ABSTRACT
Objective: To relate anticoagulant use to pregnancy complications in women with systemic lupus erythematosus (SLE) and primary antiphospholipid syndrome (APS). Methods: All ongoing pregnancies, 184, in two Dutch tertiary centers between 2000 and 2015. Results: LMWH and aspirin was prescribed in 15/109 SLE women without antiphospholipid antibodies (aPL), 5/14 with aPL, 11/13 with APS, 45/48 with primary APS. Main complications in the four treatment groups (no anticoagulant treatment, aspirin, LMWH, aspirin and LMWH) included hypertensive disorders of pregnancy (9.4%, 23.3%, 50%, 18.4%, respectively, p = 0.12) and preterm birth (16.7%, 34.3%, 75%, 36.8%, respectively, p < 0.001). Conclusion: Maternal and perinatal complications occurred frequently, despite LMWH and aspirin use.
Introduction
Women with systemic lupus erythematosus (SLE) and antiphospholipid syndrome (APS) have an increased risk for pregnancy complications like hypertensive disorders of pregnancy (HD), preterm birth, intrauterine fetal death (IUFD) and small-for-gestational age (SGA) infants (1). The risk for women with SLE is two to four times higher compared to women without SLE (2,3). In 40% of SLE patients, antiphospholipid antibodies (aPL) are found. Only after occurrence of thrombosis, IUFD, HD combined with birth <34 weeks gestation or consecutive spontaneous miscarriages, patients are diagnosed with APS (primary APS in patients without auto-immune disease and secondary APS in patients with auto-immune disease) (4,5). Secondary APS has been reported in approximately 14% of the SLE patients (6).
Despite lack of evidence from randomized controlled trials (RCTs) for the use of low-molecular-weight heparin (LMWH) to prevent pregnancy complications in women with SLE, it is often prescribed. Aspirin is also often prescribed in women with SLE and/or primary APS without evidence from an RCT, since it prevents recurrent HD in other populations (7–10). To our knowledge, differences in pregnancy outcomes in relation to the use of LMWH and aspirin between women with SLE without aPL, SLE with aPL, SLE with APS and primary APS have not been well described yet. The current Dutch obstetric guideline advises LMWH for women with SLE and APS and aspirin may be considered for all women with SLE (11). Therapeutic advice of the Dutch guideline for treatment of women with primary APS is related to the cause of APS (e.g. thrombotic or obstetric complications) (12). The Royal College of Obstetrics and Gynaecology does not provide guidelines for these populations.
We have the opportunity to pool data from the rheumatology and obstetric departments of two tertiary centers during a 16-year period. The aim of this study is to describe LMWH and aspirin use during pregnancy in women with SLE and/or APS and the association with maternal and perinatal outcomes.
Specific questions are:
What are the anticoagulant treatment strategies used in women with SLE without aPL, SLE with aPL, SLE with APS and primary APS?
What is the incidence of maternal and perinatal complications in the four treatment groups: no anticoagulant treatment, aspirin only, LMWH only or both aspirin and LMWH?
This study is undertaken to provide more insight in anticoagulant therapy strategies in daily clinical practice and enhances our knowledge on the rationale of anticoagulant treatment in this specific population.
Methods
This cohort study was facilitated by a collaboration between two tertiary centers in the Netherlands performing joined care for pregnant women with SLE and/or APS by rheumatologists and obstetricians; VU University Medical Center in Amsterdam and University Medical Center Utrecht in Utrecht. All women with SLE and/or APS and an ongoing pregnancy (>16 weeks) between 2000 and 2015 were included. Women were included when SLE or APS was diagnosed before pregnancy. The revised American College of Rheumatology criteria were used for the classification of SLE (13). APS was diagnosed according to the Sapporo criteria (4,5). Presence of aPL was based on two positive measurements of anticardiolipin antibodies, lupus anticoagulants or beta2-glycoprotein antibodies, measured at least 6 weeks or 12 weeks apart, neither during nor within ten weeks after pregnancy (when applicable) (4,5). Samples were considered positive for anticardiolipin antibodies or beta-2-glycoprotein antibodies when either above 40GPL or above the 99th percentile.
