Journal Logo

Articles

Comparison of 24-Hour Blood Pressure, Heart Rate, and Autonomic Nerve Activity in Hypertensive Patients Treated with Cilnidipine or Nifedipine Retard

Minami, Junichi; Ishimitsu, Toshihiko; Kawano, Yuhei*; Numabe, Atsushi; Matsuoka, Hiroaki

Author Information
Journal of Cardiovascular Pharmacology: August 1998 - Volume 32 - Issue 2 - p 331-336
  • Free

Abstract

Dihydropyridine calcium antagonists have been widely used in the treatment of hypertension. Cilnidipine is a novel and unique dihydropyridine derivative that was synthesized by Fujirebio Inc. (Tokyo, Japan; 1) and possesses a slow-onset, long-lasting hypotensive effect (2,3). In spontaneously hypertensive rats (SHRs), Hosono et al. (4) found that cilnidipine causes an inhibition of the pressor responses induced by acute cold stress in addition to its hypotensive effect. Fujii et al. (5) recently reported that cilnidipine has potent inhibitory action on the L-type and N-type voltage-dependent calcium channels in rat dorsal root ganglion neurons, by means of the whole-cell patch-clamp technique. However, no data are available as to the clinical advantages of cilnidipine over other dihydropyridine derivatives.

The aim of this study was to compare the effects of cilnidipine and nifedipine retard on 24-h blood pressure (BP), heart rate (HR), and autonomic nerve activity in patients with essential hypertension. The autonomic nerve activity was evaluated noninvasively by a power spectral analysis of HR variability.

METHODS

Subjects

Fourteen outpatients with essential hypertension participated in this study. Informed consent to participate in the study was obtained from all patients after they had been given a detailed explanation of the study protocol. The study protocol was approved by the Ethical Committee of our institution. The patients each had a systolic BP >160 mm Hg or a diastolic BP >95 mm Hg, or both, on at least three occasions at the outpatient clinic. The possibility of secondary causes of hypertension was excluded through a comprehensive check-up including an assessment of their medical history, physical findings, urinalysis, blood chemistry, and endocrinologic and radiologic examinations when needed. All of the patients had a normal renal function, as judged from their endogenous creatinine clearance. According to the World Health Organization criteria for organ damage, all of the patients were classified as having stage I or II hypertension. Twelve patients were taking antihypertensive medication, which consisted mainly of calcium antagonists, whereas the other two were newly diagnosed as having essential hypertension and had not received prior treatment.

Study protocol

This was a three-way, randomized crossover study conducted at one hospital. After discontinuation of any previous antihypertensive treatment, each patient underwent a 4-week treatment period with cilnidipine, a 4-week treatment period with nifedipine retard, and a 4-week drug-free period, in one of six possible sequences. The allocation of patients to a treatment sequence was carried out blindly. The dosage of the drugs was determined according to the previous antihypertensive treatment; 10 patients including two who were newly diagnosed as having essential hypertension were assigned to receive cilnidipine, 5 mg once daily orally, or nifedipine retard, 10 mg twice daily orally, and the other four to receive cilnidipine, 10 mg once daily, or nifedipine retard, 20 mg twice daily. The dosage was not changed during each treatment period. On the last day of each period, the 24-h BP was measured.

Ambulatory blood pressure measurement

The 24-h ambulatory BP was monitored every 30 min by using the cuff-oscillometric device, TM-2425 (A&D Co., Tokyo, Japan). The device satisfied the criteria of the Association for the Advancement of Medical Instrumentation (AAMI) and British Hypertension Society (BHS; 6). The usefulness of this recorder in clinical hypertension research had previously been reported by us (7) and other investigators (6,8). The daytime and nighttime BPs were defined as the average values during the awake period (06:30-22:00 h) and the sleeping period (22:30-06:00 h), respectively. The same recorder was used for each subject for the entire protocol to avoid different BP readings being caused by different recorders. To minimize the effect of the patients' physical activity on BP, the ambulatory BP monitoring was performed on the same day of the week.

Power-spectral analysis of RR intervals

The ambulatory BP recorder used in this study, TM-2425, also monitored the RR interval of the electrocardiogram. The procedures of the power-spectral analysis of RR intervals in this device were previously reported in detail by us (7).

