- Email: [email protected]
Errors in the treatment of hypertensive disorders of pregnancy and their impact on maternal mortality
International Journal of Gynecology and Obstetrics 121 (2013) 78–81
Contents lists available at SciVerse ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Errors in the treatment of hypertensive disorders of pregnancy and their impact on maternal mortality John J. Zuleta-Tobón a, b,⁎, Heleodora Pandales-Pérez a, Sandra Sánchez a, Gladis A. Vélez-Álvarez a, b, Jesús A. Velásquez-Penagos b a b
Department of Obstetrics and Gynecology, University of Antioquia, Medellín, Colombia NACER, Sexual and Reproductive Health, University of Antioquia, Medellín, Colombia
a r t i c l e
i n f o
Article history: Received 9 July 2012 Received in revised form 11 October 2012 Accepted 18 December 2012 Keywords: Maternal mortality Pre-eclampsia Substandard care
a b s t r a c t Objective: To describe the patients' characteristics and the factors that contributed to the maternal deaths associated with hypertensive disorders of pregnancy that occurred in the department of Antioquia, Colombia, from 2004 through 2011. Methods: A committee of experts conducted a retrospective descriptive study to analyze the information obtained from both mandatory reports of health facilities to the Public Health Surveillance System and interviews with family members. Results: From 2004 through 2011, there were 720 170 births and 389 maternal deaths in the Department of Antioquia, and 70 of the deaths were due to hypertensive disorder of pregnancy. The factors that most contributed to the deaths were a lack of emergency administration of antihypertensive drugs (64.6%); the inadequate administration of antihypertensive drugs (58.8%); retaining the patient at a health facility ill equipped to treat her appropriately for her clinical state (54.7%); untimely referral or inadequate conditions for transfer (50.8%); and an error in classifying the severity of the disorder, which prevented appropriate management (49.1%). Conclusion: A substandard quality of care was the determining factor in the deaths of women who presented with hypertensive disorders of pregnancy. © 2012 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The incidence of pre-eclampsia varies between 3% and 10% among pregnant women in high-income countries, and the global number of maternal deaths due to pre-eclampsia is estimated to exceed 63 000 each year [1]. Pre-eclampsia is the major cause of maternal death in Latin America and the Caribbean, with an incidence of 25.7% [2]. In Colombia, pre-eclampsia was the second-leading cause of maternal deaths in 2010, accounting for 37% of the 416 that occurred [3]. In the department of Antioquia, between 2004 and 2007, the leading cause was obstetric hemorrhage. However, by 2008, owing to a decrease in the number of deaths from hemorrhage, hypertensive disorder of pregnancy (HDP) became the leading cause, with an incidence rate of 27% [4]. Identifying risk factors that could be controlled, and enacting control, may have significantly contributed to the reduction in maternal mortality from hemorrhage in the department of Antioquia [5]. The success of the intervention later prompted the design and implementation of both an evaluation strategy and a training program for medical personnel, and resulted in the improved management of this complication [6].
⁎ Corresponding author at: Calle 70 nro. 52–72 of. 504 Medellín, Antioquia, Colombia 050010. Tel.: +57 4 2635600, +57 4 2195400; fax: +57 4 2191031. E-mail address: [email protected] (J.J. Zuleta-Tobón).
Maternal death can be associated with inadequate care as well as with risks of pregnancy and childbirth [7]. The objective of the present study was to analyze the personal characteristics and the factors related to inadequate care contributing to the deaths of women with HDP in the department of Antioquia from 2004 through 2011. In an approach similar to that developed for obstetric hemorrhage, the objective was also to identify critical benchmarks in the management of women with HDP, develop management criteria for each of these benchmarks, and implement training interventions.
2. Materials and methods A retrospective descriptive study was conducted by an investigating committee of the Department of Antioquia, Colombia. Although the healthcare delivery system of Antioquia comprises 184 facilities, some of the residents of the 126 municipalities can have difficulty accessing health services. In Antioquia, 98.8% of deliveries are institutional, 93.7% of the women have at least 1 prenatal visit, and 84.7% have at least 4 prenatal visits. All institutional births are attended by physicians. All maternal deaths with HDP as the underlying cause that occurred in the department from 2004 through 2011 were analyzed. Maternal death was defined as in the International Statistical Classification of Diseases and Related Health Problems, 10th Revision [8].
0020-7292/$ – see front matter © 2012 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2012.10.031
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
The cases were collected by reviewing the Population Registry, which contains the mandatory reports from health facilities to the Public Health Surveillance System, and death certificates for all women of childbearing age. The sources of information concerning the deaths included medical records, interviews with the families, analyses performed at the health facilities and/or by the departmental epidemiologic surveillance committee, and autopsy reports (when available). Permission for publication was obtained from the Regional Health and Social Protection Secretary of Antioquia. The 3-delays model proposed by Maine et al. [9] was used: delay 1, delay in deciding to seek care; delay 2, delay in reaching a treatment facility; and delay 3, delay in receiving adequate treatment at the facility. There is no official guideline for pre-eclampsia management in Antioquia, and each facility has its own management protocol; however, the local recommendations for the management of pre-eclampsia were consistently based on the best available scientific evidence [10–12]. The effect of the lack of adherence to the recommendations was graded by consensus. Prenatal care was defined as receiving at least 1 complete visit. The following information was recorded during family interviews, which were the primary information source: Whether the deceased woman had received any treatments from nonmedical persons; whether any financial, administrative, or transportation difficulties delayed access to medical care; and whether the woman