve for M:

M k - Cm (g/day) 2)( g) - K (day’ ) x {Cr/l. 16) (g/kg)

C) Since there are two species of tissue creatine, Cr and P-Cr, the measured fractional conversion K really reflects

a combination of breakdown of Cr at fractional rate and P-Cr at fractional rate K2, we then modify equation(2) to:

M (k ) - Cm (g/day) (3)g - K, (day’) X Cr (g/kg) � K� (day� ) x P-Cr (g/kg)

1.16 1.59

* Abbreviations: Cr = creatine concentration in skeletal muscle, Cm = total urinary creatinine, M = total muscle

mass, and P-Cr - phosphocreatine concentration in skeletal muscle.

480 HEYMSFIELD ET AL

TABLE 2

Quantification of the creatinine-muscle mass method*

TABLE 3Measured turnover rates and creatine and creatinine concentrations for equations in Table 2.

Term Unit Value Subjects Method Reference

K1K2K

Day’Day’Daf’

0.0110.02640.0164

0.014-0.017

0.015-0.026

0.0169 ± 0.006t

0.0138-0.0188

One adult man7 patients with muscular and

neurological disease and onewith refractory anemia

8 children recovering from

protein-energy malnutrition13 young men

4 adults; 2 had endocrine dis-

eases, and 2 were normal

In vitroIn vitro

‘5N-creatine

‘4C..��reatine

‘5N-creatine

‘5N-creatine

‘4C-creatinc

2525

26

27

8

289

Total Cr� g/kg WWM

g/kg NCP

3.91

22.08

21.2

5 normal subjects; on general

diet

5 normal subjects; on generaldiet

1 1 normal subjects; on general

diet

Biopsy

Biopsy

Biopsy

29

29

30

Cr/P-Cr g/kg WWM

g/kg NCP

g/kg WWM

1.53/2.39

8.3/12.9

1. 19/2.74

11 normal subjects; on general

diet

1 1 normal subjects; on generaldiet

from ref 26, assume that 70%of total Cr is P-Cr

Biopsy

Biopsy

30

30

31

* Abbreviations: Cr, creatine; K, fractional breakdown rate oftotal body creatine to creatinine; K,, K2, breakdown

rate of creatine and phosphocreatine; P-Cr, phosphocreatinc.

tMean±SDforn= 13.:1:Creatinine + phosphocreatine.

and estimated values of the quantities re-quired to apply the formulae are presented inTable 3.

In this review, the first section examinescreatine biosynthesis and metabolism in thecontext of the above four assumptions. In thesecond section, physiological and dietary fac-

tors that may influence the validity of thecreatinine-muscle mass method are evalu-ated. Section three presents studies aimed atestablishing the creatinine equivalence, andthe fmal section provides a summary andrecommendations for evaluating selected pa-tients.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

(1)

At�ns Glycins

?42

CNH+ H-lj-NH2

NH COOH

l�H2

�H2

112

H-C�NH�

cooH

E$�

NH2

CNH

NH

ENZYMES

1. g�cins �ns�au2. guanidinoscststsm.thyltransf.rss.

3. ctutin.kinsu

MW Al!

URINE

� �H2

,,-,-; NH - P1�OH)2 N-CH3 CREAT1NENHC CH2 �

��JLOS�E _� NHC\ � CREAT1NINE

-__=�zz�-=� L� CH2 �

CREATININE MUSCLE MA�� ME�T1-IOD 481

FIG 1 . Principal pathways in creatine metabolism.

Creatine-creatinine metabolism

Biosynthesis of creatine

Creatine is a nitrogenous organic com-

pound, which participates in cellular energy

metabolism (Fig 1), and is found primarily in

muscle. Bloch and Schoenheimer (32) firstproposed that creatine was derived from threeamino acids: arginine, glycine, and 5-adeno-syl methionine. Within several years ofBloch’s studies, Borsook and Dubnoff (33)elucidated the first of two biosynthetic steps:the synthesis of guanidoacetate (glycocy-amine) from glycine and arginine (Fig 1).This reaction, under control of the rate lim-iting enzyme glycine amidinotransferase(transamidinase), is found in human kidney,

liver, pancreas, brain, spleen, and mammarygland (31). The predominant biosynthetic site

in humans is the kidney. The rate of creatinesynthesis is closely regulated by feedback in-

hibition oftransamidinase. On a creatine-freevegetarian diet, this pathway is fully acti-

vated, and adequate guanidoacetate is syn-

thesized from amino acid precursors. Crea-

tine ingested from meat partially or totallyrepresses transamidinase, freeing the precur-sor amino acids for other metabolic reactions,notably growth.

The activity of transamidinase is also mod-ulated by sex hormones (3 1). Testosteronestimulates the de novo synthesis of transam-idinase, which in turn increases the produc-tion rate of guanidoacetate and creatine.Krisko and Walker (34) suggest that the

higher creatinine coefficient (mg creatinine/kg body weight) found in men as comparedto women might be explained on this basis.

Creatine is formed in the next reaction bytransfer of a methyl group from 5-adenosyl-methionine to guanidoacetate (Fig 1). Con-trary to the previous reaction, this step is

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

482 HEYMSFIELD ET AL

irreversible, not rate limiting, and occurs pri-marily in liver tissue (31).

Synthesized creatine is released into thecirculation, where the next step is active up-

take against a concentration gradient by mus-dc and other tissues (3 1, 35). During the

transport process, creatine binds to a specificmembrane site which recognizes the formam-

idine moiety of the molecule. This activeuptake of creatine replaces about 2% of thetotal amount of creatine in muscle each day(3 1). The rate of creatine uptake is influenced

by at least three factors; the process is re-tarded by cooling and anaerobiosis and en-hanced by insulin (35).

About 98% of the total body creatine pool

(120 g in a 70 kg adult man) (36) is withinmuscle. This creatine pool is slowly saturable(37) and has a relatively slow daily turnoverrate of about 1.5 to 2% (37). Although the

creatine concentration per kg ofmuscle variesfrom muscle to muscle, the average concen-

tration is 3 to 5 g/kg of wet fat-free tissue(Table 3). The functions of muscle creatineare enumerated in Table 4.

Other tissues (brain, kidney, and liver) andfluids (blood and urine) contain measurableamounts ofcreatine. Because creatine in thesetissues represent only a small percentage ofthe total body creatine pool (<2%), their over-all contribution to creatine turnover is small.

Creatine uptake by muscle and other tis-sues is relatively complete, and blood levelsrarely exceed 0.3 mg/dl (25). Urinary lossesof creatine are usually neglible, unless bloodlevels become excessively high or renal tu-bular reabsorption is impaired.