In brief, electrocardiogram tracings were obtained with a pericardial lead (V5). A total of 512 RR intervals was recorded at intervals of 30 min at a resolution of 7.8 ms. Ectopic beats or artifacts were excluded automatically, although all subjects had normal sinus rhythm. The power-spectral density of the RR variability was then analyzed and quantified by using TM-2021-05 power-spectrum analytical software (A&D Co.) on the basis of the autoregressive model. The frequency range of 0.05-0.15 Hz was computed as the low-frequency (LF) component, which is an index of both parasympathetic and sympathetic nerve activities. The frequency range of 0.15-0.40 Hz was computed as the high-frequency (HF) component, which reflects parasympathetic nerve activity exclusively. The LF/HF ratio was calculated as an index of sympathovagal balance (9).

Statistical analysis

Values are expressed as mean ± SEM. Comparisons within the three periods were performed by means of a two-way analysis of variance. When a significant overall effect was detected, the Tukey method was used for the comparison between the mean values for two variables. For the comparisons of power-spectral density, the naturally logarithmic values [i.e., ln (the LF component), ln (the BF component), or ln (the LF/HF ratio)], were used to normalize the skewness of the data. A value of p < 0.05 was considered significant.

RESULTS

All 14 of the subjects completed the study protocol. Their baseline characteristics are shown in Table 1. No patients reported side effects during any treatment period.

TABLE 1
TABLE 1

Figure 1 depicts the 24-h trendgram of BP and HR in each period. Table 2 lists the average values for the entire 24-h period, the daytime, and the nighttime. Cilnidipine and nifedipine retard reduced the systolic and diastolic BPs to similar extents, although the changes in the diastolic BP were not significant in the nighttime during either treatment period. The decreases in 24-h BP were 11 ± 3/6 ± 1 mm Hg in the treatment period with cilnidipine and 15 ± 3/6 ± 2 mm Hg in the treatment period with nifedipine retard. Cilnidipine did not change the ambulatory HR in the daytime or nighttime. However, nifedipine retard significantly increased the 24-h average HR (+3.3 ± 1.4 beats/min; p < 0.05) and the daytime HR (+3.5 ± 1.2 beats/min; p < 0.05).

FIG. 1
FIG. 1
TABLE 2
TABLE 2:
Blood pressure and heart rate during the cilnidipine, nifedipine retard, and drug-free periods

Figure 2 depicts the 24-h trendgram of the LF component, the HF component, and the LF/HF ratio in each period, and Table 3 lists the average values for the entire 24-h period, the daytime, and the nighttime. Nifedipine retard significantly increased the LF/HF ratio in both the daytime (p < 0.01) and nighttime (p < 0.05), whereas such changes were limited to the daytime in the treatment period with cilnidipine. Especially in the nighttime, the LF/HF ratio was significantly higher in the treatment period with nifedipine retard than that in the treatment period with cilnidipine. Neither cilnidipine nor nifedipine retard changed the LF or HF components.

FIG. 2
FIG. 2:
The 24-h trendgram of the low-frequency (LF) component, the high-frequency (HF) component, and the LF/HF ratio during the cilnidipine (▪), nifedipine retard (□), and drug-free (○) periods. Nifedipine retard significantly increased the 24-h average and daytime LF/HF ratio, whereas such changes were limited to the daytime in the treatment period with cilnidipine. Neither cilnidipine nor nifedipine retard changed the LF or HF components.
TABLE 3
TABLE 3:
Power spectral data during the cilnidipine, nifedipine retard, and drug-free periods

DISCUSSION

This is the first three-way, randomized crossover study to compare the effects of cilnidipine and nifedipine retard treatments in patients with essential hypertension. The results revealed that cilnidipine was effective as a once-daily antihypertensive agent and had less influence on the autonomic nervous system and HR than did nifedipine retard in these patients.

