Caloric Equivalents of Gained or Lost Weight

By MAX WISHNOFSKY, M.D., 1958

Annotated on Fermat's Library with 8 comments. Text and annotations as Markdown Original paper

## TL;DR Max Wishnofsky’s 1958 paper was the first to translate bo...
Max Wishnofsky (December 17, 1899 – August 2, 1965) was a Russian-b...
To determine the % of fat in adipose tissue, researchers use chemic...
Wishnofsky multiplies the classic 9.5 kcal · g⁻¹ figure for animal ...
The rule collapses during total fasting: catabolising protein drags...
Wishnofsky anticipates adaptive thermogenesis by noting that mainte...
The paper’s conclusion of 3,500 kcal per pound for weight lost or g...
Wishnofsky warns that low-protein diets create "spurious" rapid los...
542
Caloric Equivalents of Gained or Lost Weight
MAX WISHNOFSKY, M.D.
W HAT 15 the caloric equivalent of one pound
of body weight, gained or lost ? To put
the question in other words : How many cab-
nies in excess of the amount necessary to mriain-
tam caloric equilibrium will produce a gain of
one pound of body weight ; conversely, what
caloric deficit will determine a loss of one pound
of body weight? It is strange that a marked
diversity of opinion exists among authorities on
this subject. In this article an attempt will be
made to clarify the matter.
PHYSIOLOGIC PRINCIPLES
It was shown by m that the aver-
age fat content of human adipose tissue taken
from various parts of the bodies of well-noun-
ished subjects is 87 per cent. One pound
(454 g) of human adipose tissue, therefore,
contains 395 g of fat. The caloric value of one
g of animal fat is 9.5 ; consequently, the caloric
equivalent of one pound of human adipose
tissue may be considered to be about 3,750 cal.
Leathes2 pointed out that carbohydrate
and protein cannot be stored dry, but retain
with them three or more parts of water, while
fat can be stored in an almost pure state. Re-
tention of a gram of fat gives the body 9 cal of
reserve energy, while retention of a gram of wa-
ter-soaked glycogen or protein provides only
about one calorie. Failure to recognize this
fundamental point has resulted in the greatest
bedevilment.
What is the metabolismii of carbohydrate, pro-
tein and fat in individuals in negative or posi-
tive caloric balances? The caloric value of the
glycogen stores in the human body, as com-
pared to protein and fat, is insignificant and
undergoes but trivial changes when the indi-
vidual is in clinical negative or positive caloric
balance.
With regard to protein metabolism, the fol-
lowing may be noted. Strang et al.3 studied 13
obese patients under the especially rigid condi-
tions of the metabolic service for an average pe-
nod of 59 days. The average diet provided 58
g of protein, 14 g of carbohydrate, and S g of fat,
with a total value of 360 cal. They state:
“During the first three weeks of dieting, pa-
tients may lose nitrogen to the extent of 40-SO
per cent above the nitrogen intake. The addi-
tion of 10 to 15 g of carbohydrate to the diet
causes an abrupt drop in the nitrogen output
and the attainment of a permanent nitrogen
balance without the alteration of the nitrogen
intake.” Another group of patients studied by
Strang and Evans4 was kept approximately in
nitrogen equilibrium on diets of 600 to 650 cal.
There is a paucity of work on indirect meth-
ods for the study of body composition and the
findings are conflicting. Behnke, Osserman, and
Welham,6 in a study of “lean body mass,”
found : (1) On a restricted diet accompanied
by body weight loss constancy in the weight of
the lean body mass indicates that nitrogen bal-
ance is maintained and that the “lost” tissue is
fat. (2) In individuals who gained weight,
the changes in the lean body mass were slight
and the calculated densities of the tissues gained
were all within the range of body fat. Keys,
Anderson, and Broek,’ in a study of weight
gain from simple overeating made observations
which are at variance with those of Behnke ci
al.’ They found the tissue mass gained to be
13 to 15 per cent extracellular fluid, 61 to 64
per cent fat, 0 to 1 per cent glycogen, and the
remainder protein.