Dehydration of creatine to creatinine

Early workers searched for the enzyme“creatase” or “creatinase” that converted

creatine to creatinine. Hahn and Meyer (38)

TABLE 4Functions of muscle creatine (3)

1) High energy intermediate in excitation-contraction

coupling:

Phosphocreatine + ADP 12 creatinc + ATP

Enzyme 1 is creatine phosphokinase, and enzyme 2

is creatine phosphorylase.

2) Modulate glycolysis.

3) Increase intracellular flux of potassium and phos-

phorous.4) Acts as a cofactor in muscle growth.

subsequently showed that creatine was con-verted to creatinine in liver and kidney cx-

tracts at the same rate as in a buffered phos-phate solution of the same pH. These inves-

tigators concluded that the creatine to creat-mine conversion was brought about purelyon a physiochemical basis. This irreversible

reaction is now firmly established. Withinmuscle, creatine exists in two forms: creatineand phosphocreatine (39). Creatine dehy-drates to creatinine at a rate of 1. 1% per day,while phosphocreatine conversion is 2.64%

per day (Table 3). Because urinary creatinineis derived from these two sources, the in vivomeasurement of creatine turnover rate pro-vides a value between 1. 1 and 2.64%.

Among healthy male subjects, a rather con-stant creatine turnover rate of 1.6 to 1.7% perday was obtained (Table 3) (26, 28, 40). How-

ever, the observed range of values between1.4 and 2.6% per day in children recoveringfrom protein-energy malnutrition (8) andadult patients (9, 27) (Table 3) would limitthe accuracy in predicting muscle mass from24-h urinary creatinine in these subjects. Be-

cause the creatine pool is large, small changesin turnover rate result in large differences in

daily urinary output of creatinine. For cx-ample, using the two turnover rates of 1.5 and2.6% in equation 2, and assigning a hypo-thetical creatinine excretion and muscle crea-tine concentration of 1.6 g/day and 3.9 g/kg,respectively, calculated muscle mass is 3 1.7 gand 18.3 kg, or a difference of42%. Assumingthat intracellular pH and temperature remainconstant, one explanation for differences increatine turnover rate between subjects is thenatural differences in the creatine to phos-phocreatine ratio.

Excretion of creatinine in the urine

Once formed, creatinine diffuses from thecell and ultimately appears in the urine afterglomerular filtration, and to a small extent,

tubular secretion (41). The effect of renalinsufficiency on creatinine excretion is dis-cussed in the next section.

Physiological and pathological conditionsthat modulate output of creatinine

Folin’s “law” (42) states that on a creatine-free diet, the output of creatinine is constantfrom day to day for each individual. Somevariability is, however, recognized in the

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

derstood.

URINARYCREATININE

(mg/hr)70

60

50

40

30

20

10

00 8 16 0 8 16 0 � 16 24

CREATININE MUSCLE MASS METHOD 483

healthy subject; more importantly, some

pathoIo8�cal states can alter normal creati-

nine output.

Normal day to day variability

Normally, urinary creatinine excretion var-

ies daily, the causes of which are largelyunknown. The constancy of 24-h creatinine

excretion reported in the early Folin studies,and later reinforced by Rose (43), has been

repeatedly challenged through the years.

Careful studies in reliable subjects haveshown a daily variation from 4 to 8% in

creatinine excretion, which cannot be cx-plained by minor variations in physical activ-ity and diet (44, 45).

Exercise

Extremely strenuous exercise can increaseurinary output of creatinine by 5 to 10% (46,47). Figure 2 demonstrates this effect in nor-mal subjects marching for 3 h at two different

speeds. The increase in daily creatinine out-

put was about 10%. The mechanism(s) ofthese exercise-induced changes in urinarycreatinine output are unknown.

Emotional stress

Creatinine excretion has been studied in awide variety of stressful conditions and psy-

!

.� �

� �

� a, � Q.w� �:�::�+9.4% i+1O.4%

HOURS

�- BASELINE ‘�4 EXERCISE -04- EXERCISE -�

3 Hour March 3 HourMarch2.9-3 MPH 3.1-4 MPH

FIG 2. Exercise-induced increase in creatininc excre-

tion. Figure 2 was prepared from the data presented inReference 47.

chiatric illnesses (48, 49). Scrimshaw et al (48)

con�luI#{232}A thid th� ��h!1Dy � �1 t�absolute amount) in creatinine output in-creases during stress, but the underlyingcause(s) of this phenomenon are poorly un-

Diet

Urinary creatinine excretion is influencedto some extent by three dietary constituents:

protein (50), creatine, and creatinine.Dietary protein is the main source of the

amino acid precursors of creatine. The activ-

ity of the first enzyme in creatine biosyn-thesis, transamidinase, is influenced by die-

tary protein intake (50, 5 1). For example,Van Pilsum (5 1) fed rats a protein-free dietfor 12 days, and found an 85% reduction intransamidinase activity. The level of proteinintake per se appears to have a small effecton urinary creatinine excretion (50, 52), but

few studies have rigorously examined thisrelation independently from other dietary

factors, notably creatine and creatinine.Bleiler and Schedl (52) studied six men who

ingested a natural food, 120-g protein diet(Fig 3). The meal contained the usual amountof dietary creatine. The six men were thendivided into three groups, and switched to acreatine-free diet; the two subjects in eachgroup received 140, 70, and 20 g of protein,respectively. Over 8 to 1 1 wk, urinary creati-nine excretion fell in each of the respective

groups by 26, 32, and 40%. The men werethen switched to another creatine-free 120-gprotein diet, and the creatinine excretion in-

creased over 3 wk in the three groups by 7.7,10, and 13%, respectively. The large reductionin creatinine excretion from the basal dietoccurred because the men were ingesting a

meat-free diet (discussed below), although

the total protein intake appears to have hadsome effect. An important consideration isthat the men on the 20-g protein diet wereprobably in negative nitrogen balance, andsome loss in muscle mass may have contrib-

uted to the larger decrease in urinary creati-

nine seen in this group.Feeding the two dietary amino acid pre-

cursors of creatine, arginine, and glycine, en-hances transamidinase activity and increasesurinary excretion of guanidoacetate. The ac-

tivation of transamidinase results in a higherrate ofcreatine production, as well as enlarge-

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

28

CREATINE FREE DIET

120 140 qms = 0 10 gms 20 qms =

#{149}12 3 4 5 6 7 8 9 10 123

484 HEYMSFIELD ET AL

E

w2

2

w

C-,

>-

2

MEATINCLUC

PROTEIN GMS

WEEKSFIG 3. Influence of dietary protein and creatine on urinary creatinine excretion. Figure 3 was prepared from the

data presented in Reference 52.