In our study, autonomic nerve activity was evaluated by a power-spectral analysis of HR variability. Power-spectral analyses of HR variability have been widely accepted as a noninvasive method of assessing autonomic nervous function in patients with various cardiovascular disorders (7,10,11). We previously reported the usefulness of this method in clinical hypertension research (7,12). The spectral analysis of HR variability can partially distinguish parasympathetic from sympathetic influences on the cardiovascular system. When 24-h monitoring of the electrocardiogram is used to analyze the power-spectral density of HR variability, the HF component represents a measure of parasympathetic nerve activity, whereas the LF component reflects the mixed measure of parasympathetic and sympathetic nerve activity and, under unrestricted or ambulatory conditions, reflects more parasympathetic than sympathetic nerve activity (13,14). The LF/HF ratio is currently considered to be an index of sympathovagal balance, with high values suggesting the predominance of sympathetic nerve activity (9). Therefore our increase in the LF/HF ratio during the treatment period with nifedipine retard indicates the activation of the sympathetic nervous system.

Although cilnidipine has a biological half-life of ∼2.5 to 3 h (15), which is similar to that of nifedipine retard, cilnidipine at a dose of 5 or 10 mg once daily and nifedipine retard at a dose of 10 or 20 mg twice daily reduced the BP of these patients to similar extents after long-term treatment for 4 weeks. However, the effects of the two drugs on the autonomic nervous system differed in some respects; nifedipine retard increased the LF/HF ratio in both the daytime and nighttime, whereas such changes were limited to the daytime in the treatment period with cilnidipine. Especially in the nighttime, the LF/HF ratio was significantly higher in the treatment period with nifedipine retard compared with that in the treatment period with cilnidipine, although both agents reduced the nighttime BP to similar extents (cilnidipine, −10 ± 3/−4 ± 1 mm Hg; nifedipine retard, −13 ± 3/−4 ± 2 mm Hg). These findings suggest that nifedipine retard caused an increase in the sympathetic nerve activity through the entire 24-h period, whereas cilnidipine produced less sympathetic nerve activation than did nifedipine retard. These results also confirm the earlier experimental and clinical observations. Yoshimoto et al. (2) examined the recovery of contractility of rat superior mesenteric arteries suppressed by cilnidipine and other dihydropyridine calcium antagonists. They found that inhibition by cilnidipine persisted for >7 h after its removal, whereas the preparation treated with nifedipine recovered complete contractility 2 h after drug removal. It has been also shown that cilnidipine significantly reduced the BP in hypertensive patients, whereas the HR and plasma catecholamines at rest and exercise loading did not change with cilnidipine (16). Recently Wenzel et al. (17) demonstrated with direct measurement of sympathetic nerve activity in the peroneal nerve in humans that nifedipine was a strong stimulator of peripheral sympathetic nerve traffic independent of its drug-release formulations. Taken together, cilnidipine may, in addition to its long-lasting hypotensive effect, suppress the increase of the sympathetic tone often observed with antihypertensive treatment with calcium antagonists, especially with short-acting agents such as nifedipine capsules (18).

Hosono et al. (19) reported that cilnidipine, but not nicardipine, inhibited the release of [3H]norepinephrine from sympathetic nerve endings in the rat mesenteric vasculature. Fujii et al. (5) recently found that cilnidipine has inhibitory actions on the N-type as well as the L-type voltage-dependent calcium channels in rat dorsal root ganglion neurons. The two inhibitory actions occurred at similar concentration ranges; the median inhibitory concentration (IC50) values for cilnidipine block on the L-type and N-type calcium channel were 100 and 200 nM, respectively. Moreover, Uneyama et al. (20) examined the inhibitory effects of cilnidipine and various calcium antagonists on the N-type voltage-dependent calcium channel, by means of the conventional patch-clamp electrophysiology, and reported that these calcium antagonists including nifedipine, except cilnidipine, failed to inhibit the N-type voltage-dependent calcium channel. Therefore it is thought that the N-type voltage-dependent calcium channel-blocking action of cilnidipine may, at least in part, contribute to the diminishment of the high sympathetic tone. Some investigators have reported that the N-type voltage-dependent calcium channel is greatly involved in sympathetic neurotransmission and regulates the release of norepinephrine from the sympathetic nerve endings (21-25). For example, the intravenous administration of ω-conotoxin GVIA, a specific N-type voltage-dependent calcium channel blocker, inhibited nitroprusside-induced tachycardia in SHRs (26).