More recently Young ci al.7 studied the ni-
trogen balance in eight obese young men dur-
ing an eight-week reducing period during
which the protein intake was 115 g daily and
the mean weight loss 22.6 pounds. The mini-
mum protein requirement when an individual
is in caloric balance is between 0.5 and 0.7 g/
kg/body weight/day, so that these subjects re-
Downloaded from https://academic.oup.com/ajcn/article-abstract/6/5/542/4729975
by Washington University in St. Louis user
on 09 April 2018
September-October 19581
WISHNOFSKY 543
ceived about twice the minimnum requirement.
During this eight-week reducing period, four of
the subjects were in nitrogen equilibrium, and
four in slight miegative nitrogen balance.
The conclusions that may be drawn are as
follows : (1) Glycogen changes are insignifi-
cant when a patient is in negative or positive
caloric balance. (2) On a customary low-
calorie reducing diet, where the protein intake
is high, the patient is either in nitrogen equilib-
rium or in slightly negative caloric balance.
(3) Under the circumstances, the caloric deficit
in weight reduction is primarily made up by
the catabolism of fat.
In total fasting the metabolism is entirely
different. In contradistinction to the obese
individual on a low-calorie diet who is practi-
cally in protein and carbohydrate (glycogen)
equilibrium, here decided protein and carbohy-
drate deficits occur. The loss of one pound
. (454 g) of protein will yield about 1 ,85O cal.
There will be a concomitant loss ofover 3 pounds
of water. The loss of 4 pounds of body weight
when 1 pound of protein is catabolized, will
result from an expenditure of only 1 ,850 cal;
whereas an individual on a low-calorie diet
(negative calorie balance) but in protein equi-
librium will lose only one-half pound from the
same caloric expenditure. When carbohydrate
(glycogen) is catabolized the same obtains
as with protein. However, since the amount
is comparatively small, for practical purposes
it need not be considered.
CRITICAL ANALYSIS AND EVALUATION OF THE
PERTINENT LITERATURE
Sbonimtm states: “If a person eats more than
35 cal/kg of his ideal weight per day, he will
gain weight at the rate of 1 g of fat plus 1 g of
water for each 9 cal in excess. If he eats less
than the ideal amount, he will lose correspond-
ingly.”
According to this concept 1 g of body weight
gained or lost, has a value of 41/2 cal; 454 g
(1 ib) will therefore have a value of 2,042 cal.
The error of this concept is the idea that a
gram of water is stored with each gram of fat.
As was pointed out by Leathes,2 fat is stored
“dry” with a negligible amount of water. Fur-
ther proof of this will be strikingly demon-
strated when the work of Benedict is considered
below. It may be noted that from the teleo-
logic standpoint it is vital for the very survival
of the organism to have a very concentrated
source of energy to act as a reserve during a
scarcity of food. Thus, adipose tissue, with a
caloric value of 8.3 per g, furnishes in an ideal
manner.
Strang, McCluggage and Evans3 made a
careful study of the influence of low calorie
diets on weight reduction in obese persons.
Thirteen persons were placed on a daily diet of
360 cal for a period of 59 days during which
time the average weight loss was 0.6 pound per
day. They estimated the number of calories
necessary to maintain caloric equilibrium in
these patients to be 2,500 per day, so that the
0.6 pounds of body weight lost had an equiva-
lent of 2,100 cal. One pound would therefore
have a caloric value of 3,500. This is in strik-
ing agreement with the value of 3,700 cal ob-
tamed by the determination of the caloric value
of one pound of human adipose tissue (Bozen-
raad).
The estimated figure of 2,500 cal to maintain
caloric equilibrium decreases slowly with loss
of weight and would result in a slight reduction
in the caloric equivalent of one pound of body
weight lost. This is compensated for by the
fact that during the first two to three weeks, as
a result of a negative nitrogen balance, there
were losses of water. If these losses of water
had not occurred, the caloric equivalent of one
pound of body weight lost would be higher than
3,500. The two factors apparently neutralize
each other.
