Stopping oral contraceptives: an effective blood pressure-lowering intervention in women with hypertension

(2005) 19, 451 455 & 2005 Nature Publishing Group All rights reserved 0950-9240/05 $30.00 www.nature.com/jhh ORIGINAL ARTICLE Stopping oral contraceptives: an effective blood pressure-lowering intervention in women with hypertension JN Lubianca 1, LB Moreira 1,2, M Gus 1,3 and FD Fuchs 1,3 1 Programa de Pós-Graduação em Medicina:Ciências Médicas, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil; 2 Division of Clinical Pharmacology, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil; 3 Division of Cardiology, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil The association of combined oral contraceptives (OC) with higher blood pressure (BP) may be attenuated with pills with smaller doses of oestrogen. The effect of stopping OC on BP of patients with hypertension was not described to date. In a cohort study of patients with hypertension, we identified 72 women using OC among 2112 patients seen from 1989 to 2002. Stopping hormonal contraception was recommended to all. The main outcome measurement was BP change in women who stopped (n ¼ 44) and who did not stop (n ¼ 28) OC, adjusting for baseline BP and age. Odds ratio for having a reduction of at least 20 mmhg in systolic blood pressure (SBP) or 10 mmhg in diastolic blood pressure (DBP), adjusting for age, change in weight and prescription of BP-lowering drugs, were calculated. The mean follow-up time was 6.677.5 months. Participants who stopped and did not stop OC had similar baseline characteristics. The deltas of SBP (adjusted) were 15.172.6 mmhg in patients who stopped and 2.873.2 mmhg in patients who did not stop OC (P ¼ 0.004). The corresponding values for DBP were 10.471.8 and 2.772.2 mmhg (P ¼ 0.008), respectively. The odds ratio (adjusted) for having a decrease of at least 20 mmhg in SBP or 10 mmhg in DBP was 0.28 (95% CI 0.08 0.90) in patients who stopped OC. Stopping OC is an effective antihypertensive intervention in a clinical setting. (2005) 19, 451 455. doi:10.1038/sj.jhh.1001841 Published online 10 March 2005 Keywords: oral contraceptives; blood pressure Correspondence: Professor FD Fuchs, Serviço de Cardiologia, sala 2061, Hospital de Clínicas de Porto Alegre, Ramiro Barcelos 2350, 90035-903, Porto Alegre RS, Brazil. E-mail: ffuchs@hcpa.ufrgs.br Received 26 October 2004; revised and accepted 29 December 2004; published online 10 March 2005 Introduction Oral contraception remains as one of the contraceptive methods mostly used worldwide. In the United States, approximately 34.5 million women were using contraception in the last decade. 1 Oral contraceptives (OC) and tubal ligation correspond to 85% of the contraceptive methods used in Brazil. 2 In a population-based survey in Porto Alegre, 81.8% of the women in the reproductive age were using OC. 3 Despite several attempts to reducing the rate of undesirable effects of OC, many are unlikely to be fully eliminated. Higher blood pressure levels are among their usual adverse effects, 4,5 and uncontrolled hypertension has been acknowledged as a relative contraindication for the use of combined OC. 6 Most studies documenting the association between hypertension and use of contraceptives involved compounds of first generation, with at least 50 mg of oestrogen. It was postulated that the incidence of hypertension would be lower if smaller doses of oestrogen were used. 7,8 However, higher blood pressure (BP) levels were found even in patients using monophasic pills containing 30 mg of oestrogen. 9 In a cohort study, a two-fold increase in risk for hypertension was detected in current OC users, as compared to nonusers. 10 Recently, we found that among hypertensive patients referred to a hypertension clinic, those using OC presented the highest probability of having uncontrolled BP, mainly at the expense of diastolic pressure. 11 Thus, the use of OC may worsen the prognosis of hypertensive women. An effective measure to lowering BP and consequent morbidity might be to substitute the contraceptive method, but as far as we

452 know there are no studies showing the effectiveness of this measure. In this report, we demonstrate that stopping the use of OC is an effective antihypertensive intervention in a clinical setting. Methods Our data come from a prospectively planned cohort study of hypertensive patients that is on way since 1989 in the Hypertensive Outpatient Clinic of Hospital de Clinicas de Porto Alegre. A more detailed description and results of this study may be seen elsewhere. 