| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Chair of Internal Medicine, Department of Clinical and Biological Sciences, Faculty of Medicine (A.M.G., P.Z., R.B., L.G., A.V.), University of Insubria, 21100 Varese; and Nuclear Medicine (A.F., L.C.), Ospedale di Circolo, 21100 Varese, Italy
Address correspondence and requests for reprints to: Anna M. Grandi, M.D., via Bagaini 15, 21100 Varese, Italy.
Abstract
The aim of this study was to evaluate the influence of blood pressure
(BP) control and familial predisposition to hypertension on
longitudinal changes in insulin sensitivity in essential hypertension.
We evaluated 6 groups of subjects twice (basal: before any treatment;
2nd: after at least 18 months): 42 hypertensives (H) with a family
history of hypertension (F+) and 30 H without a family history of
hypertension (F-) successfully treated with angiotensin-converting
enzyme inhibitors and/or calcium channel blockers (2nd: 24-h
BP
130/80 mm Hg); 22 untreated (UT) HF+ and 18 UTHF- (2nd:
24-h BP >140 and/or 90 mm Hg); 18 normotensives F+ and 15
normotensives F-. The parameters evaluated were as follows: glucose,
insulin, and C-peptide (Cp) response to an oral glucose load. Glucose
was normal in all of the subjects, similar among the 6 groups, and
unchanged at the 2nd evaluation.
At the basal evaluation insulin and Cp were higher and the metabolic clearance rate (MCR) of glucose was lower in the three F+ groups compared with the corresponding F- groups.
In the 2nd evaluation insulin and Cp were reduced and the MCR of glucose increased in THF-, whereas all metabolic parameters were unchanged in THF+; in both UT hypertensive groups insulin and Cp increased and the MCR of glucose decreased, more so in F+ than in F-; in normotensive groups metabolic parameters did not change.
A familial predisposition to hypertension influences insulin sensitivity changes during successful antihypertensive therapy, with an improvement in insulin sensitivity in F- and no changes in F+. A persistently high BP has a negative influence on insulin sensitivity in F+ and F-; this influence is greater when high BP is associated with a familial predisposition to hypertension.
ESSENTIAL HYPERTENSION IS very often associated with metabolic abnormalities that are powerful risk factors for cardiovascular diseases (1, 2), and insulin resistance appears to be a key element in this association (3, 4). In fact, a decreased insulin sensitivity with consequent hyperinsulinemia is frequently found in both lean and obese patients with essential hypertension (5, 6), but the pathophysiological link between insulin resistance and hypertension is still unclear, with some authors suggesting a causal role of insulin resistance in the development of hypertension and others suggesting that hypertension per se can induce a decrease in insulin sensitivity (7, 8, 9, 10). Recently, many studies have evaluated the influence of different antihypertensive drugs on insulin sensitivity, reaching partly different conclusions (11, 12, 13, 14, 15). The discrepancies are probably due to large differences among the studies in the number of patients selected, previous antihypertensive treatment, duration of treatment, degree of blood pressure (BP) control and methods for the assessment of both BP values and insulin sensitivity. From experimental and clinical studies it appears that genetic factors play an important role in linking insulin resistance and hypertension. In fact, a decreased insulin sensitivity has been demonstrated in lean and obese normotensives (N) and hypertensives (H) with hypertensive parents compared with the offspring of N (16, 17, 18, 19, 20, 21, 22). Moreover, a genetic predisposition to hypertension seems to influence the response of BP and myocardium to increased insulin levels (23, 24, 25).
In an attempt to clarify some of these controversies we evaluated the influence of familial predisposition to hypertension and degree of BP control on longitudinal changes in insulin sensitivity in patients with essential hypertension.
Materials and Methods
Patients
From the patients referred to our outpatient hypertension clinic, we consecutively enrolled subjects with the following characteristics: possibility of directly assessing a family history of hypertension, mean 24-h BP more than 140 and/or 90 mm Hg, no previous antihypertensive treatment, body mass index (BMI) less than 27 kg/m2, fasting plasma glucose (G) less than 5.6 mmol/L, normal glucose response to an oral glucose load, no family history of diabetes mellitus. A family history of hypertension was assessed on the basis of parents history and BP. Family history without hypertension (F-) was established when both parents were living and had BP less than 140/90 mm Hg; family history of hypertension (F+) was established when at least one parent was living and had BP more than 160/95 mm Hg or one or both parents had a history of chronic antihypertensive treatment. Parental BP was measured by sphygmomanometer three times on different days, by one of the investigators. The patients were judged to have essential hypertension on the basis of history, physical examination and laboratory findings.
