The incidence of chronic hypertension varies among different populations, ranging from 0.5-4% and averaging 2.5%. Chronic hypertension in pregnancy is usually idiopathic (80%) or due to renal disease (20%), though these figures may reflect insufficient investigation. A number of renal diseases may be causative, the most common being chronic glomerulonephritis, interstitial nephritis, diabetic glomerulosclerosis, IgA nephropathy, and renal artery stenosis.
A. Symptoms and Signs
Patients with chronic hypertension tend to be over 30 years of age, obese, and multiparous, with associated medical problems such as diabetes or renal disease. The incidence is higher in black women and in women with a family history of hypertension. A woman who has delivered one or more infants and has hypertension in this pregnancy most likely has chronic hypertension. The typical patient has hypertension without other signs of preeclampsia (eg, proteinuria or nondependent edema).
The diagnosis is made on the basis of documented hypertension before conception or before 20 weeks' gestation or persistence of hypertension after the puerperium (6 weeks). The diagnosis of chronic hypertension should be confirmed by multiple measurements including home and/or out-of-office blood pressure readings as recommended in The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. If hypertension is severe (stage 3, systolic pressure 180 mm Hg or diastolic pressure 110 mm Hg), the patient should be evaluated for reversible causes. Whether worsening hypertension represents superimposed preeclampsia or hypertension associated with renal disease is sometimes difficult to determine. Preexisting renal disease alone may have all the manifestations of preeclampsia (hypertension, edema, proteinuria, and hyperuricemia). Renal biopsy would confirm the diagnosis but is usually not necessary, because the decision to deliver can be based on difficulty of blood pressure control, renal function, and fetal well-being. For the same reasons, renal biopsy is usually not performed for the work-up of proteinuria or elevated serum creatinine in pregnancy.
Hypertensive States of Pregnancy
B. Laboratory, X-Ray, and Electrocardiographic Findings
The ECG may show left ventricular hypertrophy in 5-10% of patients. Elevated serum creatinine, decreased creatinine clearance, and proteinuria are also present in about 5-10% of patients with chronic hypertension. The chest x-ray is usually normal, though it may reveal cardiomegaly. Patients with left ventricular hypertrophy or elevated serum creatinine are at increased risk for developing superimposed preeclampsia. Patients with cardiomegaly due to either hypertensive cardiovascular disease or congestive cardiomyopathy are at increased risk for superimposed preeclampsia, pulmonary edema, and arrhythmias.
A. Maternal Complications
Hypertensive states in pregnancy include preeclampsia-eclampsia, chronic hypertension (either essential or secondary to renal disease, endocrine disease, or other causes), chronic hypertension with superimposed preeclampsia, and gestational hypertension ...
The main complication of chronic hypertension is superimposed preeclampsia, which occurs in about one-third of patients. Patients tend to deteriorate faster with superimposed preeclampsia than with preeclampsia alone. There is an increased risk of abruptio placentae with chronic hypertension (0.4-10%). Associated with this condition is the risk of disseminated intravascular coagulation, acute tubular necrosis, or renal cortical necrosis.
The effect of pregnancy on chronic renal disease is uncertain. Although there are few data for patients with severe disease, limited evidence suggests that if renal function is well preserved (creatinine < 1.5 mg/dL), pregnancy does not change the course of renal disease, but if renal insufficiency exists prior to pregnancy (creatinine > 1.5 mg/dL), the decline in renal function may be more rapid than expected.
B. Fetal Complications
The fetus has a 25-30% risk of prematurity and a 10-15% risk of growth restriction. Preeclampsia tends to occur after 34 weeks' gestation, so that prematurity is not a great concern. Preeclampsia superimposed on chronic hypertension frequently occurs earlier (at 26-34 weeks), and in such cases, fetuses are at double jeopardy for prematurity and intrauterine growth retardation. In addition, there is a risk of stillbirth or intrapartum fetal distress due to abruptio placentae or chronic intrauterine asphyxia.
A. Control of Hypertension
Most authorities agree that antihypertensive therapy will decrease the incidence of stroke and heart failure in pregnant patients with diastolic blood pressures exceeding 110 mm Hg. The real controversy concerns the value of antihypertensive therapy of mild hypertension (approximately 95% of pregnant patients with chronic hypertension have mild hypertension). One study demonstrated that treatment of diastolic blood pressures of 104-115 mm Hg in men decreased cardiovascular morbidity (myocardial infarction, congestive heart failure, and stroke) in just 10 months. Patients with diastolic pressures of 94-104 mm Hg showed benefits of therapy only after 5 years had elapsed. Therefore no benefits of antihypertensive therapy for mild chronic hypertension could be expected during the 9 months of pregnancy, and therapy cannot be justified by the same arguments used in general internal medicine. Some authors claim that antihypertensive therapy for mild chronic hypertension will decrease the incidence or delay the onset of superimposed preeclampsia, thus lowering perinatal mortality and morbidity rates. Others claim there is no benefit and considerable risk. Since this issue is still unresolved, a review of some of the recent clinical studies is helpful (see Reference section).
