The SenileLung*

Comparisonwith Normaland EmphysematousLungs

@

1 . Structural

Aspects

Erik K. Verbeken, M.D.; Michel Cauberghs, B.S.; Ingrid Merten@ M.D.;

Jacques Clement, M.D.; Joseph M. Lauweryn@s@,
M.D., Ph.D.; and
Karel P Van de Woestijne,

As part

of a study

M. D. , Ph.D.

of the structural-functional

correlations

ofexcised human lungs obtained at autopsy, the parenchyma

and peripheral airways were examined by means of mor.
phometric techniques. Among the 30 lungs characterized
by the absence offibrosis, ten differed from the normal and
emphysematous lungs by a homogeneous dilatation of the
airspaces, in excess of the dimensions predicted on the
basis of age. Study of the standard deviations of the mean
linear

intercepts

showed

that

the

airspace

dilatation

was

more regular than in emphysematous lungs; in addition,

there was no clear-cut destruction, as estimated from the
number

of alveolar

attachments.

These lungs were char

acterized in addition by an increased thickening of alveolar

recent report of the National Heart, Lung, and

Blood Institute' defines emphysema as a condi
tion ofthe lung characterized by abnormal, permanent
enlargement
of airspaces, accompanied by the de
struction of their walls and without obvious fibrosis.
Destruction may be recognized by the disorder of the
respiratory airspaces. Single
airspace enlargement,
on the other hand, is described as an increase in
airspace size, distal to the terminal bronchioles, as
compared with the airspaces of normal lungs. There is
no evidence of destruction.
The process may be
acquired, as in the uniform respiratory airspace en
largement of the aging lung.
In a study of the functional-morphologic
correlates
of emphysema,
we examined a number of lungs
obtained at autopsy. Some ofthese lungs were normal;
others were clearly emphysematous. A third group of
lungs, however, demonstrated an enlargement of air
spaces, markedly larger than that expected on the
basis of aging alone. Originally, our hypothesis was
that these lungs demonstrated panacinar emphysema.
However, the absence of any disorganization by low
power magnification and of inflammation at the level
of the airspaces and of the peripheral airways at the
microscopic level prompted
us to reconsider our
*Fmm the Laboratorium voor Pneumologie en Pathologisehe Ont
leedkunde I Universitaire
Leuven, Belgium.

Ziekenhuizen

St. Rafael, Gasthuisberg,

Supported by a grant of the Fonds voor Geneeskundig Weten

schappelijkOnderzoek.
Manuscript

received

February

14; revision

accepted

July 16.

Reprint requests: Dr. Varbeken,Laboratorium voor HLstopathologie,

Ka@njnenvoer 33, B-3000 L,@uven,Belgium

septa, without inflammation or fibrosis, normal size of the

diameter, and reduced density of the membranous bron
chioles. Since these lungs were from people older than 60
years, it is assumed that they represent cases of exaggerated
airspace

enlargement

of the

aging

lung,

differing

AA = alveolar attachments;
CLE centrilobular
emphysema;
CV= coefficient of variation; J mean diameter; Lm mean
linear

intercept;

Lma

Lm air; Lmw

tion length per airspace;

intercepts; SDN,standard

Lm wall mean

n/sq cmdensity;
deviation of NI

N1

transec

number

of

opinion. We wondered if we were not dealing with

instances of exaggerated respiratory airspace enlarge
ment rather than with emphysema. To substantiate
this impression, we performed a systematic morpho
metric study ofairspaces and membranous bronchioles
in an attempt to distinguish on an objective basis these
lungs from normal and truly emphysematous lungs.
MATERIAL

AND METHODS

Studies were performed on 17 right and 12 left human lungs and

one isolated left lower lobe (V50), obtained at autopsy. Lungs were
selected on the basis of absence of marked pleural adhesions and
leaks, gross parenchymal consolidations, and/or generalized fibrosis.
Most lungs were from patients who died of cardiovascular disease
or of malignant neoplasms. The examined lungs represent a sample
that can be considered unselected at least from the population of
patients for whom an autopsy was being asked @nour general
hospital. An exception was made for three lungs (V50, V51, V52)
that were selected because of clinical severe chronic obstructive
lung disease and/or the presumption of advanced emphysema.
Following the functional measurements (see companion article),

the lungs were perfused intrabronchially for at least 72 h with a 10

percent formalin solution, at a pressure of 2.5 to 3.0 kPa. Whole
lung sections were prepared from 26 lungs following the technique
of Gough and Wentworth.2 The lung was cut into 1.5-cm-thick
sagittal slices and a 0.4-mm-thick whole lung section was prepared
from the midsagittal slice. The remaining lung tissue of that section
was also examined under the dissection microscope. The degree of
airspace enlargement was assessed on the whole lung section by
means of a plastic Ryder grid.@ This grid divides the lung in ten
equal radiating segments. To each segment an emphysemascore
from 0 (no enlargement) to 3 (severe enlargement)
is given, the
maximal score thus being 30. Finally, tis ie blocks were selected
according to the following protocol. Twelve to 15 random tissue
blocks were chosen from each lobe separately: the blocks were cut
from both upper and lower lobes parallel to and about 1 cm from
CHEST I 101 I 3 I MARCH, 1992