Data collection
Data were derived from similar databases in both centers recording all pregnancies of SLE- and/or APS-patients since 2000. First, the digital data program Mosos (BMA BV, Houten, the Netherlands) of both obstetric departments was used, which is a nationwide used program reporting deliveries above 16 weeks gestational age. A search was performed in this database for SLE, APS and aPL. The second database on SLE and/or APS and pregnancy was derived from both departments of Rheumatology, the Amsterdam longitudinal SLE cohort study at the VU University Medical Center and the SLE or APS and pregnancies database at the University Medical Center Utrecht. After identification of the participants, the medical charts were checked to obtain information. The Institutional Review Boards of both university hospitals concluded that official approval from a medical ethical committee was not needed due to the strictly observational character of this study.
Assessments
Collection of demographic and clinical data from medical charts included age, ethnicity, body mass index (BMI) and smoking habits. The medical charts also contained information about aPL status and year of diagnosis of SLE and/or APS. In VU University Medical Center, presence of anticardiolipin antibodies and lupus anticoagulant was collected. In the University Medical Center Utrecht, also data on the presence of beta2-glycoprotein antibodies were available in several cases. Information on the general history included lupus nephritis, arterial thrombosis, venous thrombosis, chronic hypertension and diabetes mellitus. Moreover, in women with SLE, occurrence of flares and thrombocyte count was assessed within six months before pregnancy (14).
The information on the obstetric history included spontaneous miscarriages, HD (defined as preeclampsia, eclampsia or HELLP-syndrome), IUFD, placental abruption, preterm birth (defined as birth <37 weeks gestational age) and SGA infant (defined as birth weight <p10). Moreover, information about medication use during and after pregnancy, with specific notice of the use of LMWH and aspirin was collected. According to the protocol of both obstetric departments, it was strived to start treatment with LMWH and aspirin in the first trimester. Treatment with LMWH was classified into two doses; prophylactic and therapeutic dose (15,16). The information on maternal pregnancy complications included HD, placental abruption and preterm birth. Perinatal complications included IUFD, SGA infants and admission to neonatal intensive care unit (NICU) or medium care.
Statistics
Baseline characteristics were examined using descriptive statistics. Numbers and percentages of LMWH and aspirin usage per patient group were presented using descriptive statistics as well. For comparison of the four treatment groups concerning continuous variables, a linear regression analysis was performed, including a post-hoc analysis with Bonferroni correction for pairwise comparisons of treatment groups in case of a significant overall p-value for comparison of the four groups. To compare dichotomous variables, a logistic regression analysis was performed, also including a post-hoc analysis with Bonferroni correction. In these regression analyses we corrected for maternal age during pregnancy, BMI ≥ 25 kg/m2 and chronic hypertension. If the total number of women with a complication was too low and therefore a logistic regression analysis could not be performed, a Fisher’s exact test was performed. When the total number per patient group was sufficient, outcomes were compared between the four treatment groups using a Fisher’s exact test, including a post-hoc analysis with Bonferroni correction.
Statistical analyses were performed with SPSS 22.0 (SPSS Inc., Chicago, IL, USA). A two-sided p-value inferior to 0.05 was considered to be statistically significant.
Results
The inclusion criteria were met by 124 women with 184 pregnancies; 51 women with 75 pregnancies at the VU University Medical Center, 73 women with 109 pregnancies at the University Medical Center Utrecht. The baseline characteristics at the start of the pregnancy, including general and obstetric history, are depicted separately for each treatment group (). In six out of twelve pregnancies in the LMWH only group the ethnicity was non-Caucasian. These six pregnancies occurred in four women. SLE disease activity was low before pregnancy in 126 out of 136 SLE pregnancies, major flares did not occur, minor flares occurred in five pregnancies and disease activity before pregnancy was unknown in five pregnancies. None of the women with SLE did have a thrombocyte count below 100 before pregnancy. Hydroxychloroquine was used during pregnancy in 22/66 of the women treated without anticoagulant treatment, 14/30 treated with aspirin without LMWH, 0/12 treated with LMWH without aspirin and in 18/76 of the women treated with both aspirin and LMWH.
Table 1. Baseline characteristics per included pregnancy.
