The prevalence of pre-eclampsia and obstetric outcome in pregnancies of normotensive and hypertensive women attending a hospital specialist clinic.

To study the prevalence of pre-eclampsia (PE) and other obstetric outcomes (growth restriction and fetal mortality) in pregnancies of normotensive and hypertensive women attending an antenatal hypertension clinic, we studied a cohort of 372 pregnancies from 267 women. The prevalence of PE in the groups of pregnancies of normotensive and chronic hypertensive women was 11.9% (19/159 cases) and 16.0% (34/213 cases) respectively (chi 2 = 1.2, p = 0.27). There were no significant differences in respect of ethnicity, being primi- or multigravida and smoking status or age. Treatment with antihypertensive drugs during pregnancy did not decrease the prevalence of PE. In pregnancies with hypertensive complications (with or without PE) there was a trend towards higher rates of pre-term delivery (< 37 weeks), caesarean section, small for gestational age babies, stillbirth and lower baby birth weight and ponderal index values. Pregnancies in women with uncomplicated hypertension had an increased risk for emergency caesarean section, pre-term delivery (< 37 weeks), birth weight < 2500 g and stillbirth (relative risks [with confidence intervals] 2.5 [1.9-3.2], 2.3 [1.8-2.9], 3.1 [2.5-3.7] and 5.5 [2.6-11.9] respectively) compared with the general hospital obstetric population. After classification according to the type of hypertensive syndrome, a progressively higher risk for fetal growth restriction and adverse perinatal outcome was shown in the hypertensive and pre-eclamptic groups. In chronic hypertension, this was irrespective of superimposed pre-eclampsia or antihypertensive therapy. The high prevalence of PE in chronic hypertensive women (16.0%) was not statistically significant to that of normotensive women (11.9%), reflecting the referral selection of 'high risk' normotensive women to our clinic.

The measurement of blood pressure and the detection of hypertension in children and adolescents.

Despite the publication of several expert committee guidelines for the measurement of blood pressure (BP) and the diagnosis of hypertension in children and adolescents, it was our perception and clinical experience that there still appeared to be a general lack of standardisation of BP measurement techniques and little consensus on the criteria for diagnosing hypertension. To investigate this further, we have conducted a postal survey of consultant-grade paediatricians who were members of the British Paediatric Association (BPA). A total of 1500 questionnaires were sent out and 708 analysable replies were received (47.1%). This showed that 68.6% of paediatricians routinely measured BP, at least on one occasion, in children or adolescents attending their outpatient clinics, 17.7% started at or soon after birth, 12.3% started at the age of 1 year, 20.0% at 3 years, 12.0% from 7 years of age and 3.5% from the age of 13. Only 60.5% reported that they had a choice of four or more different cuff sizes in their clinic. Forty-one percent of respondents reported that the BP was always or sometimes measured by nurses. Fifty-one percent of respondents measured diastolic BP at the phase of muffling of sound (Korotkoff phase IV), 31.9% used the disappearance of sound (phase V) whilst 15.9% claimed that they measured both end-points. The criteria for diagnosing a child as being hypertensive varied greatly; 17.9% reported that they responded to the systolic BP alone, 13.5% to the diastolic BP alone, 65.9% relied on both pressures, and 2.7% responded to either the systolic or diastolic pressure if it was raised. Furthermore, 12.9% diagnosed hypertension if the BP exceeded the 90th percentile in relation to age and 41.8% used the 95th percentile. However 45.3% of respondents employed a higher dividing line. In hospitalised children, leg blood pressures were measured routinely by 30.3%, although a further 44.0% would do so if aortic coarctation or other vascular diseases were suspected. Despite considerable variation in clinical practice, techniques and criteria, only 11.4% of clinicians would manage the patients themselves, with the remainder referring the child on to the appropriate specialist. The survey suggests a general lack of standardisation of BP measurement techniques and little consensus on the criteria for diagnosing hypertension amongst paediatricians. Simplified, shortened and updated guidelines on hypertension in paediatric practice and research are needed.

