Introduction
Preeclampsia and eclampsia continue to be a major cause of maternal death.  Currently ~18% of US maternal deaths are attributed to hypertensive disorders and eclampsia and several hundred women die from eclampsia and its complications every year. In the US, preeclamptic women have received magnesium sulfate as a seizure prophylaxis agent for three decades and following publication of the MAGPIE trial(1) this practice is becoming more widely accepted internationally. Based on available data(2,3), we believe that the incidence of seizures in untreated preeclamptic women is approximately 3-4%, whilst for those receiving magnesium sulfate (without concomitant benzodiazepine or other agents) the rate is 0.8-1%(1,4). Magnesium sulfate thus does not completely prevent seizure activity in preeclamptic patients and is still associated with an appreciable failure rate of ~1%. We have recently published the results of an international multicenter randomized, clinical trial involving more than 1,600 severe preeclamptic patients(4).  This study was funded by the Orphan Products Division of the FDA. The risk of seizures for preeclamptic women receiving magnesium sulfate prophylaxis was 0.8% (7/831). Given an estimated prevalence of eclampsia of 1% in women with preeclampsia who are receiving magnesium sulfate prophylaxis, and an estimated 100 000 women per year in the USA with preeclampsia, approximately 1,000 women a year in the US will convulse despite our current best efforts.  Of these patients a significant number will suffer aspiration, asphyxia, cerebrovascular complications and death.

In addition to a recognized failure rate, there are financial, logistic and safety concerns associated with the universal administration of magnesium sulfate.  While generally manageable in the US, many institutions in the developing world lack the necessary equipment and expertise to administer the medication and many preeclamptic patients thus do not receive magnesium sulfate prior to their first seizure. As effective as it has been in reducing mortality from eclampsia, magnesium sulfate is also associated with appreciable morbidity and mortality from administration errors and magnesium toxicity.  The availability of an easily administered, cheap, safe alternative to magnesium sulfate would be welcomed in this country and would provide an extremely useful alternative therapy to the current standard of care. From a global perspective, the availability of such an eclampsia prophylactic agent would be of immense benefit.  In many developing countries, magnesium sulfate is neither available nor practical as a management modality because of expense, logistics, lack of trained personnel and lack of resources. It is estimated that in these situations seizures may occur in more than 4% of severely preeclamptic patients, and in some situation as many as 25% of eclamptic women will die (Personal communication Professor Dasgupta, India).

We believe that recent advances in our understanding of the pathophysiology of preeclampsia and eclampsia, which are outlined below, will allow us to reduce the seizure rate in treated preeclamptic women even further than what is currently reported. This will be possible by employing a scientifically directed, and well tolerated, therapy that is better suited to managing the underlying cerebrovascular pathology of preeclampsia/eclampsia than the empiric administration of a poorly understood, albeit effective, drug (magnesium sulfate).

Scientific Background and Preliminary Data
Belfort et al (5) have recently published a model that explains the cerebral hemodynamic changes in preeclampsia.  In this model there is initial elevated cerebral perfusion pressure opposed by protective vasoconstriction with preservation of normal cerebral blood flow. As barotrauma supervenes there is breakthrough vasodilatation, hypertensive encephalopathy, cerebral overperfusion, and ultimately, if therapy is delayed or absent, overt vasospasm and ischemia. In our model, persistent and progressive hypertension ultimately overwhelms the protective vasoconstriction in a vascular bed such as the cerebral or renal circulation.  The arterioles and other resistance vessels are damaged by barotrauma and become incapable of containing the elevated perfusion pressure, and forced vasodilatation occurs leading to breakthrough overperfusion. This leads to physical damage of the blood vessels with loss of intimal, endothelial, and ultimately, advential integrity.  Leakage of intravascular fluid through the damaged vessel wall causes vasogenic (and sometimes cytotoxic) edema(6) around the blood vessels and disrupts the blood-brain-barrier allowing stimulatory substances into the cerebral extravascular compartment. 

Labetalol
Labetalol is a rather unique substance in that it has both selective, competitive alpha-1 and non-selective, competitive beta-adrenergic blocking actions(7).  In humans the ratio of the alpha and beta blockade are estimated to be approximately 1:3 and 1:7 for the oral and intravenous compounds respectively(7). At higher doses than required for adrenoreceptor blockade, labetalol has been shown to be a membrane stabilizer(8).

