Sexual Dysfunction and Erectile Impotence
in Chronic Obstructive Pulmonary Disease"
Eugene G. Fletcher, M.D., and Richard ]. Martin, M.D., F.G.G.P.
We studied 20 men (ages 46 to 69, mean 45 years)
with chronic obstructive pulmonary disease (FEV1
of 0.55 to 2.1 L), to determine the relative impor-
tance of pulmonary impairment VI other occult phys-
ical or psychologic factors In the genesis of sexual
dysfunction. Seven subjects bad ceased sexual activity
concomitant with worsening of their pulmonary symp-
toms; six because of erectile impotence and one due to
dyspnea. Frequency of Intercourse for the remaining 13
was 16 percent of preluDg disease levels, and h1Jido was
decreased to 25 percent of premorbid levels. Nocturnal
penile tumescence monitoring disclosed that six subjects
had organogenic erectile impotence (OEl). None of the
subjects showed signs of periphe.... ftSCUIar disease ..
assessed by Doppler eumination of peripheral pulses
(including penDe). lbe mean bulbocavernosus reftes
A Ithough there is a large population of male
patients with chronic obstructive pulmonary
disease (COPD), only anecdotal reports of sexual
dysfunction associated with lung disease haveap-
For editorial comment see page 398
peared in the literature.v" We have found that
these patients frequently complain of decreasing
sexual interest and functional ability. Many of
them are in an age group that shows an increased
incidence of cerebrovascular, cardiovascular, or
peripheral vascular impairment and adult onset
diabetes mellitus. For this reason, complaints of
sexual dysfunction may be perfunctorily attributed
to associated illnesses or to aging. When assessing
complaints of erectile impotence, newer diagnostic
methods emphasize the examination of specific
physical factors including peripheral autonomic
nerve function, adequacy of the genital vascular
supply, and hormonal function. 2,,, Psychologic dis-
turbances such as depression and anxiety may also
impair sexual function, but their role may be dif-
ficult to separate from the physical effects of as-
°From the Pulmonary Disease and Critical Care Section,
Department of Medicine, Oklahoma City Veterans Ad-
ministration Medical Center, University of Oklahoma
Health Sciences Center, Oklahoma City.
This work was supported in part by funds from the Veter-
ans Administration and US Public Health Service, National
Institutes of Health, grant HL07210-05.
Reprint requests: Dr. Fletcher, 2002 Holcombe Blvd, H ous-
ton 77211
CHEST, 81: 4, APRIL, 1982
latency (BCRL) for the om group (N = 5) was 40.2
JDSeC, wbile that for the group with faD nocturnal erec-
tions (N = 10) was 34.5 msec (P < 0.005). Four sub-
jecCs had occult diabetes mellitus evident on on! glu-
cose tolerance tests, and ODe had evidence of aD
androgen deficit. The correlation coefficient for rank by
sexual dysfunction VI pulmonary impairment aDd aae
was 0.66 (P < 0.005) and 0.24 P > 0.05), respectively.
SubjecCs with OEI tended to have the worst pulmonary
function test results and the highest T-scores on the
hypochondri_is, depression, and hysteria scales of the
Minnesota Multiphasic Penoaality Inventory. Data
suggest that sexual dysfunction worsens as lung disease
worsens and tbat chronic obstructive pulmonary dis-
ease may be aaociated with male impotence in the
absence of other commonly known causes.
sociated systemic illnesses. We have undertaken
a prospective study to further characterize these
complaints and to determine the relative impor-
tance of pulmonary impairment vs other occult
physical and psychologic factors in the genesis of
sexual dysfunction and impotence in males with
COPD.
MATERIALS AND METHODS
Sexual function was evaluated in 20 male subjects with
moderate to severe COPD, ages 46 to 69 (mean, 56) years.
The subjects were in clinically stable conditions for three
weeks prior to the study. Five subjects were self-referred
for sexual problems; the remaining 15 were volunteers re-
cruited from a pulmonary disease clinic without our prior
knowledge of their sexual function. Seventeen subjects were
living with wives or female companions, and three lived
alone but claimed having sexual contact with female part-
ners. Excluded from the study were persons with obvious
history of cardiovascular disease, diabetes mellitus, periph-
eral vascular disease, moderate to severe hypertension, or
excessive alcohol consumption within the preceding five
years. All subjects were outpatients at the Oklahoma City
Veterans Medical Center and gave informed consent prior
to participation in the study protocol.
Subjective Sexual Function
The subjects provided sexual histories to a trained in-
terviewer regarding the following: (1) frequency of sexual