Do patients with de novo hypertension differ from patients with previously known hypertension when malignant phase hypertension occurs?

Malignant phase hypertension (MHT) represents the most severe form of hypertension, and many consider that this condition only occurs in poorly managed patients with previously known hypertension. To investigate this further, we studied 350 patients with MHT on the West Birmingham MHT database: 195 (55.7%) of these presented de novo, without any known past history of hypertension (Group 1), and 146 (41.7%) were previously known hypertensives (Group 2), of whom 86 were receiving antihypertensive therapy; in 9 patients, the status was uncertain. Median duration of clinical followup was similar in both groups (36.0 v 37.5 months, Mann-Whitney test P = .795). Patients presenting de novo with MHT (Group 1) were younger, with a predominance of whites and men. Nevertheless, the clinical features, blood pressures, and renal function at presentation were similar to MHT patients with previously known hypertension. Renal function at follow-up was also similar in both groups. There was an excess of women and nonwhites in MHT patients with previously known hypertension (Group 2), who also had higher mean follow-up blood pressures. On univariate life-table analysis, there was no statistically significant difference in survival time between Groups 1 and 2 (mean 57.5 v 63.5 months, median 36.0 v 37.0 months; log-rank test, P = .456). Using a multivariate Cox analysis of baseline variables, the independent predictors of outcome (death or dialysis) were age at presentation (P = .0019), diastolic blood pressure (P = .0466), serum urea (P = .006), and serum creatinine (P < .001). Whether the patient had presented de novo, without any known history of hypertension (Group 1) or had previously known hypertension (Group 2) did not independently predict outcome (P = .6549). We suggest that MHT can occur de novo in patients without previously known hypertension, and the clinical characteristics and prognosis in such patients were similar to MHT patients with previously known hypertension.

Serum urate is associated with baseline renal dysfunction but not survival or deterioration in renal function in malignant phase hypertension.

BACKGROUND: There has been speculation whether serum uric acid levels are an independent prognostic factor in patients with hypertension. OBJECTIVE: To investigate the clinical associations and prognostic value of serum urate in patients with malignant phase hypertension (MHT), by comparing clinical features in patients with serum urate levels above and below the median levels for this population, and secondly, by performing a survival analysis in these patients. PATIENTS AND METHODS: Review of the data on 153 patients (98 males; mean age 50.3 years, SD 13.5) with MHT on the west Birmingham MHT register. Median uric acid levels in this population was 0.41 mmol/l (6.9 mg/dl), with an interquartile range of 0.34-0.50 mmol/l (5.7-8.4 mg/dl). Clinical characteristics of patients with a serum urate <0.41mmol/1 (group 1) were compared to those with levels above the median (0.41 mmol/l, group 2). RESULTS: Mean duration of follow-up was similar in both groups. The mean diastolic blood pressure at presentation and both mean systolic and diastolic blood pressures at follow-up were significantly higher in group 2 (that is, those with high serum urate levels) (unpaired t test, P= 0.039). There was also more renal dysfunction in group 2 patients with MHT, with higher mean serum urea and creatinine levels, both at presentation and at follow-up (unpaired t test, P< 0.01). The commonest causes of death were myocardial infarction (n = 7), heart failure (n = 4), stroke (n = 10) and renal failure (n = 5). There was no difference in mean survival duration between groups 1 and 2 (Kaplan-Meier, 64.6 versus 66.8 months; log-rank test, P= 0.519). Serum urate levels also did not predict the rise in serum creatinine levels (log-rank test, P= 0.84) or urea (P= 0.4033) amongst these patients. Using a multivariate Cox proportional hazards analysis, the only independent predictors of outcomes (death or the need for dialysis) were age (P = 0.007) and serum creatinine levels at presentation (P = 0.0046). CONCLUSION: Our analysis of a large series of patients with MHT shows that those with high urate levels had higher diastolic blood pressures and greater renal impairment at baseline. At follow-up, patients with median serum urate >0.41 mmol/l showed a greater deterioration in renal function and higher blood pressures, but no significant difference in survival. Serum urate levels also do not appear to be predictive of the deterioration in renal function or overall survival in patients with MHT.