ment of the creatine pool independent of

changes in muscle mass. Crim and coworkers(28, 40) fed normal volunteers a creatine-free

diet for 71 days, adding during the last 10days isonitrogenous amounts (4 g nitrogen!

day) of either an equimolar mixture of argi-nine and glycine or alanine. There was asmall but significant increase in creatinineexcretion in two of the four subjects fed ar-ginine-glycine; no significant changes wereobserved in the other two subjects, or in the

alanine fed group.The second dietary constituent that influ-

ences urinary creatinine excretion is pre-formed creatine. The amount of dietary crea-tine directly influences the size ofthe creatinepool, which in turn is proportional to theoutput of creatinine in the urine. The“average” daily American diet includes about200 g (7 oz) of meat (53), which contains 700mg creatine and 37 mg creatinine (54). Cook-ing the meat at 77#{176}Creduces creatine content

to 570 mg and increases creatinine to 160 mg.Thus, average daily dietary intake of creatine

and creatinine replaces about one-third toone-half of the daily urinary creatinine losses.

The influence of dietary creatine on uri-nary creatinine excretion has been examined

by a number of investigators. Switching froman ad libitum diet, to a meat-free proteinsource (casein), Bleiler and Schedl (52)

showed a reduction in urinary creatinine over6 wk from 1.9 to 1.5 g/day (-21%). Theseinvestigators further demonstrated a rapidincrease in urinary creatinine when subjectswere switched from a creatine-free 120 g!dayprotein diet to an ad libitum general diet (Fig3). Crim et al (40) shifted from a 9-day lowcreatine diet (0.23 g!day), to a 10-day highcreatine diet (10 g!day), and found a prompt10 to 30% increase in urinary creatmine cx-cretion (Fig 4). In a subsequent creatine-free

period, Crim showed that nitrogen and po-tassium balance remained positive despite de-dining creatinine excretion, indicating a de-gree of independence between creatmine cx-

cretion and muscle or lean body mass. Theindependence between creatinine excretionand muscle or fat free body mass is alsoobserved when switching from a general hos-pital diet which includes meat, to a creatine-free hyperalimentation formula and viceversa. The undernourished cancer patientpresented in Figure 5A was undergoing en-teral hyperalimentation. Switching from asolid food diet that included meat, to a crea-tine-free formula caused a gradual reductionin urinary creatimne excretion. The declinein urinary creatinine output, however, was

paralleled by strongly positive nitrogen bat-ance, enlarging fat free body mass and anincreasing arm muscle area. The reverse of

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

2.50

2.25

2.00

1.75

1.50

1.25(g/D)

1018

I 2

10

.1_

0 1 2 3 4 5 6 7 8 9 10 11

WEEKS

FIG 4. Influence of dietary creatine intake on urinary creatininc excretion. Figure 4 was prepared from the datapresented in Reference 40.

CRLATIN1NL MUSCLE MASS MLT1IOL) 485

LU22

LU

C-,>-

2

this effect is seen in the anorexia nervosapatient depicted in Figure 5B, who wasswitched from a creatine-free hyperalimen-tation formula to a general hospital diet that

included meat.The third component of the diet that influ-

ences creatinine excretion, mentioned above,

is creatinine. Dietary creatinine undergoesintestinal absorption and prompt renal excre-tion (25, 55).

When applying the creatinine muscle massmethod to assess muscle mass in human sub-jects, most, if not all these dietary factors canbe eliminated if subjects are placed on a lowcreatine, meat-free diet with 60 to 80 g ofprotein per day. Most relevant, however, is

that the diet be constant, and not changedfrom one type to another during the periodof study.

Menstrual cycle

Studies by Smith (56) demonstrated a riseand fall in creatine and creatinine excretionduring the menstrual cycle, but changes wereapparently unrelated to estrogen levels. Heshowed in repeated studies that creatinineexcretion increased by 5 to 10% late in thesecond half of the menstrual cycle, and thendecreased in the days preceding or duringmenstrual flow.

Aging

In the adult, fat free body mass declines

with increasing age, and most of this loss in

lean tissue can be attributed to the atrophy ofskeletal muscle (57). Creatinine excretion alsodeclines with age, and this presumably re-flects the diminution in muscle mass (57).Dietary factors might also account for someofthe decrease in urinary creatinine with age,as the elderly tend to eat less meat. It isunknown if the reduction in creatinine excre-tion found in the elderly is in proportion tothe loss in muscle mass, and it also has not

been established how each of the terms inequation 3 (Table 2) is specifically influencedby senescence. The age dependency of thecreatinine equivalence is therefore unknown.

Infection,fever, and trauma

Waterlow et al (58) found increased crea-

tine turnover in rats suffering from lung ab-scesses, and concluded that the creatinine

equivalence may no longer be valid in thepresence of infection. The severe battle cas-ualties studied by Frawley et at (59) showedan increase in creatinine excretion, but the

daily output appeared to be independent oftype, degree, or location of injury. Schiller etal (60) confirmed the studies of both Water-low and Frawley by demonstrating a rise increatinine excretion with both fever and Se-vere trauma. In some patients, an increase of

20 to 100% in urinary creatinine excretionoccurred in the immediate posttraumatic pe-riod (60-62). Therefore, in the acute phase of

injury, other available indices ofmuscle mass(Table 1) should be applied.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

1 6 11 16 21 26 31 36

B

486 HEYMSFIELD El AL

GENERAL DIET -4.-- CREATINE FREE FORMULA -#{248}.�

URINARY CREATININE g/O

r�iEL2

LU22

LU

C-)

>-

2

0.70

0.65

0.60

0.55

0.50

0.45

0.40

L�N(g/D)

+5’0F-s----

L�FFM (kg/D)

1(-1 1

L�AMA (cm�)*+10+51 -�

#{176}I‘- -�-#{149}---�---- S

-SI:-10

DAY 1 6 11 16 21

A Cumulative

26 31 36 41

DAY

FIG 5. Influence on creatininc excretion when switching to and from a creatine-free hyperalimcntation formulawhen the hospital diet contains meat. The undernourished cancer patient (A ) and the anorexia nervosa patient (B)were undergoing enteral hyperalimentation. Abbreviations: FFM, fat free body mass; N, nitrogen.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

CREATININE MUSCLE MASS METHOD 487

Renal disease

As glomerular filtration rate falls in chronicrenal failure, there is a decrease in the excre-

tion of creatinine. Goldman (63) noted thatthe reduced output of creatinine is especiallyapparent when serum creatinine exceeds 6

mg/dl. Noting that muscle mass was not sig-nificantly decreased, Goldman suggested thatthe low output was due either to reducedcreatinine production or an alternate excre-

tory pathway. Jones and Burnett (64) inves-

ligated the “creatinine deficit” in patientswith renal insufficiency, and found that 15.9

to 65.7% ofcreatinine formed in their subjectswas metabolized or excreted via extrarenal

routes. The studies of Mitch and coworkers(65, 66) also indicate metabolism rather than

excretion of creatinine. These workers iden-tified two pathways of creatimne catabolism;

a recycling of creatinine to creatine, and in-testinal degradation of creatinine to productsother than creatine. In advanced renal dis-

ease, therefore, the creatinine equivalencemethod should be avoided.