Regarding the potent advantage of cilnidipine over other dihydropyridine derivatives, it has been shown that cilnidipine, but not other dihydropyridine calcium antagonists, attenuated the increase of both BP and plasma norepinephrine concentration induced by acute cold stress in SHRs (4). Stress is known to be one of the risk factors of cardiovascular diseases such as hypertension (27). We found that acute psychological stress induced by the Hanshin-Awaji earthquake significantly increased BP self-measured at home in treated hypertensive patients in the disaster (28). It is therefore important to evaluate the effect of antihypertensive agents on cardiovascular responses to stress. Saihara et al. (29) found that the pressor response induced by acute cold stress was significantly suppressed by the administration of cilnidipine in healthy young volunteers. Moreover, the release of norepinephrine from the renal sympathetic nerve was shown to be inhibited by an N-type calcium channel blocker in perfused rat kidney (30). This finding is consistent with the recent observation that cilnidipine suppressed the increase in norepinephrine secretion rate and antinatriuretic response and the decrease in renal blood flow that were induced by renal nerve stimulation in dogs, whereas nifedipine failed to inhibit such responses (31). In addition to such benefits, it has been suggested that an N-type calcium channel blocker was useful for preventing the brain damage caused by ischemia, as described for the synthetic peptide N-type calcium channel inhibitor, SNX-111 (32). In light of these findings, cilnidipine may possess additional clinical advantages for the treatment of hypertension over other dihydropyridines. However, further studies are required in various conditions, because the clinical and therapeutic advantages of cilnidipine have not yet been clarified.

In this study, the patients were not given placebos during any period. It is known that the BP without drug treatment gradually decreases in placebo-controlled trials; this phenomenon may be due to the placebo effect or to a reduction in patient anxiety (33). However, the use of ambulatory BP monitoring may eliminate such a placebo effect. Mancia et al. (34) reported that the placebo effect of ambulatory BP monitoring is little; thus they concluded that it is not necessary to include a placebo control group in antihypertensive drug studies in which ambulatory BP monitoring is used. Moreover, it has been shown that the use of ambulatory BP monitoring can reduce significantly the number of patients required for comparisons of the antihypertensive effects of different drugs (35,36). We have also found that ambulatory BP monitoring is reliable and useful in various clinical studies (7,12,37).

There are several limitations to our study. First, the lack of respiration data may limit the interpretation of the power-spectral data. Ideally, because respiration rate and breathing depth influence the LF and the HF components (38), respiration should have been monitored as well. Therefore the next modification of the device used in this study will be designed accordingly (6). However, patients who had evident sleep disorders such as sleep apnea syndrome were not included in our study, and no patients in the study complained of disturbances of their sleep on the day of BP and electrocardiographic monitoring, judging from their questionnaires.

Furthermore, in this study, the sympathetic nerve activity was evaluated by using the LF/HF ratio obtained from a power-spectral analysis. Available data have shown unequivocally that power-spectral analyses of HR variability can provide dynamic information and represent useful tools for the study of the mechanisms involved in cardiovascular autonomic regulation in both normal and diseased conditions (7,10). However, because no single test of autonomic system adequately fulfills the goal of a global indicator of this system, it seems to be necessary to compare the effects of cilnidipine and nifedipine retard on the sympathetic nerve activity by using other methods such as microneurography and the measurement of norepinephrine spillover from the heart.

In conclusion, cilnidipine was effective as a once-daily dihydropyridine derivative and had less influence on the autonomic nervous system and HR than did nifedipine retard.

Acknowledgment: This work was partly supported by Kimura Memorial Heart Foundation Grant for Research on Sympathetic Nervous System and Hypertension (J.M.). We thank Masahiro Hosono, Ph.D., for critical reading of the manuscript. We also thank Nobuo Shirahashi, Ph.D., for assisting us in the statistical analysis and Nobuhiko Yasui for technical advice on a power-spectral analysis of heart-rate variability.