Labetalol produces rapid dose-dependent decreases in blood pressure without reflex tachycardia or significant reduction in heart rate.  Hemodynamic effects on cardiac output are minimal and the predominant effect appears to be a decrease in peripheral vascular resistance(9). Satisfactory blood pressure control has been achieved with either intravenous or oral administration. In fact, oral labetalol has been used for acute control of severe preeclampsia and has been found to be comparable to hydralazine(10).  Mabie and coworkers(11) compared bolus intravenous labetalol with intravenous hydralazine in the acute treatment of severe hypertension. They found that labetalol had a quicker onset of action and did not result in reflex tachycardia(11). In terms of fetal effects, blood pressure reduction with labetalol does not result in fetal distress(11,12) unlike acute blood pressure reduction with hydralazine where 15% of fetuses will exhibit distress frequently requiring immediate cesarean section for delivery(13).
Studies of uteroplacental blood flow indicate that no decrease in perfusion occurs despite the reduced maternal blood pressure(14,15).   As with all beta-blockers, labetalol has been associated with hypoglycemia, bradycardia and hypotension but neonatal outcome is uniformly good(11,16,17) and ACOG currently recommends labetalol as one of the first line antihypertensive medications for preeclampsia(18).

The peak effects of an oral dose of labetalol occur within 2-4 hours and last up to 8 hours after a single oral dose of 100 mg in non-pregnant individuals(7). In pregnant women with hypertension, the peak serum level of labetalol has been shown to occur  within 20-60 minutes after a 100 mg oral dose, with a terminal elimination half-life of 1.7 ± 0.27 hours(19).  The elimination half-life after intravenous administration in pregnant hypertensive patients is similar to that seen with oral administration(20).  The elimination half life is significantly less than that seen in non-pregnant individuals (6-8 hours)(21) and for this reason more frequent dosing is recommended in pregnancy (4-6 hours). There is a fetal cord blood/maternal serum ratio of 0.5 ± 0.15 and an amniotic fluid/maternal serum ratio of 0.16 ± 0.13(19).

Labetalol has some interesting, and potentially, important non-antihypertensive effects that may be beneficial in preeclampsia. Among these are an anti-platelet aggregation action(20), a thromboxane reducing effect(21), and a fetal lung maturation accelerating influence(22).
Labetalol has been successfully used for many years in pregnant women with severe preeclampsia, and although hydralazine is more commonly used for the management of acute hypertension in preeclampsia in the USA, labetalol has been shown to be equally effective(11) and is currently a first line recommendation of ACOG for blood pressure control in preeclampsia. Of great importance are the data reported by Walker(23) on the experience at the Glasgow Royal Maternity Hospital in Scotland. Over a 10-year period Walker and his colleagues managed over 36,000 pregnancies with a 10% rate of hypertension in pregnancy. They had only one patient develop eclampsia out of a total of 555 women who were receiving treatment for their hypertension, and in this patient the hypertension was inadequately controlled.  Less than 100 of these 555 patients received magnesium sulfate, and labetalol was the drug of choice for hypertension control. Thus, their rate of seizures on labetalol is in the order of 1/455 (0.2%). We have based our sample size estimates on this number.

Labetalol Lowers CPP in Preeclampsia: Current Data
We have recently published our findings on the cerebral hemodynamic effects of labetalol in pregnant hypertensive women(24). Labetalol significantly reduced the CPP, as well as the systolic, diastolic and mean BP at 60 and 180 minutes without significantly affecting the heart rate, MCA velocities, RAP or CFI. Labetalol effectively reduces CPP, without affecting cerebral perfusion, and this is thought to be primarily by a decrease in systemic blood pressure.

Rationale for Labetalol as an Anticonvulsant Agent in Preeclampsia
We hypothesize that if reduction of CPP is an important part of the anti-seizure prophylactic mechanism of action of magnesium sulfate in preeclampsia, then the use of a more specific, less toxic, agent which can be administered orally, such as labetalol, may greatly simplify the management of severely preeclamptic patients while at the same time providing equivalent or improved efficacy.  In addition, the facility of administration, reduced risk of respiratory and cardiac depression, lack of need for intensive monitoring of blood magnesium levels, suitability for use by primary care personnel, safety, and low cost of the regimen give labetalol very attractive risk/benefit and cost/benefit ratios. 