intercourse, (2) level of interest in sexual activity, (3) ob-
servation of spontaneous morning erections, and (4) quality
of penile erections (duration and firmness). Changes in these
SEXUAL DYSFUNCTION IN COPD 413
parameters were recorded as percentage of the level of func-
tion at the time of the study vs the level of function just
prior to the onset of severe exertional dyspnea (premorbid
level). The observation of morning erections was recorded
as same, decreased, or absent The quality of penile erectile
function was categorized as follows:
O. No change in firmness or duration of erection from the
premorbid state.
1. Slight or moderate decrease in firmness or duration
but no difficulty completing coitus.
2. Firm to semifirm erections capable of intromission but
with as much as 75 percent reduction in duration. Subject
is usually able to complete coitus.
3. Firm to semifirm erection that detumesces after intro-
mission, frequently before climax.
4. Semifinn erection, much shortened in duration, usually
not able to achieve intromission, occasional short-duration,
&rm erections.
5. Usually flaccid penis with occasional short-duration,
semifinn erections or no erections at all.
Objective Erectile Function
Mercury-filled Silastic strain gauges placed at the tip and
base of the penis and connected to a tumescence monitor
(Event Systems) were used during three nights of poly-
graphic morutoring to evaluate erectile function objectively.
Sleep stages were recorded by EEG and scored according
to criteria of Rechtschaffen and Kales," Tumescence studies
were scored according to criteria of Karacan et a16 , 1 and
Fisher et al. 8 Full erections required a circumferential
change exceeding 16 mm or 80 percent of a maximal erec-
tion measured at the tip. The duration of full erections
was reported as the time from initial deflection to the time
of return to baseline (Fig 1). If the subject had no full
nocturnal erections or an average of less than one full erec-
tion lasting at least five minutes per night of monitoring
he was considered to have organogenic impotence. 8
Neurovascular Function
A complete physical examination focused on signs of
neurologic impairment and peripheral vascular disease. In
addition, neurovascular integrity was assessed by measuring
bulbocavernosus reflex latency (BCRL) and Doppler blood
pressures of the extremities and penis. The BCRL was meas-
ured in 15 of the subjects according to techniques described
by Ertekin and Reel. 9 A bipolar stimulating electrode was
placed over the glans penis, and electromyogram (EMG)
(I)
~
3~
"CD
CD
~ ~
c:: 0 ~_-I------~"--''""''"",""'''-----'--
CD
~ !~
fI)
c::
~)(
0 5
w Time in Minutes
FIGURE 1. Tumescence tracing from subject with fun nocturnal erections. Maximum erection
labeled. Partial erection with minimal tip expausion at far left of tracing.
414 FLETCHER, MARTIN
potentials were recorded from a fine-needle electrode placed
transcutaneously into the bulbocavernosus muscle. The mean
latency for normal males is 36.1 (± 4.6 msec SD). 9 Doppler
blood pressures of the dorsal arteries of the penis were per-
formed and scored according to published methods. 10 - 1 2
Peni.e blood pressure was considered normal if: (1) penile
systolic pressure exceeded the mean brachial artery pressure;
( 2 ) the ratio of penile to brachial systolic pressure was
greater than 0.8; or ( 3 ) the penile systolic pressure was
not more than 30 mm Hg below the brachial artery systolic
pressure.
Close scrutiny for occult diabetes mellitus further reduced
the possibility of missing neurovascular dysfunction unde-
tected by these two techniques. Any subject with a fasting
serum glucose level above 115 mg/dl or a two-hour post-
prandial (2 hr pp) serum glucose level above 140 mg/dl
received a standard 75-g, three-hour oral glucose tolerance
test (OGTr ). If two of the values other than fasting on
the OGrr exceeded 200 mg/dl, the test was considered
diagnostic of diabetes mellltus.P
Hormonal, Cardiopulmonary, and Neurop8flchiDtric Function
Possible hormonal abnormalities were assessed in each
patient by drawing paired serum testosterone-Ieutinizing
hormone (LH) levels, follicle-stimulating hormone (FSH)
levels, and performing prolactin assays.
Each subject underwent a routine ECG, pulmonary func-
tion test (PFT), and an exercise test. PFTs included expira-
tory spirometric tests, body plethysmography, a single-breath
carbon monoxide diHusing capacity (Dsb or Dco), and at
least three resting room-air arterial blood gas analyses.
Exercise performance was measured with bicycle ergometry
using a progressive load at one-minute increments and a
five-minute steady-state technique.
