Download Left Ventricular Wall Thickness Determined by

Left Ventricular Wall Thickness Determined
by Ultrasound in
100 Subjects without Heart
Disease*
A-L. Sjogren, M.D.
Left ventricula wall thickness (LVWT) was measured by ultrasound in 100
subjects without heart disease. The data on both diastolic (LVWT,) and systolic
(LVWT,) thickness were related to sex, age, weight, height, heart volume (HV),
and ECG voltage criteria. LVWT, was greater in men (P<0.01) but LVWT,
was of the same magnitude in both sexes. In both sexes LVWT, and LVWT,
increased with age (P<0.001). In women LVWT, correlated with HV but not
in men (R=0.429; P<0.001 for women, R=0.145; P>0.05 for men). This was
caused by an increasing HV with age in women, in contrad to the steady HV
in men. There was no carelation between LVWT, and weight in either sex.
LVWT, and height was not related io men but in women there was an inverse
correlation between LVWT, and height (R= -0.102:P>0.05 for men, R=
-0.363:P<0.01 for women). No significant correlation was found between the
LVWT and ECG voltage criteria. Irrespective of sex and age the upper normal
limit for LVWT, was 12 mm and for LVWT, 18 mm.
he estimation of the absence or presence and
Tdegree of left ventricular hypertrophy (LVH) is
an integral part of cardiologic routine. This estimate has both diagnostic and prognostic significance, and it is one of the commonly used parameters in assessing the results of both therapeutic and
surgical procedures. Electrocardiography ( ECG ) ,
vectorcardiography and recently multiple dipole
electrocardiography have been used in estimating
LVH. The criteria employed are to a large extent
empirical and are based on several assumptions.
Accordingly, both false positive and negative data
are obtained. 1-7
Angiocardiography is more accurate in determining the left ventricular wall thickness (LVWT) but
it, too, is subject to many errors.8-13 The ultrasound
examination of the heart and its movements was
introduced by Edlerl4 in 1955 and in 1966 Feigenbaum and associates15 and Soulen and associates16
reported data on LVWT. Since that time several
reports have described left ventricular wall measurements by ultrasound. l 7-22 Despite an increasing amount of data on LVWT, the normal range in
relation to age, sex, body composition and heart size
has not been established. This present report de'From the Cardiovascular Laboratory, First Department of
Medicine, University Central Hospital, Helsinki, Finland.
scribes normal variations in ultrasonically determined LVWT in subjects without clinically detectable heart disease.
One hundred subjects with no clinically recognizable heart
disease, as judged by clinical findings, ECG and chest x-ray
films, were studied ultrasonically to record LVWT.
The material consisted of 58 women, aged 18 to 61 years,
mean 35 years, and 42 men aged 17 to 69 years, mean 39
years. The subjects consisted of hospital house staff, medical
students and patients admitted for minor surgical procedures,
eg, repair of inguinal hernia. They comprised both manual
and sedentary workers with an even distribution in both
sexes. The conventional 12-lead ECG was taken as the
hospital routine. A direct-writing inkjet ECG machine was
used ( Elema-Schonander ) The calibration was carefully
checked and correction was done, if necessary, in calculating
the amplitudes. In assessing LVH the QRS criteria of Manning and Smiley ( 1 9 6 4 ) s were used as follows: R in lead I
13 mm, max R in Vn and Ve 3 3 mm, R in aVL 7 mm, R in
aVF 23 mm, R in lead I
S in lead I11 16 mrn and max R in
Va or Va
S in V1 53 mm in subjects under 30 years and 13
mm, 24 mm, 9 mm, 15 mm, 17 mm and 37 mm in subjects
over 30 years respectively. The heart volume ( H V ) in erect
wsition without reference to the cardiac cycle was determined from biplane chest x-ray films by the ellipsoid approximation method24 with certain modifications." In excluding
pathologic hearts, the HV was related to body surface
area;"'vohmes over 500 ml/M2 in men and over 450 ml/M?
BSA for women were considered pathologic.27 Ultrasound
.
+
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017
+
A-L. SJOGREN
342
Table 1-The
Meamred Variables in Three Different Age Groups.