Anticoagulant treatment strategies used during pregnancy
Different anticoagulant treatment strategies in women with SLE without aPL, SLE with aPL, SLE with APS and primary APS are presented in . In the total population, LMWH was used in 47.8% and aspirin in 57.6% of the pregnancies. Aspirin was already used before pregnancy in nine women prior to 13 pregnancies (7.1% of all pregnancies). Three out of these nine women had a history of arterial thrombosis, in the other women the reason for chronic aspirin use was unknown.
Table 2. Treatment with aspirin and/or LMWH during and after pregnancy per patient group.
History of women who used LMWH are depicted in per patient group.
Table 3. History of women including prophylactic and therapeutic dose of low-molecular-weight heparin per patient group (15,16).
In the group SLE without aPL in two pregnancies, the dose and reason for LMWH was unknown. In five pregnancies in this patient group, no LMWH was used despite previous venous thrombosis.
In the group with primary APS in two pregnancies the dose of LMWH was unknown (history of previous arterial thrombosis (n = 1) and unknown (n = 1)).
Incidences of maternal and perinatal complications
Incidences of maternal and perinatal complications in the four treatment groups (no anticoagulant treatment, aspirin only, LMWH only or both aspirin and LMWH) are depicted in . Women receiving anticoagulant treatment had the highest frequency of maternal and perinatal complications, whereas women without anticoagulant treatment had least complications. The overall prevalence of HD was 18.1%, placental abruption 1.1%, preterm birth 33.2%, IUFD 4.8%, SGA infant 17.4% and admission of the infant to NICU or medium care 52.4% (25.3% and 27.1%, respectively). The delivery of one woman with primary APS occurred at 23 weeks gestational age due to HD, which resulted in the only neonatal death in this population. The analysis of SGA showed a quasi-separation of the data, caused by chronic hypertension; none of the women with chronic hypertension had an SGA infant. Sub-analysis without correction for chronic hypertension (only correction for maternal age during pregnancy and BMI ≥ 25 kg/m2) did not show any difference; the p-value remained the same (p = 0.39). Maternal and perinatal outcomes in the patient group of SLE without aPL were stratified according to treatment, which demonstrated that women with anticoagulant treatment experienced more maternal and perinatal complications compared to women without anticoagulant treatment (). The numbers of included patients in the other patient groups were too small to perform further statistical analysis.
Table 4. Maternal and perinatal outcomes per treatment group.
Table 5. Maternal and perinatal outcomes in women with SLE without aPL.
Discussion
In our cohort any anticoagulant treatment (LMWH and/or aspirin) was prescribed in 45.0% of women with SLE without aPL, in 57.1% of women with SLE with aPL and in 100% of women with SLE with APS or primary APS. The treatment group with aspirin and LMWH consisted mainly of women with SLE and APS or primary APS. The maternal and perinatal outcomes in the complete cohort showed that the subgroups with anticoagulant treatment experienced more maternal and perinatal complications compared to those without anticoagulant therapy. The same holds true for women with SLE without aPL. The overall incidence of maternal and perinatal complications was high, irrespective of treatment group and despite low SLE disease activity in the majority of the population within six months before pregnancy.
The value of anticoagulant treatment with LMWH and/or aspirin in women with SLE and/or APS is still under debate. Treatment with aspirin was recommended for all patients with SLE in a recent review article (7). This is, amongst others, derived from studies finding a beneficial effect of aspirin in other populations at high risk for HD or intra-uterine growth (10). Furthermore, treatment with LMWH in addition to treatment with aspirin was advised for women with either primary or secondary APS in review and overview articles (7,9,17,18). In women with SLE and aPL, authors of two retrospective studies suggested that LMWH treatment during pregnancy might be beneficial to reduce maternal and perinatal pregnancy complications (19,20). However, in both studies all patients with SLE and aPL used LMWH, in absence of a control group without LMWH use.
An RCT investigating the effect of treatment with aspirin and LMWH in women with SLE is lacking and a limited number of RCTs on this subject in women with primary APS have been published (21,22). To our knowledge, only two RCT’s investigating the effect of LMWH on second and third trimester pregnancy complications in which (also) women with primary APS were included have been published: the TIPPS trial (n = 22) and the FRUIT-RCT (n = 32) (21,22). In the TIPPS trial women with thrombophilia were included. Subgroup analysis in women with primary APS showed no beneficial effect of LMWH compared to treatment without LMWH (21). In the FRUIT-RCT, including only women with primary APS, also no beneficial effect of LMWH with aspirin was demonstrated compared to treatment with aspirin alone (22).