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

from

emphysema by the absence of destruction of alveolar walls.

The term senile
lungis proposed for this condition.
(Chest 1992; 101:793-99)

793

the pleural surface. Incompletely expanded lung regions were

carefully avoided. Paraffin sections, 4 @tmthick, were prepared in a
standard way, and stained with Masson's trichrome.
Since the
shrinkage from fixed lung to histologic slide is fairly constant even
in different pathologic conditions, a constant shrinkage factor of
0.61 was used.' The stained slices were examined by light micros

copy with a x 10 eyepiece and a X 10 pIano-objective (magnification

x 100).
The mean linear intercept, Lm, was measured following the

eral airways without cartilage or mucous glands in their walls) per

millimeter ofperimeter
(AA/mm).12For each lung we measured

the internal diameter and the corresponding AA from seven

membranous

bronchioles

both in the upper and lower lobes. These

14 airways, chosen at random, had to meet the following criteria:

their walls had not been sectioned

because

they were adjacent

to

the edge ofthe block, the portion bordered by accompanying blood

vessels did not exceed 25 percent of the total perimeter,
section was (almost) completely transversal.

and the

technique of Dunnill.' Briefly, crossed hair lines with known length

The internal diameter ofat least 23 membranous bronchioles per

are projected on the parenchyma, and the intersections between

lobe, with an internal diameter of 2 mm or less, was measured

following the technique ofMatsuba and Thurlbeck. Airway density

lines and alveolar

septa are counted.'

The length

ofthe

line divided

by the number of intercepts (NI) yields Lm. To partition Lm into

its two components, namely the mean transection length, which
reflects the thickness of the septal wall (Lm wall:Lmw) and the
internal diameter ofthe airspaces (Lm air:Lma), Dunnill's technique
was slightly modified by adding a point counting method. The hair
line was divided in 100 equally distant points, with a point spacing

of the lattice (Zeiss KplW10x /18) of 0.00955 mm at a magnifi

cation of 100 times. The sum of the points falling over the
parenchyma equals the total transection length of the parenchyma.
Dividing the latter by NI yields the mean transection length per
airspace (Lmw). Two hundred points were counted for each of the

120 microscopic fields investigated per lung, thus ensuring a

statistically

representative

sample.@

Since Lm is known to be a hmction of age in normal lungs,6(b0

Lm is expressed in percent predicted, using the regression of

Thurlbeck@: Lm0.0011
of Thurlbeck, airspace

age+0.2174.
enlargement

In agreement with the data

was considered to be exag

gerated when Lm exceeded 120 percent ofits predicted value.

To evaluate objectively

the uniformity

(n/sq cm) was defined as the number of membranousbronchiole

airways per square centimeter oflung tissue; n/sq cm was measured
by a Leitz image analysis system.

RESULTS

The lungs were divided into three groups on the

basis of the global aspect of the parenchyma on the
whole lung sections (examined by eye and under the
dissection microscope) and of the value of Lm . Lungs
of group A had a normal aspect, the value of Lm did
not exceed 120 percent of expected. In the lungs of
groups B and C, the value of Lm was higher than 120
percent. The enlargement ofthe respiratory airspaces
was homogeneous in group B, irregular in C. Mor
phologically

the

lungs

of group

C were

typical

for

of airspace enlargement,

centrilobular
emphysema (CLE). According to the
we determined the standard deviation of NI. The latter (SDN,)is a
clinical history, group A consisted of five nonsmokers,
function of (1) the inhomogeneity of the alveolar sizes, and (2) the
four current smokers, and one exsmoker. None had a
average alveolar size, the larger the alveoli, the less the number of
history of chronic bronchitis. In group B, three sub
intersects (provided the optical magnification remains constant),
jects were smokers, two were exsmokers, and one was
and thus the smaller SDN,. Accordingly, to estimate the uniformity
of airspace enlargement, we used the coefficient of variation of NI:
a nonsmoker. The smoking habits were not docu
SDN,divided by the mean value of NI. With the exception of three
mented in three subjects. Two subjects had a history
lungs. both NI and SDNIwere also determined separately for upper
ofchronic productive cough. Group C consisted of six
and lower lobes.
current smokers, three exsmokers, one nonsmoker,
As an index for destruction, we counted the number of alveolar
and one unknown. Eight subjects of group C had a
attachments on membranous bronchioles (the nonalveolated periph
Table 1Biometricand Morplsometric Data Oftlse Three Groups oflrnsga: A Normal,' B Senile;
C= Emphyseinatous bings (Mean ValuesStandard Deviation)5
ABCDuncan's