decided to forgo a consultation owing to a negative perception of healthcare facilities or personnel. A delay in the identification of HDP was defined as the recording of high blood pressure values or premonitory signs or symptoms of HDP in the medical chart without HDP being considered as a diagnosis. The treatment of the patient was evaluated to determine whether, after initial stabilization, the personnel retained the care of the patient even though her clinical state required treatment at a higher-complexity medical facility. The referral process was graded with respect to timeliness and the necessity for prompt action, the means of transportation used, and the suitability of such transportation for transferring sick patients. The physical examinations, interviews, and laboratory tests were evaluated to determine whether their omission or misinterpretation delayed decision making or elicited inappropriate treatment. The non-administration of an antihypertensive agent was recorded if an emergency medication was not ordered or administrated when a patient's blood pressure reached or exceeded 160/110 mm Hg. An inadequate use of an antihypertensive agent was recorded if labetalol, nifedipine, hydralazine, or sodium nitroprusside was not used at the proper time or dosage. An absence of magnesium sulfate prophylaxis was recorded if an initial intravenous bolus of 4 to 6 g magnesium sulfate was not ordered for patients with symptoms of cortical irritation before the presumptive or confirmed diagnosis of severe pre-eclampsia was reached, or if a maintenance dose of 1 to 2 g per hour of magnesium sulfate was not ordered for these patients after they received the initial bolus. An absence of prophylaxis in the postpartum was recorded if the administration of the chosen hypertensive agent was not continued for at least 24 hours postpartum. The non-administration of an additional intravenous bolus of magnesium sulfate in patients with eclampsia was also recorded, as was the lack of strict monitoring for signs of toxicity in these patients. The administration, and frequency of administration, of medications such as α-methyldopa, captopril, and diazepam, which are not recommended for use in the management of pre-eclampsia, hypertensive crises, or eclampsia, were also recorded. Finally, the unavailability of a required treatment at a healthcare facility with the appropriate self-declared level of complexity was recorded as a lack of resources. A database was created using Microsoft Access 2007 (Microsoft, Redmond, WA, USA) and the data were analyzed using SPSS version 19 (IBM, Armonk, NY, USA). Maternal age and gestational age are presented as mean and standard deviation; discrete quantitative
79
variables are presented as median and interquartile percentiles; and qualitative variables are presented as number and percentage. Denominators indicate the number of patients known as having had the relevant characteristic. Research information is at the basis of the epidemiologic surveillance system of the department, which is supported and financed by the Regional Health and Social Protection Office. Family members provided informed consent before being interviewed, and the health facilities provided consent prior to the review of clinical records. 3. Results According to the national vital statistics system, 707 018 live births occurred in Antioquia from January 1, 2004, through December 31, 2011 (official data for 2004–2010 and preliminary data for 2011) [4]. Of the 389 maternal deaths that occurred in the same period, 70 (18.0%) were due to HDP, making HDP the second-leading cause of maternal death after obstetric hemorrhage. The overall maternal mortality ratio in Antioquia was 55.01 per 100 000 live births, and the ratio for deaths caused by HDP was 9.9 per 100 000 live births. Two patients were excluded from analysis because of insufficient information. Among the 68 remaining patients, the main complications associated with HDP were eclampsia (n = 45 [66.1%]); hemolysis, elevated liver enzyme levels, and low platelet count (HELLP) syndrome (n = 39 [57.3%]); cerebrovascular disease (n = 34 [50%]); acute kidney failure (n = 21 [30.9%]); acute pulmonary edema (n = 19 [28.0%]); hemorrhagic complication (n = 13 [19.1%]); and placental abruption (n = 8 [11.8%]). Table 1 presents the characteristics of the study population. Women sought care for the following: premonitory symptoms (n= 27 [39.7%]); eclampsia (n= 19 [27.9%]); labor (n=9 [13.2%]); and other (n= 11 [16.2%]). Two more women (2.9%) were dead on arrival. Eleven of the 63 women (17.5%) for whom information was available were hospitalized for hypertension during their pregnancy and then discharged. The mean± SD gestational age at the onset of symptoms was 34±4.5 weeks, and the median time between the onset of symptoms and consultation with a physician was 23.5 hours (interquartile range, 4.8–48.0 hours). The family interviews revealed that the pregnancy was not intended in 53.7% of cases (the information was missing for 27 women). Table 2 describes the factors that contributed to the deaths according to the type of delay. Only those the investigating committee considered to have contributed directly to the deaths were taken into account. The supplies, materials, or resources that were not available at the facilities where the patients were treated, and whose lack of availability contributed to the deaths, included antihypertensive (19.1%) and hemoderivative (4.4%) agents, an intensive care unit or other place to perform close monitoring (2.9%), computed tomographic imaging (2.9%), and magnesium sulfate (1.5%). 4. Discussion In Antioquia, deficiencies in quality of care are strong determining factors for the death of women who present with HDP. In the present study, two-thirds of the patients who died from HDP presented to healthcare facilities in a timely manner, but the correct measures were not applied soon enough to prevent the progression of the condition. Some patients who arrived with an advanced stage of HDP did not receive the care required to stabilize their condition. One study quantified the reduction in mortality due to hypertensive complications of pregnancy after evidence-based practices were implemented, and reported that the in-hospital mortality of women with severe pre-eclampsia or eclampsia could be reduced by more than 84%, even when there was a delay in seeking medical attention [13].
80
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
Table 1 Characteristics of the study population.a
Table 2 Factors contributing to maternal deaths associated with HDP, using the 3-delays model.