The exact mechanism of action of LMWH and aspirin in the prevention of pregnancy complications is still unclear and is probably not limited to the anticlotting mechanism.
LMWH probably has an early effect that may occur at cellular level, by decreasing trophoblast apoptosis and increasing the production of proteases involved in the trophoblast invasion of the maternal endometrium (23,24). In vitro studies have shown an effect of LMWH on angiogenesis in the placental villi and show an influence on the dysregulation of soluble vascular endothelial growth factor (25,26). Furthermore, an inhibiting effect of heparin on complement activation is reported, which could reduce the risk of pregnancy complications (27–29). On the other hand, LMWH might have an adverse effect by increasing soluble fms-like tyrosine-kinase-1 (a splice variant of vascular endothelial growth factor receptor) which contributes to HD (30).
Aspirin is thought to improve the trophoblastic invasion of the uterine spiral arteries and might subsequently improve the development and efficacy of the placenta probably due to thrombocyte aggregation inhibition and/or an anti-inflammatory working mechanism (10,31). These results, however, have been found in experimental setting and it is unknown what the relevance is for clinical practice.
In the present study, women with anticoagulant treatment during pregnancy had most maternal and perinatal complications. Evaluating our results, we suppose this finding is most likely explained by confounding by indication. Physicians did not prescribe LMWH and/or aspirin in cases with a perceived low a priori risk for complications, concerning both obstetrical and SLE and/or APS history. Moreover, SLE is a complex multi-organ autoimmune condition and poor obstetric outcomes are probably not only caused by thrombotic mechanisms but also influenced by disease activity and renal function before pregnancy (32,33). On the other hand, as stated before, the mechanisms of action of both LMWH and aspirin are probably not limited to the anticlotting mechanism. In all treatment groups, prevalences of complications were high when compared to healthy women. The results of our study, including a high prevalence of HD, preterm birth and admission of the neonate in patients with SLE or APS, are in line with other studies (2,3,34–36).
Beneficial effects of treatment with LMWH to prevent second and third trimester obstetric complications in women with SLE have not been reported. Approximately 2% of the pregnancies with daily LMWH injections are complicated by significant bleedings either antepartum or postpartum (16). To further develop evidence-based guidelines for LMWH use during pregnancy in women with SLE, a multicenter RCT is needed to examine the possible effect of LMWH in addition to treatment with aspirin.
Likewise, in women with primary APS a beneficial effect of LMWH on second and third trimester pregnancy outcomes has not been proven (21,22). For women with APS and a thrombotic event in history, treatment with LMWH during pregnancy is recommended, since LMWH has been reported to be effective in preventing venous thrombosis in pregnancy (16).
Presently, aspirin is advised in all women with SLE and/or (primary) APS with the drawback of lack of knowledge on the exact working mechanism and the absence of an underlying RCT. Due to limited side-effects and proven beneficial effects in other high-risk populations, the recommendation to use aspirin in the prevention of pregnancy complications was extended to women with SLE and primary APS (7,10).
To our knowledge, this is the first study presenting an overview of anticoagulant treatment in daily clinical practice in women with SLE and/or APS comparing pregnancy outcomes between different anticoagulant treatment groups. We were able to describe a considerable number of pregnancies in women with either SLE or primary APS in two tertiary centers performing joined care for pregnant SLE and APS women by rheumatologists and obstetricians in a 16-year period. The data on anticoagulant treatment, although retrospective, are valuable since there is a trend to prescribe LMWH and aspirin more frequently in patients with SLE nowadays. Therefore, comparison of the results of different treatment strategies on pregnancy outcomes might be more difficult in the future, since the number of pregnancies in which anticoagulant treatment is not applied will probably decrease. Moreover, analysis in relation to maternal and perinatal outcomes per treatment modality has not been published yet. Furthermore, anticoagulant treatment was started in the first trimester, which is important since in other populations aspirin use has been proven to be beneficial in the prevention of pregnancy complications if initiated before 16 weeks gestational age (10). Limitations of our study are the nature of the retrospective set-up making missing data inherent. In addition, the number of women in the treatment group with LMWH is small despite an observation period of 16 years, which impairs the ability to compare pregnancy outcomes in this subgroup with those in the other treatment groups.