11 15 In brief, the study was planned to investigate several aspects related to the diagnosis and management of hypertension in a clinical setting. The influence of hormonal contraceptives on the control of BP was one of the specific objectives established a priori. Patients enrolled in the original cohort represented the whole spectrum of patients with hypertension, since our outpatient clinic was open to the referral of patients with uncomplicated, resistant, and potentially secondary hypertension. In a register of 2112 patients seen from 1989 to 2002, 72 hypertensive women using combined OC were identified and constituted the sample for this study. All participants were evaluated according to the protocol used in the main study, which included an extensive collection of data concerning medical history, physical examination, and laboratory testing. BP was measured according to standard techniques at three consecutive visits. During each visit, BP was measured twice at 5-min interval. The mean of six BP measurements usual BP was used to diagnose and classify hypertension. Blood pressure diagnosis and classification for this analysis followed the recommendations from 7th report of the Joint National Committee (JNC). 16 Detailed questions were asked about the use of contraceptive methods (at the time of the visit or in the past) and method employed (Ogino-Knaus, condoms, pill, IUD, or diaphragm). Almost all women under oral contraception were using pills with 30 mg of oestrogen. All patients using hormonal contraception were advised to change their method under the orientation of a gynaecologist, explaining the reasons for the measure. In the follow-up visits, the protocol included the questioning if the patient had stopped using the contraceptive. The groups for comparison consisted of patients who had stopped and patients who were still using them at the time of the follow-up visit. The main outcome measure was the BP variation between the initial evaluation and the follow-up visit. A dummy variable was computed corresponding to the reduction of at least 20 mmhg in systolic pressure or 10 mmhg in diastolic pressure between the two visits. The baseline prognostic characteristics were compared using the Student s t-test or the w 2 test when appropriate. The deltas of systolic and diastolic BP between baseline and follow-up visits were calculated and tested by a Student s t-test for independent samples. In a multivariate model, the deltas between the groups were adjusted for the baseline BP and age. The rate of patients with improved prognosis (at least 20 mmhg decreasing in systolic pressure or 10 mmhg in diastolic pressure) in the two groups was evaluated using the w 2 test. Improved prognosis was also included as a dependent variable in a logistic regression model, controlling for age, antihypertensive drug prescription, and body weight variation. All statistical analyses were performed with SPSS version 10.0. The project was approved by the Ethics Committee of our Institution. Since the participants were under usual medical care, a signed written consent was waived. Results The mean age of the patients was 37.776.2 years. The follow-up was 6.677.5 months (range ¼ 1 33 months). There was no difference in baseline characteristics between participants who stopped and did not stop the use of OC (Table 1). The antihypertensive treatment did not differ among groups. The reasons for not having stopped the use of OC were not systematically registered, but mostly comprised difficulties in getting the assistance of a gynaecologist or fear to use a less efficacious method. The severity of hypertension was reclassified according to JNC 7 and did not differ among groups. In the group of patients who stopped combined OC, 9.1% were at the prehypertensive stage, 45.5% at stage 1, and 45.5% were classified at stage 2. Among those who did not stop using the contraceptive during the follow-up, there were 17.9% of patients classified as prehypertension, 25% classified as stage 1, and 57.1% as stage 2 hypertension (P ¼ 0.18 for comparison between groups). The antihypertensive drug and nondrug treatment did not differ among groups. Systolic blood pressure (SBP) tended to be lower at follow-up visit in patients who stopped using OC, while the variation in diastolic pressure was not significantly different between groups (Table 2). After adjustment for initial BP and age, the deltas of SBP and diastolic blood pressure (DBP) were significantly higher in the patients who stopped using OC (Table 2). There was a trend towards improved prognosis, defined as a reduction of at least 20 mmhg in SBP or 10 mmhg in DBP, in patients who stopped using the OC (Table 3). After adjustment for age, variation in weight, and drug prescription at the initial visit, the