Several oral agents may be considered if hypertension is to be treated.
1. Thiazide diuretics - Thiazide diuretics have been reported to cause a number of harmful maternal and fetal side effects, the main one being plasma volume contraction. Studies in nonpregnant patients show that thiazides have acute and chronic effects. Acutely, they cause a 5-10% decrease in plasma volume, which lowers cardiac output and blood pressure in the first 3-5 days. Over the next 4-6 weeks, renal compensatory mechanisms return the plasma volume toward normal but not pretreatment levels. At the same time, cardiac output returns to pretreatment levels, but total peripheral resistance stays low. Thus, the acute blood pressure-lowering effect of thiazides is due to volume contraction. The sustained antihypertensive effect is thought to involve mobilization of excess sodium from the arterial wall. This leads to widening of the vascular lumen and possibly to a decrease in the vascular responsiveness to endogenous catecholamines. One set of researchers showed that plasma volume contraction occurs in early pregnancy in hypertensive patients on chronic thiazide therapy. When the thiazide was stopped, normal physiologic volume expansion occurred; if the thiazide was continued, plasma volume expansion was minimal (18% mean increase in patients taking thiazides versus 52% mean increase in patients in whom diuretics were discontinued early in pregnancy). Perinatal outcome was the same in both groups. Another consideration is the volume expansion caused by antihypertensive agents. It may be that the sodium and water retention produced by antihypertensive agents offsets the volume contraction caused by the thiazide. In summary, diuretics do not prevent preeclampsia or eclampsia. Thiazide diuretics are contraindicated in patients with pure preeclampsia. They may have a place in the treatment of patients with chronic hypertension; however, with the availability of more powerful antihypertensive agents such as nifedipine and labetalol, their use is declining.
2. Methyldopa - Methyldopa, a central α-adrenergic agonist, is the only antihypertensive drug whose long-term safety for mother and fetus has been adequately assessed. It reduces total peripheral resistance without causing physiologically significant changes in heart rate or cardiac output. If methyldopa is used alone, fluid retention and loss of antihypertensive effect are frequent. For this reason, methyldopa is usually combined with a diuretic for treatment of nonpregnant patients. It is usually started at a dose of 250 mg 3 times a day and increased to 2 g/d. Peak plasma levels occur 2-3 hours after administration; the plasma half-life is about 2 hours, and the maximum effect occurs 4-6 hours after an oral dose. Most of the agent is excreted via the kidney. The most commonly reported side effects are sedation and postural hypotension. With prolonged therapy, 10-20% of patients develop a positive direct Coombs' test, usually after 6-12 months of therapy. Hemolytic anemia occurs in fewer than 5% of these patients and is an indication to stop the drug. Fever, liver function abnormalities, granulocytopenia, and thrombocytopenia have occurred rarely.
3. Clonidine - Clonidine is another central α-adrenergic agonist. Treatment is usually started at 0.1 mg twice daily and increased in increments of 0.1-0.2 mg/d up to 2.4 mg/d. Blood pressure declines 30-60 mm Hg with use of clonidine, with a maximum effect in 2-4 hours and a duration of action of 6-8 hours. Renal blood flow and the glomerular filtration rate are preserved, but cardiac output falls. This is attributable to a decrease in venous return secondary to systemic vasodilatation and bradycardia. Cardiac output responds normally to exercise. Xerostomia and sedation are the most frequently encountered side effects. Withdrawal of clonidine produces a hypertensive crisis that responds well to reinstitution of the drug. There is not as much information on clonidine in pregnancy as there is on methyldopa; one large study found it to be equivalent to methyldopa.