Each subject completed a computer-scored Minnesota
Multiphasic Personality Inventory (MMPI) at the time of
the study. Two psychologists who had no knowledge of
the subjects" sexual function or severity of lung disease
classified the results according to personality patterns and
ranked subjects by overall degree of psychopathology.
Statistical analysis of correlation coefficients and signif-
icance was done by the least squares fit method and Dun-
can's multiple range test. I •
The status of each subject's sexual function was followed
for 12 to 18 months after completion of the protocol. Re-
ports were obtained on changes in sexual function and lung
function in response to general therapy for lung disease (eg,
/a----========---1
MaXimum
· Erection
~I TIP
"""""-__- - -.................._
10 15 20 25 30 35 40 45 50 55
CHEST, 81: 4, APRIL, 1982
bronchodilators, corticosteroids, and home oxygen) and to
specific therapy for sexual dysfunction (eg, counseling and
hormonal replacement).
RESULTS
Subjective Sexual Function
Sexual function histories revealed that seven
subjects had ceased sexual activity from three
months to nine years before the study, while the
remaining 13 continued to engage in coitus with a
frequency of 16 percent of the premorbid level
(Fig 2). For the entire group, the mean frequency
of interest in engaging in sexual intercourse was
25 percent of premorbid levels, and for those still
sexually active it was 40 percent. The average time
between the onset of severe exertional dyspnea
and the time of study was five years (-+- 3.4 years
SD). Five subjects noted disappearance of spon-
taneous morning erections since the onset of pul-
monary symptoms, while the remainder noted
either about the same number or fewer than the
premorbid level.
Of the seven subjects who had ceased sexual
activity, six did so because of erectile impotence
and one because of severe dyspnea (FEV 1 = 0.55
L ), although he still had occasional 6rm erections.
Ten of the 14 subjects who remained capable of
having erections reported variable decreases in
finnness and reductions in duration ranging up to
75 percent. Three of these had frequent loss of
erections after intromission but prior to climax.
Only four subjects experienced little or no change
in duration or firmness of erections.
Obfective Erectile Function
Nocturnal tumescence monitoring identified six
100
80
eo
"
S
Premorbld Premorbld
Level 40 Level 40
..
.....
-
20
....L-
0 e..-
Coitus Frequency
FIG~ 2. SU~jective sexual function. Levels of frequency of coitus, libido, and frequency of
morning erections expressed as percentage of levels prior to onset of disabling pulmonary symp-
toms.
CHEST, 81: 4, APRil, 1982
subjects who failed to attain an average of one
full erection per night. Five of these had a history
of gradual progression of erectile impotence coin-
ciding with clinical worsening of their lung dis-
ease. Each subject's partner, when questioned,
verified this fact. One additional subject, who com-
plained of reduced erectile function but denied
complete impotence, failed to have full erections
during three nights of monitoring. Verification of
his sexual activity was not possible, since he lived
alone and did not have a regular sexual partner.
One subject (FEV t = 2.1 L) among the six
who gave histories of erectile impotence had full
erections during nocturnal tumescence monitoring
and was believed to be psychogenically impotent.
Within six months of the study he began having
frequent sexual intercourse with a different partner,
thus confirming this diagnosis.
No statistically significant diHerence in age, dura-
tion of pulmonary symptoms, duration of sexual
dysfunction symptoms, FEV! percent of predicted,
PaC02, or various sleep parameters ( Table 1 )
appeared between the group with full nocturnal
erections (N = 14) and those with no nocturnal
erections ( N = 6) by tumescence monitoring.
Functional residual capacity was significantly higher
in the group without full nocturnal erections. This
latter group also tended to have a lower resting
Pa02, maximum voluntary ventilation (MVV), and
achieved a lower mean exercise level than the
group with full erections, but these parameters
failed to reach statistical significance.
Neurovascular Function
Assessment of neurovascular function did not
reveal any evidence of peripheral vascular disease.
1
80
80 Same ............
.....
- ...
- ..... - ...
20 AbMnt
--
O
Inter••t In Coltua Reported Frequency
Morning Erection
SEXUAL DYSFUNCnON IN capo 415
 Table I-JIean and SD of J'ari. . . Parameter. for Full.,. A6aerd NoelurruJl Erection Gro.". *
Full Nocturnal Erections by NPT Absent Nocturnal Erections by NPT
(N == 14) (N-6)
A A
I I

