Blood Pressure and the Cystic Fibrosis Gene
Evidence for Lower Pressure Rises With Age in Female Carriers
Maurice Super, Ayesha Irtiza-Ali, Stephen A. Roberts, Martin Schwarz, Michele Young, Alison Smith,
Theresa Roberts, Joanna Hinks, Anthony Heagerty
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
Abstract—Individuals homozygous for the autosomal recessive disorder CF are known to have low blood pressure, thought
to be caused by greatly increased sweat salt loss. We examined whether carriers of the CF gene also have low blood
pressure. Our pilot studies had suggested an effect limited to females, leading to 2 further studies in white females. In
the first, blood pressure was measured in 232 known CF mutation carriers and compared with 246 mutation-negative
control subjects. The carriers showed a significantly lower rate of increase in systolic blood pressure with age than the
controls, especially after age 50 (3.5% per decade compared with 5.4% per decade, P⫽0.010). In a small substudy,
sweat sodium and chloride levels were highest in those CF carriers with the lowest blood pressures. In the second study,
CF carrier status was investigated in 563 normotensive females and in 607 women with essential hypertension diagnosed
to test whether a lower incidence of carriers in the hypertensives suggested a protective effect. Twenty-five of the
normotensives (4.4%) were carriers compared with 21 (3.5%) of the hypertensive group (P⫽0.45). Older CF carrier
females had lower systolic and diastolic pressures than matched control subjects, with a tendency for blood pressure to
increase less with age. This could result in significant reduction in stroke and heart disease. The effect on blood pressure
is insufficient to prevent hypertension, though it remains conceivable that the severity might be ameliorated in carriers.
(Hypertension. 2004;44:878-883.)
Key Words: blood pressure 䡲 women 䡲 DNA 䡲 mutation
I
ndividuals homozygous for the autosomal recessive disorder CF (CF) are known to have low blood pressure,
explained by greatly increased sweat salt losses.1 It was
suggested that healthy carriers of the CF gene might also have
low blood pressure,1 but this was not formally investigated.
Individuals with CF have sweat sodium and chloride levels
between 2 and 5 times normal. In the days before CF carrier
detection was possible, it was shown that obligate carriers,
the parents of affected children, had, on average, levels of
sweat sodium and chloride intermediate between those with
the disease and presumed normal controls;2 similarly, it was
shown more recently that newborn carriers of major mutations of the CF gene have slightly increased sweat electrolytes when compared with negative newborn controls.3 Thus,
one might postulate that such life-long increases in sweat
sodium and chloride could reduce blood pressure.
We conducted a pilot study of 21 male CF carriers versus
26 negative male controls and found no differences between
systolic and diastolic pressures in the 2 groups. However, 25
female carriers and 38 negative controls studied at the same
time showed markedly lower systolic and diastolic pressures
in the carriers.4 Our pilot studies also suggested reduced
numbers of carriers in hypertensive females, whereas there
were no differences in hypertensive males. This prompted us
to limit our present studies to females. We investigated white
British adult women in 2 concurrent cross-sectional studies to
establish whether there was a true relationship between
carrier status of CF and blood pressure. In study 1, blood
pressures in women known to carry a major CF mutation
were compared with matched female controls known to be
negative on mutation testing. At the end of this study, sweat
tests were performed in a sample of those carriers and
negative controls with the lowest and highest blood pressures.
In study 2, the frequency of major CF mutations was
compared between white British normotensive and hypertensive women.
Methods
Both studies were confined to nonpregnant white British adult
females from North West England. The CF mutation status of this
population has been tested extensively and a high proportion of the
mutations present are known,5 whereas the mutation status of other
ethnic groups, excluded from the study, is less well-established.
Field workers (M.Y. and A.S., 2 experienced nurses) were trained to
measure seated blood pressure according to the guidelines of the
Received July 26, 2004; first decision August 11, 2004; revision accepted September 14, 2004.