We have presented evidence that demonstrates a plausible theory regarding the etiology of seizures in eclampsia.  This theory, that initial overperfusion of the brain is the cause of subsequent seizures (with or without vasospasm that may be caused by overperfusion), incorporates all of the available evidence currently in the literature and links it into a single model. We have further shown, both with Doppler data and with a large-scale clinical trial, that the theory that vasospasm is the initial cerebral event in eclampsia is unlikely to be correct.  Using Doppler data, we have demonstrated that labetalol has some important characteristics that may make it an ideal agent to limit cerebral overperfusion, control blood pressure, and prevent seizures in preeclampsia. This drug has been in use for over two decades and thousands of preeclamptic women and their babies have been safely treated with it.  The potential advantages of using this drug as a blood pressure controlling agent, as well as a seizure prophylaxis agent, in preeclamptic women are wide reaching.  By adopting labetalol, if it is proven as or more effective as magnesium sulfate, there will be a simplification of the current system in this country and a significant cost saving. Given the risks of magnesium sulfate, it is possible that the use of labetalol will reduce maternal mortality from magnesium sulfate toxicity. 

Patients who might not otherwise receive seizure prophylaxis (with magnesium sulfate) due to logistic, personnel, safety, or training issues, will be in a position to have treatment started earlier. This is because the administration of a tablet (with known safety and no need for monitoring of blood levels) prior to transport of a patient is simply easier and more accepted by nursing staff and primary care givers.

Preliminary Data from the Labetalol versus Magnesium Sulfate Study

Preliminary data for the proposed study come from a pilot study carried out at the University of Utah in Salt Lake City, Utah Valley Regional Medical Center in Provo, Utah, the University of Tennessee at Chattanooga, and the University of Cape Town, South Africa between 2002 and 2005. The protocol used was the same as that presented in this proposal. Three hundred and twenty two patients were enrolled, of which 175 women received labetalol and 147 patients were randomized to MgSO4. Demographics, BP on admission, and laboratory results were similar between the two groups (p = NS).

Two patients in each group developed eclampsia (Labetalol = 1.1% vs. MgSO4 1.4%, p = 0.7).

One patient in the Labetalol group died as a result pulmonary fat emboli and was diagnosed at autopsy as having had Fatty Liver of Pregnancy. There were no differences in the rates of abruptio placentae (2.6%-Labetalol vs. 0% - MgSO4), cesarean section for fetal distress (35%-Labetalol vs. 46% - MgSO4), postpartum hemorrhage (2.0%-Labetalol vs. 0.5% - MgSO4), or post partum pulmonary edema (0.5%-Labetalol vs. 0% - MgSO4).

The MgSO4 group required significantly more additional antihypertensive therapy and had more peripheral skin flushing. Blood pressure control in the Labetalol group was characterized by significantly lower postpartum systolic and diastolic blood pressures. Intrapartum blood control was similar between the two groups. There were no significant differences in terms of the other recorded parameters between the two groups: headache, diplopia, scotomata, hypotension, nausea, vomiting, allergic reaction, and respiratory depression. There were also no significant differences in intra/post partum complications or neonatal outcomes. The mean heart rate was significantly lower in the Labetalol group during both antepartum and postpartum time periods.

Routine administration of labetalol to patients who were not hypertensive after the first dose did not result in hypotension or fetal compromise.

The neonatal outcomes were similar with no significant differences being detected. In particular, concerns about neonatal bradycardia and hypoglycemia were not realized and these complications were not seen. The Apgar scores were not different and there were no differences in the rates of intubation and ventilation, RDS, hypotension, hypotonia, or arrhythmia.

The nursing staff and doctors were very receptive to the protocol and preferred using labetalol which they found easier, less time consuming and more comfortable for their patients than standard magnesium sulfate therapy.

Obviously the numbers in this preliminary study are not sufficient to completely address the issue of safety, and the preliminary study was undertaken to test the protocol suitability for the multicenter study, to identify any potential logistic problems, and to assess the satisfaction of the patients, nurses, and physicians with the study. The safety issues have been addressed by others and we feel comfortable that labetalol is both safe and efficacious for hypertension control in preeclampsia. ACOG has recommended this drug as an alternative to hydralazine, and Walker(23) has amassed sufficient data to show that the drug is safe as a single agent in preeclampsia and chronic hypertension. Most units around the country use labetalol as either their primary or secondary agent for the control of hypertension in preeclampsia, and no safety issues have been raised over the many years that this drug has been in use.