We enrolled 87 HF- and 103 HF+; using the same criteria, we also
enrolled 58 N (mean 24-h BP
130/80 mm Hg): 28 NF-, and 30
NF+.
All the subjects underwent an oral glucose tolerance test (OGTT). After this basal evaluation, the H were assigned to treatment with angiotensin-converting enzyme (ACE) inhibitors or calcium channel blockers; during the subsequent visits a combination of the two drugs was used when appropriate.
After at least 18 months we called back all of the subjects, including 47 patients lost to follow-up, but excluding 21 HF- and 24 HF+ who had needed diuretics and/or ß-blockers.
Among the subjects summoned we excluded the following from the 2nd evaluation: 6 HF-, 4 HF+, 5 NF-, and 6 NF+ because of body weight changes more than 1 kg; 12 HF- and 11 HF+ because of mean 24-h BP less than 140 and/or 90 mm Hg but still more than 130/80 mm Hg; and 8 NF- and 6 NF+ who refused to undergo another OGTT.
Therefore, the 2nd evaluation was performed on 145 subjects, subdivided
as follows: 30 HF- (19 men) and 42 HF+ (27 men) with BP normalized
(mean 24-h BP
130/80 mm Hg) by ACE inhibitors and/or calcium
channel blockers; 18 HF- (11 men) and 22 HF+ (14 men) who, lost to
follow-up, stopped the treatment a few weeks after the basal evaluation
because they felt well and still had mean 24-h BP more than 140 and/or
90 mm Hg; and 15 NF- (10 men) and 18 NF+ (12 men).
Among the treated (T) H, 14 F- (47%) and 21 F+ (50%) were treated with ACE inhibitors (enalapril, fosinopril, perindopril), 10 F- (33%) and 13 F+ (31%) with calcium channel blockers (amlodipine, felodipine, isradipine), and 6 F- (20%) and 8 F+ (19%) with ACE inhibitors and calcium channel blockers.
The study was approved by the Ethical Committee of the Department of Clinical and Biological Sciences and all the subjects gave their informed consent.
We have to emphasize that none of the hypertensive subjects had been treated before the basal evaluation; in describing the results we defined the groups as TH and untreated (UT)H with reference to the subsequent period between the basal and 2nd evaluation.
Protocol
At the basal and 2nd evaluation each subject underwent a 24-h ambulatory BP monitoring and an OGTT.
OGTT
Two or 3 days after the 24-h ambulatory BP monitoring, each subject underwent a 75-g OGTT at 8am after an overnight fast. Plasma G, serum insulin (I), and C-peptide (Cp) were determined before and 30, 60, 90, and 120 min after the glucose load. The values obtained during OGTT have been expressed as area under the curve, measured using the trapezoidal rule. Using BMI, the 120-min insulin concentration and the 90-min glucose concentration, insulin sensitivity was assessed as the estimated metabolic clearance rate (MCR) of glucose using the following equation: MCR = 18.8 -0.271 x BMI - 0.0052 x I120 - 0.27 x G90 (26). We also evaluated the fasting I to Cp ratio, as an index of hepatic insulin clearance. I was measured by an antibody method with a solid-phase 125I RIA (Coat-A-Count Insulin; Diagnostic Products Corp., Los Angeles, CA), as was the Cp (C-peptide; Biodata, Rome, Italy). The method for insulin measurement has a sensitivity of 6.6 pmol/L and a coefficient of variation of 7.1% at insulin values of 6240 pmol/L. For Cp determination, the method has a sensitivity of 0.03 nmol/L and a coefficient of variation of 3.5% at Cp values of 0.161.76 nmol/L.
24-h ambulatory BP monitoring
Noninvasive ambulatory BP monitoring was performed with a portable automated Takeda TM 2421 (Osaka, Japan) and a simultaneous 24-h heart rate monitoring was obtained. The unit was set to take readings every 15 min throughout the 24 h. The following parameters were evaluated: mean 24-h, day-time (from 0700 to 2200 h), and night-time (from 2200 to 0700 h) systolic and diastolic BP and heart rate.
Statistical analysis
Data are expressed as mean ± SD. The statistical evaluation of the results was carried out by means of the following: ANOVA followed by Scheffès test for the comparison among the 6 groups, paired Students t test for the evaluation of longitudinal changes within each group, Pearsons linear correlation coefficient for the evaluation of linear correlations between variables. A value of P less than 0.05 was considered statistically significant.