17-29
Age, y n
30-49
Women
N =28
-
x
Height
Weight
Heart vol
LV-Syst thickness
LV-Diast thickness
LV. O/o Diast syst
cm
kg
ml
mm
mm
Men
N=12
i S.D.
K
T
S.1).
163.95 4.90 177.67 6.87
56.67 7.67 73.08 7.88
570.00 114.50 757.27 120.09
11.96 1.26 12.33 1.61
6.11 0.99 6.92 1.08
51.01 5.72 56.50 8.09
cardiac examinations were made using an ultrasonoscope*
utilizing 500 impulses/sec repetition, wcth a 13 mm diameter,
2.00 megahertz transducer. The transducer was placed at the
left sternal border in the fourth or fifth intercostal space to
obtain the posterior left venhicular
echoes as described
earlier.28 Measurements were performed by a single observer by repeated measurements. If variation of measurement
was found, an average figure was used for each record. The
maximal and minimal wall thicknesses as compared with the
carotid pulse were taken to represent systole and diastole
resDectivelv.
- . and the measuremen6 were made with an
accuracy of 1 mm. The ultrasonic data were compared with
different parameters as described later. Conventional statistical methods and measures were used. The statistical significances were determined by Student's t-test. The tables of
Ceigy (1962) were used.
50-69
Women
N=18
X
T
Men
N=17
T
S.D.
Women
N = 13
T
S.D.
162.55 6.65 176.79 7.38
61.94 6.87 78.88 9.16
651.47 88.67 727.81 127.27
14.11 1.64 13.94 1.91
7.61 1.91 8.12 0.93
53.89 10.40 59.00 8.35
Men
N = 13
X
S.D.
S.D.
f
158.62 6.13 175.35 7.98
64.23 10.35 79.18 9.76
642.23 95.95 805.00 95.95
15.46 1.56 15.08 1.05
8.92 1.26 9.08 1.38
57.85 6.56 60 08 6.87
l Fig 1). The same was true for LVWT.. There was
between L-~
and weight in
no
either men or women ( P > 0.10 and > 0.05
respectively). The same was true for the correlation
to height in men ( P > 0.05), but LVWTd was
inversely correlated to height in women ( P < 0.01 )
(Fig 2). A plot of the sum of maximal R wave in
wave in V1 against LVWT, reveals
V5 Or V6 and
no ~ r r e l a t i o nin either men or women ( P > 0.05,
Fig 3 ) .
In men there was no sienifimnt correlation between HV and LW,
or LW,
( P > 0.05 for
both) but in women HV correlated significantly
\
"
,
-
RESULTS
Table 1 gives an account of the total series (of
subjects) including height, weight, HV, left ventricular systolic wall thickness ( LVWT. ), left ventricular diastolic wall thickness (LVWT,), and the
percentage of LVWTd/LVWT,.
In men the LVWT, varied from 5 to 12 mm and
in women from 4 to 12 mm (Table 2 ) . The LVWT,
varied from 10 to 17 mm in men and from 10 to 18
m m in women. Considering the total series the
LVWTd was greater in men ( P < 0.01) but there
was no difference in LVWT, ( P > 0.05). Both in
men and women there was a positive correlation
between LVWTd and age ( P < 0.001 for both)
'Physionic Engineering Inc., Longmont, Colorado.
MALES
0 FEMALES
s .t..
0
, o
20
,
,
,
30
,
40
,
,
50
,
,
,
,
]
70
60
AGE. YEARS
FIGURE1. Relation between age and left ventricular wall
diastolic thickness.
Table L L e f t Ventricular Wall Thickness in Diflerent Age Groups.
Systolic Wall Thickness, mm
Diastolic Wall Thickness, mm
I
Age, Sex
<
\
Range
Mean
\
Range
Mean
17-29 years
Women
Men
30-49 years
Women
Men
50-69 years
Women
Men
CHEST, VOL. 60, NO. 4, OCTOBER 1971
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017
343
ULTRASOUND DETERMINATION OF LEFT VENTRICULAR WALL THICKNESS
FEMALES
0
0
0
4w
3%
0
5w
450
0
sso
0
aw
ex,
m
TOTAL MEART
t
Y
41
L I
0 ;45
1
1
160
I
I
I
165
170
175
Height
.
FIGURE
2. Relation between body height and diastolic thickness of left ventricular wall.
with both LVWT, ( P < 0.01) and LVWTd ( P <
0.W1) ( F i g 4 and 5 ) .