Table 1 Baseline characteristics of the women who stopped and did not stop using hormonal contraception (mean7s.d. or n (%), when appropriate) 453 Characteristics Stopped n ¼ 44 Did not stop n ¼ 28 P* Age (years) 38.375.8 37.076.97 0.404 White (n, %) 36 (81.8%) 22 (78.6%) 0.734 Income (minimum wages) 2.970.6 2.870.5 0.678 Years at school 7.173.4 7.673.3 0.548 Body mass index (kg/m 2 ) 29.175.3 27.075.1 0.125 Duration of hypertension (years) 3.071.9 2.871.4 0.562 Drugs in initial evaluation (n, %) None 27 (65.9%) 12 (42.9%) 0.138 One 11 (26.8%) 11 (39.3%) Two or more 3 (7.3%) 5 (17.9%) Systolic blood pressure (mmhg) 152.7720.3 159728 0.274 Diastolic blood pressure (mmhg) 98.7710.8 103.0720.5 0.242 Drugs in follow-up (n, %) None 23 (57.5%) 10 (37%) 0.05 One 12 (30%) 7 (25.9%) Two or more 5 (12.5%) 10 (37%) Follow-up time (months) 6.9 (8.5) 5.8 (5.3) 0.543 *Student s t-test for continuous variables; w 2 test for categorical variables. Table 2 BP (mmhg) at baseline and follow-up (means7s.d.), with the corresponding deltas (means7s.e.), in women who stopped and did not stop using OC Blood pressure Stopped OC Baseline Follow-up Delta* P* Delta** adjusted P** SBP Yes 152.7720.3 139.3718 13.773.1 0.09 15.172.6 0.004 No 159.0728.0 154.0721.8 5.074.1 2.873.2 DBP Yes 98.7710.8 89.7711.2 9.371.9 0.22 10.471.8 0.008 No 103.0720.5 98.7714 4.373.7 2.772.2 SBP, systolic blood pressure; DBP, diastolic blood pressure; OC, oral contraceptives. *Student s t-test for independent samples. **Adjusted for the respective baseline BP and age. Table 3 Proportion of participants who had a reduction of at least 20 mmhg in systolic blood pressure (SBP) or 10 mmhg in diastolic blood pressure (DBP) (n and (%)) Followed the recommendation to stopping OC Had a reduction of 20 mmhg in SBP or 10 mmhg in DBP * Yes No Exposure OR Improvement Age (10 years) 1.39 (0.89 2.16) Weight variation (10 Kg) Drug prescription 1.35 (0.61 2.97) 0.66 (0.22 1.97) Worsening Yes 22 (73.3) 22 (52.4) No 8 (26.7) 20 (47.6) Total 30 (100) 42 (100) * Exact Fisher test P ¼ 0.089; odds ratio (95% CI): 0.40 (0.15 1.10). Stopping OC 0.27 (0.08 0.90) 0.0 1.0 Figure 1 Association between stopping OC and improved prognosis (reduction of at least 10 mmhg in DBP or 20 mmhg in SBP) adjusted to other variables of exposure. 2.0 association between stopping hormonal OC and improvement in prognosis turned significant (Figure 1). Controlling for other potential confounders, such as baseline BP, informed compliance with drug prescription and with adhesion to the hypocaloric diet did not change the estimates substantially. Since the period of observation was too long, we looked at the association between OC suspension and BP changing in blocks or three periods of 4 years. Despite the wide confidence intervals, the associations were similar to the observed in the entire cohort.

454 Discussion The detailed recording of information in this cohort allowed to identifying a clinically significant BP lowering effect in hypertensive women who stopped using oestrogen-based hormonal contraceptives. We are not aware of other studies that have addressed this clinical question through this design. In the same cohort, we found that compliance with a hypocaloric diet and with the use of antihypertensive drugs was accompanied by a significant reduction in BP, but not compliance with the recommendation to increase physical activity and to follow a low sodium diet. 13 Compliance with the recommendation to stop using OC may be another effective antihypertensive measure. The magnitude of the effect, in average of 12.3 mmhg for SBP and 7.7 mmhg for DBP, is higher than the effect of other nondrug interventions accepted as effective. 16 The present findings were expected to a certain extent. The association between the use of OC and BP elevation has been repeatedly demonstrated. 11,17 19 In the last Joint National Committee, JNC 7 report, it is emphasized that OC raise BP and that the risk of hypertension increases with length of use. They recommend that OC users should get their pressure levels checked regularly during the period when they are taking OC and that the onset of hypertension is a reason to consider other forms of contraception. 16 Most studies, however, have focused on the development of hypertension in women who were previously normotensive. There are only two studies 11,20 on patients with a previous diagnosis of hypertension, in whom a positive association was observed between BP and the use of OC, anticipating that the withdrawal of OC could be an appropriate measure to improve BP control in hypertensive women. The relationship between higher BP and risk for cardiovascular and renal diseases was demonstrated in a large series of cohort studies, 21,22 showing that risks are directly proportional to the usual BP levels of individuals. Pooling 61 cohort studies, with approximately 1 million of individuals (12.7 million person-years at risk), the investigators from the Prospective Studies Collaboration demonstrated that the incidence of stroke and coronary heart disease increased two-fold at each difference of 20 mmhg in usual SBP or 10 mmhg in usual DBP. 23 Accordingly, our findings show that stopping OC may lead to a reduction of almost 50% in cardiovascular risk. The present study has some limitations. Despite the large size