In conclusion, this study provides an overview of LMWH and aspirin use in daily clinical practice in a population of women with SLE and/or APS. Pregnancy complications were frequent irrespective of treatment group. Women treated with anticoagulants showed even a higher frequency of maternal and perinatal complications compared to women without anticoagulant treatment, likely confounded by indication.
With the present knowledge, use of LMWH should be limited to women with a history of venous thrombosis. To examine the additional beneficial effect of LMWH besides aspirin on pregnancy outcome in women with SLE without a history of venous thrombosis, a RCT is needed. Prescription of aspirin is advised for all pregnant women with SLE and/or (primary) APS.
Declaration of interests
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Acknowledgments
We would like to thank Prof. A.E. Voskuyl, MD, PhD, rheumatologist at VU University Medical Center for his input in the design of this study. We would like to thank Prof. R.H. Derksen, MD, PhD, rheumatologist at University Medical Center Utrecht for his effort setting up the database in the University Medical Center Utrecht. We would like to thank P.M. van de Ven, PhD, statistician at VU University Medical Center for his help with the statistical analysis. We would like to thank Dr. A. Thijs, internal medicine specialist at VU University Medical Center for his feedback on the manuscript. We also would like to thank M. Demissi and D.E. van Zanten, MD for their help entering data in the database. We would like to thank the Nationale Vereniging LE patiënten (NVLE), a Dutch SLE patient organization for supporting this study with a research grant.
Funding
The study was supported by a single grant from the “Nationale Vereniging LE patiënten (NVLE)”.
Additional information
Funding
References
- Ostensen M, Clowse M. Pathogenesis of pregnancy complications in systemic lupus erythematosus. Curr Opin Rheumatol 2013;25(5):591–596.
- Barbhaiya M, Bermas BL. Evaluation and management of systemic lupus erythematosus and rheumatoid arthritis during pregnancy. Clin Immunol 2013.
- Clowse ME, Jamison M, Myers E, James AH. A national study of the complications of lupus in pregnancy. Am J Obstet Gynecol 2008;199(2):127–6.
- Wilson WA, Gharavi AE, Koike T, et al. International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop. Arthritis Rheum 1999;42(7):1309–1311.
- Miyakis S, Lockshin MD, Atsumi T, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4(2):295–306.
- Ruiz-Irastorza G, Egurbide MV, Ugalde J, Aguirre C. High impact of antiphospholipid syndrome on irreversible organ damage and survival of patients with systemic lupus erythematosus. Arch Intern Med 2004;164(1):77–82.
- Schramm AM, Clowse ME. Aspirin for prevention of preeclampsia in lupus pregnancy. Autoimmune Dis 2014;2014:920467.
- Les I, Ruiz-Irastorza G, Khamashta MA. Intensity and duration of anticoagulation therapy in antiphospholipid syndrome. Semin Thromb Hemost 2012;38(4):339–347.
- Ruiz-Irastorza G, Crowther M, Branch W, Khamashta MA. Antiphospholipid syndrome. Lancet 2010;376(9751):1498–1509.
- Bujold E, Roberge S, Lacasse Y, et al. Prevention of preeclampsia and intrauterine growth restriction with aspirin started in early pregnancy: a meta-analysis. Obstet Gynecol 2010; 116 (2 Pt 1): 402–414.
- Dutch Obstetric Guidelines SLE and pregnancy. 9-11-2007.
- Dutch Obstetric Guidelines antiphospholipid syndrome and pregnancy. 9-11-2007.
- Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1997;40(9):1725.
- Gladman DD, Ibanez D, Urowitz MB. Systemic lupus erythematosus disease activity index 2000. J Rheumatol 2002;29(2):288–291.
- Dutch Obstetric Guidelines venous thrombo-embolism. 12-6-2009.
- Bates SM, Greer IA, Middeldorp S, et al. VTE, thrombophilia, antithrombotic therapy, and pregnancy: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141(2Suppl): e691S–e736S.