TestAge,

16.5A,BB,CBody
20.369.94.861.3
yr49.1
CEmphysemascore1.21.69.14.614.35.4ABCLm,
1116691667A,
length, cm166

B,

mm0.2890.0390.4920.0940.5840.140ABCLm%predt105.69.5165.631.6203.047.2ABCNI5.490.723.260.552.840.82AB
CSDN,3.220.282.060.242.420.39ABCSDN/NI
BCLma,x 100%59.34.264.07.487.5

11.0A,

mm0.2650.0350.4530.0840.5300.123ABCLma%predt100.08.5156.028.8189.142.5ABCLmw,
mm0.0240.0050.0390.0110.0540.017ABCLmw%pred@98.27.7150.450.9205.867.4ABC@l,
BCn/sqcm0.850.110.700.150.640.11AB,CAA/mm6.720.876.750.915.760.72A,
mm0.800.090.850.140.510.09A,
BC
*Duncan's

multiple

range

test: the dashes

NI = number of intersections

indicate

per line; SDNI

a significant

(p<0.05)

difference

standard deviation of NI; Lma

between

groups

internal diameter

A, B, or C. Lm = mean

of airspaces; Lmw

linear

intercept;

alveolar wall thickness;

@l,
n/sq cm= mean internal diameter and density of membranousbronchioles;AA/mm number ofattachmentsper millimeterofperimeter.
tPredicted values according to Thurlbeck.'
@Prediction equations calculated from 19 normal lungs (see text).
794

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

The Senile Lung. Part 1: Structural Aspects (Wrbeken eta!)

@
@

@1
@::@:@

history

ofchronic

productive

A@

cough.

significantly larger in group C (Table 1). A value of CV

The mean values

(1 SD) of the biometricandmorphometricdataof

larger

the three groups of lungs are listed in Table 1.

group B (79 percent, in a lung with an emphysema

score of2O, the highest ofthis group), a value less than
70 percent in one lung of group C (67 percent, in a
lung with an emphysema score of 6). A systematic

Respiratory

Airs paces

Liii: In agreement

with Thurlbeck,6

Lm was found

quantitative

to depend on age: a similar age dependence

was oh
served
in the group of ten normal
lungs (Lm =
0.0014

age + 0.2202,

not improved

tipper and lower lobes

was

or lung

average to 92 percent

lungs,

and in group

Airspace

C than in B (Table 1). In

Size

(Lma)

and

Septal

Wall

Thickness

of Lm . Lma is also linked to

age: to correct for this influence, we used the following

relationship:
Lma = 0.0012 age + 0.2031 , calculated

the whole lung. For a given value of NI, SDN, and,

accordingly,
the coefficient
of variation
(CV =
SD/NI x 100 percent) will be larger in the presence
ofa larger inhomogeneity
ofairspaces.
The CV turned
out to be quite similar in lungs ofgroups A and B, l)ut

addition of the latter two factors resulting in a signifi

cant improvement of the regression (r@:0.27 for age,
0.49 for age and height, and 0.78 for age, height and
lung weight). (Correlations were maintained and sig
nificance levels improved when the data from nine

t..

@,

.@

.;,
.@,.

.@

@. _.\@

@-

@- .,@.J. @
.@
)

:;@@

@: @ .

.9

. ) :

@@@@-;:
-

@tt

@;@@-@h,.-i

.@

..

.,,

.@

:-@@

related

normal

lungs

On the other

not only

and lung weight

to age,

not

hand,
but

also

(positively),

the

@.,@

@@:1:;@t_@_
,

-.,.
.&@

(negatively)

nine

study).

.v,,1

:@
C,

-,-@

,A

(adding

i.\

..

4.

was significantly

to height

\_@

J4

Lmw

lungs

in the present

.@

@1

,r.;g;@(

.1'

-.

19 normal

included

.
:t@:s@z@.._@.:.;

4)
,

from

..@
...

