Original Article

Journal of Human Hypertension (2003) 17, 181–186. doi:10.1038/sj.jhh.1001530

Blood pressure reactivity to mental stress task as a determinant of sustained hypertension after 5 years of follow-up

P Armario1, R H del Rey1, M Martin-Baranera2, M C Almendros1, L M Ceresuela1 and H Pardell1

  1. 1Department of Internal Medicine, Unit of Hypertension and Cardiovascular Risk, Consorci Sanitari de la Creu Roja a Catalunya, L'Hospitalet de Llobregat, University of Barcelona, Barcelona, Spain
  2. 2Unit of Clinical Epidemiology, Consorci Sanitari de la Creu Roja a Catalunya, L'Hospitalet de Llobregat, Barcelona, Spain

Correspondence: Dr P Armario, Department of Internal Medicine, Unit of Hypertension and Cardiovascular Risk, Consorci Sanitari de la Creu Roja a Catalunya, Avda Josep Molins 29-41, 08906 L'Hospitalet de Llobregat. Barcelona, Spain. E-mail: Pedro.Armario@chcr.scs.es

Received 5 July 2002; Revised 12 November 2002; Accepted 23 November 2002.



Previous studies have reported an increased risk of developing sustained hypertension (SH) in borderline or mildly hypertensive subjects showing an exaggerated response of blood pressure (BP) to mental stress. The aim of this study was to assess if the response of BP to mental stress tasks is an independent predictor of SH. A total of 89 patients with grade 1 hypertension, aged 18–64 years, 62% males, were included. The mean of follow-up was 5.3 years (s.d. 2.1 years). SH was defined as the development of grades 2–3 hypertension (Systolic BPgreater than or equal to160 mmHg or diastolic BPgreater than or equal to100 mmHg) or to be in antihypertensive treatment after follow-up. Two mental stress tasks: mental arithmetic stress task and a stressful interview (SI) were applied at entry. The subjects were classified as hyper-reactors when BP increase was greater than 35 mmHg for systolic BP or greater than 21 mmHg for diastolic BP, according to the results obtained previously in a normotensive control group. In the univariate analysis, the factors associated with the development of SH were age (P=0.0007), office diastolic BP (P=0.014) and hyper-reactivity of BP during a stressful interview (P=0.003). In the Cox regression model, after adjusting for gender, age, and office BP, the hyper-reactivity of BP during SI was an independent predictor of development of SH. In conclusion, the response of BP to mental stress tasks is useful in predicting SH in young and middle-aged subjects with grade 1 hypertension.


mental stress tasks, cardiovascular reactivity, predictors of sustained hypertension



It has been fairly well established that there are certain risk factors that increase the probability of developing hypertension. Blood pressure (BP), itself, is a fairly potent predictor of what blood pressure is likely to be later in life.1 Some previous studies have reported several predictors of hypertension: age (older), race (black), heart rate (HR), borderline hypertension, in particular the patients with hyperdynamic circulation, weight, specially centrally distributed obesity, left ventricular mass in children, genetic predictors or a family history of hypertension in parents, and cationic transporters.1,2,3,4,5,6,7,8

Cardiovascular reactivity has been measured in the laboratory by a wide variety of stimuli, both physical or psychological stressors such as mental arithmetic, public speaking, and video game challenges. The response of BP and HR to these stressors tests is very heterogeneous among subjects. According to the reactivity hypothesis, supported by several investigators,9 the increase of pressure response to the stimuli is considered to be a factor in precipitating and perpetuating hypertension, but the aetiological role of cardiovascular reactivity in the development of hypertension has not been resolved.10

Hypertensive subjects exhibit larger BP and HR responses during the tasks than the normotensive subjects, even when borderline or mildly hypertensives have been studied.11 A relation between an exaggerated response of BP to mental stress tasks and subsequent BP status has been observed in some prospective studies.12,13,14,15,16,17,18,19 However, other prospective studies did not find these results: a recent published study has shown that pressure reactions to stress provide only a minimal independent contribution of BP at follow-up.20

The objective of this study was to assess the independent contribution of BP response to mental stress tasks in the development of sustained hypertension (SH) in untreated mildly hypertensive subjects, after 5 years of follow-up.