J oslin,9 states: “The caloric value of 1 kg of
body weight is problematical but most impor-
tant. Benedict’s’#{176} normal subject for the last
27 days of his fast of 31 days lost an average of
0.7 per cent of his body weight daily, the equiv-
alent of 3,258 cal/kg or approximately 1,500
cal/lb/body weight. For comparison Joslin”
had an experiment with a healthy nurse who
in 1916 volunteered to go through the exact
procedure to which diabetics were subjected
that time. Her loss of weight in 20 days rep-
resented the equivalent of 3,170 cal for each
kilogram of her body lost, and this agrees quite
closely with Benedict’s fasting subject. There-
Downloaded from https://academic.oup.com/ajcn/article-abstract/6/5/542/4729975
by Washington University in St. Louis user
on 09 April 2018
544
THE AMERICAN JOURNAL OF CLINICAL NUTRITION
[\‘ol. 6, No. 5
fore, one must assumne that a 1)atient will not
lose, or conversely, gain a ])0U11(l of actual body
tissue unless he receives at least I , 500 cal less or
more food than he requires for his standard
metabolic need, computed for rest and exer-
cise.”
In his conclusions Joslin makes the error of
failing to differentiate between the metabolism
of the individual who is on a low-calorie diet
but who is in protein and carbohydrate (glyco-
gen) equilibrium, and one who is fasting and
has marked protein and glycogen deficits. The
significance of this point is strikingly demon-
strated in Benedict’s observation.
Benedict’s subject fasted for 31 days, par-
taking only of distilled water. The pertinent
findings are best expressed in Benedict’s own
words.’2 “During the 31-day fast this sub-
ject actually excreted 277.32 g of nitrogen in
the urine, thus averaging 8.95 g of nitrogen per
day. This would correspond to 1,664 g (3.7
lb) of protein or 8,319 g (18.3 lb) of flesh. Since
the entire loss of body weight of this subject
was 13.25 kg (30 lb), it can be seen that 63 per
cent of the total loss may be accounted in flesh
catabolized.’ ‘ Study of Table 63, page 412,12
shows that whereas during the 31-day fast, the
amount of water lost concomitantly from the
catabolism of flesh, was enormous (6,383 ml or
14 lb) ; the amount lost concomitantly from
the catabolism of fat was negligible (367 ml or
0.8 lb). The catabolism of 1,664 g of protein
which resulted in a loss of 18.3 pound of body
weight, yielded only 6,828 cal. If the fasting
subject had been in protein and carbohydrate
equilibrium the 6,828 calories would have been
derived from adipose tissue with a loss of only
1.82 pound (0.83 kg). The categoric conclusion
can be drawn that the caloric equivalent of one
pound of body weight lost when an individual
is fasting, is markedly less than when he is on a
low-calorie diet, but in protein and carbohy-
cirate equilibrium.
Clinical observations confirm the error in
Joslin’s thesis. If a person who requires 2,500
cal/day to maintain caloric equilibrium is on a
daily diet containing 1,000 cal, he will give a
daily deficit of 1,500 cal or a weekly deficit of
10,500 cal. According to Joslin, such a per-
son should lose 7 pounds of body weight per
week. From practical experience we know
that this does not occur. In Strang’s3 pa-
tients the loss of 0.6 pound of body weight day
resulted fromn a deficit of 2, 100 cal. If Jos-
lin’s figure of I ,500 cal for the caloric equivalent
of one pound of body weight lost were accepted,
they should have lost 1 .4 pounds instead of 0.6
pound actually lost.
Utilizing indirect methods for the estimation
of body composition Keys et a!.6 in a study of
individuals who had not previously fasted,
found that on a high calorie diet the caloric
equivalent of 1 kg of the tissue mass gained was
6, 180 cal. They also found that the character
of the tissue mass gained in rehabilitation by
previously starved men was quite different
from that gained by well-fed men who simply
ate to excess. Whereas, about two-thirds of
the weight gain of the latter was pure fat, in
the starved men the gain in early rehabilitation
was made up of only 10 to 20 per cent fat.
DISCUSSION
Certain definite conclusions may be made
from the observations and studies described
above. The average obese individual who is
placed on a low-calorie diet for the purpose of
weight reduction, is practically in protein and
carbohydrate (glycogen) equilibrium. The ca-
boric deficit in these patients is made up almost
entirely by the catabolism of fat. The theo-
retical value of one pound (454 g) of adipose tis-
sue is 3,750 ‘ In clinical studies by Strang
et a!.’ on weight reduction on obese individuals
who were on a low-calorie diet a value of 3,500
cab was obtained for each pound of body weight
lost. The conclusion can be drawn that 3,500
is the caloric value of one pound of body
weight lost.
It must be stressed that it is fundamental
that the protein content of the low-calorie diets
employed in weight reduction be kept high.