Section summary

Table 5 summarizes factors other thanmuscle mass which influence daily creatinineexcretion. The accuracy of predicting musclemass from urinary creatinine excretion foreach individual depends on the contributionof these seven factors, and on individual ge-netic differences in creatine metabolism. Lit-tle information is available on the latter sub-ject.

TABLE 5

Factors that influence daily urinary crcatinine excretion

other than changes in muscle mass

Condition Magnitude of effect

Normal daily variation ±4 to 8%

Very strenuous exercise +5 to 10%Emotional stress ±5 to 10%Diet: switching from meat diet to ± 10 to 30%

creatine-free diet or vice versaMenstrual cycle: minimum dur- + 10 to 15%

ing menstrual flow, maximum

second half of cycleRenal disease: serum creatinine

>2 mg/dl <6 mg/dl S

�6 mg/dl �tSevere infection, high fever, +20 to l00%t

trauma

S Magnitude unknown.

t Magnitude and duration variable.

Experimental studies of the creatinine-muscle mass method

Although the assumptions of the creati-

nine-muscle mass method presented previ-ously appear generally valid, the basic ques-tion has yet to be addressed: Does the 24-hurinary creatinine excretion correlate withmuscle mass? If there was a simple answer tothis question, our review would have little

purpose. Since no one has yet examined theliterature on this subject, we now proceedwith this task. Our discussion is divided intoanimal and human studies.

Animal studies

Garn (67) has emphasized the advantagesof animal studies. The investigator can createa homogeneous study group by selecting an-imals that are similar in age and geneticbackground, and who are subjected to the

same activity level and diet. These ideal con-ditions are rarely obtainable in human stud-ies. Furthermore, animal studies have an ad-

ditional advantage: muscle mass can be meas-ured very exactly by direct carcass analysis.

The murine studies of Chinn (68) are the

most relevant animal experiments to our dis-cussion. Chinn found that muscle protein(carcass analysis) could be predicted within±4. 17% from 24-h urinary creatinine in threegroups of adult rats differing in muscle mass:normally fed males and females, and malesfed 50% of their normal diet.

Similar rat studies by Meador et al (69)and Graystone (1 1) confirm that under con-trotted laboratory conditions, creatinine cx-cretion closely parallels muscle mass.

Human studies

Apart from the biological and dietary var-iability present in human studies, the mostsignificant problem is that an independent,completely valid, noninvasive measure ofmuscle mass is unavailable. The validity orreal variability of the creatinine-muscle massmethod is thus difficult if not impossible toassess.

An additional problem is the use of differ-

ent definitions of muscle mass. The latterissue deserves additional comment. In refer-ring to “creatinine equivalence,” nonmusclecreatine pools, because of their small contri-bution to creatinine output, are neglected.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

488 HEYMSFIELD ET AL

For the most part, so are nonskeletal musclesources of creatmine, such as heart, uterus,aorta, and other tissues with smooth muscle.Generally, the creatine concentration in thesemuscles is less than striated skeletal muscle(25). In nearly all studies, the creatinine out-put is determined from values appropriate forskeletal muscle, thus introducing a probablesmall bias due to creatinine which originatesoutside skeletal muscle.

The units used to express the creatinineequivalence also differ among authors. Skel-etal muscle is composed of intra- and extra-cellular fluids and solids. Some, but not allworkers refer to fat-free wet muscle, which is2 to 5% less than whole wet muscle mass(WWM). Creatinine reflects the intracellularspace of muscle tissue, so in addition someinvestigators express results per kg of intra-

cellular or non-collagenous proteins (NCP).The creatinine equivalence is therefore pre-sented per kg WWM, per kg fat-free wetmuscle, and per kg NCP.

The fmal difference between the reviewedstudies was diet. The feeding programs weredescribed as creatine-free, low creatine, orgeneral diets.

Hunter (36), in his classic monograph oncreatine and creatinine, relates Burger’s (24)

calculation of the creatinine equivalence: “Anormal man of 63. 1 kg was assumed, on thebasis ofVierordt’s (70) figures, to possess 25.4kg of muscle. The creatine content of this

mass was taken to be 88.4 g, and 2 g morewere allowed for the other organs ofthe body.The creatinine output ofthe subject averagedI .36 g daily. From these data Burger calcu-lates that one gram of urinary creatinine perday corresponds to 22.9 kg of whole wetmuscle. This result seems to involve an arith-metical error, and is admittedly an approxi-mation.” Talbot (71), applying a similar ap-proach to literature values for the creatininecoefficient and Vierordt’s data for musclemass, found “one gram of urinary creatinineper day corresponds (on a general diet) forthe infant to 17.8 kg, and for the man, to 17.9kg of (whole wet) muscle” (Table 6).

Total body muscle mass is not required tocalculate the creatinine equivalence from Ta-ble 2 equations and the data presented inTable 3. Substitution of creatine and phos-phocreatine concentrations and turnoverrates into equations 2 and 3 produce valuesof 17.55, 17.9, and 18.45 kg WWM!g ofcreatinine, and 3.12 and 3.41 kg NCP!g crc-atinine (Table 7). These estimates would ap-ply to patients on a general diet. Somewhat

TABLE 6Studies reporting the “creatinine equivalence”

AuthorMethod ofdertvong

muscle massStudy diet

Creatin inc equivalence

SubjectsAverage

Units’

(range)WWM FFWM NCP

Burger (24) Vierordt’s dissec-

tion tables (70)

Not stated; pre-

sumably gen-

eral

22.9t X None

Talbot (71)

Picou et a!

Vierordt’s dissec-

tion tables (70)

‘5Ncrcatine

General

Meat-free

17.9, 17.8

(13.5-3 1.6)18.6 (13.8-

X

X

Children, adults

Children recover-

(8)

Kreisberg ‘4C-creatine General

31.9)

16.2 (14.5- X

ing from PEMf

Adults with

et al (9) 18.6) chronic diseaseGraystone Isotopic method Low creatine 20 X Normal children

(11) basedoncon-stant developed

in rats.