Patients and methods

A total of 96 subjects with grade 1 hypertension systolic BP (SBP) between 140–159 mmHg and/or diastolic BP (DBP) between 90 and 99 mmHg)21 aged 18–64 years, 62% males were included. Patients were sequentially selected from those referred to our hypertension unit, and entered the study on cardiovascular reactivity after informed consent was obtained. Seven patients (7.3%) were lost to follow-up.

BP at entry was the average of three measurements made in the office by a previously trained nurse on two different days separated by at least 4 weeks.

Cardiovascular reactivity tasks

After confirming the diagnosis of hypertension, two mental stress tasks: mental arithmetic stress task (MAST) and stressful interview (SI) were applied in all the subjects. The methodology of this task has been previously published by our group.22,23 Subjects were instructed not to take alcohol, caffeine and cigarettes within 30 min before the test. Participants had to complete both tasks on the same day with 7-min rest periods before, between and after tasks. Both tasks were applied in a random order.

MAST: Participants were asked to make some different and standardised mental arithmetic cal-culations. First, during 2 min, they had to solve mentally some additions, subtractions and multiplications, then they repeated a sequence of numbers in inverse order during 3 min and, finally, they subtracted 7 or 13 from a number during the remaining 2 min.

SI: Participants had to answer some questions about their lifestyle and they were recorded in a cassette player during the task with a microphone.

After the application of a BP cuff, the patients rested in the supine position for 7 min in a quiet room. During the arithmetic test, the subjects were compelled to provide rapid and correct answers. The BP and HR were measured automatically by a monitor Kontrom Minimon 7137 B every minute during the tasks and during the resting and recovery periods of the same duration (7 min). According to the results obtained in a normotensive control group, we considered patients as hyper-reactors when the increased BP values during the tasks were greater than mean +2 s.d.: 35 mmHg for SBP or greater than 21 mmHg for DBP, or both.

Follow-up period

Following WHO/ISH guidelines,21 patients were visited in month 1, 3 and 6 after the diagnosis, and then every 6–12 months in the hypertension unit. SH was defined as the progression from grade 1 to grade 2 hypertension, to be on pharmacological antihypertensive treatment or both, during the follow-up period. In the follow-up visits, a protocol was applied to all the patients including clinical history, physical examination, a general blood test; echocardiography and 24-h BP monitoring.

Statistical methods

In the data analysis, the time for the development of SH was defined as the dependent variable. By means of the Kaplan–Meier method, survival curves were compared between normoreactors and hyper-reactors using the log-rank test. To assess the relation between BP hyper-reactivity during the mental stress tasks and the development of SH after follow-up, a bivariate analysis was carried out; demographic, clinical, echocardiography and laboratory factors were considered as potential confounders. Clinically relevant variables as well as variables showing a statistically significant association with the development of sustained hypertension over the follow-up period were introduced in a Cox regression model. All tests were conducted as two-sided tests and P values less than 0.05 were considered significant.



Demographic and clinical factors

Globally, 23.6% (n=21) of the patients were current smokers, 28.1% (n=25) had a body mass index greater or equal to 30 kg/m2 and 34.8% (n=31) had hypercholesterolaemia. Baseline SBP and DBP means were 146plusminus7 and 93plusminus4 mmHg respectively, and HR 75plusminus9 bpm. The main clinical characteristics of the grade 1 hypertensive subjects included in the study are shown in Table 1.

Echocardiographic findings

Left ventricular mass was 194plusminus57 g (Table 1). Left ventricular hypertrophy was detected in 18.8%, while 8.2% of the patients showed concentric remodelling.

Development of SH after follow-up

After a mean of 5.3 years of follow-up, 44 patients (49.4%) had developed SH and 24 patients had become normotensive. The rest (21 patients) did not change their initial status, that is, they still had a grade 1 hypertension after follow-up. The cumulative probability of developing SH over the follow-up period was statistically greater in hyper-reactors than in normo-reactors (Figure 1).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Cumulative probability of developing SH over time. Comparison of hyperreactors and normoreactors.

Full figure and legend (28K)

The BP values during the baseline period, during the task and after the recovery periods were higher in subjects who developed SH than in those who became normotensive (Figure 2 and Figure 3).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Changes of BP during SI in subjects who developed SH or who did not.

Full figure and legend (132K)

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Changes of BP during mental arithmetic task in subjects who developed SH or who did not.