If they are so low as to permit appreciable
negative nitrogen balances, there will be con-
siderable loss of water. This will result in a
decrease in the caloric equivalent of 1 lb of
body weight lost and consequently a more rapid
rate of loss in weight. This, is, however, a
spurious loss because it has no permanency.
After the desired weight loss has occurred and
Downloaded from https://academic.oup.com/ajcn/article-abstract/6/5/542/4729975
by Washington University in St. Louis user
on 09 April 2018
September -October 19581
WISH NOFSKY
5-15
the patiemit has been placed on a diet adequate
to maintaimi caloric equilibriumii, the protein
stores which have heemi depleted during the
period of negative nitrogen balance will be re-
plenished. For every pound of protein replen-
ished there will be an increase of four pounds in
body weight. Thus, even though the patient
is in caloric equilibrium, he will continue to
gain weight until the protein stores have been
completely restored.
The value of one pound of body weight
gained is also 3,500 cal if there is no deposition
of protein or carbohydrate (glycogen) . After
a prolonged fast, replenishment of the depleted
protein and carbohydrate stores will be asso-
ciated with a concomitant marked deposition
of water which will decidedly reduce the caloric
equivalent of one pound of body weight gained.
When the individual is again in protein and
carbohydrate equilibrium, all foods in excess of
the amount necessary to maintain caloric equi-
librium will be converted into fat and the caloric
equivalent of one pound of body weight gained
will be about 3,500.
In prolonged fasting there is a marked nega-
tive protein balance with a concomitant loss
of large amounts of water. There is also a
negative carbohydrate (glycogen) balance with
a loss of water of the same order. However,
since glycogen stores are small, this factor is not
important. The enormous losses of water,
combined to a small degree with the fact that
the caloric values of protein and carbohydrate
are less than half the value of a similar amount
of fat, resulted in a large reduction of the caloric
value of one pound of body weight lost. Thus
Joslin6 calculated that each pound of body
weight lost by Benedict’s subject, who fasted
for 31 days and lost :30 pounds had a value of
1,500 cal, which is less than half of that deter-
mined for one pound of body weight lost in
clinical weight reduction.
The following two points are relevant to this
subject. A person who weighs 300 pounds and
is putting out 3,000 cal a day in bodily activi-
ties may be placed on a reducing diet. Because
of hunger he may reduce his activities to, per-
haps, 2,000 cal/day. Thus he would lose
weight more slowly than he would on a regi-
men of activity equal to that before the diet.
As weight loss occtmrs, caloric expenditure
decreases. For example, a male, age 40 years,
:300 pounds in weight, 70 iii. in height, has a
daily basal caloric expenditure of 2,265 cal.
If the weight were reduced to 200 pounds, the
basal caloric expenditure would drop to I ,995
cal, a difference of 270 cal/day or 8, 100 cal/-
month. Thus, on the same regimen of activ-
ity and diet, he would lose two and one-third
pounds more per month when the body weight
is 3OO pounds than when it has been reduced to
200 pounds.
It is of importance to discuss the weight
curve of patients undergoing weight reduction
on low-calorie diets. Many observers have
emphasized the fact that although a patient
may burn many grams of stored fat per day he
does not necessarily lose weight every day.
The curve of the daily weight change is a series
of ups and downs with frequent periods of no
apparent weight change. As has been repeat-
edly demonstrated, these irregularities are in-
timately associated with the storage and release
of relatively large quantities of water. The in-
fluence of this phenomenon on the true weight
loss is minimized if the period of observation is
sufficiently long. It is possible to predict with
gratifying precision the weight loss to be ex-
pected from the deficiency in exogenous cab-
ries.
The fact that I g of flesh has a value of ap-
proximately 1 cal and I g of adipose tissue a
value of approximately S cal is of consider-
able practical importance. Thus, in certain
conditions where parenteral alimentation has
to be resorted to for any considerable length of
time, it is important to supply protein deriva-
tives so as to keep the negative nitrogen balance
as low as possible, and prevent large weight
losses from the loss of flesh. This is of even
greater importance in chronic diseases such as
anorexia nervosa, carcinoma of the esopha-
gus, carcinoma of the stomach, and other con-
ditions where alimentation is very difficulL
Here the primary consideration is to furnish a
diet high in protein. If a large negative nitro-
gen balance is permitted, the loss of weight
from the catabolism of flesh will be large
with a yield of only a small number of calo-
ries.