The placebo effect and white coat effect in isolated systolic hypertension and systo-diastolic hypertension.

It is well recognised from many clinical trials that there is a blood pressure lowering effect when placebo is administered to patients with essential hypertension ("placebo effect"). The reduction in blood pressure, however, may also be partly due to loss of the alerting response ("white coat effect") as a result of familiarisation with the clinical environment. To investigate the hypothesis that there may be a more marked placebo effect and white coat effect in isolated systolic hypertension (ISH) compared with systo-diastolic hypertension (SDH), we studied 78 patients with hypertension: 34 had ISH and 44 patients had SDH. The 34 patients with ISH were older (68.7 vs 54.9 years), had a higher SBP (192.2 vs 169.6 mmHfg) and lower DBP (85.5 vs 102.0 mmHg) when compared to patients with SDH. Amongst the patients with ISH, there were no significant changes in mean blood pressures pre-placebo (paired t-test, p = NS). In the placebo period, there was a significant reduction in systolic blood pressures at all three points, and a significant reduction in diastolic blood pressures after 2 and 3 months placebo (paired t-test, p < 0.05). There was a mean reduction in mean systolic blood pressure at visit 1 by 5.2%, visit 2 by 5.1% and visit 3 by 4.6%, when compared to mean pre-placebo systolic blood pressures (p < 0.05). The mean reduction in diastolic blood pressure was 5.8% at visit 2 and 3.5% at visit 3, when compared to mean pre-placebo diastolic blood pressure (p < 0.05). At the 4-week visit after receiving placebo, the mean systolic blood pressure decreased by 9.4 mmHg (p = 0.003) and mean diastolic blood pressure by 2.7 mmHg (p = NS) in the patients with ISH. In patients with SDH, there were no statistically significant changes in recorded BP readings following the introduction of placebo. We suggest blood pressures in some patients with ISH may settle with careful follow up and initiation of treatment in these patients could potentially be delayed for at least 3 months, as therapy may not prove necessary.

White coat hypertension and carotid atherosclerosis.

To investigate further the relationship between atherosclerotic vascular disease and blood pressure, and the phenomenon of white coat hypertension, we performed a cross-sectional study of patients referred for carotid Doppler scanning, to determine the relationship between ambulatory blood pressure monitoring (ABPM) and carotid atherosclerosis. We studied 79 patients (51 men, 28 women) undergoing Doppler ultrasound examination of the carotid arteries: 44 (56%) had evidence of carotid atherosclerosis on Doppler ultrasound examination ("disease group"), whilst 35 (44%) had normal carotid arteries ("controls"). "Adequate" ABPM recordings, defined by > 90% of recordings over the 24 h, were available in 51 patients (30 positive, 21 negative). There were no significant differences in mean daytime, mean night-time or mean 24 h ABPM recordings between those with and without carotid atherosclerosis. Mean manual clinic systolic blood pressure was significantly greater in those with carotid atherosclerosis than in controls (146.7 +/- 25.2 vs 131.1 +/- 35 mmHg, p < 0.005). In patients with carotid atherosclerosis, the first systolic blood pressure ABPM recording was not significantly different from the mean manual clinic recording (mean difference -1.5 mmHg, 95% confidence interval (CI) -7.9 to 4.8 mmHg). The initial diastolic blood pressure ABPM recording was significantly higher than the mean manual recording. Carotid atherosclerosis was identified in 53% of normotensive controls compared with 56% of white coat hypertensives and 75% of persistent hypertensives. One-third (9/27) of the patients with normal carotid arteries did not have nocturnal dipping ("non-dippers") compared with 50% (12/24) of the atherosclerotic patients. This study suggests that carotid atherosclerosis may be associated with white coat hypertension. Our study adds to the body of evidence that white coat hypertension is associated with end-organ damage and is not simply a benign disease. Such patients should be screened for other cardiovascular risk factors and should be monitored for the development of persistent hypertension.