- Danza A, Ruiz-Irastorza G, Khamashta M. Antiphospohlipid syndrome in obstetrics. Best Pract Res Clin Obstet Gynaecol 2012;26(1):65–76.
- de Jesus GR, Rodrigues G, de Jesus NR, Levy RA. Pregnancy morbidity in antiphospholipid syndrome: what is the impact of treatment? Curr Rheumatol Rep 2014;16(2):403.
- Mecacci F, Bianchi B, Pieralli A, et al. Pregnancy outcome in systemic lupus erythematosus complicated by anti-phospholipid antibodies. Rheumatology (Oxford) 2009;48(3):246–249.
- Ambrosio P, Lermann R, Cordeiro A, et al. Lupus and pregnancy–15 years of experience in a tertiary center. Clin Rev Allergy Immunol 2010; 38 (2–3): 77–81.
- Rodger MA, Hague WM, Kingdom J, et al. Antepartum dalteparin versus no antepartum dalteparin for the prevention of pregnancy complications in pregnant women with thrombophilia (TIPPS): a multinational open-label randomised trial. Lancet 2014;384(9955):1673–1683.
- van Hoorn ME, Hague WM, van Pampus MG, et al. Low-molecular-weight heparin and aspirin in the prevention of recurrent early-onset pre-eclampsia in women with antiphospholipid antibodies: the FRUIT-RCT. Eur J Obstet Gynecol Reprod Biol 2015;197:168–173.
- D’Ippolito S, Ortiz AS, Veglia M, et al. Low molecular weight heparin in obstetric care: a review of the literature. Reprod Sci 2011;18(7):602–613.
- Greer IA, Brenner B, Gris JC. Antithrombotic treatment for pregnancy complications: which path for the journey to precision medicine? Br J Haematol 2014;165(5):585–599.
- Sobel ML, Kingdom J, Drewlo S. Angiogenic response of placental villi to heparin. Obstet Gynecol 2011;117(6):1375–1383.
- Sela S, Natanson-Yaron S, Zcharia E, et al. Local retention versus systemic release of soluble VEGF receptor-1 are mediated by heparin-binding and regulated by heparanase. Circ Res 2011;108(9):1063–1070.
- Girardi G, Redecha P, Salmon JE. Heparin prevents antiphospholipid antibody-induced fetal loss by inhibiting complement activation. Nat Med 2004;10(11):1222–1226.
- Girardi G, Prohaszka Z, Bulla R, et al. Complement activation in animal and human pregnancies as a model for immunological recognition. Mol Immunol 2011;48(14):1621–1630.
- Oberkersch R, Attorresi AI, Calabrese GC. Low-molecular-weight heparin inhibition in classical complement activation pathway during pregnancy. Thromb Res 2010;125(5):e240–e245.
- Hagmann H, Bossung V, Belaidi AA, et al. Low-molecular weight heparin increases circulating sFlt-1 levels and enhances urinary elimination. PLoS One 2014;9(1):e85258.
- Bujold E, Roberge S, Nicolaides KH. Low-dose aspirin for prevention of adverse outcomes related to abnormal placentation. Prenat Diagn 2014; (7):642–648.
- Bramham K, Hunt BJ, Bewley S, et al. Pregnancy outcomes in systemic lupus erythematosus with and without previous nephritis. J Rheumatol 2011;38(9):1906–1913.
- Ateka-Barrutia O, Khamashta M. The challenge of pregnancy for patients with SLE. Lupus 2013;22(12):1295–1308.
- Jakobsen IM, Helmig RB, Stengaard-Pedersen K. Maternal and foetal outcomes in pregnant systemic lupus erythematosus patients: an incident cohort from a stable referral population followed during 1990–2010. Scand J Rheumatol 2015;44(5):377–384.
- Hamed HO, Ahmed SR, Alzolibani A, et al. Does cutaneous lupus erythematosus have more favorable pregnancy outcomes than systemic disease? A two-center study. Acta Obstet Gynecol Scand 2013;92(8):934–942.
- Buyon JP, Kim MY, Guerra MM, et al. Predictors of pregnancy outcomes in patients with lupus: a Cohort Study. Ann Intern Med 2015;163(3):153–163.