Current Calculated equiva- General 17.55, 17.9, X

review lence from 18.45 X

(Table 7) equations 2 and

3

3.12, 3.41

S WWM; FFWM, fat-free wet muscle; and NCP, all in g/kg.

t According to Hunter (36) there is a calculation error in Burger’s data.1:Protein-energy malnutrition.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

CREATININE MUSCLE MASS METHOD 489

TABLE 7Examples of creatinine equivalence calculated from Table 2 equations and Table 3 data

ExampleTable 2

equationK’ K’ Total Cr Cr/P�Cr

Creatinine equivalencet

Kg WWM/g Cm Kg NCP/g Cm

A 3 0.011 0.0264 1.l9/2.74f 17.9

B 2 0.0169 3.91�,22� 17.55 3.12

C 3 0.011 0.0264 1.53/2.39�8.3/12.9�

18.45t 3.41

Abbreviations: K, , and K2 are fractional breakdown rates of respectively total creatinc, creatine, and

phosphocreatinc (Cr, P-Cr); Cm is creatininc.

S Day�.

t General diet.:1:g/kg wet wt.

§g/kgNCP.

higher values would be expected on a crea-tine-free diet.

The “Talbot” creatimne equivalence pre-sented above was widely quoted throughoutthe next decade. Graystone (1 1) later modi-fled this value to 20 kg of fat-free wet skeletalmuscle per g of urinary creatininc (Table 6).The initial observation that led Graystone to

revise the creatinine equivalence was madeby Cheek et al (72), who noted that the

calculated muscle mass in children was 20%of body weight; this percentage was less thanthe figure of 25% established by the 19thcentury German anatomists. Cheek’s expla-nation for this discrepancy was that thesechildren were on a low creatine diet, and he

therefore proposed an upward adjustment ofthe creatinine equivalence in order to com-pensate for the apparent low predicted musclemass. Support for the new creatinine equiv-alence was subsequently provided by Gray-stone’s animal studies (1 1). Investigating the7-wk-old rat, she found that 56.7% of intra-cellular water was in muscle tissue, and mus-dc is about the same proportion of bodyweight in the young rat and child. This frac-tion was then applied to intracellular watermeasured in children in order to calculatetotal muscle water. Since the ratio of musclewater to muscle solids was known in vivo,muscle mass could then be calculated. Crc-atinine collections in the same children thenallowed computation of the creatinine equiv-alence for muscle mass. Results were 21.0and 20.8 for boys and girls, respectively.These values did not differ significantly fromCheek’s estimated value of 20, and thereafterthe latter remained the creatinine equivalence(1 1). The low creatine diet ingested by these

children no doubt increased the creatinine

equivalence compared to Talbot’s data.A more recent method of measuring mus-

dc mass in children and adults is by isotopicdilution of creatine. Picou et al (8), applyingthis method to children who had recoveredfrom protein-energy malnutrition, found thattotal WWM varied between 15 and 37% ofbody weight, with a mean of 28%. One g ofurinary creatinine on a meat-free diet corre-sponded to anywhere from 13.8 to 3 1.9 kg ofWWM, with a mean (±SD) of 18.6 ± 6.6 kg.

In the study of Kreisberg et al (9), total wetfat-free muscle mass was evaluated in fouradult patients on a general diet by isotopicdilution of ‘4C-creatine. The calculated crc-

atinine equivalence was 16.2 kg fat free wetmuscle per g urinary creatinine (range 14.5 to1 8.6). Greatrex et at ( 10) have criticized theisotopic creatine-muscle mass method on thegrounds that creatine concentration variesfrom muscle to muscle, and a single biopsydoes not adequately account for this variabil-ity.

All methods of assessing muscle mass pre-sented above were indirect. Studies from our

center in a wide variety of normal and un-dernourished subjects indicated that ana-tomic muscle mass correlated strongly with

urinary creatinine excretion (73). Figure 6shows arm muscle area (AMA) calculatedfrom mid-arm circumference and tricepsskinfold thickness, plotted against 24-h uri-nary creatinine excretion. The subjects (n =

25) were on low creatine balance diets for 7

days, and plotted creatinine values are theaverage of days 4 to 7. Patients with pendu-bus or excessively large skinfolds were cx-

cluded, because anthropometric measure-

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

ments of muscle mass in these individuals is

inaccurate. Although AMA correlated wellwith creatinine excretion in this study (r =

0.94, p < 0.001), no creatinine equivalence

could be calculated from this type of analysis.

E

6.)

E

I-

C-,

7

6

5

4

3

2

_O 5 10 15 20 25 30 35 40 45 50 55 60

AMA (cm�)

FIG 6. Correlation between urinary creatinme cxcre-tion (CR/HI) and AMA. The latter was measured an-

thropometrically. The data are for 25 subjects, r = 0.94,

p < 0.001, Y = 0.14 X + 1.32. From Reference 73.

% L�CR

% t�IAMA

490 HEYMSFIELD El AL

12

11

10

9

8

Figure 7 shows similar studies conducted lon-gitudinally in patients losing (n = 3) or gain-

ing weight (n = 6); again, the change in armmuscle area correlated reasonably well withthe change in urinary creatinine. The datacompiled in the Minnesota experiment (74)at 12 and 24 wk of semistarvation is alsoplotted in Figure 7, and agrees closely withour more recent data. The AMA, as calcu-lated in the Minnesota experiment, includedfat. We therefore assumed a base-line tricepsskinfold thickness of 12 mm, and at 12 and24 wk of semistarvation (S 12 and 524) of 7

. FEMALES and 4 mm, respectively. Values for AMA

. MALES were then recalculated according to themethod of Heymsfleld et at (73). Both the

original and recalculated data provided qua!-itatively similar results. In a similar type of

analysis, Standard et a! (75) have shown asignificant correlation (r = 0.74, p < 0.001)

between muscle thickness and creatinine cx-cretion in children recovering from proteinmalnutrition. The observed variability whencomparing the two measures of muscle mass(anthropometric and biochemical) might berelated to the inclusion of extracellular fluid

and interfibrillar fat in the anthropometric

FIG 7. Longitudinal changes in arm muscle area (i�.AMA) and 24-h urinary creatinine excretion (i�CR), bothexpressed as a percentage of base-line value. Subjects were losing weight secondary to cancer, or gaining weightduring hyperaliinentation. The dashed line represents the change in muscle mass observed during 12 and 24 wk of

semistarvation in the Minnesota experiment (74).

