全文

Abstract
Objective
Labetalol and methyldopa are the two antihypertensive drugs most frequently used to control blood pressure for hypertensive disorders of pregnancy. The objective of this study was to assess if labetalol is associated with poor infant outcomes.

Study design
Retrospective population-based cohort study using the linked maternal/infant databases in the Province of Saskatchewan. Women with a diagnosis of a hypertensive disorder of pregnancy who delivered a singleton in Saskatchewan from January 1, 1990 to December 31, 2005 and who were dispensed only labetalol or only methyldopa were included in the study. Occurrences of small for gestational age (SGA) < 10th percentile, SGA < 3rd percentile, preterm birth, stillbirth, hospitalization for respiratory distress syndrome (RDS), sepsis, and seizure during infancy, and infant death were compared. Multiple logistic regression analysis was performed to adjust for potential confounding.

Results
A total of 1223 eligible women were included in the final analysis. Among them, 300 received labetalol only and 923 received methyldopa only during pregnancy. For women with chronic hypertension, the rate of hospitalization for RDS, sepsis, and seizure during infancy was significantly higher for infants born to mothers who were dispensed labetalol only as compared with infants born to mothers who were dispensed methyldopa only (adjusted odds ratio (OR) 1.51, 95% confidence interval (CI) 1.02–2.22).

Conclusion
Compared with methyldopa, the use of labetalol for chronic hypertension of pregnancy may be associated with increased rate of hospitalization during infancy.

Keywords: Chronic hypertension, Pregnancy, Hospitalization, Labetalol, Methyldopa.

1. Introduction
Hypertensive disorders of pregnancy remain a leading cause of maternal, fetal and neonatal morbidity and mortality in industrialized countries, including Canada [1], [2], [3], and [4]. Although there is general consensus on the need to control blood pressure with antihypertensive medication in non-pregnant patients, whether and how to control blood pressure for hypertensive disorders of pregnancy has not been clearly defined, especially in mild or moderate cases of hypertension [5] and [6].

Labetalol and methyldopa are generally towards the top of the list of antihypertensive drugs during pregnancy recommended by most professional associations/societies [7], [8], [9], and [10]. Several studies, however, have found an association between exposure to beta-blockers in pregnancy and low birth weight infants [6], [11], and [12]. Given the combined beta and alpha blocking effects of labetalol, we hypothesized that exposure to labetalol in pregnancy may be associated with higher rates of adverse perinatal outcomes compared with other antihypertensives. The objective of the present study was to compare perinatal outcomes in infants born to mothers with pregnancy exposure to labetalol and methyldopa, two of the most frequently recommended antihypertensive drugs during pregnancy.

2. Methods
A retrospective population-based cohort was formed using the linked maternal/infant databases in the Canadian Province of Saskatchewan. The cohort included all women with a diagnosis of a hypertensive disorder in pregnancy who were eligible for coverage by the Saskatchewan prescription drug plan (out-patient prescriptions) and who delivered a singleton in Saskatchewan from January 1, 1990 to December 31, 2005. They were identified through ICD-9/ICD-10-CA codes recorded in the database. Women with co-morbidities (i.e., diabetes, renal disease, and/or cardiac disease) were excluded from the analyses to minimize confounding, as these conditions are independently associated with increased risks of adverse perinatal outcomes [13]. The study population was comprised of women with diagnoses of a hypertensive disorder during pregnancy who were dispensed either labetalol or methyldopa. Information on labetalol and methyldopa use in pregnancy was ascertained for each study participant using number of days between date of drug dispensing and date of infant birth.

Outcomes of interest included: small-for-gestational age (SGA, defined as birth weight being less than the 10th percentile of the population-based Canadian reference recommended by Kramer et al. [14]); severe SGA (birth weight less than 3rd percentile); preterm birth (live infants delivered at less than 37 completed weeks of gestation); stillbirth (stillbirth with a birth weight above 500 grams or gestational age more than 20 weeks); offspring hospitalization during infancy; and infant death (death of a live-born infant within the first year of life). For hospitalization, we counted only first hospitalization with a ICD-9 or ICD-10 code diagnostic code for respiratory distress syndrome (or RDS, a syndrome in premature infants caused by developmental insufficiency of surfactant production and structural immaturity in the lungs), sepsis (a potentially fatal whole-body inflammation (a systemic inflammatory response syndrome) caused by severe infection), or seizures (changes in the brain's electrical activity which can cause dramatic, noticeable symptoms including violent shaking and loss of control) during the first year of life, because we thought hospitalizations for many other causes (e.g. accidents) may not be related to antihypertensive drugs uses during pregnancy.