From the Department of Clinical Genetics (M.S., M.Schwarz, M.Y., A.S., T.R., J.H.), Royal Manchester Children’s Hospital Manchester, United
Kingdom; Manchester Royal Infirmary (A.I.-A.), Manchester, UK; Cardiovascular Research Group (A.H.), Department of Medicine, Manchester Royal
Infirmary, Manchester, UK; Biostatistics Group (S.A.R.), School of Epidemiology and Health Sciences, Manchester University, Manchester, UK.
Correspondence to Maurice Super, MD FRCP, FRCPCH, Consultant Paediatric Geneticist, Department of Clinical Genetics, Royal Manchester
Children’s Hospital, Manchester, M27 4HA United Kingdom. E-mail [email protected]
© 2004 American Heart Association, Inc.
Hypertension is available at http://www.hypertensionaha.org
DOI: 10.1161/01.HYP.0000145901.81989.46
878
Super et al
TABLE 1.
Blood Pressure in Female CF Carriers
879
Demographic Characteristics of the 2 Groups
Demography*
No. of participants
Age (median, IQR)
Carriers
232*
Controls
246
39 (31–53)
39 (29–50)
Height (median, IQR) cm
161 (158–165)
161 (157–166)
Weight (median, IQR) kg
66 (59–77)
BMI (median, IQR) kg/m2
Skinfold thickness (median, IQR) mm
Hormonal contraceptive use, %
Menstrual cycle (early, mid, late, NA†), %
25.8 (22.7–30.4)
31 (24–37)
66 (58–78)
25.1 (22.6–29.0)
30 (25–36)
22 (51/229)
26 (63/246)
16, 30, 12, 43
20, 29, 13, 39
*Height, weight, body mass index, and hormonal contraceptive use and cycle data were obtained
for 229 carriers (and all controls); skinfold measurements were obtained for 204 carriers.
†Not available because of irregular cycle, postmenopausal, or not recorded.
IQR indicates interquartile range.
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
British Hypertension Society.6 The average of 2 readings was
recorded after 5 minutes of rest, using a standard mercury sphygmomanometer. In no case were the 2 readings ⬎3 mm Hg different
in either systolic or diastolic pressure. In both studies the existence
of any medical disorders and the taking of drugs that might affect
blood pressure in the study subjects were recorded; also, a family
history of CF in any known relative and one of hypertension, angina,
or ischemic heart disease in any first-degree relative. All subjects
gave full, informed, signed consent.
Study 1 was designed as a cross-sectional study to investigate
levels of blood pressure and CF gene mutation carrier status.
Hypertension was suspected if the average pressure was ⬎140/
90 mm Hg, and if such readings were obtained subjects were
informed and, in accordance with the patient information leaflet,
were advised to visit their family practitioner for monitoring.
Systolic and diastolic blood pressures were measured in known CF
carriers and in a group of females matched for age and known to be
negative on carrier testing for the 12 commonest mutations in the
North West of England. Both field workers were blind to the CF
carrier status of the women. The major cohort of subjects came from
the CF Cascade program of the Royal Manchester Children’s
Hospital,7 thus a proportion of the women were members of the same
nuclear or extended families. Subjects were examined in their own
homes. The stage of the menstrual cycle and use of oral or injected
contraceptives were recorded. Height, weight, and skin-fold thickness were measured.
At the end of the study, we performed sweat tests on a subset of
the subjects taking a sample of those with the lowest blood pressures
and highest blood pressures. Sweat sodium and chloride levels were
measured by the standard Gibson and Cooke method by staff
experienced in sweat testing and blind to the carrier status of the
subjects.
Study 2 was designed as a cross-sectional study of CF gene
mutation carrier status in women with essential hypertension and in
normotensive controls. Patients were recruited from local hospital
hypertension clinics and a primary care clinic and a group of
normotensive control women with a similar age range was recruited
from other hospital environments. Again, readings in excess of
140/90 mm Hg were communicated to prospective controls with the
advice to visit the family practitioner for follow-up measurements.
Such women were excluded from further study. Mouth wash
specimens for DNA extraction were collected from both cohorts.