., @,
\@-@ .*
,u@

between

normal

-. )@..
,
rh@@r-'k::
,@. .
b
@_Tic@
,
--@-

height

difference

was found in group C: both NI and SDNI were

significantly lower (p<O.O5) in the tipper than in the
lower lobe.

all lungs of B and C, except for one lung of group C

(V16), the absolute values of Lm exceeded 0.38 mm.
The distinction between group B and emphysema
tous lungs, achieved by examination of whole lung
sections, was reevaluated by a quantitative method
based Ofl the relationship between SDN, and NI for

:..-@

relationship

was met only in one lung of

(Lniw): In normal lungs, Lma correspondson the

The

in addition,

70 percent

weight into account. Lm, when corrected for age (Lm

%pred), is significantly larger in group B than in

@
@
@

p<O.O2).

by taking,

than

,@

,..

4
FIcuKE 1. Normal lung (yb). Overview of the paren
chyma (magnification X 6) (Massor@s trichroine stain).
Lm =0.279 nun or 105 percent ofthe predicted value.@

Ficuiw 2. Example ofa lung ofgroup B (V19) (magni

fication X 6) (Masson's trichrome stain). LIT, 0.532 mm
r 181

percent

ofthe

predicted

value.

CHEST I 101 I 3 I MARCH, 1992

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

795

additional tiorinal lungs we@-cincluded iii tlu' nuhlti1)lc

regression .) Both Li@a and Lin@v are sign ifican tl@
different @tfl1OI1g
the threc g,oii@)s of slil)jccts, @tIS()
@@hen
the@' arc (orrecte(lfor age (Ln@a), ur for age,
height, and lung
@@eight
(Liii@@)(Table 1). A tight
correlation

@@ts
found

l)et\v('('ll

Lii@a and

L,ii@@ for the

three gronp@ of lungs coiisidered together (r: 0.95,

1)<0.(@1).
The correlation
@@as
statisticall@ significant
in gr@ti@@C (r=0.94,

1)<.(XK)l)

and

t;7fll)Il 1JS('fli(! Scot-c (S

Liii : The

are significantly
lungs,

larger

in

served

SC()res

B than in normal

i<().(XX)1).

within

the

The correlation

three

20

is also oh

@fl)L1@)S
separate1@

dm111 IS

especially goo(l ill the CLE giutip.

histologic I',xa7llin(ltion : Si ni@@leh ist )I()gic exanii
nation of the lungs of gnnmp B revealed a rather
lioniogeneous

di @Sl)dCCenlargenien

disorganization

()f the

al@'e )lar

@fl)ti@)
B, to examine

inagnification.

efle(l,

t thr( a mgli )n t . An@

The

a normal

alveolar

the exact nature

10

1)t1@e11@'l1Y11itt()r l)ron

:@@:J__

chioles was not seeii. Except for the fliOst affected

Itings, one lId(l the ifl@l)reSsioii @@hen
looking at lungs
lung,

thick

@@as
not detecte(1

.6

f@h1

exaI1@1)leS, respectiyeh;

3.31;

sai@e

lung

amul a lung

of

12

14

1.6 1.8 2.0mm

(liameters of me'mhranot,s

lo@ver lO)l)e5 (mean:

P<O.Ol),

d is significantly
0.47

il/Sq

cm

vs 0.56
l)eiflg

sn@aller

in the

paired

t test:

mm;

siniilar

in tipper

and

lo@verlO)l)C5.

m@;tgniflcation.

The emphvseuiiatotis

tO

by
dI1(l B. In groui@@C,

of a nornial

.8

UHE @3.
R('l@1ti\(@
(list ril)IItion ohnt('rt,al
l)ruucllu)les in gz-ui@@@s
A, B, and (.

hut at a higher

@@aIls
@@ere(lifinsely

of which

light 1iucroscop%@Fihi-osis or iiiflaniniation

@@as
iiot d
feature ofth'lungs ofgrou@) B. Figures 1 and 2 1)m@eSent
@fl)t1@)B at tlu'

destruction

@@as
distril)umte(l

There is a significantimi@'erse
relationshipbetween

irregularly throumghoumtthe lungs in groum@@C. In a(ldi

(1and 1115(1
cni in groups A and B. This relationship

tiOll

iU)t obserted

to the

thickened

air alveoli,

as seen

characterized

sel)tal

@@alls
Surrounding

in group

B, the

enlarged

Parelidlymna

by n@ik1 focal fibroSis, at tinies

@ufca;iJ)ianietcr (@1)aiui I)t'nsittj (?i/S(j (Ill

): The
i@ean values
of' d and il/sq cni of I)eril)lleral airways of'
the three
@fl)tI@)5
of' lungs are sl@o@@'ninTable 1. In

bi'oiichiules

in the three

@vith an internal

iTini and a reduction

of dian@eters
l)et'Weefl
reduced

4).