Atenolol and fetal growth in pregnancies complicated by hypertension.

Atenolol use may be associated with growth retardation when given in pregnancy, although the relationship to trimester of initiation, duration of treatment, and its use as monotherapy is still uncertain. To compare the obstetric and fetal outcome between women receiving atenolol (as monotherapy) and other antihypertensive drug monotherapies, and also to investigate the effect of duration of treatment on fetal growth, we performed a retrospective cohort study of 312 pregnancies in 223 women attending an Antenatal Hypertension Clinic. Atenolol (as monotherapy) was given in 78 pregnancies (25.0%), other types of antihypertensive drugs as monotherapy were given in 53 pregnancies (17.0%), and multiple drug combinations were given in 90 pregnancies (28.8%). In 91 pregnancies (29.2%) no antihypertensive drugs were given. Atenolol was found to be associated with lower birth weight and ponderal index values, with a trend toward a higher prevalence of preterm (<37 weeks) delivery and small-for-gestational-age babies when compared to other antihypertensive drugs as monotherapy, or to no treatment. The adverse effect of atenolol was more pronounced in women receiving the drug earlier in their pregnancy, and continuing the drug for a longer duration. In conclusion, atenolol should be avoided in the early stages of pregnancy and given with caution at the later stages, as it is associated with fetal growth retardation, which is related to duration of treatment.

Obstetric and neonatal outcome following chronic hypertension in pregnancy among different ethnic groups.

We retrospectively studied pre-eclampsia rate and obstetric outcome in a cohort of 436 pregnancies amongst 318 women of different ethnic backgrounds attending an antenatal hypertension clinic from 1980-1997, identifying 152 women (213 pregnancies) with chronic essential hypertension. The ethnic breakdown was: White, 64 (30.0%) pregnancies in 48 (31.5%) women; Black/Afro-Caribbean, 79 (37.1%) pregnancies in 56 (36.8%) women; and Indo-Asians, 70 (32.3%) pregnancies in 48 (31.6%) women. The prevalences of pre-eclampsia in White, Black and Indo-Asian women were 17.2%, 12.7% and 18.6%, respectively (p = 0.58). Pregnancies of Indo-Asian women were of shorter gestation, and babies in this group also had lower birth weight and ponderal index compared to those of White and Black women (all p < 0.05). The proportions of overall perinatal mortality were 1.6% for Whites (1/64), 3.8% for Blacks (3/79) and 10.0% for Indo-Asians (7/70), suggesting increased risk in the Indo-Asian group. Indo-Asian women with chronic essential hypertension need careful antenatal care and observation during pregnancy.

Does renal function improve after diagnosis of malignant phase hypertension?