LVWTd/LVWT, in percentage was greater in all
groups in men ( P < 0.005), but it correlated
significantly with age in women only ( P < 0.001,
Fig 6 ) .
In ultrasonic studies the diastolic thickness is
easier and more accurate to measure than the
systolic thickness due to disturbing echoes from
papillary muscles, trabeculations and chordae tendinae. Accordingly the diastolic data are more
reliable and worth major emphasis in the analysis.
Moreover, in the correlation with HV, when x-ray
6lms are exposed without reference to the cardiac
$
,
!
i
,
o
o
;;
08,808
C
z
9
e&o
8.
o 0.
11
y. - O . ( A ~ X .UOZ
rm-0-
1
P MS.
C
LL
2 10
0 10
dbo
20
30
ECG
7so
W
E
em
,
rnl
FIGURE
4. Correlation between heart volume and left ventricular wall systolic thickness in women.
0
I
I
iCK)
em
y.0007sx.
r- 03W
PCOOl
0 0 0
40
50
RZ;'~.
sV' , m m
VOLTAGE.
FIGURE
3. Graph showing no correlation between ECG criteria and systolic thickness of left ventricular wall.
cycle, it is more likely that the exposures are in
diastole than in the shorter systole. Measurements
of LVWT at autopsy vary according to the place
and obliquity of the cut, the inclusion of trabeculae
or papillary muscles and the state of myocardial
tone. The data on normal values from different
autopsy series are compiled into Table 3.**37 The
information from angiocardiographic studies in normal hearts is collected into Table 4. Table 3 reveals
the scarcity in autopsy materials of LVWT evaluations taking age and sex into account. Various
textbooks and authors give normal values for
LVWT ranging from 10 to 15 mm, however without
regard to age or sex. Angiocardiographic observations (Table 4)"s-44are limited to small series and
give normal values for LVWTd ranging from 5 to
15 mm. Kennedy and associates (1966)48 have
related LVWTd to age and sex in their series of 22
subjects.
Even if the above listed data are obtained by
direct approaches they are subject to errors owing
to methodologic daculties. Analogous problems
existing in ultrasonocardiography have recently
-1
d
7 - 0
0
00000
@
00
a
0
$6-
EE
5-
E
41
2
8
0
0
0
0 0 0 0
o
m
o o o c e w
y-QOOBx.305
0
0
r-04a0
PcOOOI
0
0
,,
OL3M
4w
4%
500
sso
ow
650
TOTU
700
HW
7%
em
VOCWE, rnl
FIGURE5. Correlation between heart volume and left ventricular wall diastolic thickness in women.
CHEST, VOL. 60, NO. 4, OCTOBER 1971
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017
344
A-L. SJOGREN
0
20
30
40
I
M
80
70
AGE. YEARS
FIGURE
6. Contractile changes in left venMcular wall thickness in relation to age in women.
been reviewed by Ilmurzynska and associates.45
One of the drawbacks is that LVWT is measured
only in a relatively localized part of the posterior
wall of the ventricle. This has bearing upon the
appraisal of LVH, since the thickening of the wall
may not be uniform, although it presumably is so in
the majority of cases. The limited part of the left
ventricular wall delineated by ultrasound as compared to different regions measured by angiocardiography renders the comparison of the data
obtained by these two methods less satisfactory but
reproducibility of the ultrasonic data is excellent.28
With due emphasis on the above limitations the
present data show that L V W d and LVWT, increase in both sexes with age. This has not been
found in autopsy studies.:{O The increasing thickness cannot be due to changes in body weight, since
there is no correlation to weight. The same applies
to height, even if there is an unexplained negative
correlation in the women, since the only significant
difference in the height of women (Table 1 ) was
seen between the age groups of 17 to 29 and 50 to
69 years ( P < 0.02), and no change in height in
men within the years studied. In women both
LVWT, and LVWTd increased with an increasing
heart volume (Fig 4 and 5 ) ; in men there was no
correlation to HV. This apparent discrepancy is
explained by the fact that in women the HV
increased significantly with age: P < 0.02 between
the first and second age groups, and P < 0.05
between the first and third age groups (Table l ) ,
whereas HV did not increase in men with age
( Table 1: P > 0.05 in all age group comparisons ).