In

addition,

an

is

inverse

dmm
1,1
V
V

0.7-

V
A

dianieter

@fl)tI@)5B dn(l
in groups

of less

thati

B and

lungs
C.

The

C (Table

0.5-

A
AA
A

A A

: 0.87

P<0.000i

being

intern@e(liate

(1ensit\@
1). Both

1115(1 cn'i,
d and

is

mi/sq

cn@are siniilar in tip@@eran(l lower lobes in grOti@)sA

796

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

0.30.4

0.5

0.6

07

08

0.9
n/cm2

0.6

of larger air@@';tys,the (listril)ntion

d of normal

A
A

gn@ups of lungs

is shown in Figure 3: with resl)ect to the other two

groups, @fl)U@)
C is characterized by a relative increase
in air\vavs

(Fig

A and B, the values of'd are sii@ilar@oii

the average slightly lai'ger in group B, @vlwreasd is

markedly smaller in CLE. This difference is statisti
cally sigiiificauit. The distribution
of dianieters of'
rnenil)ranons

0.9-

?ufeml)rafl@USB roiu'Ii jolt's

lungs ofgronp

in

was

ac'coii@

panied 1)%a iiiot@@uiuicieuiinflai@n@atorv infiltrate and

varying tfl1OtmI@tS
of fibrosis, especially around the
respiratory bronchioles.

is f )I1I1(ll)et\veefl Vmnph@sei@;t51'(@redm111

Lni (r = 0.95;

of

B
C

dfl(l in gr@ti@@
C thami in B (fable 1). A satisfactory

correlation

30

B (r=0.92.

ei@ph vsenia

in group

40

FI(;tIIF 4. Relatinnsliip l)('tWceIl sn@all airsva@ (lian@eter, @T,and

(l(@I,Sit@,
@@1
(111. 1l1(@r(lati()nshiJ) is highly significant Lfld sin@ilar
(r. rcspecti@rl@@O.S2 tI)<(@.(@@@)
atl(l (1.94 (P<0tM@04)) in tiorinal
lungs (circles) anil li,tigs of gnnip B (triangles, 1})iflt do'@vn).The line
is the

ouauan

r('gr(@ssioI1 line

for

hotly

groups

A and

B.

No

(orrelatu)llis ohst'rvt@d 1,, group ( (tria,,glcs. point up).

The Senile Lung, Part 1: StructuralAspects(Verbekenet a!)

relationship

is found in emphysematous

lungs between

d and Lma (r: 0.68; p<O.O3) or Lmw (r: 0.63; p<O.04).

Similar relationships were found in both upper and
lower lobes, considered separately.
Alveolar Attachments (AA): The number of AA is
related to the diameter of the corresponding airways.
Accordingly, AA was expressed per millimeter of
bronchiolar perimeter (AA/mm). AA/mm was similar
in groups A and B, significantly lower in group C

(Table1).Thoughthe lowestvaluesaremet in group

C only, there is a large overlapping between individual
lungs. There is a negative correlation between AA/mm
and Lma in group A (r:
0.73, p<O.O2). The latter
correlation, not observed in group B, is at the limit of
statistical significance in group C (r
0.57, p = 0.06).
AA/mm

is similar

in both

upper

and

lower

lobes,

also

in C (in the latter group: 5.76 1.38 in tipper lobes vs

5.77 0.72 in lower lobes).
Histologic Examination: In contrast to normal and
group B lungs, where most airways are histologically
normal, the walls of the membranous bronchioles in
group C are characterized by the presence of inflam
mation and some fibrosis, extending into the walls of
respiratory bronchioles.
DIscussIoN
The use of morphometric techniques allowed us to
delineate an anatomic entity oflungs characterized by
an exaggerated
but homogeneous
enlargement
of
airspaces without destruction (group B), distinct on
the one hand from normal (group A), and on the other
hand from emphysematous lungs (group C).
The estimation of the enlargement of the airspaces
was achieved on the basis ofthe mean linear intercept,
Lm, the uniformity ofthe airspaces by comparing Lm
in upper and lower lobes and by computing the
coefficient ofvariation ofNI, the number of intercepts
counted to measure Lm. The amount of destruction
was evaluated by determining the number of AA on
membranous bronchioles (AA/mm).
It is known that Lm increases with aging also in
normal lungs.68m@To appreciate the effect of age on
Lm, we used the regression equation of Thurlbeck,6
which is very close to that calculated by Greaves and
Colebatch8 and to the regression we computed in our
group of normal lungs. According to Thuribeck, the
dispersion of normal values around Lm corrected for
age is 20 percent. A value of Lm exceeding 120
percent of predicted
may thus be considered
as
abnormal. This limit was used to separate the lungs of
groups B and C, from those of group A, and in this
respect appeared a better criterion than an absolute
upper limit for Lm.'5 For instance, taking 0.38 mm as
an upper limit for nonemphysematous
lungsi6 would
result in accepting the Lm value (0.32 mm) of one
lung ofgroup C as normal which is unlikely considering