BACKGROUND: The nature of the relationship between high blood pressure and kidney damage is controversial. There is a view that essential non-malignant hypertension does not cause renal damage in patients with normal renal function and no proteinuria at first presentation. In contrast, patients with malignant hypertension (MHT) not due to underlying renal disease often exhibit progressive renal impairment, although there can be a dissociation between short-term changes in renal function and the long-term renal outcome. It has been established that the association of MHT and renal impairment is associated with a grave prognosis. OBJECTIVE: To investigate further the changes in renal function after diagnosis of MHT. METHODS: We studied the clinical features, renal function and survival of patients on the West Birmingham MHT Register. Patients presenting with no or only mild-to-moderate renal impairment (defined as an initial serum creatinine level < 300 micromol/L) were categorized as group 1 patients. Patients with severely impaired renal function, defined as a serum creatinine level > or = 300 micromol/l at presentation, were categorized as group 2 patients. Patients who subsequently suffered a deterioration in renal function were defined as those whose serum creatinine level had risen at follow-up (group A). Patients with an invariant or improving renal function at follow-up were those whose serum creatinine level neither rose nor fell (group B). Clinical characteristics and survival of patients in group 1 were therefore compared with those of patients in group 2; features of patients in group A were compared with those of patients in group B. RESULTS: We studied a total of 169 patients with MHT [107 men, aged 49.4 +/- 13.3 years (mean +/- SD)]. Of these, 136 (80.5%) patients had an initial serum creatinine level < 300 micromol/l (group 1). After a median follow-up of 53 months (interquartile range 15-103), the serum creatinine level of 56.8% (96 of 169) patients had risen (group A). Patients with MHT with only mild-to-moderate renal impairment at presentation (creatinine level < 300 micromol/l, group 1) had lower serum urea, creatinine and blood pressure levels at follow-up compared with those of patients with initially impaired renal function. Although there was no significant change in follow-up serum urea and creatinine levels compared with initial levels in patients in group 2 (paired Wilcoxon test, NS), there was a significant deterioration in renal function in patients in group 1 (P = 0.001). Group 2 patients had a shorter median survival time than did group 1 patients (15 versus 59 months, Lee-Desu statistic 15.4, P = 0.0001). Patients whose serum creatinine level had risen (group A) had significantly higher initial serum urea and creatinine levels and follow-up blood pressures than did those whose renal function had remained invariant or improved (group B). However, there was no significant difference between the median survival times of patients in groups A and B (51 versus 58 months, Lee-Desu statistic 0.377, P = 0.54). CONCLUSION: We suggest that renal function continues to deteriorate in some patients with MHT, despite a good degree of control of their blood pressures having been achieved at follow-up. However, the renal function of 16 of the 33 patients with severe renal impairment at presentation either remained invariant or was found to have improved at follow-up. There was no evidence that those cases whose renal function remained invariant were confined to those who had presented with less renal impairment. The severity of MHT at presentation and the amount of renal impairment did not predict the outcome. In contrast the quality of control of the blood pressure that had been obtained at follow-up did predict the outcome. Careful monitoring of renal functioning and effective treatment of the blood pressure is mandatory in patients with MHT.

Effect of atenolol on birth weight.

To investigate the possible harmful effects of early antihypertensive drug therapy with atenolol versus other therapies on pregnancy outcome, we reviewed the records of 398 women referred to our antenatal hypertension clinic between 1980 and 1995. Babies born to women taking atenolol were significantly lighter than babies born to women taking other beta blockers, other antihypertensive drugs, or no therapy, suggesting that atenolol might be detrimental in early pregnancy.

Malignant hypertension in young women is related to previous hypertension in pregnancy, not oral contraception.

Previous studies have suggested that one-third of women of childbearing age who develop malignant phase hypertension (MHT) are likely to be taking oral contraceptives (OC). We surveyed 104 women with a history of MHT. None of the 65 aged > 45 years were taking OC or other sex hormones. Thirty-nine (mean age 34.9 years, SD 8.0) were aged 15-44 years at presentation: 22 Caucasian, 10 Black/Afro-Caribbean and seven Indo-Asian. Of these 39, 22 had a history of hypertension in pregnancy (group 1), and 17 did not (group 2). Three of group 1 also had a history of OC-induced hypertension. None were pregnant, but one was taking an OC at presentation with MHT. Blood pressures at presentation and follow-up, and mean serum urea and creatinine at presentation were similar between groups, as was median survival (96 vs. 47 months, Lee-Desu statistic 0.75, p = 0.38). There was a trend towards poorer renal function at follow-up in group 1 patients, with higher mean serum urea and creatinine levels. The causes of death were renal failure (5), stroke (4) and heart disease (2). The OC was not a common cause of MHT-amongst our sample of women of childbearing age, but a past history of hypertension in pregnancy was important. Such patients also had a longer duration of hypertension and poorer renal function at follow-up.

Ambulatory blood pressure monitoring in atrial fibrillation.

Ambulatory blood pressure monitoring (ABPM) devices are increasingly used in the assessment of hypertension, but patients with atrial fibrillation are usually excluded because device accuracy in this patient group is unproved. The present study investigates the clinical use of the Spacelabs 90207 oscillometric ABPM device in outpatients with chronic atrial fibrillation and suggests that such devices can be used in clinical practice to assess blood pressure in stable outpatients with chronic atrial fibrillation.