Thus age seems to be the only factor relevant to
both sexes that affects LVWT independently of
other measured variables. No comparison was made
between blood pressure readings and UCG measurements as the subjects were normotensive with
blood pressure values not exceeding 160/100 mm
Hg. The lack of any significant correlation between
LVWT and the corresponding ECG criteria is
worth emphasizing. Although the usefulness of the
ECG in the diagnosis of LVH is established, it is
generally agreed that false positives and false negatives tend to be common. The ECG data on LVH
have been correlated with postmortem findings but
there are many difficulties in assessing the upper
limit of a normal left ventricular wall thickness at
autopsy. Therefore it is not surprising that there
was no significant correlation between LVWT, and
the corresponding ECG criteria which presumably
are too insensitive to reveal significant differences in
normal hearts.
The expression of LVWTn in percentage of
LVWT, is a measure of dynamic changes in the
wall during a cardiac cycle. The considerable variability found in earlier angiocardiographic studies is
discussed by Hugenholtz and co-workers.47 In the
present work this ratio did not change with age, but
it was significantly higher in men in all age groups
(Table 1: P < 0.005) due to a significantly greater
LVWTd in men and an equal LVWT, in both
sexes. Theoretically it seems, therefore, that the
Table S A u t o p s y Findings of Z e f t Ventricukrr Wall Thickness.
No. Cases
W
M
Age, yrs
Total
LVWT,
W
M
mm
Total
Latimer"
Reiner et al"
I'ppcr l i ~ ~ i i of
t s normal LVW'r, ticrording to
Peacocks0
1,inzhachsI
Boyd32
Hudson33
Schlants'
Liu and De ~nstofaross
Svott et ~ 1 %
.Allenstein and hforisl
CHEST, VOL. 60, NO. 4, OCTOBER 1971
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017
345
ULTRASOUND DETERMINATION OF LEFT VENTRICULAR WALL THICKNESS
Table LAngioeardiographic Finding8 o f Left Ventricular Wall Thickneu.
Author
W
No. Cases
M Total
Age, yrs
W
M
Range*
W
LVWTd, mm
M
Range
L V b T., mm
9,8-15,O
Kennedy et al4s
Castellanos and Hernandez"
Hermann et also
Gault et al"
Eber et a112
Levine et a18
Hattori41
Falsetti et a142
Hood et a110
Grant et al4s
Soloff and Lawrence44
5,212,o
7-13
6,9-!I. 1
8,2-10,O
5,043,s
7,.3-10.8
9,3-13,O
14,3-20,O
7 and 9
8,47
7,7
I)
W =women; M =men; *=total series without regard t,o sex.
dynamic change in LVWT is less in men than in
women, ie, the contractile pattern would be different. This line of reasoning is not supported by the
angiographic data, which show no difference between men and women in end-diastolic and endsystolic chamber volumes." Additional data on
changes in contractility and chamber geometry
would presumably clarify the difference between
the sexes.
As the normal upper limit of LVWT,, irrespective
of age and sex we suggest 12 mm (Table 2). The
respective LVWT. is 18 mm (Table 2). The age
related LVWT in mm should be derived with the
aid of the following formulae: where x denotes age
in years and SE the standard error of the estimate.
Men :
LVWT. = 0.072~ 11.046, SE = 1.59
LVWT, = 0.055~ 5.95 , SE = 1.16
Women :