Full figure and legend (109K)

In the univariate analysis, the clinical factors associated with the development of SH were: age (P=0.0007), SBP in the office (P=0.057), DBP in the office (P=0.014) and hyper-reactivity of BP to SI (P=0.003).

In the Cox regression model, after adjusting for age, gender and BP in the office, the BP hyper-reactivity to SI was an independent predictor of SH (OR 4.5, 95% Cl 1.4–14.7) (Table 2).



In the present study, BP at entry did not reach statistical significance as an independent predictor of future SH, possibly because of a lack of power owing to a small sample size. Skarfors et al studied a group of 2232 middle-aged men, examined at a health screening, and they were re-examined approximately 10 years later. In the first survey, 19.6% of the participants had hypertension (defined as a DBPgreater than or equal to95 mmHg or were receiving drug treatment for hypertension), and in a follow-up survey, the corresponding figure was 34.7%. They observed that baseline BP were the strongest predictors of future development of hypertension.2 In the Framingham Heart Study, it was observed that subjects with BP high-normal increased the risk of future SH in comparison with subjects with initial normal BP.3

Another interesting result of our study was that an exaggerated response of BP to MAST can identify a group with high risk of SH after a mean of 5 years of follow-up. The hyper-reactors had a 4.5-fold risk of developing SH in comparison with normoreactors. In the multivariate analysis we observed that the reactivity of BP to an SI was an independent predictor of SH at follow-up. A few prospective studies have used mental stress tasks requiring an active behavioural response. Most of them have observed a relation between an exaggerated response of BP to mental stress tasks and subsequent BP status after a few years of follow-up.12,13,14,15,16,17,18,19 An important study published by Carroll et al,20 found that the response of BP to mental stress task presented only a minimal independent prediction of SH.

Such results do not necessarily support the so-called reactivity hypothesis. The exaggerated response of BP to mental stress tasks could only be a marker of high risk of established hypertension. Light et al24 carried out a study on men tested at age 18–22 years and was reassessed 10 years later, and they observed that men with a combination of high stress response and hypertensive parents demonstrated higher systolic and diastolic levels at follow-up; they showed a 7.5-fold increase in relative risk of change in BP status vs men who also had hypertensive parents. These results suggest that stress response of BP predictor is modulated by both genetic and environmental factors.

Recovery can be defined as a post-stress period that provides information about the degree to which the elevation during the task persists after the stressor has ended.25 The stress test recovery time has been investigated poorly. In the present study, postrecovery changes of BP during an SI were significantly lower in subjects who developed SH. After adjusting by baseline BP, the difference was still significant. Our results are consistent with those of Borghi et al;14,15 they observed that DBP recovery from a cognitive challenge was a more useful predictor of long-term BP changes than was reactivity itself. Some authors26,27 have not found significant differences in blood pressures variations during the recovery period when they compared the reactions of hypertensive and normotensive subjects. In our study we did not compare hypertensive vs normotensive subjects; we compared the variations of BP in hyper-reactors subjects vs normoreactors subjects, independently of the levels of baseline BP. Some authors27 have reported the hypothesis that the recovery time could be an important index in describing the sustained 24-h DBP. We can speculate that an initial impairment of vasodilator capacity might be the cause of a lower recuperation of baseline BP during the recovery period observed in our study in subjects who developed SH.

In summary, the BP hyper-reactivity to SI was observed to be an independent predictor of future SH after 5 years of follow-up of subjects with mild hypertension.