Ethics of not Treating Women with Preeclampsia with Magnesium Sulfate
In many European and Scandinavian countries magnesium sulfate is used sparingly, if at all, in the prophylaxis of eclampsia, and the only agents used are those for blood pressure control.  Even after the MAGPIE study there has not been a significant change in many developed European countries.  Magnesium sulfate is used in cases where there is “imminent eclampsia” and not in cases of mild preeclampsia or even severe preeclampsia where the patient is not deemed to have cerebral irritation.  The reported eclampsia rate in these countries is no different from that reported from countries with universal use of magnesium sulfate.  This statistic is borne out by the MAGPIE study where the majority of the adverse outcomes and all of the maternal mortality occurred in developing countries. It is felt in the developed European countries that control of the hypertension with drugs such as hydralazine and labetalol provides sufficient protection against eclampsia, and there is evidence to support this.  Some authors have theorized that magnesium sulfate is no more effective in the prevention of seizure activity than antihypertensive medication alone (hydralazine or labetalol)(23).  Moodley et al.(26) claimed no additional benefit of magnesium sulfate over antihypertensive medication alone in their study of 228 patients, of whom 112 received both antihypertensive therapy as well as magnesium sulfate, and 116 received antihypertensive therapy only. The authors concluded that patients with hypertensive crisis in pregnancy can be managed safely by aggressive lowering of the blood pressure and expeditious delivery. Walker et al.(23) showed that in Scotland, where prophylactic magnesium sulfate is very infrequently used in preeclampsia, no patient with adequately controlled blood pressure suffered a seizure in their institution in 10 years (over 500 preeclamptic patients). Given our cerebral hemodynamic findings, the capability of these investigators in preventing seizures with antihypertensive drugs alone can be explained.

References:

1. The Magpie Trial Collaboration Group. Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial. Lancet 2002;359:1877-90.

2. Coetzee EJ, Dommisse J, Anthony J. A randomised controlled trial of intravenous magnesium sulphate versus placebo in the management of women with severe pre-eclampsia. Br J Obstet Gynaecol 1998;105:300-3.

3. Burrows RF, Burrows EA. The feasibility of a control population for a randomized control trial of seizure prophylaxis in the hypertensive disorders of pregnancy. Am J Obstet Gynecol 1995;173:929-35.

4. Belfort MA, Anthony JA, Saade GR, Allen JC. “Magnesium sulfate versus nimodipine for the prevention of eclampsia.” New England Journal Medicine, 2003; 348:304-11.

5. Belfort MA, Varner MW, Dizon-Townson DS, Grunewald C, Nisell H. Cerebral perfusion pressure, and not cerebral blood flow, may be the critical determinant of intracranial injury in preeclampsia: A new hypothesis. Am J Obstet Gynecol 2002;187:626-34.

6. Koch S, Rabinstein A, Falcone S, Forteza A. Diffusion-weighted imaging shows cytotoxic and vasogenic edema in eclampsia. Am J Neuroradiol 2001;22:1068-70.

7. van Zwieten PA. An overview of the pharmacodynamic properties and therapeutic potential of combined alpha- and beta-adrenoceptor antagonists. Drugs 1993;45:509-17.

8. Blakeley AG, Summers RJ. The pharmacology of labetalol, an alpha- and beta-adrenoceptor blocking agent. Gen Pharmacol 1978;9:399-402.

9. Lund-Johansen P. Pharmacology of combined alpha-beta-blockade. II. Haemodynamic effects of labetalol. Drugs 1984;28 Suppl 2:35-50.

10. Walker JJ, Greer I, Calder AA. Treatment of acute pregnancy-related hypertension: labetalol and hydralazine compared. Postgrad Med J 1983;59 Suppl 3:168-70.

11. Mabie WC, Gonzalez AR, Sibai BM, Amon E.  A comparative trial of labetalol and hydralazine in the acute management of severe hypertension complicating pregnancy.  Obstet Gynecol 1987;70(3):328-33.