Characteristicb
Value
Contributing factor
Age, y (n = 68) Women with prenatal visits (n = 63) Median number of prenatal visits Median parity Multiple gestation Relevant medical history Chronic hypertension (n = 63) Pre-eclampsia (n = 55) Diabetes (n = 61) Autoimmune disease (n = 60) Area of residence (n = 56) Rural residence Urban residence Marital status (n = 63) Married Common union Single Separated Education level (n = 51) None Primary, incomplete Primary, complete Secondary, incomplete Secondary, complete University or technological Place of death (n = 68) Home Institution Primary Secondary Tertiary During transfer In a public place Time of death relative to delivery (n = 68) Before b24 hours postpartum 2–7 days postpartum 8–42 days postpartum
29.1 ± 8.8 49 (77.8) 4 (1–5) 2 (1–4) 2 (2.9)
Delay 1 Negative attitude of the woman toward healthcare services Initial consultation with a traditional healer Delay in consulting a physician for economic reasons Delay 2 Delay in reaching a healthcare facility for an initial consultation Delay 3 Misdiagnosis Mistake in classifying HDP Delay in suspecting HDP Delay in performing laboratory tests The laboratory results were not returned Substandard physical examination Substandard medical interview Poor interpretation of laboratory results Inadequate clinical monitoring Inadequate monitoring of warning signs in the postpartum Inadequate monitoring of magnesium sulfate administration, with non-detection of magnesium sulfate poisoning Incomplete laboratory tests Postpartum bleeding not quantified Non-quantification of diuresis Treatment failure Non-administration of emergency antihypertensive agents Administration of inadequate antihypertensive agents No prophylaxis with magnesium sulfate to prevent eclampsia Non-administration of magnesium sulfate in patients with eclampsia Administration of medications not recommended for the treatment of eclampsia Inadequate management of intravenous fluids Non-admission to an intensive care unit when indicated Non-administration of blood products when indicated No prophylaxis with magnesium sulfate to prevent eclampsia in the postpartum Inadequate decision Retaining management of a patient when the facility was inadequate for her clinical state Untimely referral Inadequate decisions after receiving the laboratory results Discharge before 48 hours Other Resources not matching the facility's level of complexity Substandard teamwork Inadequate referral conditions Administrative difficulties that hampered care
17 (27.0) 9 (13.2) 1 (1.6) 0(0.0) 29 (42.6) 39 (57.4) 17 (27.0) 35 (55.6) 10 (15.9) 1 (1.6) 4 (7.4) 7 (13.0) 15 (27.8) 9 (16.7) 12 (22.2) 4 (5.9) 4 (5.9) 7 (10.3) 14 (20.6) 39 (57.4) 3 (4.4) 1 (1.5) 17 23 18 10
(25.0) (33.8) (26.5) (14.7)
a
Values are given as mean ± SD, number (percentage), or median (interquartile range). b The indicated number of patients is the number for whom the information was available.
In the present study, some women were not instructed to seek the care of a physician when warning signs or symptoms presented, and some of those who consulted a physician did not receive appropriate treatment—a situation also noted by others [14]. Moreover, because there were problems assessing the severity of HDP, the potential complications were probably underestimated. Consequently, available treatment may not have been administered at an opportune time or at all, or the necessary human and technological resources were not available at the facilities reached by the patients. For women who consulted a physician at an early stage of the disease, the most influential factors of mortality were omissions and errors in the pharmacologic management. Studies that have evaluated the implementation of evidence-based practices for the treatment of pre-eclampsia insist that actions at the primary care level should be directed toward the early identification, stabilization, and timely referral of women with complications or who are at risk for complications [13]. A significant number of women did not receive magnesium sulfate for the prevention or initial management of eclampsia. A work group cited by the Program for Appropriate Technology in Health described the administration of magnesium sulfate as the most important action to prevent death from eclampsia [15]. The situation observed in the present study is common elsewhere as well. A study performed in hospitals in Argentina and Uruguay between 2003 and 2005 reported the use of magnesium sulfate in only 33% of women with pre-eclampsia to prevent eclampsia, and in only 58.3% of women with eclampsia to treat the condition [16].
No. (%)a 14/59 (23.7) 7/62 (11.3) 6/56 (10.7) 17/59 (28.8)
34/64 25/61 12/63 6/52 6/62 5/57 4/49
(49.1) (41.0) (17.6) (11.5) (9.7) (8.8) (8.2)
5/46 (10.9) 3/46 (6.5) 4/48 (5.9) 2/48 (4.2) 2/61 (3.3) 42/65 30/51 17/48 14/43 10/46
(64.6) (58.8) (35.4) (32.6) (21.7)
12/62 11/61 5/44 3/45
(19.4) (18.0) (11.4) (6.7)
35/64 (54.7) 32/63 (50.8) 15/50 (22.1) 1/22 (4.5) 19/63 14/57 11/47 7/64
(30.2) (24.6) (23.3) (10.9)
a Denominators indicate the number of patients for whom the information was available or relevant.
Two-thirds of the maternal deaths from HDP were considered secondary to eclampsia and its complications. The frequency of pre-eclampsia has remained stable in both high-income and low-income countries. However, there have been much larger decreases in the incidence of eclampsia in the former than in the latter, with the incidence 10-fold lower in high-income countries [1]. Likewise, mortality from preeclampsia is 10-fold higher in low-income than in high-income countries [1]. In Antioquia, during the period studied, the maternal mortality ratio for HDP was lower than it was for Latin American and the Caribbean combined for 2005, which was 33.4 per 100 000 live births, but well above the 1.4 per 100 000 for high-income countries [13]. It is believed that inadequate practices directly contribute to more than one-third of all maternal deaths [17]. In a study conducted in Tanzania, only 63% of patients with eclampsia received treatment for severe hypertension, the patients were poorly monitored during labor, and they were inadequately prescribed magnesium sulfate and other intravenous fluids [14]. A multicenter study in Argentine about maternal mortality found that women experienced misdiagnosis, inappropriate treatment, and delays in referral, and that there was an overall shortage of supplies and skilled personnel [18]. These situations are less severe but not unknown in high-income countries. The underestimation of clinical symptoms by physicians, and the non-administration or insufficient
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
administration of required antihypertensive agents and magnesium sulfate have been reported in the Netherlands [19]. In the Report of the Confidential Enquiries into Maternal Deaths in the UK covering the 2006 through 2008 period, 20 of the 22 women who died from HDP had received substandard care [20]. Consistent with the present study, the absence of an adequate antihypertensive treatment was identified in that report as the most frequent treatment failure. Although basic interventions to reduce maternal mortality are well known, simple to perform, and even profitable, vast segments of the world population do not benefit from them, as verified by the findings of the present study [7]. It is not difficult to determine what practices should be observed to reduce maternal mortality due to pre-eclampsia but the difficulty lies in knowing how to implement these practices. In 2001, Villar et al. [21] suggested that factors influencing the non-implementation of established measures in low-income countries included the attitude of the healthcare provider, the use of outdated information, the passive transmission of knowledge, and a loss of interest in updating received knowledge. Grol and Grimshaw [22] noted that sufficient evidence indicates that a change in behavior is possible but that such a change would require comprehensive approaches adapted to different scenarios and target groups. The present study has several strengths. First, it transcends the description of features and non-modifiable risk factors [23,24]. Moreover, it provides an analysis of the medical care factors that directly contribute to maternal mortality from HDP, and can be modified at all levels of care. It was recommended that information obtained in the study of maternal mortality be suited to a purpose [17]. In the present case, the obtained information precisely identifies the clinical aspects that must be prioritized. The second strength of the present study is that it decreases variability in the judgments made by the investigating committee, as the findings concerning clinical history were compared with recognized evidence-based clinical practice guidelines. Its third strength is that it includes all the maternal deaths from HDP that occurred in an entire department of Colombia over 7 years. The extended period enables more reliable conclusions and a greater possibility of extrapolation. It is likely that the knowledge obtained will also be useful in other regions of Colombia, and probably also in other low-income countries. One limitation is that the study does not rely on a baseline number of HDP cases in Antioquia to calculate the incidence and lethality of the condition. Additionally, some subjectivity in the judgments made may persist because pre-eclampsia is a dynamic disease, and many variations in the care process hinder analysis. In conclusion, because pre-eclampsia cannot be prevented or predicted, it is necessary to rely on trained personnel for the early detection and adequate management of the disease, to have emergency resources available for conducting initial treatment at all levels of care, and to terminate the pregnancy at the right time under safe conditions. One of the steps required for ensuring that the personnel are qualified is to design and implement training and retraining programs based on innovative and practical methodologies. Acknowledgments The study was financed by the Secretaría Seccional de Salud y Protección Social de Antioquia. Conflict of interest The authors have no conflicts of interest.