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

CREATININE MUSCLE MASS METHOD 491

method, the inaccuracies inherent in the an-

thropometric method (6, 73), and between

subject differences in diet and creatinc turn-

over rate.

From these human investigations it is con-

cluded that: 1 ) studies differed in diet, healthstatus of study subject, definition, andmethod of measuring muscle mass, 2) thecreatinine equivalence in humans is not de-

finitvely established, 3) even ifsuch an equiv-alence can be agreed upon, the SD aroundthis average tentatively appears quite large,

and 4) reliable estimates of the creatmineequivalence and associated variability will be

forthcoming only when a definitive methodof measuring muscle mass in vivo is available

to clinical investigators. Such methods arenow becoming available, but at the present

are limited to only a few centers (2, 16).Unlike muscle, lean body mass (LBM) can

be measured accurately in the clinical setting;because muscle is roughly 30 to 50% of LBM,many workers have focused on the correla-tion of creatinine with accurately and easilymeasured LBM (76-79).

When LBM is correlated with creatininc,there is almost universal agreement that thesetwo indices are closely related. Some varia-tion in the predictive value of this relation isfound by some authors, but the principaldifference appears to be the precision withwhich urine samples were collected, and thenumber of urines analyzed.

Why should a precise relation between crc-atinine and LBM be found in the abovementioned human studies, while so muchvariability is found in the studies relating

muscle mass directly to creatinine? Two pos-sible explanations are offered. The first is thatcreatinine excretion may be a more accuratereflection of the total creatine pool (muscleplus other soft lean tissues), rather than themuscle creatine pool. The small contribution

of nonskeletal muscle creatine to urinary crc-atinine has been discussed above. A second,and more likely explanation, is that LBM canbe measured more accurately and reproduc-ibly than muscle mass. A much larger numberof subjects have therefore been examined,and usually under much more carefully con-trolled dietary conditions. It would thereforeseem likely that a very carefully controlledstudy (diet, activity, absence of stress or dis-

ease, adequate number of urine collections)correlating creatinine with some as yet un-

developed or unverified accurate in vivomeasure of muscle mass would provide a

creatinine equivalence with minimum varia-

bility.

Conclusions and summary

The collective review of the literature al-lows us to draw the following conclusions.

1 ) If an individual consumes the sameamount of a constant composition diet, crc-atininc output in the urine is proportional to

muscle mass.

2) Changing from a meat diet to a creatine-free diet and vice versa causes readjustmentover time in the size of the creatine pool. The

result is a change in urinary creatinine excre-tion that is independent from changes in mus-dc mass. Diet therefore must remain reason-ably constant.

3) Other than dietary factors, the vanabi!-ity in the creatminc-musclc mass relationshipis due to normal daily variation in creatinineexcretion; variability is increased by severeemotional stress, by sampling during differentphases in the menstrual cycle, and by veryvigorous exercise. Collecting several consec-

utive 24-h specimens, and ensuring a mini-mum ofemotional stress and physical activitycan eliminate some of the variation in creat-inine excretion.

4) Severe renal insufficiency, acute infec-tion, or the early phases of major injury are

conditions in which creatinine output is notproportional to muscle mass. Quantifying

muscle mass in these patients should be ac-complished by other methods. A notableproblem however, is that not all of the meth-

ods listed in Table 1 are clinically available,and of those that are, most have limitationsthat also limit their accuracy.

5) The creatinine equivalence is roughly17 to 20 kg muscle per g of creatinine, butthe variability and precision of these esti-mates remain uncertain. The value is higher

on a creatine-free diet, and lower for a meatdiet. Converting creatinine to muscle massmay thus serve as an approximation, butshould not be considered a definitive ap-praisal of muscle mass.

6) In the evaluation of selected patients,

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

492 HEYMSFIELD El AL

creatinine can be compared to serial samplescollected over time to establish trends (assum-ing diet remains fairly constant), or compared

directly to “standard” tables based on age,sex, stature, and diet (80, 81). The comparisonto reference tables provides an estimate of the

“percent of standard” muscle mass.

In summary, the use of urinary creatinineexcretion as an index of muscle mass is con-ceptually interesting, and the method mayprovide useful information in carefully se-lected subject groups. There remains, how-

ever, a large group of individuals in whomthe method cannot be reliably applied. Forexample, what set of reference values shouldbe used in the athlete who ingests an unusu-

ally large amount of meat in his!her diet, orin the cancer patient with reduced food (ic,creatine) intake secondary to anorexia? Howcan change in muscle mass in the latter pa-tient be accurately followed over time, whenfood intake may vary with exacerbation andremission of disease? There is little question

that the assumptions of the creatinine-musclemass method are invalid in the early phase ofinjury, and in the presence of fever and in-

fection. Since undernutrition and muscleatrophy are often found in patients in whomthe creatinine-muscle mass method is mac-curate, a priority research focus should be thedevelopment and validation of accurate andpractical methods of measuring muscle massin these individuals.

References

1. Behnke AR, Wilmore JH. Application ofthc variousfield methods. In: Evaluation and regulation of bodybuild and composition. Englewood Cliffs, NJ: Pren-

lice-Hall, mc, 1974:58-9.

2. Cohn SH, Sawitsky A, Vartsky D, Yasumura 5,Zanzi I, Ellis KJ. Body composition as measured by

in vivo activation analysis. In: Nutritional assess-

ment-present status, future directions and pros-pects. Columbus, OH: Ross Laboratories, 198 1:99-102.

3. Heymsfield SB, Stevens V, Noel R, ct al. Biochemi-

cal composition of muscle in normal and protein-

energy starved human subjects: relevance to anthro-pometric measurements. Am J Clin Nutr

1982;36: 131-42.4. Bistrian BR, Blackburn GL, Sherman M, Scrimshaw

NS. Therapeutic index of nutritional depletion in

hospitalized patients. Sum Gynecol Obstet

1975; 141:5 12-16.

5. Viteri FE, Alvarado J. The creatininc height index:

its use in the estimation of the degree of protein

depletion and repletion in protein calorie malnour-

ished children. Pediatrics 1970;46:696-706.

6. Heymsfield SB, Olafson RP, Kutner MH, Nixon

DW. A radiographic method ofquantifying protein-

energy undernutrition. Am J Clin Nutr 1979;32:693-

702.7. Stouffer JR, Westervelt MS. A review of ultrasonic

applications in animal science. J Clin Ultrasound

l976;5: 124-8.