4. Calcium channel blockers - Currently available calcium channel blockers include nifedipine, verapamil, diltiazem, nicardipine, isradipine, amlodipine, and felodipine. They cause direct arteriolar vasodilation by selective inhibition of slow inward calcium channels in vascular smooth muscle. Since calcium channel blockers affect such a fundamental cellular response, their therapeutic applications are wide-ranging, from angina pectoris to premature labor. Nifedipine is the calcium channel blocker most widely used in pregnancy. Ninety percent of oral nifedipine is absorbed from the gastrointestinal tract. After moderate first-pass liver metabolism, the bioavailability is 65-70%. Onset of action after bite-and-swallow administration is about 3 minutes. The drug has an initial fast half-life of 2.5-3 hours and a terminal slow half-life of 5 hours. It is almost completely metabolized by the liver and excreted 90% by the kidney and 10% by the liver. Side effects include hypotension, headache, flushing, tachycardia, and ankle edema. Since magnesium sulfate is also a calcium channel blocker, the use of both nifedipine and magnesium sulfate together could be potentially hazardous (eg, hypotension). Nifedipine has been used in several human studies comparing its tocolytic effect with ritodrine. It has also been used both acutely and chronically as an antihypertensive agent in pregnancy. In the largest study to date, one set of researchers randomly treated 200 preeclamptic patients with nifedipine and bed rest or bed rest alone. There was no prolongation of pregnancy or improved perinatal outcome in the nifedipine group, but uncontrolled hypertension as an indication for delivery was reduced. Because it is such a powerful and dependable agent, nifedipine is becoming increasingly popular for antihypertensive therapy in pregnancy.
5. Prazosin - Prazosin is a competitive blocker of the postsynaptic α1-adrenergic receptor. It causes vasodilatation of both the resistance and capacitance vessels, reducing cardiac preload and afterload. It lowers blood pressure without significantly lowering heart rate, cardiac output, renal blood flow, or the glomerular filtration rate. It is almost exclusively metabolized in the liver. Approximately 90% of the drug is excreted via bile into the feces. It appears to be more slowly absorbed and its half-life slightly prolonged during pregnancy. In one study, the median time to peak concentration was 165 minutes in pregnant women and 120 minutes in men of similar age. The mean elimination half-life was 171 minutes in pregnant women and 130 minutes in men. Prazosin may cause a first-dose phenomenon characterized by sudden hypotension 30-90 minutes after the initial dose. This can be avoided by limiting the first dose to 1 mg given just prior to bedtime. Animal studies have demonstrated no teratogenic effects. Prazosin is not a very powerful agent and has usually been combined with the beta blocker oxprenolol in obstetric studies. In one study prazosin was used alone or in combination with oxprenolol in 44 pregnant women. No fetal abnormalities or adverse effects were noted. Another set of researchers used prazosin with or without oxprenolol to treat pregnancy hypertension beginning before 34 weeks' gestation. None of the 22 patients had significant maternal or fetal side effects attributable to drug therapy. Although available since 1976, prazosin had not been widely used in pregnancy.
6. Hydralazine - Hydralazine is an excellent drug for intravenous therapy of hypertension in pregnancy. It is an arteriolar vasodilator that causes a secondary baroreceptor-mediated sympathetic response, increasing heart rate and cardiac output. However, it is poorly tolerated orally as a single agent. Prominent side effects are headache, tachycardia, palpitations, fluid retention, and a lupuslike syndrome when chronic dosage exceeds 200 mg/d. Many of the unwanted side effects are minimized when it is used with a diuretic, methyldopa, or a beta blocker; however, use of multiple agents is discouraged in pregnancy. Dosage is initiated at 10 mg 4 times daily and increased to 200 mg/d.
7. Beta blockers - Beta blockers were introduced in the 1960s and have been used in pregnancy to treat migraine headache, hypertrophic obstructive cardiomyopathy, mitral valve prolapse, Graves' disease, and hypertension. Beta blockers are usually not adequate to control severe hypertension and are frequently combined with a diuretic, a vasodilator, or both. Beta blockers have been associated with neonatal bradycardia, hypoglycemia, hyperbilirubinemia, intrauterine growth retardation, respiratory depression, blocking of tachycardiac response to hypoxia, and increase in uterine muscle tone causing decreased uterine blood flow. The frequency of these side effects is unknown. Though clinical experience is accumulating, the safety of beta blockers in pregnancy has not yet been clearly established. Their use requires thoughtful risk-benefit analysis and clinical judgment. Infants of mothers taking beta blockers should be placed in an intermediate care unit after delivery to be monitored for side effects.
8. Labetalol - The low incidence of side effects, lack of teratogenicity, maintenance of uterine blood flow, and low propensity to cross the placenta make labetalol attractive for treatment of pregnant women. One randomized study found it offered no advantages over methyldopa in hypertensive pregnancy. Another study compared the use of labetalol plus hospitalization versus hospitalization alone in the management of 200 mildly preeclamptic women. No benefit was demonstrated in the labetalol-treated group; in addition, the incidence of SGA infants was higher in that group. Labetalol is started at 100 mg 3-4 times daily and increased to a maximum dosage of 2400 mg/d. Side effects are minor and include tremulousness and headache.
