Download Unlikely Association between Clinically Apparent Herpesvirus

Unlikely Association between Clinically Apparent
Herpesvirus Infection and Coronary
Incidence at Older Ages
The Framingham Heart Study
Richard J. Havlik, William C. Blackwelder, Richard Kaslow, and William Castelli
Downloaded from http://atvb.ahajournals.org/ by guest on May 10, 2017
Experimental studies in chickens have shown a relationship of a herpesvirus to
atherosclerosis. The hypothesis of an association in humans was tested by using
data on the history of cold sores and other manifestations of herpes Infections
reported by 658 male and 919 female participants (ages 58 to 89) In the Framingham
Heart Study from 1977 to 1979 and on the prevalence and subsequent 6-year
Incidence of coronary heart disease (CHD). Approximately 40% of the men and 52%
of the women reported a history of ever having "fever blisters or cold sores." Overall,
there was no association between a history of such oropharyngeal manifestations
and prevalent CHD. Only In the subgroup of women with recurrent infections was
there a suggestion of a possible relationship (relative risk = 1.5, 95% confidence
Interval 1.0 to 2.1). Among members of the cohort without CHD at baseline there was
no association between the history of cold sores, chicken pox, shingles, or Infectious
mononucleosls and 6-year CHD Incidence. However, a possible interaction among
women with recurrent herpes, lower levels of serum cholesterol, and Incidence of
angina pectoris without myocardial infarction was suggested In post hoc analyses.
These data from the Framingham cohort do not support the notion that any
self-reported clinically manifest herpesvirus Infection has a strong etlologlcal role In
older persons, but they do raise Issues to be addressed In any further research.
(Arteriosclerosis 9:877-880, November/December 1989)
A
mong the 246 possible causes of atherosclerosis
summarized by Hopkins and Williams,1 viremia,
but not specifically herpesvirus infection, is mentioned as a suggested risk factor for coronary heart
disease (CHD). During the past 15 years there have been
reports of other viruses associated with atherosclerosis,2
and a report of the specific relationship between one type
of herpesvirus in chickens and atherosclerosis appeared
in 1978.3 The herpesvirus of Marek's disease was inoculated in young chickens with or without the addition of an
elevated cholesterol diet; these were compared with control chickens given normal feed or a cholesterol-enriched
diet. Chickens with combined virus inoculation and cholesterol feeding showed the greatest number of atherosclerotic lesions. These findings prompted the editorial
comment, "Surprisingly, there seems to be a great lack of
epidemiological data on infections in relation to CHD and
this might be a profitable area of study."4 More recently
there have been further animaJ experiments 567 and a
relevant epidemiologicai study.8 In the latter study the
prevalence of cytomegalovirus antibodies was higher in
cardiovascular surgery patients than in controls, but the
prevalence of other herpesvirus antibodies was not different in the two groups.
In 1977, investigators at the National Institute of Allergy
and Infectious Diseases began to explore the relationships of HLA types and clinical expressions of certain
herpes infections through a special study in the Framingham population.9 Using specific data collected for that
study on the history of manifest oral/labial herpes simplex
virus type (HSV-1) and other herpes infections on approximately 1600 participants, along with routine data for
ascertainment of CHD, we searched for a possible association between a history of herpes infections and the
prevalence and incidence of CHD.
Methods
Population
The Framingham Heart Study is a longitudinal investigation of risk factors for cardiovascular and other diseases. It has been in operation since 1948 and is based in
Framingham, Massachusetts. Biennial examinations are
performed on the cohort, which originally numbered 5209.10
At the time of the 15th round of examinations, occurring
primarily during the period between 1977 and 1979, a
questionnaire was administered to approximately 1600
participants who were ages 58 to 89 as part of a study of
the potential relationship of herpesvirus infections to HLA
Richard J. Havlik Is at the National Center for Health Statistics,
Hyattsvllle, Maryland. William C. Blackwelder and Richard Kaslow
are at the National Institute of Allergy and Infectious Diseases,
Bethesda, Maryland. William Castelli is at the Framingham Heart
Study, National Heart, Lung, and Blood Institute, Framingham,
Massachusetts.
Address for reprints: Richard J. Havlik, National Center for
Health Statistics, 3700 East-West Highway, Room 2-12, Hyattsvllle, MD 20782.
Received August 5, 1986; revision accepted June 29, 1989.