We first described maternal characteristics and occurrence of adverse perinatal outcomes between the two study groups, women who received labetalol only and women who received methyldopa only during pregnancy. Multiple logistic regression analyses were then performed to assess the independent association between the type of antihypertensive drugs and adverse perinatal outcomes, with methyldopa only as the reference. Potential confounding variables included in the regression models were: maternal age (<25 years, 25–29 years, and ≥30 years; <25 years as the reference); year of childbirth; Saskatchewan assistance plan coverage (no versus yes; no as the reference); parity (1 versus ≥2; 1 as reference); and type of hypertensive disorders in pregnancy (chronic hypertension [including chronic hypertension with superimposed preeclampsia], gestational hypertension, and preeclampsia/eclampsia; gestational hypertension as the reference). Confounding variables and their categorizations included in the regression model were selected a priori by an expert panel based on literature review. Those variables that are possibly associated with both drug selection and the examined outcomes were considered, but we were restricted by factors that cannot be controlled by ourselves. Some important variables were not available in the database. A full model consisting of all independent variables was used in the analysis.

As there are major differences in the clinical indications and duration of treatment of hypertensive disorders of pregnancy we performed two pre-planned analyses as follows. First we compared perinatal outcomes between the two study groups in women with a diagnosis of chronic hypertension and in those women with a diagnosis of gestational hypertension or preeclampsia/eclampsia separately. Second, women with chronic hypertension may be treated with antihypertensive medications for longer durations during pregnancy and the initiation time of treatment may be associated with clinical outcomes. As such, we performed an additional multiple logistic regression analysis for women with a diagnosis of chronic hypertension to adjust for timing of initiation of antihypertensive treatment (pre-conception, first trimester, second trimester, and third trimester) as it reflects duration of therapy. Supplementary analysis restricted to first pregnancies was also performed.

3. Results
A total of 1223 women were included in the final analysis. Among them, 300 received labetalol only and 923 received methyldopa only during pregnancy. Table 1 describes the baseline characteristics of the two study groups. Women who were dispensed labetalol only tended to be of older age, to initiate treatment in later gestation, to be diagnosed with gestational hypertension/preeclampsia, and to be delivered in more recent years, but were less likely to depend on government assistance.

Table 2 compares odds ratios of various adverse perinatal outcomes between the two study groups. None of the differences between the two study groups reached statistical significance.

Table 3 compares outcomes between the two study groups in women with a diagnosis of chronic hypertension and in those women with a diagnosis of gestational hypertension or preeclampsia/eclampsia separately. For women with chronic hypertension, the odds of hospitalization for RDS, sepsis, and seizures during infancy was significantly higher (odds ratio (OR) 1.51, 95% confidence interval (CI) 1.03–2.22) in infants born to mothers who were dispensed labetalol only, compared with infants born to mothers who were dispensed methyldopa only. For women with gestational hypertension or preeclampsia/eclampsia, no difference in the incidence of adverse perinatal outcomes was observed between the two study groups.

PE: Preeclampsia.

Table 4 presents the results in women with chronic hypertension, after additional adjustment for time of initiation of antihypertensive drugs. For women with chronic hypertension, the odds of hospitalization for RDS, sepsis, and seizures during infancy was significantly higher (OR 1.51, 95% CI 1.02–2.22) among infants born to mothers who were dispensed labetalol only, compared with infants born to mothers who were dispensed methyldopa only.

Results from supplementary analysis restricted to first pregnancies were similar, although statistically less stable because of the reduced number of subjects after the restriction (data available upon request).

4. Comment
This contemporary population-based study assessing the association between antihypertensives dispensed for hypertensive disorders in pregnancy and adverse perinatal outcomes found that for women with chronic hypertension, the odds of hospitalization for RDS, sepsis, and seizures during infancy were increased among infants born to mothers who were dispensed labetalol only compared with methyldopa only. Our study is unique in that it compares the outcomes of mothers who were exposed to specific antihypertensive drugs.