the age of the subject (23 years). In comparison, the

scores attributed on whole lung sections, by means of
a Ryder grid,3 appeared to correlate satisfactorily with
Lm. This is in agreement with the findings of others
in emphysematous lungs.@7@
In the present study, a
correlation was found also in group B, because the
score

was based

on airspace

enlargement,

whether

regular or not.
A characteristic of airspace enlargement in emphy
sema as compared with that in aging lungs is its
irregular distribution.3 The airspace enlargement
in
the lungs of group B was remarkably homogeneous in
upper and lower lobes. In contrast, in group C, besides
the presence ofbullae, also the enlargement of smaller
airspaces was irregular and differed topographically:
Lm was systematically larger in the upper (0.607
0. 123 mm) than in the lower lobes (0.545 0. 157 mm).
This is in agreement with the findings of Bignon et
aP' and was not observed in the lungs of groups A and
B. Within the acinus, the airspaces varied also more
in size in group

C than in B. As an estimate

for this

variability, we determined the CV of the number of

intercepts counted in the determination
of Lm. The
latter CV was indeed significantly larger in group C,
and almost identical in groups A and B (Table 1).
Finally, the amount of airspace destruction
was
estimated by counting the number of AA on the
membranous bronchioles. Since this figure depends
on the size of the bronchioles in normal lungs,12 the
number of attachments was expressed per millimeter
of bronchiolar perimeter (AA/mm). The AA/mm was
similar in normal lungs and lungs of group B and
significantly

smaller

in C. The number

of attachments

decreases, indeed, with the amount ofemphysema.@

There was no difference in the number of attachments
between upper and lower lobes.
Summarizing, a number oflungs were characterized
by an increase in airspace size out of proportion to the
increase expected on the basis of age. This increase
was probably not a consequence
of emphysema.
Indeed, the airspace enlargement
was regular and
there was no loss ofAA. In addition, there was neither
conspicuous chronic inflammation nor postinflamma
tory fibrosis. These lungs should thus be classified in
the category of simple airspace enlargement. Because
all subjects of this group were older than 60 years, we
are probably dealing with an exagj@erated form of aging
lung, which we suggest to designate by the term
senile
lung.
Senile lungs demonstrate two other characteristics,
so far unexplained, which they share with CLE: a
thickening of alveolar walls and a reduction of the
number of peripheral airways.
A partitioning of Lm into its airspace (Lma) and
wall components (Lmw) shows that alveolar walls are
significantly thicker in groups B and C than in normal
CHEST I 101 I 3 I MARCH, 1992

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

797

lungs. The latter increase in thickness, also described

by Anderson et al,'@'@in emphysematous lungs is not
easily observed by eye: it is only in advanced emphy
sema that a diffuse thickening ofthe walls surrounding
large holes is noticed. Simple histologic examination
does not reveal any differences between the increase
in Lmw in groups B and C.
The density ofmembranous bronchioles is generally
reduced in emphysematous
lungs2@': this is in
agreement

with our findings

in CLE.

A similar

reduc

tion in density is observed in the lungs ofgroup B. On

the other hand, bronchiolar diameter is normal in
group B and decreased in emphysema.
The latter
decrease accompanied by an excess of airways with a
diameter of 0.6 mm or less has been documented
previously by others.'3'2' @@@@
Bronchiolar diameter
and density are negatively correlated in both groups
A and B. This relationship, observed by Matsuba and
Thurlbeck'4 in nonemphysematous
lungs and by Nagai
et aP@ in patients with airflow obstruction
is not
observed in emphysematous
lungs (Fig 4). In the
latter, we found in agreement

with Mitchell

et aP'and

Hale et aFa negative correlation between bronchiolar

diameter and the severity of emphysema, expressed
by the size of Lm.
The reduction of bronchiolar diameter is more
pronounced in the lower than in the upper lobes in
CLE . In contrast, Lm is larger in the upper lobes.
This is in agreement with the data of Berend,37 who
observed more bronchiolar inflamfnation and narrow
ing in the lower lobes in CLE and suggested that
different mechanisms may he responsible for small
airways narrowing and alveolar enlargement
and/or
destruction.
On the other hand, we observed an
inverse relationship in CLE between Lm and diameter
in both upper

and lower lobes, considered

separately.