877
878
ARTERIOSCLEROSIS
VOL 9, No 6, NOVEMBER/DECEMBER 1989
Table 1. Recurrence of Cold Sores and Coronary Heart Disease Prevalence In Framlngham
History of cold sores or recurrence
Males
Coronary heart disease
prevalence at exam 15
None
Yes
No
80
313
Total
% wfth coronary heart disease
Relative risk
(compared to none)
393
20.4
95% confidence interval
—
—
Any
history
47
218
265
17.7
0.9
(0.6,1.2)
Females
(Recurrent)*
(28)
(145)
(173)
(16.2)
(0.8)
(0.5,1.2)
None
49
389
438
11.2
—
—
Any
history
Total
(Recurrent)*
69
412
(50)
(251)
481
14.3
(301)
(16.6)
(1.5)
1.3
(0.9,1.2)
Any
history
None
129
116
630
702
831
746
15.5
15.5
1.0
—
(1.0,2.1)
—
(0.8,1.3)
(Recurrent)*
(78)
(396)
(474)
(16.5)
(1.1)
(0.8,1.4)
*The "Recurrent" group Is a £lubset of the "Any history" group.
Table 2. Recurrence of Cold Sores and Coronary Heart Disease Prevalence in Framlngham
History of cold sores or recurrence
Males
Downloaded from http://atvb.ahajournals.org/ by guest on May 10, 2017
Coronary heart disease
Incidence at exam 15
Yes
No
Total
% with coronary heart disease
Relative risk
(compared to "None")
95% confidence interval
Females
(Recurrent)*
None
(17)
(234)
(251)
(6.8)
(0.9)
64
638
702
9.1
—
28
384
412
6.8
0.9
—
(0.6,1.5)
(0.5,1.6)
—
None
(Recurrent)*
None
35
278
313
11.2
32
186
218
14.7
(19)
(126)
(145)
(13.1)
29
360
389
7.5
1.3
(1.2)
(0.7,2.0)
—
—
(0.8,2.1)
Total
Any
history
Any
history
—
Any
history
60
570
630
9.5
1.0
(0.7,1.5)
(Recurrent)*
(36)
(360)
(396)
(9.1)
(1.0)
(0.7,1.5)
•The "Recurrent" group Is a subset of the "Any history" group.
types.9 At the time of the examination, blood pressure,
cholesterol level, and smoking history were ascertained.
Trters of antibodies to HSV-1, HSV-2, varicella-zoster
virus, and cytomegalovirus were determined for 402 persons. At each examination, the individual participants
have been routinely questioned and examined for the
presence of CHD (angina pectoris, coronary insufficiency,
or myocardia) infarction). Cases of the disease have been
reviewed by a panel of Framingham personnel for consistency with a series of criteria.10 This group of participants was followed for the onset of CHD, either fatal or
nonfatal, during the subsequent 6 yeans.
Questionnaire
The participants were asked questions on their history
of herpes labialis (fever blisters or cold sores), whether the
lesions were recurrent, and how often they recurred. Questions on a history of zoster (shingles), varicella (chicken
pox), and infectious mononucleosis were also asked.9
Statistics
We compared CHD prevalence and incidence rates for
those with and those without a history of herpesvirus
infection including the subset with recurrent manifestations by using relative risks (RRs) and 95% confidence
intervals.11 To adjust simultaneously for the possible
influences of age, blood pressure, cholesterol level, and
smoking, we used the logistic regression model and
estimated coefficients by the method of Walker and
Duncan.12 In addition, we did further analyses which
stratified on the type of CHD endpoint and the serum
cholesterol levels.
Results
Of the total of 1657 persons ages 58 to 89, 1024 were
58 to 69 years old, 450 were 70 to 79 years old, and
183 were 80 to 89 years old. The proportion of women
increased in the three age groups: 55%, 61%, and 66%,
respectively.
A total of 265 of 658 men (40%) and 481 of 919 women
(52%) reported a history of fever blisters or cold sores.
Among the 746 persons reporting such a history, 116 (15.5%)
had CHD at the time of Examination 15. The same
proportion (15.5%) of the 831 participants with no history
of blisters or cold sores had CHD (Table 1).
Among the 301 women reporting recurrent cold sores,
50 (16.6%) had CHD at the time of Examination 15,
compared to 49 of 438 (11.2%) with no history of cold
sores (relative risk=1.5, 95% confidence interval 1.0 to
2.1). Information on recurrences was unavailable for three
men and three women. Differences in age and other CHD
risk factors (cholesterol, blood pressure, and smoking) did
not explain the relationship. In contrast, the proportion of
men with prevalent CHD was somewhat lower in those
reporting recurrent cold sores than in those with no
history. When the analysis was limited to those with the
onset of nonfatai disease at or before 50 years of age,
CHD prevalence results were similar for those with and
those without a cold sore history (data not shown).