  1. Horan MJ, Lanfant C. Epidemiology of blood pressure and predictors of hypertension. Hypertension 1990; 15 (Suppl. 1): I-20–I-24.
  2. Skarfors ET, Lithell HO, Selinus I. Risk factors for the development of hypertension: a 10-year longitudinal study in middle-aged men. J Hypertens 1991; 9: 217–223.
  3. Leitschuh M et al. A. High-normal blood pressure progression to hypertension in the Framingham Heart Study. Hypertension 1991; 17: 22–27. | PubMed | ISI | ChemPort |
  4. Vasan RS et al. Assessment of frequency of progression to hypertension in nonhypertensive participants in the Framingham Heart Study: a cohort study. Lancet 2001; 358: 1682–1686. | Article | PubMed | ISI | ChemPort |
  5. Jamerson K, Julius S. Predictors of blood pressure and hypertension. General principles. Am J Hypertens 1991; 4: 598S–602S.
  6. Watt GCM, Foy CJW, Holton DW, Edwards HE. Prediction of high blood pressure in young people. J Hypertens 1991; 9: 55–58.
  7. Hunt SC, Stephenson SH, Hopkins PN, Williams RR. Predictors of an increased risk of future hypertension in Utah. A screening analysis. Hypertension 1991; 17: 969–976. | PubMed | ISI | ChemPort |
  8. Nakanishi N et al. Risk factors for the development of hypertension: a 6-year longitudinal study of middle-aged Japanese men. J Hypertens 1998; 16: 753–759. | PubMed |
  9. Shapiro AP. Hypertension and Stress. A Unified Concept. Lawrence Erlbaum Associates: Mahwah, NJ, 1996.
  10. Bedi M, Varshney VP, Babbar R. Role of cardiovascular reactivity to mental stress in predicting future hypertension. Clin Exp Hypertens 2000; 22: 1–22. | Article |
  11. Pickering TG, Gerin WG. Cardiovascular reactivity in the laboratory and the role of behavioral factors in hypertension: a critical review. Ann Behav Med 1990; 12: 3–16.
  12. Falkner B, Onesti G, Hamstra B. Stress response characteristics of adolescents with high genetic risk for essential hypertension: a five-year follow-up. Clin Exp Hypertens 1981; 3: 583–591. | PubMed |
  13. Falkner B, Kushner H, Onesti G, Angelakos ET. Cardiovascular characteristics in adolescents who develop essential hypertension. Hypertension 1981; 3: 521–527.
  14. Borghi C, Costa FV, Mussi A, Ambrosioni E. Predictors of stable hypertension in young borderline subjects: a five-year follow-up study. J Cardiovasc Pharmacol 1986; 8 (Suppl 5): S138–S141.
  15. Borghi C et al. Factors associated with the development of stable hypertension in young borderline hypertensives. J Hypertens 1996; 14: 509–517.
  16. Light KC, Dolan CA, Davis MR, Sherwood A. Cardiovascular responses to an active coping challenge as a predictors of blood pressure patterns 10 to 15 years later. Psychosom Med 1992; 54: 217–230.
  17. Murphy JK, Alpert BS, Walker SS. Ethnicity, pressor reactivity, and children's blood pressure: five years ob observations. Hypertension 1992; 20: 327–332.
  18. Matthews KA, Woodall KL, Allen MT. Cardiovascular reactivity to stress predicts future blood pressure status. Hypertension 1993; 22: 479–485. | PubMed | ChemPort |
  19. Markovitz JH et al. Cardiovascular reactivity to video game predicts subsequent blood pressure increases in young men: the Cardia Study. Psychosom Med 1998; 60: 186–191.
  20. Carroll D et al. Pressor reactions to psychological stress and prediction of future blood pressure: data from the Whitehall II Study. BMJ 1995; 310: 771–776.
  21. 1999 World Health Organization-International Society of Hypertension. Guideliness for the Management of Hypertension. J Hypertens 1999; 17: 151–183. | Article | PubMed | ISI |
  22. Armario P et al. Blood pressure response to mental stress in young subjects with high or slightly high blood pressure. Does it reflect the changes in blood pressure observed during outpatient monitorization? Med Clin (Barc) 1994; 102: 547–652. [in Spanish].
  23. Armario P et al. Relationship between cardiovascular reactivity to mental stress and early target organ damage in untreated mild hypertension. Hospitalet Study. Med Clin (Barc) 1999; 113: 401–406. [in Spanish].
  24. Light KC et al. High stress responsitivity predicts later blood pressure only in combination with positive family history and high life stress. Hypertension 1999; 33: 1458–1464. | PubMed | ISI | ChemPort |
  25. Linden W, Earle TL, Gerin W, Christenfeld N. Physiological stress reactivity and recovery: conceptual siblings separated at the birth. J Psychosom Res 1997; 42: 117–135. | Article |
  26. Köhler TH, Fricke M, Ritz Th, Scherbaum N. Psychophysiological reactivity of borderline hypertensives and their recovery after mental stress. Psychother Psychosom 1997; 66: 261–267.
  27. Guasti L et al. Reactivity of blood pressure to mental arithmetic stress test stress-test, recovery time, and ambulatory blood pressure in hypertensive and normotensive subjects. Blood Press Monit 1998; 3: 275–280.

Extra navigation