Measurement Mean SD Mean SD

Mean age, yr 55 (6.5) 57 (6.6)
Duration of pulmonary symptoms, yr 10.3 (6) 8.5 (5.5)
Duration of sexual dysfunction, yr 2.8 (2.9) 4.3 (3.3)
FEVl , % pred 37.6 (15) 34.5 (17.6)
FRC, % predf 169 (30) 222 (53)
MVV, %pred 40 CIS) 27 (12)
PaCOs, mm Hg 39.6 (6) 42.8 (7.3)
PaOt, mm Hg 62.2 (10) 54.7 (12)
Dco, % pred] 67 (24) 42 (16)
Exercise level achieved, kpm 400 (225) 250 (130)
% Change VE resting to VE max during exercise] 197 (109) 62 (56)

Bulbocavernosus reflex latency, m/secj 34.5 (3.0) 40.3 (2.8)

Sleep period time, min 408 (35) 394 (45)

Sleep efficiency index .84 (0.1) .83 (0.1)

Total REM sleep, min 66.7 (20) 65.9 (25)

Total maximum tumescence episodes per night] 1.6 (1.22) 0.1 (.15)
Total maximum tumescence time, min f 60.7 (50) 2.6 (4.1)
*Abbreviations: FEV1-forced expiratory volume in first second; MVV -maximal voluntary ventilation; NPT-nocturnal penile
tumescence; FRC-functional residual capacity; PaCOs==arterial tension of carbon dioxide; PaOs=-arterial tension of oxygen;
Dco-diffusion of carbon monoxide; kpm==kilopond-meters; VB-minute ventilation; and REM-rapid eye movement.
fSignificant at P <0.05.
Each subject fulfilled at least two of the three tion of FSH, although his LH and testosterone
criteria for adequate penile arterial blood pressure levels were normal. One subject complaining of
relative to brachial artery pressure. decreased libido and decreased firmness and dura-
While none of the subjects had clinical signs of 44
peripheral neuropathy, there was a statistically *
significant difference between the BCRL for the 42 ••
full nocturnal erection group (34.5 msec), and
five subjects without full nocturnal erections (40.2 40
msec, Fig 3). However, two subjects from each *
group did have abnormal OGTT responses. I
Five subjects had elevated 2-hr pp serum glucose
BULBOCAVERNOSUS 38 *•
•
-.-
REFLEX LATENCY
levels, ranging from 144 to 262 mg/dl. Four of (m sec) 36
these demonstrated abnormal responses to an
OGTT. Of these four, three were taking 15 to 34
20 mg of prednisone per day. Although all were
diet-controlled, one did require insulin for a brief 32
f•
time while his dosage of prednisone was being •
tapered. 30~---. ......- - - - -.....- -
FULL ABSENT
Hormonal Function NOCTURNAL NOCTURNAL
ERECTIONS ERECTIONS
Hormonal dysfunction other than diabetes did
not appear to be a major problem in this group. * DIABETIC OGTT
There were no subjects with elevated prolactin FiGURE 3. Bulbocavernosus reflex latency. Mean BCRL for
each group (P < 0.005; solid bars). Diabetic subjects de-
assays, and only one subject had a persistent eleva- picted by stars.

418 FLETCHER, MARTIN CHEST, 81: 4, APRIL, 1982

 80
Y
• 70 I
80

-
• •
•
•• -l-
----
1.5 10 50
• I
Llt.r. mmHg mmHg
•
1.0
•
50 •
••
40
T-
o.a y
40 30 •
•
• FIGURE 4. Pulmonary function and blood gases
FEV1 Pa02 PaC02 for all subjects. Solid bars, mean values.
tion of erections had low to normal testosterone (Fig 4). Five subjects were severely hypoxemic,
levels (4.6 MIU I ml; normal, 4 to 10 MIU I ml) with with PaOt below 55 mm Hg, and four were hyper-
elevated LH levels (139 MIU I ml; normal, 3 to 30 carbic, with PaC02 over 44 mm Hg. Three subjects
MIU I ml). Replacement doses of testosterone cipro- were unable to perform exercise tests because of
nate for six months improved his libido but failed severe dyspnea.
to correct his erectile dysfunction. Based on their historic responses, the subjects
were ranked 1 through 20 in each of four areas
Cardiopulmonary Function of sexual function: frequency of coitus, libido,
The mean FEV! for the group was 1.1 L (range, Table 2--Correladon Coef/ide.... and P Yalua lor
0.55 L to 2.09 L). The mean resting Pa02 was Esel"eUe and Pulmonary Funcdon Parameier. f t
60 mm Hg, while the mean PaC02 was 40 mm Hg Sesrud Dy_function RfIlIlt·

80 Correlation P
• • • Measurement Coefficient Va.lue
• • • • ••
..
Sexual 80 Pulmonary function tests (N ==20)
Function • FEV., L 0.46 0.040

.
40
Rank FEV., % pred 0.40 0.085
20 .. * R=.66
FEV., % pred after bronchodilator
FEV./FVC
0.32
0.33
0.169
0.158
P c 0.005 TLC, % pred -0.62 0.006
0
0 20 40 eo 80
FRC, % pred -0.68 0.002
CO PO Rank (FEV. FRC. P02.0CO) RV/TLC -0.60 0.008
FVC, % pred 0.32 0.169
MVV, %pred 0.40 0085
MVV, % pred after bronchodilators 0.44 0.061
80
0.42 0.065

..
• Dco, % pred
PaC02, mm Hg 0.54 0.013
Sexual 60 • PaOt, mm Hg 0.52 0.019
FunCtion
Rank 40 • Exercise tests (N == 16)
Maximum level rea.ched, kpm 0.66 0.002
20 • •
•* R= .24 % Change, resting to maximum
NS Heart rate 0.64 0.003
0 Respiratory rate 0.54 0.015
45 50 55 60 65 70 Ventilation 0.63 0.004
Age in years Oxygen consumption 0.59 0.010
CO2 production 0.52 0.028
Psychogenic Impotence ..
No nocturnal erection • VD/VT ratio -0.23 0.354
Full nocturnal erection • O2 saturation 008 0.732
FIGURE 5. Sexual function rank vs pulmonary function rank
*Abbreviations: kpm==kilopond meter; Vn,IVT==dead space/
and age. Top: Increasing sexual dysfunction correlates posi-
tidal volume; FEV. ==forced expiratory volume in first
tively with worsening pulmonary function tests (P < 0.005).
(Bottom): Increasing sexual dysfunction appears to be unre- second; TLC == total lung capacity; FRC == functional residual
lated to age (P > 0.05). Solid triangles, subjects with full capacity; FVC ==forced vital capacity; RV ==residual volume;
erections during nocturnal monitoring; solid circles, sub- MVV == maximal voluntary ventilation; D c o == diffusion of
jects without full erections during monitoring; soUd star, carbon monoxide; Pae02 == arterial tension of carbon dioxide;
single subject with psychogenic impotence. and PaO:!=arterial tension of oxygen.

CHEST, 81: 4, APRil, 1982 SEXUAL DYSFUNCnON IN capo 417

morning erections, and quality of erections. The
rank scores from each area were added to give
a composite sexual function rank with a minimum
score of 4 and a maximum score of 80. The higher
scores represented those subjects with the least
impairment of sexual function.P In a similar man-
ner, using the four pulmonary function parameters
that individually correlated best with sexual dys-
function, subjects were ranked using the Pa02 and
the percent of predicted values of FEV1, FRC, and
Dco (Fig 5); a high score represented the least
amount of pulmonary functional impairment. The
correlation coefficient between rank by sexual func-
tion vs pulmonary function was 0.66 (IP < 0.005),
while the correlation coefficient for sexual function
rank vs age (Fig 5) was 0.24 (P > 0.05). Individ-
ual pulmonary and exercise function test results
with their correlation coefficients against sexual
function rank by history are listed in Table 2.
Neuropsychiatric Function
Seven subjects scored greater than 2 SD (T-
score = 70) above the mean on the hypochondriasis
( US ), depression (D), and hysteria (HY ) scales
of the MMPI, which was interpreted to mean de-
pression mixed with anxiety and concern over
body function. However, the scales tended to be
higher for HS than for D. Similar findings have
been reported previously among patients with se-
vere COPD.1S Four of the six subjects who had
organogenic impotence were among those with high
T-scores on these three scales. The means of each
scale according to group (full nocturnal erection vs
absent nocturnal erection) are shown in Figure 6.
low number of subjects in the study, we could not
eliminate corticosteroids by objective data as a
contributory factor to sexual dysfunction in the
five subjects taking them. However, based on
testosterone levels, lack of correlation between
impotence by monitoring and steroid use, and in
all cases, sexual dysfunction history antedating the
use of corticosteroids, we concluded that the use
of oral prednisone was not the cause of sexual
dysfunction in these subjects.
Clinical F oUow-up
Six subjects reported some improvement in sexual
function during the study period. In four cases,
these changes were confirmed by partner ques-
tioning.
One subject with an FEV1 of 0.87 L ( stable
over eight weeks prior to study entry) at the begin-
ning of the study, improved over a three-month
period to an FEV1 of 1.38 L. This was in apparent
response to the use of short-term corticosteroids
in addition to his prestudy regimen of oral the-
ophylline and inhaled metaproterenol. He reported
(partner confirmed) that both libido and erectile
function had improved significantly with his change
in pulmonary status.
Three subjects were given home oxygen during
the study because of resting hypoxemia «55 mm
Hg) or nocturnal hypoxemia «45 mm Hg). All
80

.----.
Although the HS, D, and HY scales tended to be 80
higher for the erection-absent group, the only
scales that showed significant differences were ............
those of psychasthenia and social introversion.
However, absolute values for both of these scales
were within 1 SD of the mean for normal subjects.
The correlation coefficients for rank by sexual
dysfunction vs rank by severity of psychopathology
done by each psychologist were 0.35 and 0.30, re- 30 V IIdlt
spectively. The correlation for psychopathologic ~c.l.1 ·Clinlcal Scales
severity rank between the two psychologists was
0.76.
Medication regimens recorded at the time of
the study included theophylline, metaproterenol,
terbutaline, various antibiotics, prednisone, and
home oxygen. No single medication correlated
significantly with degree of sexual dysfunction * Significant difference No nocturnal erection •
(established by history) or erectile impotence (es- to P < .05 level Full nocturnal erection •
tablished by tumescence monitoring). We specif. FIGURE 6. MMPI results for all subjects. Solid circles, mean

ically examined the possible effect of corticosteroids T -scores for subjects with full nocturnal erections; Solid tri-
angles, subjects without full erections by tumescence moni-
on sexual function. Because of the design of and toring.

418 FLETCHER, MARTIN CHEST, 81: 4, APRIL, 1982

three ( partner confirmed) had improvement in
libido, and two stated erectile function had im-
proved on therapy. These two subjects are men-
tioned in more detail later.
Two subjects responded to counseling regarding
their pulmonary disease and sexual function. One
had improved libido after reassurance that sexual
intercourse would not be harmful to lung function.
The other was the only subject with proved
psychogenic impotence who responded by seeking
a different sexual partner. One subject, previously
discussed, in whom hypogonadism was diagnosed,
failed to respond to high doses of intramuscular
testosterone for erectile dysfunction. The remain-
ing 11 subjects noted no change in libido or erectile
function during the course of the study.
DISCUSSION
Sexual dysfunction has been associated with
long-standing diabetes mellitus,6,16 cerebrovascular
disease," after myocardial Infarction," and several
other chronic diseases. IS To our knowledge, no
prospective studies of sexual dysfunction in COPD
have previously been undertaken. The care of pa-
tients with COPD has centered on symptomatic
treatment of airway infections, bronchospasm, and
respiratory failure, while related physical and psy-
chosocial problems have received little attention.f
Sexual function may be an important aspect of life
to many patients who have lost jobs, earning pow-
er, traditional roles in society, and authority in
the family because of their pulmonary impairment.
In our experience, the routine questioning of male
patients with moderate to severe COPD commonly
yields a history of sexual dysfunction ranging from
decrease in libido to complete impotence, general-
ly coinciding with progression of pulmonary symp-
toms. Many physicians may not be familiar with
recently developed methods of evaluating sexual
dysfunction and may attribute declining sexual
ability to aging or other disorders common to this
population, such as cardiovascular disease, diabetes,
or peripheral vascular disease. Although it has
been shown that some decrease in sexual drive
and ability may accompany aging, more than 80
percent of males surveyed continued to have sexual
intercourse at least once per month through the
sixth decade."
It is difficult to quantify sexual dysfunction for
purposes of data analysis. In reviewing the litera-
ture and talking with numerous COPD patients,
we established what we believed was a represen-
tative scale of progressive penile erectile dysfunc-
tion. Combining this with historic changes in
frequency of sexual activity, libido, and self-observa-
CHEST, 81: 4, APRil, 1982
tion of morning erections, we ranked the 20 sub-
jects according to severity of sexual dysfunction.
This ranking correlated significantly with pulmo-
nary impairment as measured by the combined
parameter of FEV1, FRC and Dco percent pre-
dicted, and Pa02. Table 2 shows that eight of 15
parameters of pulmonary function correlated sig-
nificantly (P < 0.05 ) with sexual function rank,
while an additional four almost reached statistical
significance (P = 0.061 to 0.085). Six of eight param-
eters of exercise function correlated Significantly
(P < 0.05) with sexual dysfunction. Thus, the ma-
jority of commonly used parameters of pulmonary
mechanics, gas exchange, and exercise that define
the severity of dysfunction in COPD correlate with
the degree of severity of sexual dysfunction.
With the exception of the five patients who
were self-referred for sexual problems (none of
whom failed to have erections by tumescence
monitoring), subjects were selected on the basis
of severity of their lung disease and absence of
historic or physical evidence of accompanying, im-
potence-causing disease. Because the sample may
not be representative, this study makes no attempt
to assess the incidence of sexual dysfunction among
COPD patients. We were successful in collecting
a group of COPD patients without major accom-
panying illnesses, since none showed evidence of
significant peripheral vascular disease, and only one
had a hormonal abnormality that may have contrib-
uted directly to his sexual complaints.
We are uncertain of the effect of glucose intoler-
ance on sexual function in four of our subjects.
Although two of the four eventually proved by
tumescence monitoring to have organogenic im-
potence, one was taking 20 mg/ day of prednisone
and the other had had a normal 2-hr pp glucose
(75-g load) three years earlier at a time when
he was already experiencing sexual dysfunction.
The other two subjects, who had full erections
during monitoring, were both taking oral prednisone
at the time of the OCTI.
In a previous study," the mean BCRL for 14
normal control subjects was 36.1 msec. Twenty-two
patients with peripheral neuropathy (mostly dia-
betic patients) showed a mean BCRL of 47.8 msec
(SD, 4.9). While this test may not be suitable for
determining mechanisms of impotence in individual
subjects unless values are greatly increased, it ap-
pears to be a useful epidemiologic tool. Our study
group is too small to make a strong conclusion.
However, the higher latency times in the subjects
without nocturnal erections suggest peripheral
autonomic nerve damage as one possible mecha-
nism causing erectile dysfunction.
SEXUAL DYSFUNCTION IN COPO 419
 In addition to a functional autonomic nerve sup-
ply, voluntary penile erection depends on proper
cerebral cortical input. 8,4 Cerebrovascular disease
has been implicated as a causative factor in sexual
dysfunction." Since chronic hypoxemia may be
associated with neuropsychologic dysfunction, in-
cluding organic brain syndromes,22 it is conceivable
that it might also affect erectile function through
central cortical impairment, just as it can affect
reasoning and perceptual-motor integration. Al-
though there was no statistically significant differ-
ence between the mean Pa02 in either group (Ta-
ble 1), there was a trend toward a lower resting
Pa02 in the subjects without nocturnal erections.
We restudied nocturnal tumescence in two impo-
tent subjects with hypoxemia (Pa02 = 51, 48 mm
Hg) at eight and 20 weeks, respectively, after
beginning long-term low-How home oxygen. Neither
subject showed full erections on follow-up studies,
but the latter subject (after 20 weeks of 02) stated
that he was having semi6rm erections and was
able to achieve intromission and complete coitus
about one time per month. This history was con-
firmed by his sexual partner.
We have avoided use of the term psychogenic
to label those subjects who reported sexual dys-
function but whose tumescence monitoring records
were normal. The penile strain gauge measures
size changes only and the absence of full nocturnal
erections usually confirms organogenic impotence.
Conversely, the presence of nocturnal penile ex-
pansion does not necessarily mean that the penis
is functionally erect. Since we did not awaken the
subjects to examine erections nor measure rigidity
with pressure devices," we cannot comment on
what portion of the 14 subjects with "full" erections
by strain gauge records might have had insufficient
rigidity for vaginal intromission. Other researchers
have theorized that organogenic impotence follows
a progressive course of gradually deteriorating
erectile function: decreasing rigidity, failure to sus-
tain, and decreasing response to masturbation. 6,24
Psychogenic erectile dysfunction is believed to
follow this same sequence. Since both types may
continue to show full erections by nocturnal tumes-
cence monitoring, differentiation of the etiology
may be difficult. Only one of the 14 subjects with
full nocturnal erections claimed to have complete
impotence; the other 13 all complained of gradual-
ly progressive, partial loss of erectile function.
This subject was the only one in the study in
whom we could clearly diagnose psychogenic im-
potence. Since several of the subjects who were
impotent by nocturnal penile tumescence monitor-
ing had high T-scores on the HY, D, and HS scales
of the MMPI, psychogenic factors may have con-
420 A.ETCHER, MARTIN
bibuted to their sexual dysfunction. However,
underlying organic disease also could have caused
both the psychogenic and erectile function dis-
turbances.
Organogenic impotence has been studied most
extensively in diabetes mellitus. Although the fre-
quency and duration of nocturnal erections have
been shown by strain gauge monitoring to decrease
gradually with aging," a similar regression of tumes-
cence has not been demonstrated in patients with
diabetes mellitus. Perhaps one reason is the dif-
ficulty of clinical staging of diabetic patients.
COPD, on the other hand, is amenable to staging
using parameters such as pulmonary function, blood
gases, and exercise studies. Thus, loss of erectile
function could theoretically be correlated with
severity of the disease. We tested these parameters
against the objective results of strain gauge moni-
toring but were unable to show statistically sig-
nificant correlations. In a larger population, how-
ever, tumescence monitoring may show a significant
correlation between the number and duration of
full erections and objective parameters of lung
function.
Data from this study suggest that sexual dys-