LVWT, = 0.111~ 9.48 , SE = 1.40
LVWTd = 0.096~ 3.83 , SE = 1.24
+
+
+
+
1 Massie E, Walsh TJ: Clinical Vectorcardiography and
Electrocardiography. The Year Book Publishers, Inc, Chicago, 1960
2 Johnston FD: What is a normal electrocardiogram ( e d ) .
Circulation 24 :707, 1961
3 Bram I: The significance of the electrocardiographic
pattern for assessment of the degree and type of left
ventricular hypertrophy. Cardiologia ( Basel) 46:3, 1965
4 Dower GE, Horn HE, Ziegler WG: On electrocardiographic-autopsy correlations in left ventricular hypertrophy. A simple postmortem index of hypertrophy proposed. Amer Heart J 74:351, 1967
5 Holt JH, Barnard ACI, Lynn MS: A study of the human
heart as a multiple dipole electrical source. 11. Diagnosis
and quantitation of left ventricular hypertrophy. Circulation 40:697, 1969
6 Romhilt DW, Bove KE, No& RJ, et al: Electrocardiographic criteria for the diagnosis of left ventricular hypertrophy. Circulation 40: 185, 1969
7 Skjaeggestad 0, Kierulf P: The electrocardiographic diagnosis of left ventricular hypertrophy. Acta Med Scand
185:539, 1969
8 Levine ND, Rockoff SD, Braunwald E: An angiocardiographic analysis of the thickness of the left ventricular
wall and cavity in aortic stenosis and other valvular
lesions. Circulation 28:339, 1963
9 Dodge HT, Sandler H, Baxley WA, et al: Usefulness and
limitations of radiographic methods for determining left
ventricular volume. Amer J Cardiol 18: 10, 1966
10 Hood WP, Racklev CE. Rolett EL: Wall stress in the
normal and hype&ophied human left ventricle. Amer J
Cardiol 22 :550, 1968
11 Pech HJ, Porstmann W: Nahen~ngsweiseintravitale Bestimmung der Muskelrnasse des linken Ventrikels. RoFo
108:425, 1968
12 Eber Lhl, Greenberg HM, Cooke JM, et al: Dynamic
changes in left ventricular free wall thickness in the
human heart. Circulation 39:455, 1969
13 Gootman N, Rudolph AM, Buckley NM: Effects of angiocardiographic contrast media on cardia function. Amer J
Cardiol 25:39, 1970
14 Edler L: Diagnostic use of ultrasound in heart disease.
Acta Med Scand 32: ( Suppl ) 308, 1955
15 Feigenbaum H, Zaky A, Waldhausen JA: Use of ultrasound in the diagnosis of pericardial effusion. Ann Intern
Med 65:443, 1966
16 Soulen RL, Lapayowker MS. Gimenez JL: Echocardiography in the diagnosis of pericardial effusion. Radiology
86:1047, 1966
17 Feigenbaum H, Popp LR, Chip JN, et al: Left ventricular
wall thickness measured by ultrasound. Arch Intern Med
121:391, 1968
18 Klein JL, Shimada GRH, Kingsley B, et al: Evaluation of
induced pericardial effusion by reflected ultrasound.
Amer J Cardiol 22:49, 1968
19 Popp RL, Wolfe SB, Hirata T, et al: Estimation of right
and left ventricular size by ultrasound. Amer J Cardiol
24:523, 1969
20 Schroeder JS, Popp RL, Stinson EB, et al: Acute rejection
following cardiac transplantation. Phonocardiographic
and ultrasound observations. Circulation 40: 155, 1969
21 Wirth J, Wenzelides K: Herzmuskeldickenmess~~ngmit
Ultraschall. Cor e Vasa 12: 112, 1970
22 Askanas A, Stopczyk hl: Ultrasounds for the measurement of the thickness of the heart wall. Cor e Vasa
10:223, 1968
CHEST, VOL. 60, NO. 4, OCTOBER 1971
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017
A-L. SJOGREN
23 Manning GW, Smiley JR: QRS-voltage criteria for left
ventricular hypertrophy in a normal male population.