Soluble adhesion molecule P-selectin and endothelial dysfunction in essential hypertension: implications for atherogenesis? A preliminary report.

OBJECTIVE: Patients with essential hypertension are at high risk of atherosclerotic vascular disease. To investigate this further, we measured levels of the soluble adhesion molecule P-selectin, which is associated with platelet activity/function and atherosclerosis, von Willebrand factor, which is a marker of endothelial dysfunction, and plasma fibrinogen. PATIENTS AND METHODS: We studied 104 consecutive patients (47 males, 57 females; mean +/- SD age 54.8 +/- 14.1 years) with essential hypertension compared with 47 normotensive healthy controls (55.0 +/- 19.2 years). Levels of soluble adhesion molecule P-selectin and von Willebrand factor were measured by enzyme-linked immunosorbent assay, and plasma fibrinogen by a clotting method (CLAUSS). RESULTS: Compared with normotensives, the hypertensives showed significant increases in soluble P-selectin (300 versus 228 ng/ml; median difference 55 ng/ml, Mann-Whitney test P = 0.03), von Willebrand factor (114 versus 96 IU/I; unpaired t-test P < or = 0.001) and fibrinogen (3.3 versus 2.9 g/l; unpaired t-test P < or = 0.001). There were significant correlations between fibrinogen and P-selectin (r = 0.16; P = 0.02) and von Willebrand factor (r = 0.39; P<0.001), but not between P-selectin and von Willebrand factor. There were no differences in these factors between patients with (n = 53) and without (n = 51) antihypertensive therapy or between those with good blood pressure control (systolic/diastolic < or = 160/90 mmHg; n = 17) and those with poor control. A stepwise multiple regression analysis showed that diastolic blood pressure was a significant predictor for soluble P-selectin levels; diastolic blood pressure and von Willebrand factor levels were significant predictors for fibrinogen levels (P<0.05). CONCLUSIONS: This study suggests that hypertensives have high plasma fibrinogen levels, platelet dysfunction (which could contribute to atherogenesis, as indicated by raised soluble P-selectin levels) and endothelial dysfunction (as indicated by high von Willebrand factor levels), which are related to diastolic blood pressure. These factors may act synergistically to increase atherogenesis and may explain the high risk of atherosclerotic vascular disease in hypertensives.

Severe hypertension with lone bilateral papilloedema: a variant of malignant hypertension.

Patients with severe hypertension with retinoscopic bilateral papilloedema only are not classically regarded as having malignant hypertension (MHT). We have encountered 23 such patients between 1965-1993, whilst over a similar period we have seen 315 patients who fulfilled the conventional criteria for MHT with bilateral retinal haemorrhages, exudates with or without papilloedema. We hypothesised that patients with "lone" papilloedema and severe hypertension were suffering from a disease which was identical in aetiology and outcome to conventional MHT. There were no significant differences in age, mean blood pressure, proteinuria or renal function at presentation, ethnic composition, smoking status and followup blood pressure control between the papilloedema group and those presenting with conventional MHT. Clinical features at presentation in the papilloedema only group included strokes in 4, visual disturbance in 2, headaches in 3 and heart failure in 1 patient. Many patients however had no complications at presentation. After a mean followup of 59.8 months, of the "lone" papilloedema group, 7 patients (30.4%) were still alive, 1 patient was on renal dialysis therapy, 13 were dead (56.5%) and 2 (8.7%) were lost to followup. The commonest causes of death were stroke in 4 patients, renal failure in 4 and heart disease in 2. This was a similar pattern of mortality to those patients with "conventional" MHT. Lifetable analyses showed a median survival of 35.9 months for the papilloedema group which was significantly worse than the 108.7 months for the conventional MHT group (Lee-Desu statistic 4.04, p = 0.045). We suggest that patients with high blood pressure and lone bilateral papilloedema may comprise a hitherto unrecognised subgroup of patients with MHT. Once intracerebral pathology has been excluded, these patients need to be treated as aggressively as those with MHT.