9. ACE inhibitors - The FDA-approved angiotensin-converting enzyme inhibitors include captopril, enalapril, lisinopril, fosinopril, ramipril, benazepril, and quinapril. They are widely used as first-line therapy for hypertension because they decrease systemic vascular resistance and have few side effects. Most of the reported experiences with ACE inhibitors in pregnancy are with captopril or enalapril. In human pregnancy these agents have been associated with several fetal and neonatal complications including hypotension, growth retardation, oligohydramnios, anuria, renal failure, malformations, stillbirth, and neonatal death. Although they have been successfully used in pregnancy, they should be avoided in pregnant women.
B. Effects of Antihypertensives on Breastfeeding
Little is known about the pharmacokinetics of antihypertensive drugs in human breast milk. In general, drugs that are lipid-soluble, un-ionized, and not protein bound are found in significant levels in breast milk. Specific recommendations concerning some of the more important agents are as follows: Thiazide diuretics should be avoided, since they decrease milk production and have been used in the past to suppress lactation. However, no electrolyte abnormalities have been found in infants of mothers taking thiazides. Methyldopa is probably safe during breastfeeding, since low plasma levels are found in the infants. Except for propranolol, the other beta-blocking agents are found in higher concentrations in breast milk than in maternal plasma. Therefore, propranolol would probably be the drug of choice if a beta blocker was needed. Nevertheless, accumulated experience with various beta blockers has shown only very low drug concentrations in breast milk. Clonidine is found in very small amounts in breast milk. Captopril appears to be safe during breast-feeding because only small amounts are found in breast milk. Data on the other drugs are insufficient to serve as a basis for recommendations.
C. General Obstetric Management
In taking the medical history of the hypertensive pregnant patient, particular attention should be paid to the duration of hypertension, use of antihypertensive medications, history of renal or heart disease, and the outcome of previous pregnancies. Physical examination should include a careful funduscopic examination, listening for renal artery bruit, and checking the dorsalis pedis pulses for coarctation of the aorta. The blood pressure should be measured in the sitting position. If the bladder of the blood pressure cuff does not completely encircle the arm, a falsely high blood pressure reading may be obtained. In this situation, a thigh cuff should be used.
At the first prenatal visit, baseline laboratory studies should be obtained for organ systems likely to be affected by chronic hypertension or to deteriorate during pregnancy. Tests should include (but are not limited to) urinalysis; complete blood count; measurements of blood urea nitrogen, creatinine, serum electrolytes, uric acid, calcium, and phosphorus; liver function tests; ECG; and 24-hour urine collection for creatinine clearance and total protein. If significant heart disease is suspected, a chest x-ray (with the abdomen shielded) or echocardiogram should be obtained. A 3-hour oral glucose tolerance test is desirable, since as many as one-fourth of patients may have unrecognized diabetes. If hyperglycemia or wide blood pressure swings are evident, 24-hour urine testing for vanillylmandelic acid and metanephrines is recommended to rule out pheochromocytoma. The patient may be given a regular diet without salt restriction and should be followed every 2-3 weeks until 30 weeks' gestation and then weekly thereafter. Gestational age can be documented and an SGA infant detected with serial ultrasound examinations started early in pregnancy. Fetal well-being may also be assessed with the nonstress test and amniotic fluid index starting at 34 weeks or whenever the patient develops superimposed preeclampsia. Superimposed preeclampsia is diagnosed on the basis of worsening hypertension (30 mm Hg systolic or 15 mm Hg diastolic rise) together with either nondependent edema or proteinuria. Some of the more frequent indications for early delivery include superimposed preeclampsia, underlying medical problems such as diabetes or renal insufficiency, abnormal antepartum fetal heart rate, and a growth restricted fetus. A patient with worsening hypertension may be given betamethasone to accelerate fetal lung maturity if the lecithin:sphingomyelin ratio is less than 2 and if delivery can be delayed for 48-72 hours after the first dose.
Pregnancy outcome is usually favorable in patients with mild chronic hypertension, and perinatal survival rates of 95-97% can be expected. The main complications are superimposed preeclampsia, abruptio placentae, prematurity, and intrauterine growth retardation. If the patient has severe hypertension in the first trimester, onset of superimposed preeclampsia before 28 weeks' gestation, renal insufficiency prior to pregnancy, hypertensive cardiovascular disease, or congestive cardiomyopathy, the prognosis is more guarded. These patients require close follow-up of multiple clinical and laboratory parameters. The physician must be certain that these patients can be relied upon to take their medication. They may require a long period of hospitalization and are more likely to require cesarean delivery. Their fetuses are at significant risk for prematurity, growth retardation, and death.