This study's results of more adverse neonatal outcomes in women exposed to labetalol are consistent with published data on beta-blockers as a whole group. First, in a systematic review of 13 trials involving a total of 854 women, Magee and Duley [11] compared birth outcomes between pregnancies exposed to oral beta-blockers with methyldopa, and found that overall the effect of beta-blockers on the risk of SGA was significantly higher than methyldopa (relative risk (RR): 1.36, 95% CI: 1.02–1.82). Second, animal experiments have shown that the somewhat selective beta-blocker metoprolol and the nonselective beta-blocker propranolol might adversely affect fetal circulation [15] and [16]. For example, Houlihan et al. [17] compared the in vitro effects of labetalol, hydralazine, alpha-methyldopa, nifedipine and magnesium sulphate on human umbilical artery resistance, and found that all compounds investigated, except alpha-methyldopa, exerted a significant relaxant effect on umbilical arterial tone. Alpha-methyldopa was a significantly less potent vasodilator when compared to all other drugs (mean maximal inhibition value 20.89 ± 7.99% versus all other agents (range 63.15 ± 8.70 − 84.12 ± 3.84%), P < 0.01) [17]. Of note, labetalol is a unique antihypertensive drug that exerts combined beta-and alpha-blocking effects, whereas methyldopa acts as an alpha agonist. This may, in part, explain the increased adverse outcomes due to labetalol's blockade of the sympathetic nervous system. On the other hand, a small clinical trial in pregnant women with gestational hypertension (systolic blood pressure of 140–160 mmHg and/or diastolic blood pressure of 90–105 mmHg between 20 and 38 weeks of gestation without proteinuria; 50 with labetalol and 50 with methyldopa) [18] found that the rate of admission to neonatal unit was higher in the methyldopa group than in the labetalol group. Whether this isolated finding from a small study was due to the play of chance or through a different mechanism (e.g., increased risk of preterm birth of methyldopa users) deserves further investigation.

While there are very few studies in the literature directly comparing perinatal outcomes between labetalol and methyldopa, emerging data suggest an association between labetalol use in pregnancy and several clinically important maternal and neonatal complications. First, Immanni et al. [19] reported a case of a 42-year-old multigravida with severe preeclampsia who had an emergency caesarean section under spinal anaesthesia. Perioperatively, her arterial pressure was controlled with oral methyldopa and an intravenous infusion of labetalol. Postoperatively, in the intensive care unit, she had recurrent episodes of hypoglycaemia which required treatment with intravenous glucose. These episodes resolved when the labetalol infusion was stopped. [19] Similar effects of labetalol-induced hypoglycemia in the neonates have been reported [20]. Second, Heida et al. [20] compared the outcomes of infants of mothers with severe preeclampsia (55 infants exposed to labetalol in utero versus 54 control infants with no treatment). Neonatal hypotension occurred significantly more in infants exposed to labetalol (29.1%) compared with controls (7.4%; P = 0.003) [20]. In the multivariate regression model, labetalol exposure, the need for intubation, and presence of a patent ductus arteriosus were independently associated with neonatal hypotension (P < 0.001) [20]. Like others, we hypothesize that the combined effects of labetalol on fetus (e.g. SGA, hypoglycemia, hypotension, etc., from blunted autonomic responsiveness) may lead to increased rate of hospitalization in this group of infants.

This study has a number of strengths. First, the study was designed to test an a priori hypothesis based on a strong biological rationale of the potential adverse fetal effects of labetalol, as animal and in vitro human experiments [15], [16], and [17] have shown that beta-blockers may adversely affect fetal circulation and several clinical studies have suggested an association between beta-blockers use during pregnancy and low birth weight infants [6] and [11]. Second, we employed extensive adjustment for known clinical predictors of adverse outcomes by using multiple regression analyses and data restriction to reduce the potential for bias and confounding. Third, the Saskatchewan Ministry of Health's health services database [21] used in our study contains a unique identifier in all health services covered by the provincial government and can be reliably used to link different databases. This rich health service database is population-based and covers the majority of the population in the Province of Saskatchewan. Validation, mostly by hospital chart review, has been built into several studies using such data. A recent study found high sensitivity, specificity, positive predictive, and negative predictive values for diagnosis based on ICD-9 codes alone for depression [22], which support the validity of Saskatchewan's health services database.