REFERENCES

1. Iida H, Fujiyoshi H, Ikeda K, et al. Antihypertensive and cardiovascular effects of a new dihypropyridine derivative, 2-methoxyethyl(E)-3-phenyl-2-propen-1-yl-1,4-dihydro-2,6-dimethyl-4-(3-nitrophenyl)pyridine-3,5-dicarboxylate (FRC-8653). Jpn J Pharmacol 1987;43(suppl):296.
2. Yoshimoto R, Dohmoto H, Yamada K, Goto A. Prolonged inhibition of vascular contraction and calcium influx by the novel 1,4-dihydropyridine calcium antagonist cinaldipine (FRC-8653). Jpn J Pharmacol 1991;56:225-9.
3. Tominaga M, Ohya Y, Tsukashima A, et al. Ambulatory blood pressure monitoring in patients with essential hypertension treated with a new calcium antagonist, cilnidipine. Cardiovasc Drugs Ther 1997;11:43-8.
4. Hosono M, Hiruma T, Watanabe K, et al. Inhibitory effect of cilnidipine on pressor response to acute cold stress in spontaneous hypertensive rats. Jpn J Pharmacol 1995;69:119-25.
5. Fujii S, Kameyama K, Hosono M, Hayashi Y, Kitamura K. Effect of cilnidipine, a novel dihydropyridine Ca2+-channel antagonist, on N-type Ca2+-channel in rat dorsal root ganglion neurons. J Pharmacol Exp Ther 1997;280:1184-91.
6. Tochikubo O, Ikeda A, Miyajima E, Ishii M. Effects of insufficient sleep on blood pressure monitored by a new multibiomedical recorder. Hypertension 1996;27:1318-24.
7. Minami J, Kawano Y, Ishimitsu T, Takishita S. Blunted parasympathetic modulation in salt-sensitive patients with essential hypertension: evaluation by power-spectral analysis of heart-rate variability. J Hypertens 1997;15:727-35.
8. Kawano Y, Tochikubo O, Watanabe Y, Miyajima E, Ishii M. Doxazosin suppresses the morning increase in blood pressure and sympathetic nervous activity in patients with essential hypertension. Hypertens Res 1997;20:149-56.
9. Pagani M, Lombardi F, Guzzetti S, et al. Power spectral analysis of heart rate and arterial pressure variabilities as a marker of sympathovagal interactions in man and conscious dog. Circ Res 1986;59:178-93.
10. Parati G, Saul JP, Di Rienzo M, Mancia G. Spectral analysis of blood pressure and heart rate variability in evaluating cardiovascular regulation-a critical appraisal. Hypertension 1995;25:1276-86.
11. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology: heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Circulation 1996;93:1043-65.
12. Minami J, Ishimitsu T, Kawano Y, Matsuoka H. Effects of amlodipine and nifedipine retard on autonomic nerve activity in hypertensive patients. Clin Exp Pharmacol Physiol (in press).
13. Bigger JT Jr, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE, Rottman JN. Correlations among time and frequency domain measures of heart period variability two weeks after acute myocardial infarction. Am J Cardiol 1992;69:891-8.
14. Petretta M, Bonaduce D, Marciano F, et al. Effect of 1 year of lisinopril treatment on cardiac autonomic control in hypertensive patients with left ventricular hypertrophy. Hypertension 1996;27:330-8.
15. Ishii M, Kakuo M, Shishido R, Okuno T, Itida S, Matsumoto M. Phase I clinical study of FRC-8635 (cilnidipine): single-dose study. Jpn Pharmacol Ther 1993;21(suppl 1):S7-22.
16. Kohno M, Yasunari K, Murakawa K, Yokokawa K, Horio T, Takeda T. Effect of FRC-8653 (cilnidipine) on hemodynamics, the renin-angiotensin system, the sympathetic nervous system, and atrial natriuretic polypeptides at exercise loading in patients with essential hypertension [in Japanese]. Jpn Pharmacol Ther 1993;21(suppl):S227-32.
17. Wenzel RR, Allegranza G, Binggeli C, et al. Differential activation of cardiac and peripheral sympathetic nervous system by nifedipine: role of pharmacokinetics. J Am Coll Cardiol 1997;29:1607-14.
18. Ruzicka M, Leenen FHH. Relevance of 24 h blood pressure profile and sympathetic activity for outcome on short- versus long-acting 1,4-dihydropyridines. Am J Hypertens 1996;9:86-94.