function and erectile impotence can accompany
COPD in the absence of other known causes of
sexual problems. Furthermore, sexual dysfunction
tended to be worse in those subjects with more
severe pulmonary function impairment as assessed
by PFTs, blood gases, and exercise tests. Nocturnal
penile tumescence monitoring was successful in
establishing organogenic erectile impotence as the
cause for sexual dysfunction in six subjects, but
did not demonstrate an objective gradual loss of
erectile function corresponding to the deterioration
in PITs. While there was a significant difference
between the mean BCRL in the organogenic erec-
tile impotence vs the full nocturnal erection group,
further work may demonstrate that other factors,
such as hypoxemia, may play a role in erectile dys-
function accompanying COPD.
ACKNOWLEDGMENT: The authors wish to thank Nguyen
N. Thong, M.D., for performing the BCRL reflex studies;
Robert L. Kane, Ph.D., and George P. Prigatano, Ph.D., for
evaluating MMPI results, Paul Costiloe, Ph.D., for statistical
analysis, and Mrs. Fern Brandt, R.N., and the nurses of the
clinical research unit of the Veterans Administration Med-
ical Center at Oklahoma City for their help in conducting
this study.
REFERENCES
1 Kass I, Updegraff K, MufHy RB. Sex in chronic obstruc-
tive pulmonary disease. Med Aspects Human Sexuality
1972; 63:33-42
2 Agle DP, Bawn GL. Psychologic aspects of chronic ob-
structive pulmonary disease. Med Clin North Am 1977;
61:749-57
CHEST, 81: 4, APRil, 1982
 3 Malloy TR, Wein AJ. The etiology, diagnosis and surgical
treabnent of erectile impotence. J Reprod Med 1978;
20:183-94
4 Barry JM, Hodges CV. Impotence: a diagnostic ap-
proach. J Urol 1978; 119:575-78
5 Rechtschaflen A, Kales A. A manual of standardized
terminology, techniques, and scoring system for sleep
stages of human subjects. Bethesda, Md: US Dept of
HEW, Public Health Service, NIH, National Institute
of Neurologic Disease and Stroke, Neurological Infor-
mation Network, 1968; PHS publication no. 204
6 Karacan I, Salis PJ, Ware C, et ale Nocturnal penile
tumescence and diagnosis in diabetic impotence. Am J
Psychiatry 1978; 135: 191-97
7 Karacan I, Salis PJ, Thornby JI, Williams RL. The on-
togeny of noctural penile tumescence. Waking Sleeping
1976; 1:21-44
8 Fisher C, Gross J, Zuch J. Cycle of penile erections
synchronous with dreaming ( REM ) sleep. Arch Gen
Psychiatry 1965; 12:29
9 Ertekin C, Reel F. Bulbocavernosus reflex in normal men
and patients with neurogenic bladder and/or impotence.
J Neurol Sci 1976; 28:1-15
10 Abelson D. Diagnostic value of the penile pulse and
blood pressure: a Doppler study of impotence in dia-
betics. J Urol 1975; 113:636-39
11 Gaskell P. The importance of penile blood pressure in
cases of impotence. Can Med Assoc J 1971; 105:1047-51
12 Malvar T, Baron T, Clark S. Assessment of potency with
the Doppler flow meter. Urology 1973; 2:396-400
13 Classification and diagnosis of diabetes mellitus and
other categories of glucose intolerance-National Dia-
CHEST, 81: 4, APRil, 1982
betes Data Group. Diabetes 1979; 28:1039-57
14 Statistical Analysis System. Cary, NC: Statistical Analysis
Syd~,Inc, 1979;361
15 McSweeny AJ, Heaton RK, Grant I, Cugell D, Solliday
N, Timms R. Chronic obstructive pulmonary disease,
socioemotional adjustment and life quality. Chest 1980;
77 ( suppl ) :309
16 Rubin A, Babbott D. Impotence and diabetes mellitus.
JAMA 1958; 168:498-500
17 Kalliomaki JL, Markkanen TK, Mustonen VA. Sexual
behavior after CVA: a study on patients below the age
of sixty years. Fertil Sterill961; 12: 156-58
18 Hellerstein H, Friedman EH. Sexual activity and pod-
coronary patient. Arch Intern Med 1970; 125:987-99
19 Sadoughi W, Lesher M, Fine H. Sexual adjustment in a
chronically ill and physically disabled population: a pilot
study. Arch Phys Med Rehabil 1971; 52:311-17
20 Dudley DL, Glaser EM, Jorgenson BN, Logan DL.
Psychosocial concomitants to rehabilitation in chronic
obstructive pulmonary disease. Chest 1980; 77 :413-20
21 Pfeiffer E, Verwoerdt A, Davis G. Sexual behavior in
middle life. Am J Psychiatry 1972: 128:82-89
22 Grant I, Heaton RK, McSweeney AI, Adams KM, Timms
RM. Brain dysfunction in COPD. Chest 1980; 77
( suppl ) :308
23 Karacan I, Daria RL. Nocturnal penile tumescence
( NPT): the phenomenon and its role in the diagnosis of
impotence. Sexuality Disability 1978; 1:260-71
24 Fisher C, Schiavi RC, Edwards A, Davis OM, Reitman
M, Fine J. Evaluation of noctwnal penile tumescence in
the differential diagnosis of sexual impotence. Arch
Cen Psychiatry 1979; 36:431-37
SEXUAL DYSFUICnOl II COPD 421