This, in turn, supports the classic idea of a direct

causal relationship between small airway abnormality
and the occurrence of emphysema.@'
In a study of 80 lungs from nonsmokers, Anderson
et al@ observed the presence of mild to moderate
emphysema of the panacinar type in 20 elderly sub
jects (60 years or older). There was a diffuse dilatation
of alveolar spaces, distributed
more or less evenly
throughout secondary lobules and lobes or lung. Al
veolar walls were often thickened by subtle, diffuse
interstitial

inflammation.

There

was rupture

of alveolar

structures with occasional simple dilatation. We sub

mit that a number of these lungs might be categorized
as senile lungs. The distinction between mild panaci
nar emphysema and senile lungs is often very difficult
by visual inspection.37 It is important that quantita
tive techniques are developed further for separation
ofborderline cases (equivocal
emphysematous areas
ofThurlbeck@). We suggest that the determination of
the

CV

of the

number

of intercepts,

counted

798

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

to

determine

Lm, is useful for this purpose:

70 percent

appears

to separate

a value of

quite well emphysem

atous from normal and senile lungs.

REFERENCES

1 Snider GL, Kleinerman J, Thurlheck WM, Bengali ZH. The

definition ofemphysema:
report ofa National Heart, Lung, and
Blood Institute, Division of Lung Diseases Workshop. Am Rev
Respir Dis 1985; 132:182-85
2 Gough J, Wentworth JE. Thin sections ofentire organs mounted
on paper. In: Recent advances in pathology London: J & A
Churchill Ltd, 1960:80-6, chap 3
3 Ryder RC, Thurlbeck WM, Cough J. A study of interobserver
variation in the assessment of the amount of pulmonary emphy
sema in papee.mounted whole lung sections. Am Rev Respir Dis

1969; 99:354-64
4 Van Brabandt H, Cauberghs M, Verbeken E, Moerman P.
Lauweryns JM, Van de Woestijne KP Partitioning of pulmonary
impedance in excised human and canine lungs. J AppI Physiol
1983; 85:1733-42
5 Dunnill MS. Quantitative methods in the study of pulmonary
pathology. Thorax 1962; 17:320-28
6 Thurlbeck WM , The internal surface area of nonemphysematous
lungs. Am Rev Respir Dis 1967; 95:765-73
7 Weibel ER. Stereological methods. London: Academic Press,
1980

8 Greaves IA, Colebatch HJH . Elastic behavior and structure of

normal and emphysematous lungs post mortem. Am Rev Respir
Dis 1980; 121:127-36
9 Niewoehner DE, Kleinerman J, Liotta L. Elastic behavior of
postmortem human lungs: effects ofaging and mild emphysema.

J ApplPhysiol1975;39:943-49
10 Willems LNA, Kramps JA, Stijnen T, Sterk PJ, Weening JJ,
Dijkman JH. Relation between small airways disease and
parenchymal destruction
in surgical lung specimens. Thorax
1990; 45:89-94
11 Thurlbeek WM. Internal surface area ofnormal and emphysem
atous lungs. Tenth Aspen Emphysema Conference, 1967:379-93
12 Saetta M, Chezzo H, Kim WD, King M, Angus GE, Wang NS,
et al. Loss of alveolar attachments in smokers: a morphometric
correlate oflung function impairment. Am Rev Respir Dis 1985;
132:894-900
13 Anderson AEJr, Foraker AG. Relative dimensions of bronchioles
and parenchymal spaces in lungs from normal subjects and
emphysematous patients. Am J Med 1962; 32:218-26

14 Matsuba K, Thurlbeck WM. The number and dimensions of

small airways in nonemphysematous
1971; 194:516-24

lungs. Am Rev Respir Dis

15 Saetta M, Shiner RJ,Angus GE, Kim WD, WangNS, King M,

et al. Destructive index: a. measurement of lung parenchymal
destruction in smokers. Am Rev Respir Dis 1985; 131:764-69
16 Matsuba K, Ikeda T, Nagai A, Thurlbeck WM. The National
Institutes of Health Intermittent Positive-Pressure Breathing

Trial: pathology studies: IV. the destructive index. Am Rev

Respir Dis 1989; 139:1439-45
17 Saito K, Cagle P. Berend N, Thurlbeck WM. The destructive
index' in nonemphysematous and emphysematous lungs: mor
phologic observations and correlation with function. Am Rev
Respir Dis 1989; 139:308-12
18 Thurlbeck WM . Internal surface area and other measurements
in emphysema. Thorax 1967; 22:483-96
19 West WW, Nagai A, Hodgkin JE, Thurlbeck WM. The National
Institutes

of Health

Intermittent

Positive-Pressure

Breathing

Trial: pathology Studies: III. the diagnosis of emphysema. Am

Rev Respir Dis 1987; 135:123-29
20 Berend W, Woolcock AJ, Marlin GE. Correlation between the
The Senile Lung, Part 1: Structural Aspects (Wrbeken eta!)

function

21

22

23

24

and structure

of the lung in smokers.

Am Rev Respir

Dis 1979; 119:695-705

Bignon J, Andre-Bougaran J, Brouet C. Parenchymal, bronchi
olar and bronchial measurements
in centrilobular emphysema.
Thorax 1970; 25:556-67
@.
Petty TL, Silvers GW, Stanford RE. Radial traction and small
airways disease in excised human lungs. Am Rev Respir Dis
1986; 133:132-35
Linhartova A, Anderson AE Jr. Foraker AG. Radial traction and
bronchiolar obstruction in pulmonary emphysema. Arch Pathol
Lab Med 1971; 92:384-91
Linhartova A, Anderson AE Jr, Foraker AG. Aflixment arrange
ments of peribronchiolar alveoli in normal and emphysematous
lungs. Arch Pathol

Lab Med

1982; 106:499-502

25 Anderson AE Jr, Furlaneto JA, Foraker AG. Bronchopulmonary

derangements in nonsmokers. Am Rev Respir Dis 1970; 101:51827
26 Depierre A, Bignon J, Lebeau A, Brouet G. Quantitative study
of parenchyma and small conductive airways in chronic nonspe
cific lung disease. Chest 1972; 62:699-706
27 Matsuba K, Thurlbeck WM. Disease of the small airways in
chronic bronchitis. Am Rev Respir Dis 1973; 107:552-58
28 Anderson AE Jr. Foraker AG. Populations of nonrespiratory
bronchioles in pulmonary emphysema. Arch Pathol Lab Med
1967; 83:286-92
29 Matsuba K, Thurlbeck WM. The number and dimensions of
small airways in emphysematous
lungs. Am J Pathol 1972;
67:265-75
30 Linhartova A, Anderson AE Jr, Foraker AG. Further observa
tions on luminal deformity and stenosis of nonrespiratory

bronchioles in pulmonary emphysema. Thorax 1977; 32:53-9

31 Petty TL, Silvers GW, Stanford RE. Mild emphysema is
associated with seduced elastic recoil and increased lung size

but not with air'flow limitation. Am Rev Respir Dis 1987;

136:867-71
32 Cosio MG, Hale KA, Niewoehner
DE. Morphologic
and
morphometric
effects of prolonged cigarette smoking on the
small airways. Am Rev Bespir Dis 1980; 122:265-71
33 Hale KA, Ewing SL, Cosnell BA, Niewoehner
DE. Lung
disease in long-term cigarette smokers with and without chronic
air.flow obstruction. Am Rev Respir Dis 1984; 130:716-21
34 Nagai A, West WW, Paul JL, Thurlbeck WM. The National
Institutes of Health Intermittent Positive-Pressure Breathing
Trial: pathology studies: I. interrelationship
between morpho
logic lesions. Am Rev Respir EMs 1985; 132:937-45
35 Petty TL, Silvers GW, Stanford RE. The morphology and
morphometry

of small

airways

disease

(relevance

to chronic

obstructive pulmonary disease). Trans Clin Climat Assoc 1983;

94:130-40
36 Mitchell RS, Stanford RE, Johnson JM, Silvers GW, Dart C,
George MS. The morphologic features ofthe bronchi, bronchi
oles, and alveoli in chronicairway obstruction: aclinicopathologic
study. Am Rev Respir Dis 1976; 114:137-45
37 Berend N. Lobar distribution of bronchiolar inflammation in

emphysema. Am Rev Respir Dis 1981; 124:218-20

38 Leopold JC, Cough J. The centrilobular

form of hypertrophic

emphysema and its relation to chronic bronchitis. Thorax 1957;

12:219-35
39 Thurlbeck WM. Chronic airflow obstruction
Philadelphia: WB Saunders Co, 1976

in lung disease.

CHEST I 101 I 3 I MARCH, 1992

Downloaded From: http://journal.publications.chestnet.org/ on 10/06/2013

799