Downloaded from https://academic.oup.com/ajcn/article-abstract/6/5/542/4729975
by Washington University in St. Louis user
on 09 April 2018
546 THE AMERICAN JOURNAL OF CLINICAL NUTRITION
[Vol. 6, No. 5
615 Williams Avenue
Brooklyn 7, New York
SUMMARY AND CONCLUSIONS
The low-calorie diets on which individuals
are placed for the purpose of weight reduction
should be high in protein so that protein and
glycogen will be approximately in equilibrium.
The calorie deficit will be made up chiefly by
the catabolism of fat. Under the circum-
stances (high-protein intake) the caloric equiva-
lent of one pound of body weight lost is approx-
mately 3,500 cal.
The caloric equivalent of one pound of body
weight gained is dependent on the state of the
protein and carbohydrate (glycogen) stores.
If the protein and glycogen stores have been
depleted as the result of fasting, their replenish-
ment will be associated with a concomitant
deposition of large quantities of water. During
this period the caloric equivalent of one pound
of body weight gained will be markedly less
than 3,500.
When a state of protein and glycogen equi-
librium is reached, all food ingested in excess of
the amount necessary to maintain caloric
equilibrium will be converted into fat (adipose
tissue). The caloric equivalent of one pound
of body weight gained will then be 3,500.
In fasting there are always decided negative
protein and carbohydrate (glycogen) balances.
The catabolism of protein and glycogen is asso-
ciated with losses of large quantities of water.
The caloric equivalent of one pound of body
weight lost in fasting (negative nitrogen bal-
ance) is always markedly less than 3,500.
REFERENCES
1. BOZENRAAD, 0. : Ueber den Wassergehalt des
mnenschlichen Fettgewebes unter verschiedenen
Bedingungen, Deutsch. Arch. f. kl-in. Med. 103:
120, 1911.
2. LEATHES, J. B. and RAPER, H. S. , quoted by
PETERS J. and VAN SLYKE, D. D. : Quantitative
Clinical Chemistry: Vol. I: Interpretations.
Williams and Wilkins, Baltimore, 1931, p. 222.
3. STRANG, J. M., MCCLUGGAGE, H. B. , and EVANS,
F. A. : Further studies in the dietary correction
of obesity. Am. J. Al. Sc. 179: 687, 1930.
4. STRAN’G, J. M. and EVANS, F. A. : The energy
exchange in obesity. J. Oin. Investigation 6:
277, 1928.
5. KEYES, A. , ANDERSON, J. T. , and BRO.EK, J.:
Weight gain from simple overeating. Metabolism
4:427, 1955.
6. BEHNKE, A. R., OSSERMAN, E. F., and WELHAM,
W. C. : Lean body mass. A rch. In!. Med. 91:
585, 1953.
7. YOUNG, C. M., EMPEY, E. L. , SERRAON, V. U.,
and PIERCE, Z. H. : Weight reduction in obese
young men: Metabolic studies. J. Nutrition
61:437, 1957.
8. SLONIM, R. J., JR.: A simple aid in calculation of
diets. J.A.M.A. 162: 1233, 1956.
9. JOSLIN, E. P., RooT, H. F., WHITE, P., and
MARBLE, A : Treatment of Diabetes Meilitus.
ed. 9. Lea and Febiger, Philadelphia, 1952, p.
236.
10. BENEDICT, F. G.: A Study of Prolonged Fasting.
Carnegie Institute of Washington, pub. #203,
1915.
11. JosLIN, E. P.: Diabetic Metabolism with High and
Low Diets. Carnegie Institute of Washimigtomi,
pub. #323, 1923.
12. BENEDICT, F. G.: A Study of Prolonged Fasting.
Carnegie Institute of \Vasliingtomi, pub. #203.
1915, p. 251.