of the original cohort, just a few patients fulfilled the inclusion criteria. Owing to the relatively low statistical power, an alpha error may not be fully discarded (chance finding), but the differences in BP behaviour between participants who stopped and did not stop using OC seem to be consistent. Because of the non-experimental design, it cannot be ruled out that part of the BP lowering effect in patients who followed the recommendation of stopping the OC is due to other prognostic characteristics of these women. The most likely confounders, however, were controlled in the multivariate analysis and showed to be in the conservative side. This observation was made among a population attending to a hypertension clinic of a hospital offering secondary and tertiary care in Brazil, thus limiting generalizability to other populations. In conclusion, we demonstrated that when women with hypertension treated in an outpatient setting stop using hormonal contraceptives, there is a clinically relevant reduction of BP and presumably an improvement in prognosis. The replacement of the contraceptive method is required in women with hypertension. Acknowledgements This work was supported by the grants CAPES, CNPq, and FAPERGS, Brazil. References 1 Peterson LS. Contraceptive use in the United States: 1982 90. Advance Data, No. 260, February 14, 1995. Division of Vital Health Statistics, National Center for Health Statistics, Centers for Disease Control and Prevention. 1995; 260: 1 15. 2 Moreno L. Residential mobility and contraceptive use in northeastern Brazil. DHS Working Papers No. 9. Macro International Inc., Calverton, MD, 1994, pp 1 36. 3 Fuchs FD et al. Prevalence of systemic arterial hypertension and associated risk factors in the Porto Alegre metropolitan area. Arq Bras Cardiol 1994; 63: 473 479. 4 Nichols M et al. Effect of four combined oral contraceptives on blood pressure in the pill-free interval. Contraception 1993; 47: 367 376. 5 Shen Q et al. Blood pressure changes and hormonal contraceptives. Contraception 1994; 50: 131 141. 6 MacKay HT. Gynecology. In: Tierney LM, McPhee SJ, Papadakis MA (eds) Current Medical Diagnosis & Treatment, 43th edn. McGraw-Hill: New York, 2004, pp 694 728. 7 Malatino LS, Glen L, Wilson ES. The effects of low dose estrogen progestogen oral contraceptives on blood pressure and the renin angiotensin system. Curr Ther Res 1988; 43: 743 749. 8 Kaplan NM. The treatment of hypertension in women. Arch Intern Med 1995; 155: 563 567. 9 Wilson ESB, Cruickshank J, McMaster M, Weir RJ. A prospective controlled study of the effect on blood pressure of contraceptive preparations containing different types and dosages of progestagen. Br J Obstet Gynaecol 1984; 91: 1254 1260. 10 ChasanTaber L et al. Prospective study of oral contraceptives and hypertension among women in the United States. Circulation 1996; 94: 483 489.

11 Lubianca JN, Faccin CS, Fuchs FD. Oral contraceptives: a risk factor for uncontrolled blood pressure among hypertensive women. Contraception 2003; 67: 19 24. 12 Fuchs FD et al. Study of the usefulness of optic fundi examination of patients with hypertension in a clinical setting. J Hum Hypertens 1995; 9: 547 551. 13 Fuchs FD et al. Blood pressure effects of antihypertensive drugs and changes in lifestyle in a Brazilian hypertensive cohort. J Hypertens 1997; 15: 783 792. 14 Lubianca Neto JF et al. Is epistaxis evidence of endorgan damage in patients with hypertension? Laryngoscope 1999; 109: 1111 1115. 15 Fuchs FD et al. Headache is not more frequent among patients with moderate to severe hypertension. J Hum Hypertens 2003; 17: 787 790. 16 Chobanian AV et al. The Seventh Report of Joint National Committee on prevention, detection, evaluation and treatment of high blood pressure. The JNC 7 Report. JAMA 2003; 289: 2560 2572. 17 Weir RJ et al. Blood pressure in women taking oral contraceptives. BMJ 1974; 1: 533 535. 18 Cook NR et al. Regression analysis of changes in blood pressure with oral contraceptive use. Am J Epidemiol 1985; 121: 530 540. 19 Curtis KM, Chrisman CE, Peterson HB. WHO programme for mapping Best Practices in Reproductive Health. Contraception for women in selected circumstances. Obstet Gynecol 2002; 99: 1100 1112. 20 Narkiewicz K et al. Ambulatory blood pressure in mild hypertensive women taking oral contraceptives a case control study. Am J Hypertens 1995; 8: 249 253. 21 Pooling Project Cooperative Group. Relationship of blood pressure, serum cholesterol, smoking habit, relative weight and ECG abnormalities to incidence of major coronary events: final report of the National Cooperative Pooling Project. J Chron Dis 1978; 31: 201 306. 22 MacMahon S et al. Blood pressure, stroke, and coronary heart disease. Part 1, prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990; 335: 765 774. 23 Lewington S et al. Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360: 1903 1913. 455