8. Picou D, Reeds PJ, Jackson A, Poulter N. The

measurement of muscle mass in children using crea-

tine-”-N. Pediatr Res 1976; 10:184-8.9. Kreisberg RA, Bowdoin B, Meador CK. Measure-

ment of muscle mass in humans by isotopic dilution

ofcreatine-’4-C. J App! Physiol l970;28:264-7.10. Greatrex G, Morgan AP, Mcore FD. Evaluation of

muscle mass measurement by creatine-’4-C dilution

in the dog. Metabolism l972;21:757-60.

1 1. Graystone JE. Creatinine excretion during growth.

In: Cheek DB. Human growth: body composition

cell growth, energy and intelligence. Philadelphia,

PA: Lea & Febiger, l968;12: 182-97.12. Anderson EC. Three component body composition

analysis based on water determinations Ann NYAcad Sci 1963; 110:189-210.

13. Burkinshaw L, Hill GL, Morgan DB. Assessment of

the distribution of protein in the human body by invivo neutron activation analysis. Nuclear Activation

Techniques in the Life Sciences (Proc Symp Vienna,

1978). Vienna: LAEA, 1979:787-9.

14. Cohn SH, Vartsky D, Yasumura 5, et a!. Compart-

mental body composition based on total-body nitro-

gen, potassium, and calcium. Am J Physiol

1980;239:524-30.

15. Cohn SH, Cartenhaus W, Sawitsky A, et a!. Corn-

partmental body composition of cancer patients by

measurement of total body nitrogen, potassium, andwater. Metabolism 198 1;30:222-9.

16. Mallard J, Hutchinson JMS, Edelstein WA, Ling

CR, Foster MA, Johnson G. In vivo NMR imagery

in medicine. The Aberdeen approach, both NMR

physical and biological. Phil Trans R Soc Lond B

l980;280:5 19-33.

17. Lukaski H, Mendez J. Relationship between fat-freeweight and urinary 3-methyl histidine excretion in

man. Metabolism 1980;29:758-6l.

18. Lukaski HC, Mendez J, Buskirk ER, Cohn SH. A

comparison of methods of assessment of body corn-position including neutron activation analysis of to-

tal body nitrogen. Metabolism 198 1;30:777-82.19. Chinn KSK. Prediction of muscle and remaining

tissue protein in man. J AppI Physiol l967;23:713-715.

20. van Dobeln W. Estimation of muscle mass of the

human body from potassium-40 determinations.Medd Flygg Navalmed Namnd 1962; 11:1-4.

21. Heymsfield SB, McManus C, Smith J, Stevens V.Muscle mass: reliable indicator of protein calorie

malnutrition. Am J Clin Nutr l982;35:1 192-1199.

22. Myers VC, Fine Mi. The creatine content of muscleunder normal conditions: its relation to the urinary

creatinine. J Biol Chem 1913;14:9-26.

23. Palladin A, Wallenburger L. Contribution a l’ctude

de Ia formation de la creatine dans l’organismc

animal. Compt Rend Soc de Biol 1915:111-113.

24. Burger MZ. Bcitrage zum Krcatininstoffwechsel: I.Die Bedeutung des Kreatininkoefizienten f#{252}rdie

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

CREATININE MUSCLE MASS METHOD 493

quantitative Bewertung der Muskulatur als K#{246}rpcr-

� II. Di� kf�litIh. �nausscheidung: die StOrungen des Muskelstoffwech-sels. Z Ges Exp Med l9l9;9:361-99.

25. Hunter A. The physiology ofcreatine and creatinine.

Physiol Rev 1922;2:586-626.

26. Hoberman HD, Sims EAH, Peters JH. Creatinc and

creatinine metabolism in the normal male adult

studied with the aid of isotopic nitrogen. J Biol

Chem 1948; 172:45-58.27. Fitch CD, Sinton DW. A study of crcatine mctabo-

lism in diseases causing muscle wasting. J Clin Invest

1964;43:444-52.28. Crim MC. Creatine metabolism in men: crcatine

pcol size and turnover in relation to creatine intake.J Nutr l976;106:37l-8l.

29. Baldwin D, Robinson PK, Zierler KL, Lilienthal JL.

Interrelations of magnesium, potassium, phospho-

rus, and creatinc in skeletal muscle of man. J Clin

Invest l952;3 1:850-8.

30. Vignos PH Jr, Warner JL. Glycogen, creatinc, and

high energy phosphate in human muscle disease. JLab Clin Med 1963;62:579-90.

3 1. Walker JB. Creatine: biosynthesis, regulation, and

function. Adv Enzymo! l979;50: 177-242.

32. Bloch K, Schoenheimer R. The biological precursorsofcrcatinc. J Biol Chem 1941;l38:167-91.

33. Borsook H, Dubnoff JW. The hydrolysis of phos-

phocreatine and the origin of urinary creatinine. JBiol Chem 1947;168:493-510.

34. Krisko I, Walker JB. Influence of sex hormones on

amidinotransferasc levels. Metabolic control of crca-

tine biosynthesis. Acta Endocrinol 1966;53:655-62.

35. Haugland RB, Chang DT. Insulin effect on crcatinetransport in skeletal muscle. Proc Soc Exp Biol Mcd

1975; 148:1-4.

36. Hunter A. Creatinc and creatininc, monographs on

biochemistry. New York, NY: Longmans, Green

and Co, Ltd, 1928.

37. Fitch CD, Lucy DD, Bomhofen JH, Dalrymple GV.

Creatine metabolism in skeletal muscle: creatinc ki-

netics in man. Neurology l968;l8:32-42.

38. Hahn A, Meyer G. On the mutual transformation ofcreatine and creatinine. Ztschr Biol 1928;78: 1 1 1-15.

39. Saks VA, Rosenshtraukh LV, Smirnov VN, Chazov

El. Role of crcatine phosphokinase in cellular func-

tion and metabolism. Can J Physiol Pharmocol1978;56:69 1-706.

40. Crim MC, Calloway DH, Margen S. Creatinc me-

tabolism in men: urinary creatine and creatinine

excretions with creatine feeding. J Nutr

1975; 105:428-38.

4 1 . Bjornsson ID. Use of serum creatinine concentra-tions to determine renal function. Clin Pharmaco-kinetics l979;4:200-22.

42. Folin OH. Laws governing the chemical compositionof urine. Am J Insanity l904;60:709-l0.

43. Rose WC. The metabolism of creatine and creati-

nine. Ann Rev Biochem 1933;2: 187-207.

44. Cryer PE, Sode J. Variation in urinary creatinineexcretion and its relationship to measurement of

urinary 17-hydroxy-corticosteroids. Clin Chem1970; 16:1012-15.

45. Greenblatt DJ, Rausil BJ, Harmatz JS, Smith 1W,

Duhme DW, Koch-Weser J. Variability of 24-hoururinary creatinine excretion by normal subjects. J

Clin Pharmacol l976;l6:32l-8.