12. Mahmoud TZ, Bjornsson S, Calder AA. Labetalol therapy in pregnancy induced hypertension: the effects on fetoplacental circulation and fetal outcome. Eur J Obstet Gynecol Reprod Biol 1993 Jul;50(2):109-13.

13. Spinnato JA, Sibai BM, Anderson GD. Fetal distress after hydralazine therapy for severe pregnancy-induced hypertension. South Med J 1986 May;79(5):559-62.

14. Jouppila P, Kirkinen P, Koivula A, Ylikorkala O. Labetalol does not alter the placental and fetal blood flow or maternal prostanoids in pre-eclampsia. Br J Obstet Gynaecol 1986 Jun;93(6):543-7.

15. Lunell NO, Nylund L, Lewander R, Sarby B.  Acute effect of an antihypertensive drug, labetalol, on uteroplacental blood flow. Br J Obstet Gynaecol 1982 Aug;89(8):640-4.

16. Michael CA.  Use of labetalol in the treatment of severe hypertension during pregnancy.  Br J Clin Pharmacol 1979;8(Suppl 2):211S-215S.

17. Riley AJ.  Clinical pharmacology of labetalol in pregnancy.  J Cardiovasc Pharmacol 1981; 3(Suppl 1):S53-9.

18. Management of Preeclampsia, ACOG Technical Bulletin, No. 219, February, 1996. The American College of Obstetricians and Gynecologists.  Washington, D.C.

19. Rogers RC, Sibai BM, Whybrew WD. Labetalol pharmacokinetics in pregnany-induced hypertension. Am J Obstet Gynacol 1990;162:362-6.

20. Rubin PC, Butters L, Kelman AW, Fitzsimons C, Reid JL. Labetalol disposition and concentration-effect relationships during pregnancy. Br J Clin Pharmacol 1983;15:465 –70.

21. McNeill JJ, Anderson AE, Louis WJ, Morgan DJ,. Pharmacokinetics and pharmacodynamic studies of labetalol in hypertensive subjects. Br J Clin Pharmacol 1979;8:157S.

20. Greer IA, Walker JJ, McLaren M, Calder AA, Forbes CD. A comparative study of the effects of adrenoreceptor antagonists on platelet aggregation and thromboxane generation. Thromb Haemost 1985;54:480 – 4.

21. Greer IA, Walker JJ, Maclaren M, Calder AA, Forbes CD. Inhibition of thromboxane and prostacyclin in whole blood by adrenoreceptor antagonists. Prostaglandins Leukotrienes Med 1985;19:209 – 17.

22. Michael CA. Early fetal lung maturation associated with labetalol therapy. Singapore J Obstet Gynaecol 1980;11:2 – 5.

23. Walker JJ. Hypertensive drugs in pregnancy. Antihypertension therapy in pregnancy, preeclampsia, and eclampsia. Clin Perinatol 1991;18(4):845-73.

24. Belfort MA, Tooke-Miller C, Allen JC, Dizon-Townson D, Varner MA. Labetelol decreases cerebral perfusion pressure without negatively affecting cerebral blood flow in hypertensive gravidas. Hypertension in Pregnancy. 2002:21(3):185-97.

25. Warren J, Lacoursiere Y, Varner MW, Silver R, Anthony J, Belfort MA. First Interim Report on the Labetalol versus Magnesium Sulfate for the Prevention of Eclampsia Trial (LAMPET). Hypertension in Pregnancy (Abstract in press).

26. Moodley J, Moodley VV. Prophylactic anticonvulsant therapy in hypertensive crises of pregnancy - the need for a large, randomized trial. Hypertension in Pregnancy 1994;13:245-252.

27. Friedman LM, Furrberg CD, DeMets DL. Fundamentals of Clinical Trials. Mosby Publishing, New York, New York, 1996.

28. Lachin JM. Introduction to sample size to determination and power analysis for clinical trials. Controlled Clinical Trials 2,93-113; 1981.

29. Phillips KF. Power of the two 1-sided tests procedure in bioequivalence. J Pharmacokinetics Biopharmaceutics 1990;18:137-44.

30. Blackwelder WC. Proving the null hypothesis in clinical trials. Controlled Clinical Trials 1982, 3:345-353.

31. Piaggio G, Pinol AP. Use of the equivalence approach in reproductive health clinical trials. Stat Med. 2001;20(23):3571-7.