Downloaded from https://academic.oup.com/ajcn/article-abstract/6/5/542/4729975
by Washington University in St. Louis user
on 09 April 2018

Discussion

Tap the paper where you want to comment
## TL;DR Max Wishnofsky’s 1958 paper was the first to translate body-mass change into a concrete number: about 3,500 kcal for every pound (454 g) gained or lost. He showed that when protein and glycogen are kept in balance - e.g., on a high-protein, very-low-calorie diet - almost all the energy deficit is drawn from adipose tissue, whose energy density (~3 750 kcal·lb⁻¹) minus small water/protein corrections yields the tidy 3,500 kcal figure. Yet Wishnofsky also mentioned that real-world weight loss is "a series of ups and downs" sensitive to observation length, fasting state, and hydration shifts. Because his conclusion was based on short-term data from a small sample and a then-limited grasp of metabolic adaptation, the simple 3,500 kcal rule often over-promises, leaving dieters puzzled when progress lags. In conclusion, Wishnofsky gave us a useful first-order estimate, but one that must be tempered by the body's complex, dynamic biology. Max Wishnofsky (December 17, 1899 – August 2, 1965) was a Russian-born American physician based in Brooklyn, New York, where he earned an M.D. and practiced medicine. Wishnofsky’s rule (1 pound of body weight corresponds to approximately 3,500 kilocalories) became deeply embedded in diet and health guidelines for decades, even as later research revealed its limitations by overlooking dynamic metabolic and tissue-composition changes during weight loss. Wishnofsky multiplies the classic 9.5 kcal · g⁻¹ figure for animal fat by the 395 g of lipid in a 454 g adipose-tissue “pound” to obtain 3 750 kcal · lb⁻¹. Modern bomb-calorimetry rounds fat to ≈ 9.4 kcal · g⁻¹ and adipose fat fraction to ≈ 85 %, so the theoretical value is closer to 3 500 kcal · lb⁻¹, remarkably close to the rule of thumb that became popular decades later. To determine the % of fat in adipose tissue, researchers use chemical analysis of tissue samples. Here's how it’s typically done: 1) Sample collection: A biopsy is taken from human adipose tissue. 2) Drying: The tissue is dried to remove water (adipose tissue is ~15% water). 3) Fat extraction: Solvents like ether, chloroform-methanol, or hexane are used to dissolve and remove lipids from the dried tissue. 4)Weighing: The extracted fat is weighed and compared to the original tissue mass. To determine the caloric value of 1 gram of animal fat, scientists use a method called bomb calorimetry, which measures the total energy released when the fat is completely combusted in a high-pressure, oxygen-rich chamber. A known mass of purified animal fat is placed in the calorimeter and ignited, and the heat released is absorbed by a surrounding water bath. By measuring the increase in water temperature and knowing the specific heat capacity of water, the total energy released (expressed in kilocalories) can be calculated. This "gross energy" value typically ranges around 9.4 to 9.6 kcal per gram for animal fat, depending on its specific fatty acid composition. The commonly cited value of 9.5 kcal/g is an average derived from such experiments and reflects the chemical energy content of the fat, not accounting for digestion or metabolic losses. The rule collapses during total fasting: catabolising protein drags out three times its weight in water, so a pound of scale weight may cost barely 1 500 kcal  This is an early quantification of the “fasting diuresis” well known to clinicians today. Fasting diuresis refers to the increase in urine output that occurs during periods of fasting, particularly in the early stages. As the body shifts from using glucose to burning fat for energy, glycogen stores in the liver and muscles are depleted. Since glycogen binds water (approximately 3–4 grams of water per gram of glycogen), its breakdown releases significant amounts of water, leading to increased urination. Additionally, insulin levels drop during fasting, which reduces sodium reabsorption in the kidneys and promotes natriuresis (loss of sodium in urine), further enhancing diuresis. This process can result in rapid initial weight loss due to water loss, not fat loss, and may also cause temporary electrolyte imbalances if fluids and salts are not adequately replenished. Wishnofsky anticipates adaptive thermogenesis by noting that maintenance needs “decrease slowly with loss of weight,” an idea firmly validated 60 years later (e.g., “Biggest Loser” follow-ups). The paper’s conclusion of 3,500 kcal per pound for weight lost or gained when protein & glycogen are in equilibrium, became a cornerstone of dietary counselling, though modern dynamic models show the value rises as weight decreases and overestimates long-term loss. Wishnofsky warns that low-protein diets create "spurious" rapid losses via water, emphasising high protein to stay near nitrogen balance, precisely the rationale behind today’s High-protein Very Low-Calorie Diets (VLCDs).