UNLIKELY HERPESVIRUS-CORONARY ASSOCIATION
There was no association between a history of fever
blisters or cold sores, whether recurrent or not, and 6-year
CHD incidence in either women or men (Table 2). The
CHD rate in those with a history of cold sores was 9.5%
compared to 9.1% for those with no such history. Stratification by recurrence in the last year or by age 70 and over
did not materially alter the results. However, stratification
by serum cholesterol level (less than or equal to 240 mg/dl)
suggested a possible interaction. The relative risk for CHD
incidence in women with the lower cholesterol level and
recurrent cold sores was 2.3 (95% confidence interval 0.9
to 5.9). When the endpoint, angina pectoris without myocardial infarction, was selected from the total CHD endpoints, the relative risk was 5.2 (95% confidence interval
1.1 to 24.7).
Downloaded from http://atvb.ahajournals.org/ by guest on May 10, 2017
We found no association of other clinically apparent
herpesvirus infections (shingles, chicken pox, or infectious
mononucleosis) with CHD prevalence or incidence (data
not shown). Only 167 of 1603 persons reported having had
shingles, and 13 of 1612 persons reported having had
mononucleosis. The number who reported having had
chicken pox was substantial (662 of 1135), but there were
many unknowns. Incidence rates for both men and women
combined for those with and without a baseline history of
shingles were 8.9% compared to 9.2% (RR=1.0, 95%
confidence interval 0.5 to 1.7); for a baseline history of
chicken pox, 8.6% and 9.6%, respectively (RR=0.9, 95%
confidence interval 0.6 to 1.3); and for a baseline history of
mononucleosis, 8.3% versus 9.2%, respectively (RR=0.9,
95% confidence interval 0.1 to 5.9).
Discussion
An attempt was made to demonstrate an association
between self-reported clinical manifestations of HSV-1
infection and CHD in a human population-based study.
Although prevalence data at baseline showed a crosssectional relationship in a subgroup of women, no consistent prospective relationship was found for either gender
in this group of older persons in Framingham. The older
age structure of the population precluded considering
issues related to the initiation of disease in a younger
population.
Various analytic approaches were used to discover
relationships that may have been less obvious. At baseline, questions were asked about recurrences and about
the frequency of such recurrences. Since the pathogenesis of arterial lesions in chickens depended on active viral
infection, the human subgroup with recurrent activity
might have been expected to be at greatest risk. Other
than the modest association between CHD prevalence
and reported recurrent disease in women, there was no
indication of an association between risk of CHD and
recurrent HSV-1 infection. However, stratification by risk
of cholesterol level and use of various CHD endpoints did
sometimes give "significant" results. However, such subgroup analyses done in a post hoc manner could result in
spurious findings.13
The presence of an apparent relationship between
recurrent HSV-1 infection in women only and the prevalence, but not incidence, of CHD is not easily explained.
Havlik et al.
879
There is no a priori reason why any association should be
limited to either men or women, although women may
have a higher frequency of HSV-1 infection. However, if a
different physiologic or emotional response to stress in
women generally leads to a different clinical expression of
HSV-1 infection, then recurrent exacerbations of herpesvirus infection in women may have been a consequence,
rather than an etiological factor, in the development of
CHD. In Framingham, most of the CHD, even in older
women, took the form of angina pectoris.10 Perhaps the
reaction to the recurrent chest pain accounted for recurrent expression of HSV-1. More extensive information on
the timing of recurrence and CHD status would be necessary to test such an explanation.
Only clinical measures and endpoints were used in this
study. We may have missed an association with clinically
inapparent, but biologically important, infection. It is worth
noting that data from this population have previously
demonstrated a clear association between a positive
history of cold sores and higher antibody titers to HSV-1 9 ;
approximately 90% (190 of 211) of participants with a
positive history also had detectable titers to HSV-1. Furthermore, there was a dose-response relationship, i.e.,
those reporting more frequent recurrence of cold sores
had higher geometric mean titers.9 Nevertheless, a relationship might be more readily recognized by distinguishing serologically or virologically between infected and
uninfected persons and by documenting CHD with pathologic and angiographic data. For example, in one such
study, herpesvirus probes were used on specimens of
aortic wall taken from patients undergoing coronary bypass
surgery.7 The herpes simplex viral mRNA was found in
some of the specimens. There was no evidence for the
presence of other herpes-like viruses from hybridization
studies with either cytomegaJovirus or the Epstein-Barr
viral probes. However, in another study, cytomegalovirus
antigen was evident within arterial smooth muscle cells.14
It has been speculated that viruses can reach the intimal
smooth muscle cells and may initiate or enhance cell
proliferation. This process might produce the nidus for
increased concentrations of cholesterol and cholesterol
esters. However, in a secondary analysis of data from our
study, we actually found a stronger relationship in those
with serum cholesterol levels equal to or less than 240 mg/dl.
It is possible that an effect of herpesvirus infection might
be more evident in those without other risk factors.
Studies of individuals at younger ages with less atherosclerosis might also be more revealing. Again, in the
Framingham Study, however, when the analysis was
limited to those with the onset of disease at an early age,
no relationship was found.
Our investigation in a population-based study provided little support for a relationship between CHD and
previous occurrence of clinically active herpesvirus infection. The negative overall results of both cross-sectional
and longitudinal analyses based on clinical history are
consistent with the antibody prevalence study in cardiovascular surgery patients.8 However, our subgroup analyses did identify women with prevalent CHD, lower
cholesterol levels, and incident angina as having more
frequently recurring herpes infections. Since both her-
880
ARTERIOSCLEROSIS
VOL 9, No 6, NOVEMBER/DECEMBER 1989
pesvirus infections and atherosclerosis are ubiquitous in
this population, the importance of any possible relationships should not be minimized. The results of this study
should not dissuade other investigators from performing
more specific etiologic studies.
Acknowledgments
The authors thank Diane Yerg and Maria Deloria for programming assistance.
References
Downloaded from http://atvb.ahajournals.org/ by guest on May 10, 2017
Hopkins PN, Williams RR. A survey of 246 suggested
coronary risk factors. Atherosclerosis 1981 ;40:1-52
Burch GE, Giles TD. The role of viruses in the production of
heart disease. Am J Cardlol 1972;20:231-240
3. Fabrlcant CG, Fabricant J, Lttrenta MM, Mlnlck CR.
Virus-induced atherosclerosis. J Exp Med 1978;148:335-340
4. Virus Infection and atherosclerosis. Lancet 1978;2:312-322
5. Fabrlcant CG, Fabrlcant J, Mlnlck CR, Lttrenta MM.
Herpesvirus-lnduced atherosclerosis in chickens. Fed Proc
1983;42:2476-2479
Index Terms:
6. Mohr FC, Branson RT, Hurt RD. Failure of herpesvirus
salmlri to enhance atherogenesls in owl monkeys. Atherosclerosis 1983;46:173-179
7. Bendltt EP, Barrett T, McDougall JK. viruses in the
etiology of atherosclerosis. Proc Nat) Acad Scl USA 1983;
80:6386-6389
8. Adam E, Probtufleld TJ, Burek J, et al. High levels of
cytomegalovirus antibody In patients requiring vascular surgery for atherosclerosis. Lancet 1987;2:291-293
9. Blackwelder WC, Dolln R, Mtttal KK, McNamara PM,
Payne FJ. A population study of herpesvirus infections and
HLA antigens. Am J Epidemiol 1982; 115:569-576
10. Kannel WB. Some lessons in cardiovascular epidemiology
from Framingham. Am J Cardiol 1976:37:269-282
11. Klelnbaum DG, Kupper LL, Morgenstem H. Epidemlologic
research. Belmont, CA: Lifetime Learning Publications,
1982:283-311
12. Walker SH, Duncan DB. Estimation of the probability of an
event as a function of several independent variables. Blcmetrika 1967;54:167-179
13. Felnsteln AR. Scientific standards in epidemiologic studies
of the menace of daily life. Science 1988;242:1257-1263
14. Melnlck JL, Dreesman GR, McCollum CH, Petrle BL,
Burek J, DeBakey ME. Cytomegalovirus antigen within
human arterial smooth muscle cells. Lancet 1983;2:644-647
coronary heart disease • herpesviruses • HSV-1 • Framingham Heart Study
Downloaded from http://atvb.ahajournals.org/ by guest on May 10, 2017
Unlikely association between clinically apparent herpesvirus infection and coronary
incidence at older ages. The Framingham Heart Study.
R J Havlik, W C Blackwelder, R Kaslow and W Castelli
Arterioscler Thromb Vasc Biol. 1989;9:877-880
doi: 10.1161/01.ATV.9.6.877
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1989 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/9/6/877
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/