The DNA specimens were tested for 12 of the most common
mutations in the UK by means of Elucigene kits (Orchid Biosciences), which cover ⬇87% of UK CF mutations, ⌬F508, G551D,
G542X, 621⫹1 g⬎t, R560T, W1282X, 1717⫺1 g⬎a, 3849⫹10Kb
c⬎t, R553X, R117H, R1162X, and R334W. Given the benign nature
of the 7t polymorphism, a finding of R117H was followed by a reflex
test of the intron 8 exon acceptor site and, for the purposes of this
study, only those with the 5t polymorphism were regarded as having
a mutation.8
The results of carrier testing were divulged to women who wished
to know them and, with their permission, to their medical care givers.
The letter giving the result offered genetic counseling. All studies
received ethical approval from the Central Manchester, Salford, and
South Manchester Local Research Ethics Committees.
Data Analysis
The primary analysis for study 1 was prespecified as an analysis of
covariance allowing for age. Because the blood pressure data showed
a skewed distribution, log-transformed blood pressures were used in
the analysis. Inspection of the data showed some curvature in the age
effect and therefore a quadratic term was included in the analysis.
Likelihood ratio tests were used to compare models including group
and group by age interactions with models excluding the group
effects. To give more interpretable parameters, models excluding the
quadratic term were used, giving an overall age-averaged slope for
comparison between groups. For presentation, this log-scale slope is
expressed as a percentage change in blood pressure per decade.
Because there was evidence of influential outliers, a robust regression was used to confirm the significance of the main findings.9
Additionally, the data were divided arbitrarily into 3 age groups,
based on the tertiles of the age distribution, and blood pressure
compared between groups using Mann–Whitney U tests and their
associated 95% confidence intervals. The choice of 3 equal-sized
groups was prespecified, but the actual tertiles were derived from the
data.
The proportions of mutation carriers in the 2 groups of study 2
were compared using Fisher exact tests and by computing odds ratios
and their exact confidence intervals. Logistic regression was used to
produce age-adjusted odds ratios.
Results
Study 1
The 232 CF mutation carriers and 246 mutation-negative
controls were well-matched for age and anthropometric and
menstrual cycle stage characteristics (Tables 1 and 2). Both
systolic and diastolic pressures showed the expected increase
with age in both groups of women (Figure). However, the CF
mutation carriers showed a smaller rise with a more gradual
slope, especially after 50 years. A quadratic regression model
of log-transformed blood pressure shows that the 2 groups are
significantly different for systolic pressure (P⫽0.017) but not
diastolic (P⫽0.22).
A simpler linear model gives age-averaged rates of change
in blood pressure (Table 3). The data show a decreased slope
for carriers compared with controls (3.5%/decade compared
880
TABLE 2.
Hypertension
December 2004
Age Distribution of the 2 Groups in Study 1
Age (y)
Controls
Carriers
Total
⬍20
18
14
32
20–29
48
37
85
30–39
62
68
130
40–49
56
37
93
50–59
26
44
70
60–69
25
22
47
ⱖ70
11
10
21
Total
246
232
478
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
with 5.4%/decade), again statistically significant for systolic
(P⫽0.010) but not diastolic pressure (P⫽0.28) (Table 3).
The data show a few potential outliers (Figure), particularly one of the older carriers, which shows the lowest blood
pressure in the study and has symptomatic hypotension. A
sweat test in this subject showed sodium 69 mmol/L and
chloride 50 mmol/L, which are high levels and a possible
cause of her low blood pressure. This subject was included in
the sweat test study. There is no objective reason to exclude
any participants. Therefore, we undertook a robust regression
analysis that downweights the outliers in a systematic way,
and in this analysis the difference in systolic blood pressure–
age relationship remained significant (P⫽0.037).
The increasing effect of CF carrier status with age is
readily seen if the data are divided into 3 age bands using the
tertiles of the age distribution (Table 4). Here both systolic
and diastolic blood pressure are significantly lower in CF
carriers than controls in the older age group, although the
diastolic difference would not be considered significant if an
adjustment for multiple testing were used.
TABLE 3. Slopes of Blood Pressure With Age as Obtained
From Fitting a Linear Analysis of Covariance Model to the
Log-Transformed Blood Pressure
Systolic
Diastolic
Controls
5.4 (4.3–6.4)
4.9 (3.6–6.1)
Carriers
3.5 (2.4–4.5)
3.9 (2.6–5.2)
Difference
1.9 (0.4–3.2)
1.0 (⫺0.8–2.7)
0.010
0.28
Significance of difference
Slopes are expressed as percentage change per decade with their associated 95% CI.
Within the ANCOVA framework, there is no evidence of a
differential effect on blood pressure between the most common CF mutation ⌬F508 and the other mutations, although
by far the biggest number carried ⌬F508.
We have recomputed the effect sizes adjusting for each of
the potential confounders in addition to age: body mass index,
menstrual cycle stage, hormonal contraceptive usage, and use
of medication, all of which can affect blood pressure (data not
shown). These adjustments make no substantive difference to
the estimated effect sizes nor do their confidence intervals,
and the difference in the slope with age remains statistically
significant.
The median sweat sodium in 9 carriers with especially low
blood pressure was 49 mmol/L (range, 19 to 107) compared
with a median of 30 mmol/L (range, 22 to 90) in 7 hypertensive carriers with the highest blood pressures and compared
with a median 28 mmol/L (range, 16 to 33) in 7 negative
controls, of whom 4 had especially low blood pressure, 1 had
normotension, and 2 had high blood pressure. Sweat chloride
levels showed very similar trends. Three carriers with especially low blood pressures had very high sweat sodium and
chloride levels, as did 1 hypertensive carrier (see discussion).
None of the subjects had any symptoms of CF, and in all
cases sweat sodium exceeded sweat chloride, whereas in the
disease the converse occurs. Given the small number of
recruits, these results are not significant in themselves.
Study 2
We recruited 563 normotensive females and 607 women with
essential hypertension. The median age range was 60, with a
range of 17 to 89 (interquartile range, 50 to 69). Twenty-five
(4.4%) of the normotensive group were found to be carriers of
a CF mutation, compared with 21 (3.5%) of the hypertensive
group This corresponds to an odds ratio of 0.77 (95%
confidence interval [CI], 0.41 to 1.45), which was not
statistically significant (P⫽0.45). Twenty-two (3.9%) of the
controls were positive for the ⌬F508 mutation compared with
17 (2.8%) of the hypertensives (odds ratio, 0.71; 95% CI,
0.35 to 1.41; P⫽0.33). There was a small age difference
between the groups, but adjusting for age using logistic
regression made little difference to the effect of carrier status
(odds ratio, 0.71; 95% CI, 0.39 to 1.31). CFTR mutations
found in studies 1 and 2 are contained Tables V and VI (see
online supplement, available at http://hyper.ahajournals.org).
Systolic and diastolic blood pressure as a function of age for CF
mutation carriers (closed symbols and solid lines) and controls
(open symbols and dashed lines). The lines represent the fit
from a quadratic regression model of log-transformed blood
pressure.
Discussion
This is the first full study to our knowledge to test the
hypothesis of a link between the CF carrier state and reduced
Super et al
TABLE 4. Blood Pressure for Age Using 3 Equal-Size Groups
Based on the Tertiles of the Age Distribution (33.7 and 48.5 Years)
Age Tertile
Controls
Median (IQR)
CF Carriers
Median (IQR)
Difference
(95% CI)
114 (105–120)
113 (105–121)
0 (⫺3 to 4)
0.87
P
Systolic
⬍33.7
33.7–48.5
118 (106–128)
116 (108–131)
⫺1 (⫺5 to 4)
0.62
⬎48.5
133 (123–147)
129 (114–138)
7 (2 to 13)
0.009
65 (58–74)
63 (57–72)
2 (⫺1 to 4)
0.32
Diastolic
⬍33.7
33.7–48.5
69 (62–80)
68 (62–78)
0 (⫺3 to 3)
0.91
⬎48.5
79 (69–85)
73 (67–80)
4 (0 to 8)
0.037
Median and IQRs are shown along with the CI and significance level of the
difference using a Mann–Whitney U test.
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
blood pressure. Blood pressures in our study group agree
closely with those of the 1946 British birth cohort10 for
female blood pressure. Blood pressures in those women
negative on CF mutation testing also compared well with the
results from the 1996 Health Survey for England.11
Our findings of lower systolic blood pressures and a slower
rise in pressure with age in white female CF mutation carriers
are of potential public health significance. Diastolic pressure
showed a similar trend, but the association with carrier status
was statistically weaker. Much has been written on the causes
of the rise in blood pressure in postmenopausal females. Our
data suggest a role for the CF gene in modulating this
age-associated rise.
The mechanism for the blood pressure differences is likely
to reside in changes in sodium loss, with the sweat gland
more likely than the kidney to account for this. Increased
sweat salt losses in carriers would contribute to lower blood
pressure. This effect could be confined to females if it is more
than offset by factors such as the increased blood volume and
red cell count,12 androgens causing stimulation of renal
sodium reabsorption,13 and increased muscle bulk,13 all of
which result in higher blood pressure of males. Differences in
aldosterone responsiveness in postmenopausal females may
explain the gender difference and why our findings are
confined to older women. However, we do intend to study a
male cohort.
The kidney is the main organ implicated in genetic abnormalities of sodium associated with low blood pressure.14
Thus, in Gitelman syndrome, mutations in the Na⫹Cl⫺ transporter result in reduced blood pressure by diminishing renal
sodium reabsorption. Different mutations in the ␣, ␤, and ␥
subunits of ENaC cause pseudohypoaldosteronism with reduced renal and sweat gland sodium reabsorption and lower
blood pressures. It is interesting to note that Lifton14 postulates that heterozygotes for these rare disorders might also
have low blood pressure. Although the CF transmembrane
conductance regulator (CFTR) is expressed in the kidney, and
proximal convoluted tubule urine is sodium-rich, with the
adrenal–renal axis intact, reabsorption of this sodium occurs
and no net sodium and chloride renal losses occur in CF
patients.15
Blood Pressure in Female CF Carriers
881
The same does not apply to the sweat gland in which
chloride loss is the basic defect caused by mutations of the CF
gene CFTR16 Further, there is a close interaction between
CFTR and ENaC in the sweat gland. Quinton and Reddy17
have shown clearly that ENaC function in the sweat gland
depends critically on intact CFTR in reabsorbing sodium and
that failure to absorb NaCl in CF is not only a result of loss
of intact CFTR but also a result of secondary inability to
activate ENaC. Under resting conditions, serum sodium and
chloride levels in CF patients are the same as in healthy
controls, although they decrease on salt depletion.18 There
have been no such studies on CF carriers. Legris et al18 found
reduced mean blood pressure in CF patients, which decreased
significantly during a salt depletion phase of their experiments, despite markedly increased plasma renin, aldosterone,
and angiotensin. They ascribe their findings to sweat salt loss
with vigorous renin, aldosterone, and angiotensin responses
helping to maintain blood pressure. The lower blood pressures and further drop on salt depletion show that the kidney
is not able to compensate fully for the excessive sweat
sodium and chloride losses. Clearly, effects would be less
marked in healthy carriers than in those affected by CF, but
there is likely to be slight chronic hyperstimulation of the
renin-angiotensin-aldosterone axis in them, too. It is interesting that differences in blood pressure between our carriers
and negative controls only manifested in middle age. Although ours was a cross-sectional and not a longitudinal
study, it is tempting to speculate whether, with chronic
hyperstimulation of the rennin-angiotensin-aldosterone axis,
relative failure of these mechanisms in older CF carriers
occurs, leading to lowering of blood pressure with continuing
sweat hyperelectrolytemia. Postmenopausal females have
been shown to have low basal aldosterone levels and a
reduced increment in aldosterone in response to angiotensin
II on a low-salt diet.19 This is in contrast to males of any age
and premenopausal women. Thus, men and younger women,
including CF carriers, may maintain blood pressure despite
sweat salt losses Further studies in elderly male and female
carriers are indicated.
An inverse relationship between sweat sodium and blood
pressure was found by Quintero-Acenzio20 both in normotensives and hypertensives, although Levi in a separate study21
could not confirm this. It is interesting to note that 47 of the
53 who Quintero-Azenzio studied were female, whereas only
12 of 31 were female in Levi’s study. Our small sweat test
study is consistent with the suggestion that the sweat sodium
and chloride losses account for lower blood pressure in CF
carriers. Although we studied rather few subjects, there were
remarkably high sweat sodium and chloride levels in 3
women with especially low blood pressures (all
⬍100 mm Hg systolic and 50 mm Hg diastolic) and sodium
(107, 85, and 69 mmol/L) and chloride (93, 63 and
50 mmol/L) levels, respectively. One hypertensive carrier had
sweat sodium of 90 and sweat chloride of 74 mmol/L. None
of the subjects had any symptoms to suggest CF. None of the
controls with low or high blood pressure had raised sweat
electrolytes.
882
Hypertension
December 2004
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
It should be noted that CFTR forms part of a macromolecular complex22 and interacts with a number of other
pathways that regulate blood pressure.23,24
Lower systolic and diastolic pressure in older women
carrying a CF gene is likely to have significant protective
effect against the development of stroke and of heart disease.
Lowering blood pressure is associated with reduced incidence
of both conditions.25 In this context, our finding of a reduction of 7 mm Hg in systolic and 4 mm Hg in diastolic
pressure may be viewed as highly significant and could be
associated with a 30% reduction in stroke and 20% reduction
in myocardial infarction. Because in many Western populations as many as 1 in 25 carry a CF gene, this is an
observation of considerable significance. Indirectly, our study
supports the findings of the Inersalt Trial in which rise in
blood pressure with age was correlated with sodium intake.26
Macek27 found a significant excess of the B haplotype, the
background on which most important CF mutations occur in
191 elderly (older than age 75 years; mean age, 83 years)
Czechoslovakian women, when compared with younger
women and elderly or young men. There was a steady rise in
this haplotype from female infancy to old age. Macek
suggests that the selection process for this haplotype might
have been subtle and extending from prereproductive to
postreproductive years in these women but gives no explanation as to a possible cause of his finding. A lower incidence
of stroke and heart disease might be the explanation. The B
haplotype has been more closely associated with actual levels
of sweat sodium and chloride than specific CFTR mutations,
including ⌬F508 in CF patients,28 and it is possible that there
are genes just outside the CFTR gene but in strong linkage
equilibrium with it, which are as important in regulating
sweat electrolytes as is CFTR.
Our findings suggest that the cardiovascular benefit gained
from lower blood pressure in female CF carriers would confer
a survival advantage. This could increase public interest in
being carrier-tested.
Although study 2 did not show a statistically significant
reduction in the number of CF carriers in the hypertensive
versus normotensive groups, the estimate of the odds ratio of
0.71 is in the expected direction and the 95% CIs are wide
(0.35 to 1.41) and consistent with a strong protective effect
against hypertension. With relatively low carrier rates of 1 in
25, a much larger study (⬇10 000 participants) would be
needed to confirm the modest effect found. Although other
factors resulting in clinical hypertension are likely to be more
powerful than any protection afforded by the CFTR gene, it
remains conceivable that the level of blood pressure observed
in carriers is not as severe as in noncarriers, perhaps with
hypertension easier to control. To demonstrate this would
require a longitudinal study of blood pressure levels and drug
dosages. If it were shown to be true, then new avenues of
hypertension treatment based on modulation of CFTR effects
could emerge.29
Perspectives
We have demonstrated that older CF carriers have lower
blood pressures than noncarrier women and a substantially
attenuated rise in pressure with age. This could confer
considerable protection against circulatory diseases and be of
considerable public health significance. It could explain the
increased life expectancy reported in other European cohorts.
A cohort of males needs to be studied to ensure whether the
phenomenon is truly confined to females
Acknowledgments
We thank Orchid Biosciences (originally Cellmark Diagnostics) for
providing us with the CF 12 kits. We are also grateful to consultants
at Hope Hospital, Manchester Royal Infirmary, and Macclesfield
General Practice for allowing us access to their patients. We thank
Dr Paul Quinton for helpful discussions. We thank Dr Mike Addison
for arranging the sweat tests. Professor Nicola Cherry advised on the
original planning of the study. We thank the British Heart Foundation for supporting this project.
References
1. Lieberman J, Rodbard. Low blood pressure in young adults with cystic
fibrosis. Ann Int Med 1975;82:806 – 808.
2. Lieberman J, Kellog F. Evaluation of the sweat chloride assay in adults:
use of pilocarpine iontophoresis. Am J Med Sci. 1963;246:537–549.
3. Farrell PM, Koscik RE. Sweat chloride concentrations in infants
homozygous and heterozygous for F508 cystic fibrosis. Pediatrics. 1996;
97:524 –528.
4. Super M, Heagerty AM, Irtiza -Ali A, Schwarz MJ, Roberts T, Corbett.
Blood pressure and cystic fibrosis carriers Pediatr Pulmonol 1998;suppl
19:208.
5. Super M Schwarz MJ. Mutations of the cystic fibrosis gene in the
population of north-west England. Eur J Pediatr. 1992;151:108 –111.
6. Dillon M, Littler WA, Mee F, O’Brien ET, Petrie JC, de Swiet M. Blood
Pressure Measurement: Recommendations of the British Hypertension
Society. London: BMJ Publishing Group; 1997.
7. Super M, Schwarz MJ, Malone G, Roberts T, Haworth A, Dermody G
Active cascade testing for carriers of cystic fibrosis gene BMJ. 1994;308:
1462–1467.
8. Kiesewetter S, Macek M, Jr Davis C, Curristin SM, Chu C-S, Graham C,
Shrimpton AE, Cashman SM, Tsui L-C, Mickle J, Amos J, Highsmith
WE, Shuber A, Witt DR, Crystal RG, Cutting GR. A mutation in CFTR
produces different phenotypes depending on chromosomal background.
Nat Genet. 1993;5:274 –278.
9. Hamilton LC. How robust is robust regression? Stata Tech Bull 1999;2:
21–26. Reprinted in Stata Tech Bull Reprints 1991;1:169 –175.
10. Hardy R, Kuh D, Lanenberg C, Wadsworth MEJ. Birthweight, childhood
social class and change in adult blood pressure in the 1946 British birth
cohort. Lancet. 2003;362:1178 –1183.
11. Dong W, Primatesta P, Bost L Health Survey for England and Wales
1996. Accessed at http://www.archive.official-documents.co.uk/
document/doh/survey96/tab6 –19.htm (Figures 6C, 6G, 6 – 4, 6 – 8).
12. Haslett, Chilvers ER, Boon NA, Colledge NR. Davidson’s Principles and
practice of medicine, 19th ed. Edinburgh: Churchill Livingstone;
2002:1219.
13. Genuth SM. The Endocrine System. In: Berne M, Levy MN, Koeppen
BM, Stanton BA, eds. Physiology, 5th ed. St. Louis: Mosby; 2004:946.
14. Lifton RP. Molecular genetics of human blood pressure variation.
Science. 1996;272:676 – 680.
15. Donckerwolcke RA, Van Diemen-Steenvoorde R, van der Laag J,
Koomans HA, Boer WR. Impaired diluting segment chloride reabsorption
in patients with cystic fibrosis. Child Nephrol Urol. 1992;12:186 –191.
16. Quinton PM Chloride impermeability in cystic fibrosis Nature. 1983;301:
421– 422.
17. Reddy MM, Light MJ, Quinton P. Activation of the epithelial Na⫹
channel (ENaC) requires CFTR Cl- channel function. Nature. 1999;402:
301–304.