Results
Basal evaluation
The 6 groups were not significantly different with regard to
gender, BMI (THF- 23.8 ± 2.5, THF+ 23.9 ± 2.4, UTHF-
24.1 ± 2.2, UTHF+ 23.9 ± 1.9, NF- 23.7 ± 2.3, NF+
23.6 ± 2.2 kg/m2, ANOVA NS), and
24-h heart rate (data not shown); BP values from 24-h monitoring were
similar among the four H groups and significantly lower in N groups
(Table 1
and Fig. 1
). Age was similar among UTH and N,
whereas T H were significantly older (THF- 47 ± 6, THF+ 48
± 5, UTHF- 40 ± 3, UTHF+ 40 ± 4, NF- 39 ± 4, NF+
38 ± 3 yr, ANOVA, P < 0.0005); the known
duration of hypertension was significantly longer in TH than UTH (THF-
27 ± 8, THF+ 25 ± 10, UTHF- 15 ± 8, UTHF+ 16 ±
9 months, ANOVA, P < 0.001). Glucose (fasting and
during OGTT) was normal in all the subjects and not significantly
different among the 6 groups (Table 2
),
as was the fasting insulin to Cp ratio (data not shown). I and Cp
(fasting and stimulated) were significantly higher and the MCR of
glucose lower in each F+ group compared with the corresponding F-
group (Table 3
and Fig. 2
). Among the F+ groups and among the F-
groups, insulin and Cp (fasting and stimulated) were higher and the MCR
of glucose was lower in TH compared with UTH and N (Table 3
and Fig. 2
). Insulin and Cp parameters did not correlate with BP values in any
of the six groups, nor in the total subject population.
|
|
|
|
|
The mean interval between basal and 2nd evaluation was
not significantly different among the 6 groups (THF- 26 ± 7,
THF+ 30 ± 9, UTHF- 31 ± 11, UTHF+ 28 ± 9, NF-
31 ± 10, NF+ 28 ± 9 months, ANOVA NS); 24-h heart
rate did not change (data not shown). As a consequence of selection
criteria, BMI was unchanged in all of the groups, BP was normalized in
both TH groups and remained similar to basal values in UTH groups and
in N groups (Table 1
and Fig. 1
). BP was similar in THF- and THF+ and
in UTHF- and UTHF+. The percentage of patients treated with ACE
inhibitors, calcium channel blockers, or both was not statistically
different between HF+ and HF-.
Glucose (fasting and during OGTT) was normal in all the subjects and
similar to basal values in each group (Table 2
), as was the fasting
insulin to Cp ratio (data not shown). Insulin and Cp during OGTT were
reduced and the MCR of glucose was increased in THF-, whereas all
metabolic parameters were unchanged in THF+; in UTHF- and UTHF+ groups
insulin and Cp (fasting and stimulated) increased and the MCR of
glucose decreased; in both N groups the metabolic parameters did not
change (Table 3
and Figs. 2
and 3
). The
percent changes of metabolic parameters in UTH groups were
significantly (0.05 < P < 0.02) greater in F+
than in F- (Fig. 3
). In THF- the decrease in BP did not correlate
with changes in insulin parameters.
|
|
To evaluate the influence of familial predisposition to hypertension and BP control on insulin sensitivity changes in H, we planned a pattern of selection criteria that aimed to avoid some potentially confounding factors. For the assessment of family history of hypertension, we did not trust parents history, with the exception of parents on chronic antihypertensive treatment, and we selected patients with living parents, whose BP was measured directly at least three times. From previous studies, despite some differences among the results, many calcium channel blockers and ACE inhibitors appear to be metabolically neutral or to induce a small improvement in insulin sensitivity, whereas nifedipine, many ß-blockers, and diuretics seem to have a detrimental effect on insulin sensitivity (11, 12, 13, 14, 15, 27, 28, 29, 30, 31, 32, 33). Therefore, we selected patients who had never been treated before the basal evaluation to avoid the possible influence of previous therapy on the basal assessment of insulin sensitivity and we enrolled for the 2nd evaluation only subjects continuously treated for at least 18 months with drugs that do not seem to affect insulin sensitivity (ACE inhibitors and/or calcium channel blockers, with the exception of nifedipine). This choice was made to avoid the mixing of subjects treated with drugs with potential detrimental or potential positive action on insulin sensitivity. Because obesity and insulin resistance are strictly related and weight loss can influence insulin sensitivity (34, 35, 36), we only selected lean or slightly overweight (BMI < 27 kg/m2) patients for the basal evaluation and also excluded the patients who developed weight changes more than 1 kg between the two evaluations. To study the possible influence of the degree of BP control on the changes of insulin sensitivity, we assessed BP by means of 24-h ambulatory monitoring, using a high cutoff BP value for the enrolment at basal evaluation (mean 24-h BP >140 and/or 90 mm Hg); for the 2nd evaluation we selected H with BP normalized by treatment (mean 24-h BP after treatment < 130/80 mm Hg) or with continuing high BP (24-h BP >140 and/or 90 mm Hg) because of lack of treatment. These selection criteria substantially reduced the number of patients eligible for the study but strengthen the reliability of the results.