Limitations of our study should be acknowledged. First, there may be some gaps between drug prescription and drug utilization, which we have no direct means of evaluating as there is no compliance information in Saskatchewan's prescription drug file. In general, the only source of information on actual drug consumption in pregnancy is that reported by the patient herself. A European study found that mothers were able to report only 76% of drug groups, and 52% of exact drug names, and that the non-compliance rate was about 6% for the drugs dispensed by the pharmacy [23]. Second, drugs dispensed during hospitalizations or given as samples in physicians’ offices are not available from the Saskatchewan Prescription Drug Plan; however, this likely represents a small fraction of antihypertensive medication use. Third, some important variables such as non-prescription drugs and tobacco and alcohol use that may affect neonatal outcomes were not available in this dataset. Fourth, administrative data are prone to certain degree of coding errors. Misclassification of outcomes caused by either coding error or non-compliance may have occurred randomly, which would tend to attenuate the observed effects [24]. Fifth, because of confidentiality issues, we have relied on pre-coded variables (e.g. preterm birth as defined as <37 complete weeks of gestation) instead of the original records (e.g. gestational age). As a result, we cannot perform analysis on several commonly used outcome measures (e.g. very preterm birth as defined as <32 complete weeks of gestation). The research file we received was in mother–baby pair, and there is no unique identifier in the data to for the same mothers who may have delivered more than one baby. The impact of failure to adjust for potential “cluster” effect may be limited, because for infants born to the same mother, most of the study outcomes, exposure status, and confounding variables (e.g., age, parity, poverty, and year of birth) would have changed over time. Supplementary analysis restricted to first pregnancies yielded similar although statistically less stable results because of the reduced number of subjects after the restriction. Sixth, although the overall study sample size was large, the numbers of subjects diminished rapidly in subgroup analyses (e.g. by type of hypertensive disorder in pregnancy with specific types of antihypertensive drugs). Finally, residual confounding may still exist even after extensive adjustment by regression analysis and restrictions.

This study adds to the body of scientific evidence on the association between antihypertensive medications and adverse perinatal outcomes, with an emphasis on the potential adverse impacts of labetalol on hospitalization of infants for RDS, sepsis, and seizures. Physicians caring for women with hypertensive disorders in pregnancy should be aware of this possibility when choosing among different antihypertensive medications, and balance the benefits and risks of different specific mediations on maternal and neonatal outcomes when treatment during pregnancy is indicated. Given that the use of labetalol to treat hypertensions in pregnancy is rising rapidly [25], there is an urgent need to further study the potential adverse effect of labetalol. Ideally, randomized controlled trials making head-to-head comparison between labetalol and methyldopa with perinatal outcomes as the primary outcomes of interest should be conducted to definitively prove or disprove of the association between labetalol use in pregnancy and increased risks of adverse perinatal outcomes as observed in this study and previous studies.

Competing interests
The authors declared no conflict of interest.

Funding
This study was supported, in part, by a grant from the Ontario Ministry of Health and Long-term Care through a Drug Innovation grant (grant # 2008-007) and a grant from Canadian Institutes for Health Research (CIHR; grant # MOP 86537). Dr. Wen is a recipient of Mid-Career Award from CIHR's Institute for Gender-Ontario Women's Health Council. Dr. Krewski is the Natural Sciences and Engineering Research Council of Canada Chair in Risk Science at the University of Ottawa. Dr. Mark Walker is supported by a University of Ottawa Tier 1 Chair in Perinatal Epidemiology.

Ethical approval
This study has been approved by Research Ethics Board of the Ottawa Hospital (file # 2008471-01H).

Contributors
SWW and MW designed the study and obtained the funding. RHX and YG did the literature review, performed the statistical analysis, and drafted the manuscript. DK, DM, KN, MW, and SWW revised the manuscript critically.

Acknowledgements
The authors thank Dr. Xi-kuan Chen for his contribution at the early stage of this project. This study was based on de-identified data provided by the Saskatchewan Ministry of Health. The interpretation and conclusions contained herein do not necessarily represent those of the Government of Saskatchewan or the Saskatchewan Ministry of Health. We thank Dr. Monica Taljaard of the Ottawa Hospital Research Institute for assistance in statistical analysis of the data.