81
References [1] Goldenberg RL, McClure EM, Macguire ER, Kamath BD, Jobe AH. Lessons for low-income regions following the reduction in hypertension-related maternal mortality in high-income countries. Int J Gynecol Obstet 2011;113(2):91–5. [2] Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet 2006;367(9516):1066–74. [3] National Institute of Health. National Epidemiological Report 2010. http://190.26. 202.205/index.php?idcategoria=92089#. [Published 2011. Accessed December 13, 2012]. [4] National Bureau of Statistics DANE. Vital Statistics. http://www.dane.gov.co/ index.php?option=com_content&view=article&id=73&Itemid=119. [Updated November 22, 2012. Accessed September 2, 2012]. [5] Vélez Álvarez GA, Gómez Dávila JG, Zuleta Tobón JJ. Analysis of maternal deaths due to hemorrhage in the department of Antioquia, Colombia. Years 2004 and 2005. Rev Colomb Obstet Ginecol 2006;57(3):147–55. [6] Vélez Álvarez GA, Agudelo-Jaramillo B, Gómez Dávila JG, Zuleta Tobón JJ. Code Red: guidelines for the management of obstetric hemorrhage. Rev Colomb Obstet Ginecol 2009;60(1):34–48. [7] Freedman LP, Waldman RJ, de Pinho H, Wirth ME, Chowdhury AM, Rosenfield A. Transforming health systems to improve the lives of women and children. Lancet 2005;365(9463):997–1000. [8] WHO Collaborating Centres for Classification of Diseases. International Statistical Classification of Diseases and Related Health Problems, 10th Revision: Instruction manual. International Statistical Classification of Diseases and Related Health Problems, vol. 2. Geneva: World Health Organization; 1992. [9] Maine D, Akalin MZ, Ward VM, Kamara A. The Design and Evaluation of Maternal Mortality Programs. http://www.rhrc.org/resources/DesignandEvalofMMPrograms. pdf. [Published June 1997]. [10] ACOG Committee on Practice Bulletins–Obstetrics. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Number 33, January 2002. Obstet Gynecol 2002;99(1):159–67. [11] Lowe SA, Brown MA, Dekker GA, Gatt S, McLintock CK, McMahon LP, et al. Guidelines for the management of hypertensive disorders of pregnancy 2008. Aust N Z J Obstet Gynaecol 2009;49(3):242–6. [12] National Collaborating Centre for Women's and Children's Health (UK). Hypertension in Pregnancy: The Management of Hypertensive Disorders During Pregnancy. London: RCOG Press; 2010. [13] Ronsmans C, Campbell O. Quantifying the fall in mortality associated with interventions related to hypertensive diseases of pregnancy. BMC Public Health 2011;11(Suppl. 3):S8. [14] Tsu VD, Shane B. New and underutilized technologies to reduce maternal mortality: call to action from a Bellagio workshop. Int J Gynecol Obstet 2004;85(Suppl. 1): S83–93. [15] Karolinski A, Mazzoni A, Belizán JM, Althabe F, Bergel E, Buekens P. Lost opportunities for effective management of obstetric conditions to reduce maternal mortality and severe maternal morbidity in Argentina and Uruguay. Int J Gynecol Obstet 2010;110(2):175–80. [16] Kidanto HL, Mogren I, Massawe SN, Lindmark G, Nystrom L. Criteria-based audit on management of eclampsia patients at a tertiary hospital in Dar es Salaam, Tanzania. BMC Pregnancy Childbirth 2009;9:13. [17] Ronsmans C, Graham WJ, Lancet Maternal Survival Series steering group. Maternal mortality: who, when, where, and why. Lancet 2006;368(9542):1189–200. [18] Ramos S, Karolinski A, Romero M, Mercer R, Maternal Mortality in Argentina Study Group. A comprehensive assessment of maternal deaths in Argentina: translating multicentre collaborative research into action. Bull World Health Organ 2007;85(8):615–22. [19] Schutte JM, Schuitemaker NW, van Roosmalen J, Steegers EA, Dutch Maternal Mortality Committee. Substandard care in maternal mortality due to hypertensive disease in pregnancy in the Netherlands. BJOG 2008;115(6):732–6. [20] Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al. Saving Mothers' Lives: Reviewing maternal deaths to make motherhood safer: 2006-2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118(Suppl. 1):1–203. [21] Villar J, Carroli G, Gülmezoglu AM. The gap between evidence and practice in maternal healthcare. Int J Gynecol Obstet 2001;75(Suppl. 1):S47–54. [22] Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients' care. Lancet 2003;362(9391):1225–30. [23] Moodley J. Maternal deaths associated with hypertensive disorders of pregnancy: a population-based study. Hypertens Pregnancy 2004;23(3):247–56. [24] MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001;97(4):533–8.
Contents lists available at SciVerse ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
Errors in the treatment of hypertensive disorders of pregnancy and their impact on maternal mortality John J. Zuleta-Tobón a, b,⁎, Heleodora Pandales-Pérez a, Sandra Sánchez a, Gladis A. Vélez-Álvarez a, b, Jesús A. Velásquez-Penagos b a b
Department of Obstetrics and Gynecology, University of Antioquia, Medellín, Colombia NACER, Sexual and Reproductive Health, University of Antioquia, Medellín, Colombia
a r t i c l e
i n f o
Article history: Received 9 July 2012 Received in revised form 11 October 2012 Accepted 18 December 2012 Keywords: Maternal mortality Pre-eclampsia Substandard care
a b s t r a c t Objective: To describe the patients' characteristics and the factors that contributed to the maternal deaths associated with hypertensive disorders of pregnancy that occurred in the department of Antioquia, Colombia, from 2004 through 2011. Methods: A committee of experts conducted a retrospective descriptive study to analyze the information obtained from both mandatory reports of health facilities to the Public Health Surveillance System and interviews with family members. Results: From 2004 through 2011, there were 720 170 births and 389 maternal deaths in the Department of Antioquia, and 70 of the deaths were due to hypertensive disorder of pregnancy. The factors that most contributed to the deaths were a lack of emergency administration of antihypertensive drugs (64.6%); the inadequate administration of antihypertensive drugs (58.8%); retaining the patient at a health facility ill equipped to treat her appropriately for her clinical state (54.7%); untimely referral or inadequate conditions for transfer (50.8%); and an error in classifying the severity of the disorder, which prevented appropriate management (49.1%). Conclusion: A substandard quality of care was the determining factor in the deaths of women who presented with hypertensive disorders of pregnancy. © 2012 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction The incidence of pre-eclampsia varies between 3% and 10% among pregnant women in high-income countries, and the global number of maternal deaths due to pre-eclampsia is estimated to exceed 63 000 each year [1]. Pre-eclampsia is the major cause of maternal death in Latin America and the Caribbean, with an incidence of 25.7% [2]. In Colombia, pre-eclampsia was the second-leading cause of maternal deaths in 2010, accounting for 37% of the 416 that occurred [3]. In the department of Antioquia, between 2004 and 2007, the leading cause was obstetric hemorrhage. However, by 2008, owing to a decrease in the number of deaths from hemorrhage, hypertensive disorder of pregnancy (HDP) became the leading cause, with an incidence rate of 27% [4]. Identifying risk factors that could be controlled, and enacting control, may have significantly contributed to the reduction in maternal mortality from hemorrhage in the department of Antioquia [5]. The success of the intervention later prompted the design and implementation of both an evaluation strategy and a training program for medical personnel, and resulted in the improved management of this complication [6].
⁎ Corresponding author at: Calle 70 nro. 52–72 of. 504 Medellín, Antioquia, Colombia 050010. Tel.: +57 4 2635600, +57 4 2195400; fax: +57 4 2191031. E-mail address: [email protected] (J.J. Zuleta-Tobón).
Maternal death can be associated with inadequate care as well as with risks of pregnancy and childbirth [7]. The objective of the present study was to analyze the personal characteristics and the factors related to inadequate care contributing to the deaths of women with HDP in the department of Antioquia from 2004 through 2011. In an approach similar to that developed for obstetric hemorrhage, the objective was also to identify critical benchmarks in the management of women with HDP, develop management criteria for each of these benchmarks, and implement training interventions.
2. Materials and methods A retrospective descriptive study was conducted by an investigating committee of the Department of Antioquia, Colombia. Although the healthcare delivery system of Antioquia comprises 184 facilities, some of the residents of the 126 municipalities can have difficulty accessing health services. In Antioquia, 98.8% of deliveries are institutional, 93.7% of the women have at least 1 prenatal visit, and 84.7% have at least 4 prenatal visits. All institutional births are attended by physicians. All maternal deaths with HDP as the underlying cause that occurred in the department from 2004 through 2011 were analyzed. Maternal death was defined as in the International Statistical Classification of Diseases and Related Health Problems, 10th Revision [8].
0020-7292/$ – see front matter © 2012 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijgo.2012.10.031
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
The cases were collected by reviewing the Population Registry, which contains the mandatory reports from health facilities to the Public Health Surveillance System, and death certificates for all women of childbearing age. The sources of information concerning the deaths included medical records, interviews with the families, analyses performed at the health facilities and/or by the departmental epidemiologic surveillance committee, and autopsy reports (when available). Permission for publication was obtained from the Regional Health and Social Protection Secretary of Antioquia. The 3-delays model proposed by Maine et al. [9] was used: delay 1, delay in deciding to seek care; delay 2, delay in reaching a treatment facility; and delay 3, delay in receiving adequate treatment at the facility. There is no official guideline for pre-eclampsia management in Antioquia, and each facility has its own management protocol; however, the local recommendations for the management of pre-eclampsia were consistently based on the best available scientific evidence [10–12]. The effect of the lack of adherence to the recommendations was graded by consensus. Prenatal care was defined as receiving at least 1 complete visit. The following information was recorded during family interviews, which were the primary information source: Whether the deceased woman had received any treatments from nonmedical persons; whether any financial, administrative, or transportation difficulties delayed access to medical care; and whether the woman decided to forgo a consultation owing to a negative perception of healthcare facilities or personnel. A delay in the identification of HDP was defined as the recording of high blood pressure values or premonitory signs or symptoms of HDP in the medical chart without HDP being considered as a diagnosis. The treatment of the patient was evaluated to determine whether, after initial stabilization, the personnel retained the care of the patient even though her clinical state required treatment at a higher-complexity medical facility. The referral process was graded with respect to timeliness and the necessity for prompt action, the means of transportation used, and the suitability of such transportation for transferring sick patients. The physical examinations, interviews, and laboratory tests were evaluated to determine whether their omission or misinterpretation delayed decision making or elicited inappropriate treatment. The non-administration of an antihypertensive agent was recorded if an emergency medication was not ordered or administrated when a patient's blood pressure reached or exceeded 160/110 mm Hg. An inadequate use of an antihypertensive agent was recorded if labetalol, nifedipine, hydralazine, or sodium nitroprusside was not used at the proper time or dosage. An absence of magnesium sulfate prophylaxis was recorded if an initial intravenous bolus of 4 to 6 g magnesium sulfate was not ordered for patients with symptoms of cortical irritation before the presumptive or confirmed diagnosis of severe pre-eclampsia was reached, or if a maintenance dose of 1 to 2 g per hour of magnesium sulfate was not ordered for these patients after they received the initial bolus. An absence of prophylaxis in the postpartum was recorded if the administration of the chosen hypertensive agent was not continued for at least 24 hours postpartum. The non-administration of an additional intravenous bolus of magnesium sulfate in patients with eclampsia was also recorded, as was the lack of strict monitoring for signs of toxicity in these patients. The administration, and frequency of administration, of medications such as α-methyldopa, captopril, and diazepam, which are not recommended for use in the management of pre-eclampsia, hypertensive crises, or eclampsia, were also recorded. Finally, the unavailability of a required treatment at a healthcare facility with the appropriate self-declared level of complexity was recorded as a lack of resources. A database was created using Microsoft Access 2007 (Microsoft, Redmond, WA, USA) and the data were analyzed using SPSS version 19 (IBM, Armonk, NY, USA). Maternal age and gestational age are presented as mean and standard deviation; discrete quantitative
79
variables are presented as median and interquartile percentiles; and qualitative variables are presented as number and percentage. Denominators indicate the number of patients known as having had the relevant characteristic. Research information is at the basis of the epidemiologic surveillance system of the department, which is supported and financed by the Regional Health and Social Protection Office. Family members provided informed consent before being interviewed, and the health facilities provided consent prior to the review of clinical records. 3. Results According to the national vital statistics system, 707 018 live births occurred in Antioquia from January 1, 2004, through December 31, 2011 (official data for 2004–2010 and preliminary data for 2011) [4]. Of the 389 maternal deaths that occurred in the same period, 70 (18.0%) were due to HDP, making HDP the second-leading cause of maternal death after obstetric hemorrhage. The overall maternal mortality ratio in Antioquia was 55.01 per 100 000 live births, and the ratio for deaths caused by HDP was 9.9 per 100 000 live births. Two patients were excluded from analysis because of insufficient information. Among the 68 remaining patients, the main complications associated with HDP were eclampsia (n = 45 [66.1%]); hemolysis, elevated liver enzyme levels, and low platelet count (HELLP) syndrome (n = 39 [57.3%]); cerebrovascular disease (n = 34 [50%]); acute kidney failure (n = 21 [30.9%]); acute pulmonary edema (n = 19 [28.0%]); hemorrhagic complication (n = 13 [19.1%]); and placental abruption (n = 8 [11.8%]). Table 1 presents the characteristics of the study population. Women sought care for the following: premonitory symptoms (n= 27 [39.7%]); eclampsia (n= 19 [27.9%]); labor (n=9 [13.2%]); and other (n= 11 [16.2%]). Two more women (2.9%) were dead on arrival. Eleven of the 63 women (17.5%) for whom information was available were hospitalized for hypertension during their pregnancy and then discharged. The mean± SD gestational age at the onset of symptoms was 34±4.5 weeks, and the median time between the onset of symptoms and consultation with a physician was 23.5 hours (interquartile range, 4.8–48.0 hours). The family interviews revealed that the pregnancy was not intended in 53.7% of cases (the information was missing for 27 women). Table 2 describes the factors that contributed to the deaths according to the type of delay. Only those the investigating committee considered to have contributed directly to the deaths were taken into account. The supplies, materials, or resources that were not available at the facilities where the patients were treated, and whose lack of availability contributed to the deaths, included antihypertensive (19.1%) and hemoderivative (4.4%) agents, an intensive care unit or other place to perform close monitoring (2.9%), computed tomographic imaging (2.9%), and magnesium sulfate (1.5%). 4. Discussion In Antioquia, deficiencies in quality of care are strong determining factors for the death of women who present with HDP. In the present study, two-thirds of the patients who died from HDP presented to healthcare facilities in a timely manner, but the correct measures were not applied soon enough to prevent the progression of the condition. Some patients who arrived with an advanced stage of HDP did not receive the care required to stabilize their condition. One study quantified the reduction in mortality due to hypertensive complications of pregnancy after evidence-based practices were implemented, and reported that the in-hospital mortality of women with severe pre-eclampsia or eclampsia could be reduced by more than 84%, even when there was a delay in seeking medical attention [13].
80
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
Table 1 Characteristics of the study population.a
Table 2 Factors contributing to maternal deaths associated with HDP, using the 3-delays model.
Characteristicb
Value
Contributing factor
Age, y (n = 68) Women with prenatal visits (n = 63) Median number of prenatal visits Median parity Multiple gestation Relevant medical history Chronic hypertension (n = 63) Pre-eclampsia (n = 55) Diabetes (n = 61) Autoimmune disease (n = 60) Area of residence (n = 56) Rural residence Urban residence Marital status (n = 63) Married Common union Single Separated Education level (n = 51) None Primary, incomplete Primary, complete Secondary, incomplete Secondary, complete University or technological Place of death (n = 68) Home Institution Primary Secondary Tertiary During transfer In a public place Time of death relative to delivery (n = 68) Before b24 hours postpartum 2–7 days postpartum 8–42 days postpartum
29.1 ± 8.8 49 (77.8) 4 (1–5) 2 (1–4) 2 (2.9)
Delay 1 Negative attitude of the woman toward healthcare services Initial consultation with a traditional healer Delay in consulting a physician for economic reasons Delay 2 Delay in reaching a healthcare facility for an initial consultation Delay 3 Misdiagnosis Mistake in classifying HDP Delay in suspecting HDP Delay in performing laboratory tests The laboratory results were not returned Substandard physical examination Substandard medical interview Poor interpretation of laboratory results Inadequate clinical monitoring Inadequate monitoring of warning signs in the postpartum Inadequate monitoring of magnesium sulfate administration, with non-detection of magnesium sulfate poisoning Incomplete laboratory tests Postpartum bleeding not quantified Non-quantification of diuresis Treatment failure Non-administration of emergency antihypertensive agents Administration of inadequate antihypertensive agents No prophylaxis with magnesium sulfate to prevent eclampsia Non-administration of magnesium sulfate in patients with eclampsia Administration of medications not recommended for the treatment of eclampsia Inadequate management of intravenous fluids Non-admission to an intensive care unit when indicated Non-administration of blood products when indicated No prophylaxis with magnesium sulfate to prevent eclampsia in the postpartum Inadequate decision Retaining management of a patient when the facility was inadequate for her clinical state Untimely referral Inadequate decisions after receiving the laboratory results Discharge before 48 hours Other Resources not matching the facility's level of complexity Substandard teamwork Inadequate referral conditions Administrative difficulties that hampered care
17 (27.0) 9 (13.2) 1 (1.6) 0(0.0) 29 (42.6) 39 (57.4) 17 (27.0) 35 (55.6) 10 (15.9) 1 (1.6) 4 (7.4) 7 (13.0) 15 (27.8) 9 (16.7) 12 (22.2) 4 (5.9) 4 (5.9) 7 (10.3) 14 (20.6) 39 (57.4) 3 (4.4) 1 (1.5) 17 23 18 10
(25.0) (33.8) (26.5) (14.7)
a
Values are given as mean ± SD, number (percentage), or median (interquartile range). b The indicated number of patients is the number for whom the information was available.
In the present study, some women were not instructed to seek the care of a physician when warning signs or symptoms presented, and some of those who consulted a physician did not receive appropriate treatment—a situation also noted by others [14]. Moreover, because there were problems assessing the severity of HDP, the potential complications were probably underestimated. Consequently, available treatment may not have been administered at an opportune time or at all, or the necessary human and technological resources were not available at the facilities reached by the patients. For women who consulted a physician at an early stage of the disease, the most influential factors of mortality were omissions and errors in the pharmacologic management. Studies that have evaluated the implementation of evidence-based practices for the treatment of pre-eclampsia insist that actions at the primary care level should be directed toward the early identification, stabilization, and timely referral of women with complications or who are at risk for complications [13]. A significant number of women did not receive magnesium sulfate for the prevention or initial management of eclampsia. A work group cited by the Program for Appropriate Technology in Health described the administration of magnesium sulfate as the most important action to prevent death from eclampsia [15]. The situation observed in the present study is common elsewhere as well. A study performed in hospitals in Argentina and Uruguay between 2003 and 2005 reported the use of magnesium sulfate in only 33% of women with pre-eclampsia to prevent eclampsia, and in only 58.3% of women with eclampsia to treat the condition [16].
No. (%)a 14/59 (23.7) 7/62 (11.3) 6/56 (10.7) 17/59 (28.8)
34/64 25/61 12/63 6/52 6/62 5/57 4/49
(49.1) (41.0) (17.6) (11.5) (9.7) (8.8) (8.2)
5/46 (10.9) 3/46 (6.5) 4/48 (5.9) 2/48 (4.2) 2/61 (3.3) 42/65 30/51 17/48 14/43 10/46
(64.6) (58.8) (35.4) (32.6) (21.7)
12/62 11/61 5/44 3/45
(19.4) (18.0) (11.4) (6.7)
35/64 (54.7) 32/63 (50.8) 15/50 (22.1) 1/22 (4.5) 19/63 14/57 11/47 7/64
(30.2) (24.6) (23.3) (10.9)
a Denominators indicate the number of patients for whom the information was available or relevant.
Two-thirds of the maternal deaths from HDP were considered secondary to eclampsia and its complications. The frequency of pre-eclampsia has remained stable in both high-income and low-income countries. However, there have been much larger decreases in the incidence of eclampsia in the former than in the latter, with the incidence 10-fold lower in high-income countries [1]. Likewise, mortality from preeclampsia is 10-fold higher in low-income than in high-income countries [1]. In Antioquia, during the period studied, the maternal mortality ratio for HDP was lower than it was for Latin American and the Caribbean combined for 2005, which was 33.4 per 100 000 live births, but well above the 1.4 per 100 000 for high-income countries [13]. It is believed that inadequate practices directly contribute to more than one-third of all maternal deaths [17]. In a study conducted in Tanzania, only 63% of patients with eclampsia received treatment for severe hypertension, the patients were poorly monitored during labor, and they were inadequately prescribed magnesium sulfate and other intravenous fluids [14]. A multicenter study in Argentine about maternal mortality found that women experienced misdiagnosis, inappropriate treatment, and delays in referral, and that there was an overall shortage of supplies and skilled personnel [18]. These situations are less severe but not unknown in high-income countries. The underestimation of clinical symptoms by physicians, and the non-administration or insufficient
J.J. Zuleta-Tobón et al. / International Journal of Gynecology and Obstetrics 121 (2013) 78–81
administration of required antihypertensive agents and magnesium sulfate have been reported in the Netherlands [19]. In the Report of the Confidential Enquiries into Maternal Deaths in the UK covering the 2006 through 2008 period, 20 of the 22 women who died from HDP had received substandard care [20]. Consistent with the present study, the absence of an adequate antihypertensive treatment was identified in that report as the most frequent treatment failure. Although basic interventions to reduce maternal mortality are well known, simple to perform, and even profitable, vast segments of the world population do not benefit from them, as verified by the findings of the present study [7]. It is not difficult to determine what practices should be observed to reduce maternal mortality due to pre-eclampsia but the difficulty lies in knowing how to implement these practices. In 2001, Villar et al. [21] suggested that factors influencing the non-implementation of established measures in low-income countries included the attitude of the healthcare provider, the use of outdated information, the passive transmission of knowledge, and a loss of interest in updating received knowledge. Grol and Grimshaw [22] noted that sufficient evidence indicates that a change in behavior is possible but that such a change would require comprehensive approaches adapted to different scenarios and target groups. The present study has several strengths. First, it transcends the description of features and non-modifiable risk factors [23,24]. Moreover, it provides an analysis of the medical care factors that directly contribute to maternal mortality from HDP, and can be modified at all levels of care. It was recommended that information obtained in the study of maternal mortality be suited to a purpose [17]. In the present case, the obtained information precisely identifies the clinical aspects that must be prioritized. The second strength of the present study is that it decreases variability in the judgments made by the investigating committee, as the findings concerning clinical history were compared with recognized evidence-based clinical practice guidelines. Its third strength is that it includes all the maternal deaths from HDP that occurred in an entire department of Colombia over 7 years. The extended period enables more reliable conclusions and a greater possibility of extrapolation. It is likely that the knowledge obtained will also be useful in other regions of Colombia, and probably also in other low-income countries. One limitation is that the study does not rely on a baseline number of HDP cases in Antioquia to calculate the incidence and lethality of the condition. Additionally, some subjectivity in the judgments made may persist because pre-eclampsia is a dynamic disease, and many variations in the care process hinder analysis. In conclusion, because pre-eclampsia cannot be prevented or predicted, it is necessary to rely on trained personnel for the early detection and adequate management of the disease, to have emergency resources available for conducting initial treatment at all levels of care, and to terminate the pregnancy at the right time under safe conditions. One of the steps required for ensuring that the personnel are qualified is to design and implement training and retraining programs based on innovative and practical methodologies. Acknowledgments The study was financed by the Secretaría Seccional de Salud y Protección Social de Antioquia. Conflict of interest The authors have no conflicts of interest.
81
References [1] Goldenberg RL, McClure EM, Macguire ER, Kamath BD, Jobe AH. Lessons for low-income regions following the reduction in hypertension-related maternal mortality in high-income countries. Int J Gynecol Obstet 2011;113(2):91–5. [2] Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet 2006;367(9516):1066–74. [3] National Institute of Health. National Epidemiological Report 2010. http://190.26. 202.205/index.php?idcategoria=92089#. [Published 2011. Accessed December 13, 2012]. [4] National Bureau of Statistics DANE. Vital Statistics. http://www.dane.gov.co/ index.php?option=com_content&view=article&id=73&Itemid=119. [Updated November 22, 2012. Accessed September 2, 2012]. [5] Vélez Álvarez GA, Gómez Dávila JG, Zuleta Tobón JJ. Analysis of maternal deaths due to hemorrhage in the department of Antioquia, Colombia. Years 2004 and 2005. Rev Colomb Obstet Ginecol 2006;57(3):147–55. [6] Vélez Álvarez GA, Agudelo-Jaramillo B, Gómez Dávila JG, Zuleta Tobón JJ. Code Red: guidelines for the management of obstetric hemorrhage. Rev Colomb Obstet Ginecol 2009;60(1):34–48. [7] Freedman LP, Waldman RJ, de Pinho H, Wirth ME, Chowdhury AM, Rosenfield A. Transforming health systems to improve the lives of women and children. Lancet 2005;365(9463):997–1000. [8] WHO Collaborating Centres for Classification of Diseases. International Statistical Classification of Diseases and Related Health Problems, 10th Revision: Instruction manual. International Statistical Classification of Diseases and Related Health Problems, vol. 2. Geneva: World Health Organization; 1992. [9] Maine D, Akalin MZ, Ward VM, Kamara A. The Design and Evaluation of Maternal Mortality Programs. http://www.rhrc.org/resources/DesignandEvalofMMPrograms. pdf. [Published June 1997]. [10] ACOG Committee on Practice Bulletins–Obstetrics. ACOG practice bulletin. Diagnosis and management of preeclampsia and eclampsia. Number 33, January 2002. Obstet Gynecol 2002;99(1):159–67. [11] Lowe SA, Brown MA, Dekker GA, Gatt S, McLintock CK, McMahon LP, et al. Guidelines for the management of hypertensive disorders of pregnancy 2008. Aust N Z J Obstet Gynaecol 2009;49(3):242–6. [12] National Collaborating Centre for Women's and Children's Health (UK). Hypertension in Pregnancy: The Management of Hypertensive Disorders During Pregnancy. London: RCOG Press; 2010. [13] Ronsmans C, Campbell O. Quantifying the fall in mortality associated with interventions related to hypertensive diseases of pregnancy. BMC Public Health 2011;11(Suppl. 3):S8. [14] Tsu VD, Shane B. New and underutilized technologies to reduce maternal mortality: call to action from a Bellagio workshop. Int J Gynecol Obstet 2004;85(Suppl. 1): S83–93. [15] Karolinski A, Mazzoni A, Belizán JM, Althabe F, Bergel E, Buekens P. Lost opportunities for effective management of obstetric conditions to reduce maternal mortality and severe maternal morbidity in Argentina and Uruguay. Int J Gynecol Obstet 2010;110(2):175–80. [16] Kidanto HL, Mogren I, Massawe SN, Lindmark G, Nystrom L. Criteria-based audit on management of eclampsia patients at a tertiary hospital in Dar es Salaam, Tanzania. BMC Pregnancy Childbirth 2009;9:13. [17] Ronsmans C, Graham WJ, Lancet Maternal Survival Series steering group. Maternal mortality: who, when, where, and why. Lancet 2006;368(9542):1189–200. [18] Ramos S, Karolinski A, Romero M, Mercer R, Maternal Mortality in Argentina Study Group. A comprehensive assessment of maternal deaths in Argentina: translating multicentre collaborative research into action. Bull World Health Organ 2007;85(8):615–22. [19] Schutte JM, Schuitemaker NW, van Roosmalen J, Steegers EA, Dutch Maternal Mortality Committee. Substandard care in maternal mortality due to hypertensive disease in pregnancy in the Netherlands. BJOG 2008;115(6):732–6. [20] Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al. Saving Mothers' Lives: Reviewing maternal deaths to make motherhood safer: 2006-2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118(Suppl. 1):1–203. [21] Villar J, Carroli G, Gülmezoglu AM. The gap between evidence and practice in maternal healthcare. Int J Gynecol Obstet 2001;75(Suppl. 1):S47–54. [22] Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients' care. Lancet 2003;362(9391):1225–30. [23] Moodley J. Maternal deaths associated with hypertensive disorders of pregnancy: a population-based study. Hypertens Pregnancy 2004;23(3):247–56. [24] MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001;97(4):533–8.