19. Hosono M, Fujii S, Hiruma T, et al. Inhibitory effect of cilnidipine on vascular sympathetic neurotransmission and subsequent vasoconstriction in spontaneous hypertensive rats. Jpn J Pharmacol 1995;69:127-34.
20. Uneyama H, Takahara A, Dohmoto H, Yoshimoto R, Inoue K, Akaike N. Blockade of N-type Ca2+ current by cilnidipine (FRC-8653) in acutely dissociated rat sympathetic neurons. Br J Pharmacol 1997;122:37-42.
21. Hirning LD, Fox AP, McCleskey EW, et al. Dominant role of N-type Ca2+ channels in evoked release of norepinephrine from sympathetic neurons. Science 1988;239:57-61.
22. Clasbrummel B, Osswald H, Illes P. Inhibition of noradrenaline release by ω-conotoxin GIVA in the rat tail artery. Br J Pharmacol 1989;96:101-10.
23. Pruneau D, Angus JA. ω-Conotoxin GIVA is a potent inhibitor of sympathetic neurogenic responses in rat small mesenteric arteries. Br J Pharmacol 1990;100:180-4.
24. Rittenhouse AR, Zigmond RE. ω-Conotoxin inhibits the acute activation of tyrosine hydroxylase and the stimulation of norepinephrine release by potassium depolarization of sympathetic nerve ending. J Neurochem 1991;56:615-22.
25. Fabi F, Chiavarelli M, Argiolas L, Chiavarelli R, Del Basso P. Evidence for sympathetic neurotransmission through presynaptic N-type calcium channels in human saphenous vein. Br J Pharmacol 1993;110:338-42.
26. Pruneau D, Belichard P. Haemodynamic and humoral effects of ω-conotoxin GIVA in normotensive and spontaneous hypertensive rats. Eur J Pharmacol 1992;211:329-35.
27. Widgren BR, Wikstrand J, Bergland G, Andersson OK. Increased response to physical and mental stress in men with hypertension parents. Hypertension 1992;20:606-11.
28. Minami J, Kawano Y, Ishimitsu T, Yoshimi H, Takishita S. Effect of the Hanshin-Awaji earthquake on home blood pressure in patients with essential hypertension. Am J Hypertens 1997;10:222-5.
29. Saihara S, Kakuo, M, Hosono M, Hayashi H, Kitamura H, Hashimoto S. Effect of FRC-8653 (cilnidipine) on blood pressure changes during the cold pressor test. Jpn Pharmacol Ther 1993;21(suppl):S271-6.
30. Mohy El-Din M, Malik K. Differential effect of ω-conotoxin on release of the adrenergic transmitter and the vasoconstrictor response to noradrenaline in the rat isolated kidney. Br J Pharmacol 1988;94:355-62.
31. Takahara A, Dohmoto H, Hisa H, Satoh S, Yoshimoto R. Cilnidipine attenuates renal nerve stimulation-induced renal vasoconstriction and antinatriuresis in anesthetized dogs. Jpn J Pharmacol 1997;75:27-32.
32. Valientino K, Newcomb R, Gadbois T, et al. A selective N-type calcium channel antagonist protects against neuronal loss after global cerebral ischemia. Proc Natl Acad Sci U S A 1993;90:7894-7.
33. Bienenfeld L, Frishman W, Glasser SP. The placebo effect in cardiovascular disease. Am Heart J 1996;132:1207-21.
34. Mancia G, Omboni S, Parati G, Ravogli A, Villani A, Zanchetti A. Lack of placebo effect on ambulatory blood pressure. Am J Hypertens 1995;8:311-5.
35. Mallion JM, Baguet JP, Boutelant S, Tremel F, Siche JP, de Gaudemaris R. Second-generation calcium antagonists and ambulatory blood pressure monitoring. J Cardiovasc Pharmacol 1995;25(suppl 3):S17-S22.
36. Coats AJS, Radaelli A, Clark SJ, Conway J, Sleight P. The influence of ambulatory blood pressure monitoring on the design and interpretation of trials in hypertension. J Hypertens 1992;10:385-91.
37. Minami J, Kawano Y, Ishimitsu T, Yoshimi H, Takishita S. Seasonal variations in office, home, and 24 h ambulatory blood pressure in patients with essential hypertension. J Hypertens 1996;14:1421-5.
38. Hayano J, Mukai S, Sakakibara M, Okada A, Takata A, Fujinami T. Effects of respiratory interval on vagal modulation of heart rate. Am J Physiol 1994;267:H33-40.
Keywords:

Cilnidipine; N-type voltage-dependent calcium channel; Nifedipine retard; Calcium antagonist; Heart-rate variability; Hypertension

© Lippincott-Raven Publishers