18. Legris GJ, Dearborn D, Stern R, Geiss CL, Hopfer U, Douglas JG,
Doershuk CF. Sodium space and intravascular volume:dietary sodium
effects in cystic fibrosis and healthy adolescent subjects. Pediatrics.
1998;101:48 –56.
19. Williams GW. Genetic factors associated with volume sensitive hypertension. Molecular and Cellular Endocrinology. 2004;217:41– 44.
Super et al
20. Quintero-Acenzio J, Vasquez-Leon M, Pino-Quintero LM. Association of
sweat sodium with arterial blood pressure. N Engl J Med. 1966;274:
1224 –1228.
21. Levi J, Rosenfeld LB, Hansel Z, Elian E. Relationship of sweat sodium
concentration to blood pressure in control subjects and in patients with
essential and renal hypertension. Isr J Med Sci. 1970;6:665– 668.
22. Naren AP, Cobb B, Li C, Roy K, Nelson D, Heda GD, Liao J, Kirk KL,
Sorcher EJ, Hanrahan J, Clancy JP. A macromolecular complex of ␤
adrenergic receptor, CFTR, and Ezrin/radixin/moesin-binding phosphoprotein 50 is regulated by PKA. PNAS. 2003;100:342–346.
23. De Lemos JA, McGuire D, Drazier MH B. B-type natiuretic peptide in
cardiovascular disease. Lancet. 2003;362:316 –322.
24. Kulaksiz H, Schmid A, Honscheid M, Eissels R, Klempnauer J, Cetlin Y.
Guanylin in the human pancreas: a novel luminocrine regulatory pathway
of electrolyte secretion via cGMP and CFTR in the ductal system. Histochem Cell Biol. 2001;115:131–145.
Blood Pressure in Female CF Carriers
883
25. Warlow C, Sudlow C, Dennis M, Wardlaw J. Sanderxcock C. Stroke.
Lancet. 2003;362:1211–1214.
26. INTERSALT Cooperative Research Group. Intersalt: An International
Study of Electrolyte Excretion and Blood Pressure. Results for 24 hour
Urinary Sodium and Potassium Excretion. BMJ. 1988;297:319 –328.
27. Macek M Jr Macek M Sr Krebsova A, Nash E, Hamosh A, Reis A,
Varon-Mateeva R, Schmidtke J, Maestri NE, Sperling K, Krawczak M,
Cutting GR. Possible association of the allele status of CS. 7/Hhal
polymorphism 5⬘ of the CFTR gene with postnatal female survival. Hum
Genet. 1997;99:565–572.
28. Witt M, Erickson RP, Ober C, Howatt WF Farber R. Correlation of
phenotypic and genetic heterogeneity in cystic fibrosis: variability in
sweat electrolyte levels contributes to heterogeneity and is increased with
the XV-2C/KM19 B haplotype. Am J Med Genet. 1991;39:137–143.
29. Super M. CFTR and disease implication for drug development. Lancet.
2000;355:1840 –1842.
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
Blood Pressure and the Cystic Fibrosis Gene: Evidence for Lower Pressure Rises With
Age in Female Carriers
Maurice Super, Ayesha Irtiza-Ali, Stephen A. Roberts, Martin Schwarz, Michele Young, Alison
Smith, Theresa Roberts, Joanna Hinks and Anthony Heagerty
Downloaded from http://hyper.ahajournals.org/ by guest on June 16, 2017
Hypertension. 2004;44:878-883; originally published online October 11, 2004;
doi: 10.1161/01.HYP.0000145901.81989.46
Hypertension is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2004 American Heart Association, Inc. All rights reserved.
Print ISSN: 0194-911X. Online ISSN: 1524-4563
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://hyper.ahajournals.org/content/44/6/878
Data Supplement (unedited) at:
http://hyper.ahajournals.org/content/suppl/2004/12/02/44.6.878.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Hypertension 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 Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Hypertension is online at:
http://hyper.ahajournals.org//subscriptions/