References
[1] SOGC. Hypertension and pregnancy guidelines. J Obstet Gynaecol Can. 2008;30:1-52
[2] J. Zhang, J. Zeisker, M.C. Hatch. Epidemiology of pregnancy-induced hypertension. Epidemiol Rev. 1997;19:218-232 Crossref.
[3] B.K. Wittmann, K.J. Murphy, J.F. King, B.H. Yuen, D. Shaw, A.G. Wittman. Maternal mortality in British Columbia in 1971-86. CMAJ. 1988;139:37-40
[4] National High Blood Pressure Education Working Group. Report on high blood pressure in pregnancy. Am J Obstet Gynecol. 1990;163:1691-1712
[5] L.A. Magee, E. Abalos, P. von Dadelszen, et al. How to manage hypertension in pregnancy effectively. Br J Clin Pharmacol. 2011;72:394-401 Crossref.
[6] T. Podymow, P. August. Update on the use of antihypertensive drugs in pregnancy. Hypertension. 2008;51:960-969 Crossref.
[7] E. Abalos, L. Duley, D.W. Steyn, D.J. Henderson-Smart. Antihypertensive drug therapy for mild to moderate hypertension during pregnancy. Cochrane Database Syst Rev. 2007;1:CD002252
[8] ACOG Practice Bulletin. Chronic hypertension in pregnancy. Obstet Gynecol. 2012;119:396-407
[9] A.V. Chobanian, G.L. Bakris, H.R. Black, et al. National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, and Treatment of high blood pressure. The Seventh Report of the Joint Committee. JAMA. 2003;289:2560-2572 Crossref.
[10] L. Magee, M. Helewa, J.M. Moutquin, P. von Dadelszen, Hypertension Guidelines Committee. Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy. JOGC. 2008;30(Suppl. 1):S1-S48
[11] L.A. Magee, L. Duley. Oral beta-blockers for mild to moderate hypertension during pregnancy. Cochrane Database Syst Rev. 2003;3:CD002863
[12] A.M. El-Qarmalawi, A.H. Morsy, A. Al-Fadly, A. Obeid, M. Hashem. Labetalol vs. methydopa in the treatment of pregnancy-induced hypertension. Int J Gynaecol Obstet. 1995;49:125-130 Crossref.
[13] K. Duckitt, D. Harrington. Risk factors for pre-eclampsia at antenatal booking: systematic review of controlled studies. BMJ. 2005;330:565 Crossref.
[14] M.S. Kramer, R.W. Platt, S.W. Wen, et al. A new and improved population-based Canadian reference of birth weight for gestational age. Pediatrics. 2001;108(2):e35 Crossref.
[15] K. Karlsson, Y. Lundgren, U. Ljungblad. The acute effects of a non-selective beta-adrenergic blocking agent in hypertensive pregnant rats. Acta Obstet Gynecol Scand Suppl. 1984;118:81-84 Crossref.
[16] I. Kjellmer, A. Dagbjartsson, A. Hrebk, K. Karlsson, K.G. Rosen. Maternal beta-adrenoceptor blockade reduces fetal tolerance to asphyxia. A study in pregnant sheep. Acta Obstet Gynecol Scand Suppl. 1984;118:75-80 Crossref.
[17] D.D. Houlihan, M.C. Dennedy, N. Ravikumar, J.J. Morrison. Anti-hypertensive therapy and the feto-placental circulation: effects on umbilical artery resistance. J Perinat Med. 2004;32:315-319
[18] S.N. Molvi, S. Mir, V.S. Rana, F. Jabeen, A.R. Malik. Role of antihypertensive therapy in mild to moderate pregnancy-induced hypertension: a prospective randomized study comparing labetalol with alpha methyldopa. Arch Gynecol Obstet. 2012;285:1553-1562 Crossref.
[19] S. Immanni, E.I. Khan, M. Staunton. Hypoglycaemia secondary to labetalol infusion. J Coll Physicians Surg Pak. 2011;21:297-298
[20] K.Y. Heida, G.G. Zeeman, T.R. Van Veen, C.V. Hulzebos. Neonatal side effects of maternal labetalol treatment in severe preeclampsia. Early Hum Dev. 2012;88:503-507 Crossref.
[21] Health databases in Saskatchewan. L.M. Strand, W. Downey, B.L. Strom (Eds.) Pharmacoepidemiology 2nd ed. (John Wiley & Sons, Chichester, 1994) 217-230
[22] S.L. West, A. Richter, C.A. Melfi, M. McNutt, M.E. Nennstiel, J.A. Mauskopf. Assessing the Saskatchewan database for outcomes research studies of depression and its treatment. J Clin Epidemiol. 2000;53:823-831 Crossref.
[23] L.T. De Jong van den Berg, N. Feenstra, H.T. Sorensen, M.C. Cornel. Improvement of drug exposure data in a registration of congenital anomalies. Pilot study: pharmacist and mother as sources for drug exposure data during pregnancy. Teratology. 1999;60:33-36 Crossref.
[24] K.J. Rothman, S. Greenland. Modern epidemiology. 2nd ed. (Lippincott Raven, Philadelphia, 1998)
[25] R.H. Xie, Y. Guo, D. Krewski, et al. Trends in using beta-blockers and methyldopa for hypertensive disorders during pregnancy in a Canadian population. Eur J Obstet Gynecol Reprod Biol. 2013;171:281-285 Crossref.