4�. llo�� W. Uf�fiAf�k�liE�U1of #{233}f�ALIB#{234}Ah� �fe�li.nine associated with physical exertion and its rela-tionship to carbohydrate metabolism. Biochem J

l939;33: 1425-31.

47. Srivastava SS, Mani KV, Soni CM, Bhati J. Effect

ofmuscular exercises on urinary excretion of creatineand creatinine. Ind J Med Res 1957;55:953-60.

48. Scrimshaw NS, Habicht JP, Piche ML, Cholakos B,

Arroyave G. Protein metabolism of young men dur-

ing university examinations. Am J Clix Nutr

1966; 18:321-4.

49. Schottstaedt WW, Grace WJ, Wolff HG. Life situ-

ations, behavior, attitudes, emotions and renal excre-

tion of fluids and electrolytes: V. Variations in cx-

cretion of endogenous creatininc. J Psychosom Res

1956;1:292-8.

50. Lykken GI, Jacob RA, Munoz JM, Sandstead HH.A mathematical model of creatine metabolism innormal males-comparison between theory and cx-

periment. Am J Clin Nutr l980;33:2674-85.5 1. Van Pilsum JF. Creatin and creatine-P in normal

and protein depleted rats. J Biol Chem

1957;228: 145-8.

52. Bleiler RE, Schedl HP. Creatinine excretion: varia-

bility and relationships to diet and body size. J Lab

Clin Mcd l972;59:945-55.

53. USDA Nationwide Food Consumption Survey

1977-78, (preliminary report no 2). Washington, DC:

the Office of the Director, Science and Education

Administration, US Department of Agriculture,

Spring, 1977.

54. Macy RL Jr, Naumann HD, Bailey ME. Water-

soluble flavor and odor precursors of meat. 5. Influ-ence of heating on acid-extractable nonnucleotidechemical constituents ofbeef, lamb and pork. J Fcod

Sci 1970;35:83-7.

55. Shaffer P. The excretion of kreatinin and kreatin in

health and disease. Am J Physiol l908;23: 1-17.

56. Smith OW. Creati.nine excretion in women: datacollected in the course of urinalysis for female sex

hormones. J Clin Endocrinol l942;2: 1-12.57. Tomlinson BE, Walton JN, Rebeiz JJ. The effects of

aging and of cachexia upon skeletal muscle-a his-

topathological study. J Neurol Sci l969;9:32l-46.

58. Waterlow JC, Neale RJ, Rowe L, Palm I. Effect of

diet and infection on creatine turnover in the rat.

Am J Clin Nutr 1972;25:371-5.

59. Frawley JP, Aria CP, Howard JM. Muscle metabo-

lism and catabolism in combat casualties. Arch Surg

1955;7 1:612-16.

60. Schiller WR, Long CL, Blakemore WS. Creatininc

and nitrogen excretion in seriously ill and injured

patients. Surg Gynecol Obstet 1979; 149:561-6.

61 . Geiger JW, Long CL, Sills LM, Blakemore WS.

Creatine, creatinine, and urinary nitrogen excretion

in traumatized males. Fed Proc 198 l;40:852(abstr).

62. lhrclfall Ci, Stoner HM, Galasko CSB. Patterns in

the excretion of muscle markers after trauma and

orthopedic surgery. J Trauma 198 l;2l:140-7.

63. Goldman R. Creatinine excretion in renal failure.Proc Soc Exp Biol Mcd 1954;85:446-8.

64. Jones JD, Burnett PC. Creatinine metabolism inhumans with decreased renal function: creatininedeficit. Clin Chem 1974;20: 1204-12.

65. Mitch WE, Walser M. A proposed mechanism for

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from

494 HEYMSFIELD El AL

reduced creatininc excretion in severe chronic renal

failure. Nephron l978;21:248-54.66. Mitch WE, Collier VU, Walser M. Creatininc me-

tabolism in chronic renal failure. Clin Sci

1980;58:327-55.

67. Garn SM. Human biology and research in bodycomposition. Ann NY Acad Sci 1963;l 10:429-46.

68. Chinn KSK. Potassium and creatinine as indexes of

muscle and nonmusc!e protein in rats. J Nutrl966;90:323-30.

69. Meador CK, Kreisberg RA, Friday JP, et a!. Muscle

mass determination by isotopic dilution of creatine-

‘4-C. Metabolism 1968; 17:1104-8.70. Vierordt H. Date und labellen fur Mediziner. Ed 3.

Jena, Gustau Fischer 1906:36-43.71. Talbot NB. Measurement of obesity by the creati-

nine coefficient. Am J Dis Child 1938;55:42-50.72. Cheek DB, Brasel JA, Elliott D, Scott RE. Muscle

cell size and number in normal children and in

dwarfs (pituitary, cretins and primordial) before and

after treatment. Bull Johns Hopkins Hosp1966;1 19:46-62.

73. Heymsfield SB, McManus C, Smith J, Stevens V,Nixon DW. Anthropometric measurement of musclemass: revised equations for calculating bone-free

arm muscle area. Am J Clin Nutr 1982;36:680-90.

74. Keys AJ, Brozak A, Henschel 0, Michelson 0, lay-br HL. In: The biology of human starvation. Mm-

neapolis, MN: Minnesota Press 1950:130-60.75. Standard KL, Willis VG, Waterlow JC. Indirect

indicators of muscle mass in malnourished infants.

Am J Clin Nutr l959;7:271-9.

76. Muldowney FP, Crooks J, Bluhm NM. The relation-

ship of total exchangeable potassium and chloride tolean body mass, red cell mass and creatinine excre-

tion in man. J Clin Invest 1957;36:l375-81.

77. Edwards KDG, Whyte HM. Creatininc excretionand body composition. Chin Sci 1959;18:361-6.

78. Forbes GB, Bruining GJ. Urinary creatinine excre-tion and lean body mass. Am J Chin Nutrl976;29: 1359-66.

79. Miller Al, Blyth CS. Estimation of lean body massand body fat from basal oxygen consumption and

creatinine excretion. J Appl Physiol l952;5:73-8.

80. Blackburn GL, Bistrian BR, Maini BS, et al. Nutri-tional and metabolic assessment of the hospitalized

patient. JPEN 1977;l:l 1-22.

81. Turner WJ, Cohn S. Iota! body potassium and 24

hour-creatinine excretion in healthy males. ClinPharmacol Iher 1975; 18:405-12.

by on Septem

ber 23, 2009 w

ww

.ajcn.orgD

ownloaded from