Insulin sensitivity was assessed by means of glucose and insulin response to an oral glucose load. This method does not allow a direct measurement of insulin sensitivity, but recently Stumvoli et al. (26) have validated the use of data obtained from OGTT, finding a highly significant correlation between the values of MCR of glucose measured by the euglycemic clamp and the values of MCR estimated using BMI, 120-min insulin, and 90-min glucose concentration during OGTT.
The results at the basal evaluation confirmed the previously demonstrated (16, 17, 18, 19, 20, 21, 22) association between familial predisposition to hypertension and reduced insulin sensitivity; in fact, despite normal glucose levels (fasting and stimulated) in all the subjects and similar levels among the six groups, insulin levels (fasting and stimulated) were higher, and the MCR of glucose lower in HF+ and NF+ compared with H and N with normotensive parents. It should be pointed out that the difference in insulin parameters between F+ and F- were not accounted for by differences in gender, BMI, or 24-h BP; moreover, the hepatic clearance of insulin was similar among the groups, as demonstrated by the fasting I to Cp ratio.
Comparing insulin values from basal and 2nd evaluation, we found a
significant different metabolic behavior in successfully THF+ and
THF-. In fact, after antihypertensive therapies that were similar with
regard to drugs used, length of treatment, and BP values attained (24-h
BP
130/80 mm Hg in all), insulin sensitivity significantly
improved in HF-, as shown by reduced insulin area during OGTT and
increased MCR of glucose, whereas insulin parameters did not change at
all in HF+. Considering only H treated with ACE inhibitors or calcium
channel blockers, both kinds of treatment induced a small but
significant improvement in insulin sensitivity, but this improvement
was due only to F- patients. In UTH, the persistence of high BP was
associated with a decrease in insulin sensitivity in both F+ and F-
groups, with a greater decrease in F+. The changes in insulin
parameters between basal and 2nd evaluation in the four groups of H
were not accounted for by changes in body weight, as pointed out above,
and the difference in insulin sensitivity behavior between F+ groups
and corresponding F- groups cannot be ascribed to differences in age,
BMI, or degree of BP control. The hypothesis that the decrease in
insulin sensitivity in UTH is related to the persistence of high BP is
also strengthened by the results obtained in F+ and F- normotensive
groups; these two groups, similar to UTH groups in age, gender, BMI,
and interval length between basal and 2nd evaluation, showed no changes
in insulin parameters. The detrimental effect of persistently high BP
on insulin sensitivity could also explain the significantly lower
insulin sensitivity found at basal evaluation in T HF- and HF+
compared with normotensive groups and UTH groups, who had a shorter
known duration of hypertension than TH.
To the best of our knowledge, this is the first study testing the hypothesis that a familial predisposition to hypertension and degree of BP control can influence the longitudinal changes in insulin sensitivity in patients with essential hypertension and we think that our results support this hypothesis. Both familial predisposition to hypertension and high BP in itself appear to have a detrimental effect on insulin sensitivity. In fact, the persistence of high BP was accompanied in both F+ and F- by a decrease in insulin sensitivity and this decrease was significantly greater in F+ than in F-; on the other hand, BP well controlled by successful therapy was associated with an improvement in insulin sensitivity in F-, whereas metabolic parameters did not change in F+, meaning that in this group BP control only prevented a further deterioration of insulin sensitivity. Considering the previous studies on the effect of antihypertensive drugs on insulin sensitivity, it is likely that some of the discrepancies among the results can be accounted for by a different degree of BP control and a different prevalence of F+ and F- among the hypertensive groups.
Obviously many other factors influence the relationship between insulin sensitivity and hypertension, as indicated by the lack of correlation between insulin and BP values at the basal evaluation and between percent changes of insulin parameters and percent changes of BP at 2nd evaluation.
In conclusion, a familial predisposition to hypertension influences insulin sensitivity changes during successful antihypertensive therapy and the persistence of high BP in itself has a negative influence on insulin sensitivity, this influence being greater when high BP is associated with a familial predisposition to hypertension.
Received July 19, 2000.
Revised January 22, 2001.
Accepted March 21, 2001.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |