全文

Objective

We evaluated the trends of the last decades in maternal mortality in Hungary and compared Hungarian results with those of other European countries.

Study design

Cases of maternal death in Hungary during the study period from calendar year 1978 to 2010 were analyzed in a retrospective manner to characterize mortality distribution and to identify potential clinical or demographic predictors. Data in all cases were extracted both from the national Obstetric Registry operated by the National Institute of Gynecology and Obstetrics, from the Hungarian Central Bureau of Statistics and from the National Public Health and Medical Officer Service. Detailed clinical data were obtained based on obligatory reporting by individual clinical institutions.

Results

The annual maternal mortality rate (MMR) was 26.7 per 100,000 live births in the period 1978–1987 and declined significantly to 10.9 per 100,000 live births in the period 1997–2010. In the period 1988–1996 (with missing associated clinical and demographic data) the MMR was 16.4 per 100,000 live births. The proportion of delivery-associated causes of death increased significantly between the two study periods from 49.4% to 62.9% (p < 0.05). Among obstetric causes of death, the rate of thromboembolism showed a significant increase, while there was a trend toward a decline in rate of maternal deaths attributable to hemorrhagic shock. Among medical causes of death not directly attributable to obstetric complications, the rate of renal and gastrointestinal etiologies declined significantly throughout the study periods.

Conclusions

We observed a marked decline in maternal mortality during the last few decades in Hungary. Recent changes in mortality distribution highlight current characteristics of pregnancy care in Hungary and may help identify strategies for future improvement.

Keywords: Maternal mortality, Obstetric causes of maternal death, Indirect causes of maternal death, Thromboembolic complications of pregnancy.

 

1. Introduction
Maternal mortality is the death of women while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and the site of pregnancy, from any cause related to or aggravated by the pregnancy or its management, but not from accidental or incidental causes [1]. Delayed maternal mortality is defined as maternal death after the 42nd postnatal day but within the first postnatal year.

In order to further refine this measure from the point of view of medical statistics, the maternal mortality ratio (MMR) has been introduced defined as the number of maternal deaths per 100,000 live births. Maternal mortality is classified by etiology as either directly attributable to obstetric conditions or attributable to other medical conditions not directly associated with obstetric complications. Maternal mortality due to other medical conditions may either result from a medical condition present before gestation exacerbated by the physiological processes occurring during pregnancy or it may develop de novo during the gestation period [2], [3], and [4].

The Millennium Development Goals (MDG) program was established to find concrete solutions regarding five specific community health issues, three of which relate to improving maternal and perinatal fetal health [5] and [6]. The program highlighted the urgency of reducing the MMR within countries with less developed infrastructure. This recognition of urgency gave birth to the Countdown to 2015 Initiative [5], which focuses exclusively on community health issues directly related to maternal and fetal health.

It was recognized that in order to protect the health of the pregnant mother and to reduce maternal mortality, a medical database of obstetric data is necessary with precise parameters capable of being statistically evaluated. In 2005 the World Health Organization (WHO), United Nations Children's Fund (UNICEF), United Nations Population Fund (UNFPA) and the World Bank established global parameters in order to standardize data reporting relating to maternal and perinatal fetal mortality in countries where these parameters remain unacceptably high [7] and [8]. At the same time, in more developed European countries the EURO-PERISTAT data processing system handles all obstetric statistical parameters including the ones related to maternal mortality [9].

The first Hungarian obstetric registry was founded in 1931 by Vilmos Tauffer, the medical director of the Second Department of Gynecology and Obstetrics in Budapest. As the first such data registry in the world, the Hungarian Obstetric Registry included data relating to several obstetric complications and described trends in maternal mortality [10]. In this study we analyzed trends in maternal mortality in the last few decades in Hungary. We also compared Hungarian results with those of other European countries.

 
2. Materials and methods
During the study period of 1978–2010, we analyzed the Hungarian obstetric registry data in a retrospective manner to evaluate trends in maternal mortality and of associated clinical and demographic parameters. There was a gap in data reporting between 1988 and 1996. In these years very significant political changes happened in Hungary, which have had also serious consequences on changes of the health policy of our country. Probably a suitable data reporting system of maternal death cases was not available in these years, and that may be the reason for the lack of adequate information from this period, when only the total number of maternal deaths was reported, with relevant clinical and demographic data missing. Due to this gap, our descriptive data are dichotomous, consisting of two separate periods, i.e. one before (1978–1988) and one after (1997–2010) the gap. Data in all cases were extracted both from the national Obstetric Registry operated by the National Institute of Gynecology and Obstetrics, from the Hungarian Central Bureau of Statistics and from the National Public Health and Medical Officer Service [11]. Detailed clinical data were obtained based on obligatory reporting by individual clinical institutions.

The following clinical parameters were included in our analysis: maternal age, parity, neonatal birth weight, cause of death and time of death. Time of death was defined as gestational age at death if the mother expired during pregnancy (in weeks), or time after delivery (in weeks) if maternal death occurred in the postnatal period.

Cause of maternal deaths was identified based on clinical diagnosis reported through diagnostic coding as per the 10th International Classification of Diseases (ICD10) system [1] and [12]. Cause of death was defined as a clinical condition occurring during gestation or during the first 42 days postnatally either associated with an obstetric complication or attributable to a pre-existing medical problem exacerbated during pregnancy. In the determination of cause of death, no consideration of length of gestation or localization of pregnancy (ectopic pregnancy) was made. Adverse events arising during an obstetric procedure or failure of performing an obstetric procedure in a timely manner were regarded as valid causes of maternal death [7] and [13]. Procedures associated with trauma cases were not included in this category.

Classification of causes of maternal mortality according to the ICD10 coding system was based on whether that death was the result of an obstetric complication or other medical condition [4] and [8]. The definitions are as follows. Direct obstetric cause: death of the mother results from obstetric complications of the pregnant state (pregnancy, labor, and puerperium), from interventions, omissions, incorrect treatment, or from a chain of events resulting from any of the above [1]. Indirect obstetric cause: maternal death results from previous existing disease or diseases that developed during pregnancy and which was not due to direct obstetric causes, but which was aggravated by the physiologic effects of pregnancy [1].

In the group of direct obstetric causes, the obstetric conditions hemorrhagic shock, sepsis, preeclampsia, thromboembolism and death associated with general anesthesia were included. Deaths attributable to a pre-existing medical condition or a medical condition arising during pregnancy but not directly attributable to an obstetric cause were subsequently classified by organ systems in which the specific condition resulting in death was thought to originate as cardiovascular, neurological, pulmonary, gastrointestinal, renal, tumor-associated and miscellaneous organ system. Deaths associated with trauma and deaths from an unknown cause were classified as a separate category from either obstetric or medical causes and labeled as miscellaneous cause of death.

In situations where the cause of death could be attributed to multiple obstetric or medical conditions, the case was reviewed by a panel of clinicians to determine the primary cause of death, but in certain cases we had to take into consideration two or more causes of death due to the lack of a clinical basis to make a preference among them.

Cases of maternal mortality were divided into three groups based on maternal age: group 1 <30 years; group 2, maternal age 31–36 years; group 3, maternal age >37 years. In cases where maternal death occurred during pregnancy, three gestational age categories were created: <28 weeks; 28–37 weeks; >37 weeks.

SPSS statistical software was used for analysis of the data. Parameters of descriptive statistics and other clinical variables were categorized by calendar year. Only those cases were included in a given analytical procedure in which all relevant data could be extracted for the procedure.

Our study protocol was approved by the local research ethics committee at Semmelweis University (approval number: TUKEB 43349–2/2013/EKU(519/2013)).

 
3. Results
In the study period 1978–1987, a total of 358 maternal deaths occurred compared to only 151 in the study period 1997–2010; in both of these periods associated clinical and demographic data were available for evaluation. In the study period 1988–1996, however, the 167 deaths that were reported could not be analyzed due missing clinical and demographic data.

In the period 1978–1987, the median age of maternal mortality cases was 30.7 years with a range of 19–46 years. In the study period 1997–2010, the median age was 32.7 years with a range of 20–44 years. In the group of maternal mortality cases where death occurred during pregnancy, the median gestational age was 31.8 weeks with a range of 28–41 weeks in the former study period, whereas in the latter the median gestational age was 31.6 weeks with a range of 25–40 weeks. There was no significant difference between median values in the two study periods. In contrast, a significant difference in parity between the two study periods was found, with the period 1978–87 having a higher parity (2.62) compared to the study period 1997–2010 (2.23; p < 0.05). No significant difference was found in neonatal birthweight between the two study periods (2675 ± 416 g vs. 2594 ± 353). A significant reduction in MMR from 26.7/100000 live births in 1978–87 to 10.9/100000 live births 1997–2010 study period (p < 0.05) was observed (Table 1). In the period 1988–1996, i.e. the period with missing associated clinical and demographic data, MMR was 16.4/100000 live births (MMR1978–1987 vs. MMR1988–1996; p < 0.05; MMR1988–1996 vs. MMR1997–2010; p > 0.05).

Regarding maternal mortality during or following delivery, in both study periods deaths were most common within 24 h after delivery (approx. 40%), whereas mortality was least common after the 30th postnatal day (Table 2).

Regarding the distribution of causes of death, the proportion of deaths due to delivery-related complications rose significantly from the 1978–1987 study period to the 1997–2010 period from 49.4% to 62.9% (p < 0.05); while at the same time, the absolute number of deaths due to delivery-related complications decreased from 180/364 to 88/140 in the respective study periods. Conversely, with regard to spontaneous abortion and ectopic pregnancy, both related death rates and their relative contribution to death declined. In the case of spontaneous abortion, the total number decreased from 60/364 to 10/140 cases, with corresponding rates of 16.5–7.1% (p < 0.05). In the case of ectopic pregnancy, the numbers were 18/364 cases vs. 4/140, while the rates were 5% vs. 2.9% (NS) (Table 3).

Regarding the number of mortality cases, there was a significant increase from the first to the second study period in the rate of thromboembolism (19.4% vs. 29.8%) directly attributable to obstetric causes (p < 0.05), though the absolute number of cases with a thromboembolic event declined from 35/187 to 27/87 cases between the study periods. There was a clear decline in the rate of hemorrhagic shock between the two study periods, but the difference proved to be not significant (1978–1987: 70/187; 38.9% vs. 1997–2010: 23/87; 27.4%) (p > 0.05). There was not a single case of death associated with general anesthesia in the second study period (Table 4).

Between the study periods there was a non-significant rise in the rate of non-obstetric cardiovascular deaths from 30.9% to 47.2% (p > 0.05), while their absolute number decreased from 43/139 to 25/53 cases. There was also a non-significant rise in the rate of non-obstetric tumor-related deaths from 10.1% to 17.0% (p > 0.05), and a decrease in absolute number from 14/139 to 9/53. A significant decline from 10.1% to 1.9% (p < 0.05) was noted in the rate of renal mortality between the study periods, while the absolute number decreased from 14/139 to 1/53 cases. A similar significant decrease from 10.8% to 1.9% (p < 0.05) was detected in the rate of gastrointestinal mortality between the study periods, with absolute numbers decreasing from 15/139 to 1/53 cases (Table 5).

 4. Comments
During the study period, the median age of maternal mortality cases in Hungary was similar to that found in other European countries [14], [15], [16], [17], [18], [19], and [20]. In spite of continuous improvements in obstetric medical care in Hungary over the last three decades, it is clear that increased maternal age, a common trend throughout Europe, increases the risk of certain obstetric complications. Among others, these risks include delivery-related maternal death, maternal death due to thromboembolic complications and death due to cardiovascular diseases in pregnancy. Clearly advanced maternal age requires a different approach in obstetric care, compared to care regimens required for younger obstetric patients.

In both study periods the median gestational age was close to 32 weeks, suggesting that the third trimester poses the greatest risk for maternal mortality. A French study reported that in their population two-thirds of maternal mortality cases occurred after the 37th gestational week, suggesting that this risk further increases toward the end of pregnancy [13]. Management of the third trimester requires close follow-up of patients, with a focus on prevention and early recognition of complications which may lead to maternal mortality.

It is important to note that most maternal deaths attributable to a complication after an obstetric event (delivery or miscarriage) occurred within the first 24 h and least commonly after the 30th day. This time dynamic suggests that close monitoring during hospitalization is especially important within the first 24 h after delivery.

A global report by the WHO in 2008 provides international data regarding recent trends in maternal mortality in multiple countries [11]. In this report, countries are divided into three categories (A, B, C) based on the reliability of their registry data; Hungary belongs to category A, signaling a high degree of reliability in terms of registry data.

The MMR in Hungary between 1997 and 2010 proved to be 10.9 per 100,000 live births. The last Europeristat Report (2010) states a different value (13.4 per 100,000 live births) for the period 2006–2010. The discrepancy between the values may be due to the differing lengths of the study periods [21], but also the lower MMR in 2003–2004 (7.4 per 100,000 live births) may stand in the background of this phenomenon [22]. Among neighboring countries, the MMR was found to be lower in Austria and Slovakia, while in Romania and the Ukraine, results were higher than those reported in Hungary (Fig. 1) [23]. Regarding the time dynamics of MMR decline, a significant decline similar to the one reported in the present study was seen in the last twenty years both in Eastern European countries formerly belonging to the communist bloc and in European countries with high gross domestic product (GDP) such as Sweden or Germany [13] and [23].

 During the study period, a notable decline of deaths associated with either abortion or ectopic pregnancy was seen, while the proportion of deaths attributable to delivery-related complications rose. In terms of abortion, the more widespread use of the vacuum aspiration technique and quality improvement in anesthesia procedures likely played a role. We suggest that the decreased death rate associated with ectopic pregnancy may be a result of improved diagnostic procedures and consequent prompt surgical interventions.

While in the study period of 1978–1987 hemorrhagic shock was the most common cause of death, followed by sepsis and thromboembolism, deaths attributable to hemorrhagic shock declined (without significance) by the study period 1997–2010 and these three most important causes of death became approximately equally common. Historical reports from Austria and from Switzerland revealed that in these countries hemorrhagic shock remained the most common obstetric complication leading to maternal deaths, followed by thromboembolism and preeclampsia [19] and [24].

In their recent study published earlier this year, Rossi and Mullin reported that in their study population of 9750 cases from different European countries, hemorrhagic shock and preeclampsia were the most common causes of death, with a relative percent distribution of around 22% each [23].

Another study described the distribution of causes of maternal deaths in several countries in the period 2003–2004 [25]. Although the short span of this study does not allow far-reaching conclusions, it is noteworthy that in economically developed countries hemorrhagic shock, thromboembolism and preeclampsia contributed nearly equally to maternal deaths.

In our investigations the only obstetric cause of death which showed a significant difference between the study periods was thromboembolism, with a rate increasing from 19.4% to 29.8%. Due to this increase, genetic testing for predisposition for thromboembolic complications among patients having a positive personal or family history is an important task of the obstetrician, as it may aid in the prevention of this life-threatening complication.

Concerning medical causes of death not directly associated with obstetric complications, we report a relatively high frequency of cardiovascular deaths in our population, followed by central nervous system and tumor-related deaths. Our data conform to similar reports from Austria, in contrast to Dutch results showing a nearly equal incidence of cardiovascular and central nervous system-related deaths [15] and [20]. Regarding the statistical value of the changes of rates of medical causes of death between the two study periods, only the decline of the incidence of renal (10.1% vs. 1.9%) and gastrointestinal diseases (10.8% vs. 1.9%) are worth mentioning.

A different approach is necessary to reduce maternal mortality associated with obstetric complications. To improve survival, early diagnosis of obstetric conditions known to increase risk for maternal mortality appears necessary, together with continuous quality improvement in pregnancy care.

Our study has several potential limitations. One limitation may have been the potential failure of reporting delayed maternal mortality cases to the registry. This could be caused by the failure of medical professionals to report such cases, especially under circumstances when maternal death was not associated with an obstetric condition. Another study limitation may have been related to a potential pitfall known to occur in assigning the appropriate cause of death to maternal mortality cases. This arises from the difficulty in determining whether death occurred due to an obstetric condition or as a result of a specific complication associated with that condition.

Due to international experience with under-reporting maternal mortality, an international system has developed where data provided by national registries are adjusted by the estimated degree of underreporting for that particular country [13] and [26]. With regard to the Hungarian national registry, an adjustment factor of 1.5 has been established which reflects a relatively favorable reliability of Hungarian data reporting.

In summary, our study suggests that maternal mortality in Hungary has significantly declined in the last several decades. Among causes of death, delivery-associated complications have increased in importance while deaths associated with either abortion or ectopic pregnancy declined. Among deaths directly attributable to obstetric complications, a significant decline in hemorrhagic shock cases was identified. On the other hand, a trend toward rising rates of thromboembolic complications was observed.

 
Conflict of interest
The authors declare no conflict of interest.

 
References
[1] World Health Organization. International statistical classification of diseases and related health problems. 10th Revision. 2nd ed. (WHO, Geneva, 2004)
[2] A.M. Gülmezoglu, R.C. Pattinson, G.J. Hofmeyr, P. Lumbiganon. Global maternal and perinatal health issues. D.K. James, P.J. Steer, C.P. Weiner, B. Gonik (Eds.) High risk pregnancy – management options 4th ed. (Elsevier, St. Louis, USA, 2011) 1-7
[3] A.P. MacKay, C.J. Berg, C. Duran, et al. An assessment of pregnancy-related mortality in the United States. Paediatr Perinat Epidemiol. 2005;19:206-214 Crossref.
[4] K.S. Khan, D. Wojdyla, L. Say, et al. WHO analysis of causes of maternal death: a systematic review. Lancet. 2006;367:1066-1074 Crossref.
[5] www.countdown2015mnch.org.
[6] United Nations General Assembly. United Nations Millennium Declaration. A/RES/55/2. (United Nations, New York, 2000)
[7] S. Alexander, K. Wildman, W. Zhang, et al. Maternal health outcomes in Europe. Eur J Obstet Gynecol Reprod Biol. 2003;111(Suppl. (1)):S78-S87 Crossref.
[8] K. Wildman, M.H. Bouvier-Colle, MOMS Group. Maternal mortality as an indicator of obstetric care in Europe. BJOG. 2004;111:164-169
[9] www.europeristat.com/images/doc/EPHR/european-perinatal-health-report.pdf.
[10] L. Lampé, I. Batár, Á. Szállási. Semmelweis Egyetem Általános Orvostudományi Kar. II. Számú Szülészeti és Nőgyógyászati Klinika. L. Lampé, I. Batár, Á. Szállási (Eds.) The history of the Hungarian Obstetrics and Gynecology (Medicina Kiadó, Budapest, 2009) 69-82
[11] www.ksh.hu.
[12] WHO, UNICEF, UNFPA. The World Bank. Trends in maternal mortality: 1990–2010. (, 2013) www.unfpa.org
[13] M.H. Bouvier-Colle, A.D. Mohangoo, M. Gissler, et al. What about the mothers? An analysis of maternal mortality and morbidity in perinatal health surveillance systems in Europe. BJOG. 2012;119:880-889
[14] M. Saucedo, C. Deneux-Tharaux, M.H. Bouvier-Colle. Understanding regional differences in maternal mortality: a national case–control study in France. BJOG. 2012;119:573-581 Crossref.
[15] J.M. Schutte, L. de Jonge, N.W. Schuitemaker, et al. Indirect maternal mortality increases in the Netherlands. Acta Obstet Gynecol Scand. 2010;89:762-768 Crossref.
[16] N. Vrachnis, S. Iliodromiti, E. Samoli, et al. Maternal mortality in Greece, 1996–2006. Int J Gynaecol Obstet. 2011;115:16-19 Crossref.
[17] S. Donati, S. Senatore, A. Ronconi. Regional maternal mortality working group. Maternal mortality in Italy: a record-linkage study. BJOG. 2011;118:872-879 Crossref.
[18] B. Bødker, L. Hvidman, T. Weber, et al. Maternal deaths in Denmark 2002–2006. Acta Obstet Gynecol Scand. 2009;88:556-562
[19] P. Bollini, P. Wanner, S. Pampallona. Trends in maternal mortality in Switzerland among Swiss and foreign nationals, 1969–2006. Int J Public Health. 2011;56:515-521 Crossref.
[20] D. Karimian-Teherani, G. Haidinger, T. Waldhoer, et al. Under-reporting of direct and indirect obstetrical deaths in Austria, 1980–98. Acta Obstet Gynecol Scand. 2002;81:323-327
[21] www.tno.nl/downloads/european_perinatal_health_report_2010.pdf.
[22] www.europeristat.com/reports/european-perinatal-health2004.html.
[23] C. Rossi, P. Mullin. The etiology of maternal mortality in developed countries: a systematic review of literature. Arch Gynecol Obstet. 2012;285:1499-1503
[24] M. Fässler, R. Zimmermann, K.C. Quack Lötscher. Maternal mortality in Switzerland 1995–2004. Swiss Med Wkly. 2010;140:25-30
[25] C. Deneux-Tharaux, C. Berg, M.H. Bouvier-Colle, et al. Underreporting of pregnancy-related mortality in the United States and Europe. Obstet Gynecol. 2005;106:684-692 Crossref.
[26] S. Kao, L.M. Chen, L. Shi, M.C. Weinrich. Underreporting and misclassification of maternal mortality in Taiwan. Acta Obstet Gynecol Scand. 1997;76:629-636 Crossref.