Circulation 29: 224, 1964
24 Jonsell S: A method for determination of the heart size by
teleroentgenography ( a heart volume index ) . Acta Radio1
(Stockholm ) 20:325, 1939
25 Heikkilii J: Mitral incompetence as a complication of
acute myocardial infarction. Acta Med Scand Suppl 1967,
475
26 DuBois D, DuBois EF: A formula to estimate the approximate surface area if height and weight be known. Arch
Intern Med 17:863, 1916
27 Weens HS, Gay BB: Radiologic examination of the heart.
In The Heart, edited by Hurst JW, Logue RB, New York,
McCraw-Hill, 1966
28 Sjogren, A-L, Hytonen I, Frick MH: Ultra-onic measurements of left ventricular wall thickness. Chest 57:37,
1WO
29 Latimer, HB: The weight and thickness of the ventricular
walls in the human heart. Anat Rec 117:713, 1953
30 Peacock (cited by Busch): Neue Ergebnisse der Messung und Wagung der Herzkammern bei den venchiedenen Hypertrophieformen mit besonderen Berucksichtigung der Histologie. Arch Kreisl Fonch 22:267, 1955
31 Linzbach AJ: Die quantitative Anatomie des Herzens, in
Hand buch der Inneren Medizin edited by Mohr L,
Staehelin R, von Bergmann G, et al. Berlin, Springer,
1960, vol9, ed 4
32 Boyd W: A Textbook of Pathology. Philadelphia, Lea and
Febiger, 1961
33 Hudson REB: Cardiovascular Pathology. London, Arnold,
1965, vol 1
34 Schlant RC: Normal anatomy and function of the cardiovascular system. In The Heart edited by Hurst JW, Logue
RB, New York, McCraw-Hill, 1966
35 Liu CK, DeCristofaro D: Sensitivity and specificity of
electrocardiographic evaluation of LVH in 364 unselected
autopsy cases. Amer Heart J 76:596, 1968
36 Scott RC, Seiwert VJ, Simon DL, et al: Left ventricular
hypertrophy. A study of the accuracy of current electrocardiographic criteria when compared with autopsy findings in 100 cases. Circulation 11:89, 1955
37 Allenstein BJ, Mori H: Evaluation of electrocardiographic
diagnosis of ventricular hypertrophy based on autopsy
comparison. Circulation 21 :401, 1960
38 Castellanos A, Hernandez FA: Angiocardiographic determinations of the thickness of the free wall of the left ventrical in congenital aortic and subaortic lesions. Cardiologia (Basel) 49: 1, 1966
39 Hermann HJ, Singh R, Dammann JF: Evaluation of
myocardial contractility in man. Amer Heart J 77:755,
1969
40 Gault JH, Ross J, Braunwald E: Contractile state of the
left ventricle in man. Instantaneous tension-velocitylength relations in patients with and without disease of
the left ventricular myocardium. Circulation 22:451, 1968
41 Hattori M : Correlative study between electrocardiographic findings and ventricular wall thickness based on cardiosynchronous angiocardiography. Jap Circ J 30:655, 1966
42 Falsetti HL. Mates RE. Grant C. et al: Left ventricular
wall stress 'calculated from one-plane cineangiography.
Circulation Res 26:71, 1970
43 Grant C, Creene DG, Bunnel IL: Left ventricular enlargement and hypertrophy. A clinical and angiocardiographic study. Amer J Med 39:895, 1965
44 Soloff LA, Lawrence JW: The electrocardiographic findings in left ventricular hypertrophy and dilatation. Circulation 26:553, 1962
45 Ilmurzynska K, Segal B: Measuring the thickness of the
posterior wall of the left ventricle using ultrasounds. Pol
Arch Med Wewnet 41:351, 1968
46 Reiner L, Mazzoleni A, Rodriquez FL, et al: The weight
of human heart. I. Normal cases. Arch Path 68:68, 1959
47 Hugenholtz PC, Kaplan E, Hull E: Determination of left
ventricular wall thickness by angiocardiography. Amer
Heart J 78:513, 1969
48 Kennedy JW, Baxley WA, Figley MM, et al: Quantitative
angiocardiography. I. The normal left ventricle in man.
Circulation 34:272, 1966
Reprint requests: Dr. Sjogren, Helsinki University Central
Hospital, Haartmaninkatu 4, Helsinki 29, Finland
Food Resources of the Oceans
Food in the form of fish and plants was first of the
riches man obtained from the seas. Over the course of
countless years many of the world's peoples have made
seafood a major element in their diet. Many seafoods are
rich iri vitally important protein. T h e Japanese, for example, consume a high proportion of seafood, as d o
many of the people living in the Mediterranean region.
Man's food needs account for a world fish catch of about
110 billion pounds a year. This is a n enormous amount
measured by any standard. Yet marine biologists say it is
really only a fraction of what the ocean is capable of
supplying. Out of some 25,000 species of fish, only a
handful of different species are actually caught and used
for food.
Ross F Jr: Undersea Vehicles and Habitats
-The peaceful Uses of the Ocean. New York,
T Y Crowell, 1970
CHEST, VOL. 60, NO. 4, OCTOBER 1971
Downloaded From: http://publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21522/ on 06/16/2017