Objective
The objective of the investigation was to study the association of fetal death with maternal age by length of gestation.
Study Design
This was a population study including all ongoing pregnancies after 16 weeks of gestation in Norway during the period 1967-2006 (n = 2,182,756).
Results
The risk of fetal death was 1.4 times higher in women 40-44 years old than in women aged 20-24 years in midpregnancy but 2.8 times higher at term. In term pregnancies the relative importance of maternal age increased by additional pregnancy weeks. In gestational weeks 42-43, the crude risk was 5.1 times higher in mothers 40 years old or older. In the recent period, the elevated risk of fetal death in elderly mothers at term has been attenuated.
Conclusion
Women 40 years old or older had the highest risk of fetal death throughout pregnancy, particularly in term and postterm pregnancies. Improved obstetric care may explain the attenuation of risk associated with age in recent time.
In the Western world, it has become increasingly common to postpone child-bearing. The mean age of primiparous women has increased during the last decades. According to the Medical Birth Registry of Norway, more than 19% of child-bearing women were 35 years old or older in 2008. For comparison, only 8% were 35 years old or older 20 years ago. The mean age of all women giving birth was 29.8 years in 2006, more than 4 years older than women delivering 30 years earlier.
The negative impact of high maternal age on fetal death risk is well known. Recent knowledge suggests that the risk of fetal death varies considerably according to length of gestation. The fetal death rate seems to be high at 20-22 weeks of gestation and lowest at 27-33 weeks, before it increases rapidly from 37 through 43 gestational weeks. The knowledge of the impact of maternal age on fetal death at different gestational ages is limited. Such knowledge, however, may be important in understanding causes of fetal death. It has been suggested that infectious causes play a greater role in midpregnancy than in fetal death at term, whereas pregnancy-related maternal disease and placenta insufficiency seem to be increasingly important as the pregnancy proceeds.
In a study comprising births from 36 states in the United States during 2001-2002, women of advanced maternal age had a higher risk of stillbirth across all gestational ages, with a peak risk postterm. This study is essential; however, the lack of follow-up after the 41st gestational week and the high level of obstetrical interventions may have caused an underestimation of the fetal death risk in term and postterm pregnancies.
Obstetrical ultrasonographic examination was gradually introduced in the public obstetric health in Norway by the mid-1980s, and since then almost all pregnancies have been examined at pregnancy weeks 17-19. This may have improved the term prediction in pregnancy and also the diagnosis of fetal failure to thrive when used later in pregnancy. Because elderly mothers are at increased risk of fetal death, they may in particular have gained from improved obstetrical diagnostic tools, if such tools are advantageous.
In all ongoing pregnancies after 16 weeks of gestation in Norway during the years 1967-2006, we estimated the risk of fetal death according to maternal age at different lengths of gestation. We also studied whether the fetal death risk at term in elderly mothers has changed after ultrasonographic examinations were introduced in obstetrical care.
Materials and Methods
Data were obtained from the Medical Birth Registry of Norway. This registry contains information on all births after 16 weeks of gestation. Compulsory notification of birth is made on standardized forms by the midwife or attending physician at the delivery, and the data elements in the notification form have been almost unchanged since the start of the registration in 1967 ( http://www.mfr.no ).
The study population comprised deliveries after 16 weeks of gestation in Norway during the period 1967-2006, a total of 2,337,392 births. A total of 28,595 pregnancies were recorded to last longer than 43 weeks. A proportion of these were miscoded, and we could not with certainty determine whom, and we therefore decided to exclude all pregnancies recorded to last longer than 43 weeks. Information on length of gestation at birth was missing for 125,997 offspring, leaving 2,182,800 births for the crude data analyses. An additional 44 births lacked information on potentially confounding variables; hence, 2,182,756 births were included in the multivariable data analyses.
We defined fetal death as birth of a dead fetus after 16 weeks of gestation. We studied fetal death risk at the following gestational lengths; 16-22, 23-29, 30-36, and 37 or more weeks of gestation. Pregnancies lasting 38 weeks or longer were in addition subdivided in groups of 2 week intervals: 38-39, 40-41, and 42-43 weeks of gestation.
Information on length of gestation at delivery was based on women’s reporting of first day of the last menstrual period in standardized antenatal care forms or date of term estimated from ultrasonographic examination if available.
Maternal age at delivery was our main explanatory variable and coded as less than 20, 20-24 (reference), 25-29, 30-34, 35-39, 40-44, and 45 years and older. As potential confounders we included parity, plurality, year of delivery, paternal age, and preeclampsia ( Table 1 ). Parity was defined as the number of previous births after the 16th week of gestation coded as 0, 1, 2, 3, and 4 or greater. Plurality was divided into 2 categories: 1 and 2 or more offspring. Year of delivery (period) was coded as 1967-1971, 1972-1976, 1977-1981, 1982-1986, 1987-1991, 1992-1996, 1997-2001, and 2002-2006. Paternal age was categorized as younger than 30, 30-39, 40 years old or older, and missing. Preeclampsia was defined as maternal blood pressure above 140/90 mm Hg and proteinuria.
| Characteristic | Total births | Percent | Fetal deaths, n | Fetal deaths per 1000 births, n |
|---|---|---|---|---|
| Parity | ||||
| 0 | 901,171 | 41.3 | 9508 | 10.6 |
| 1 | 758,413 | 34.7 | 6456 | 8.5 |
| 2 | 356,952 | 16.4 | 3965 | 11.1 |
| ≥3 | 166,220 | 7.6 | 2825 | 17.0 |
| Plurality | ||||
| 1 | 2,125,470 | 97.4 | 20,633 | 9.7 |
| ≥2 | 57,286 | 2.6 | 2121 | 37.0 |
| Paternal age, y | ||||
| <30 | 1,058,341 | 48.5 | 8661 | 8.2 |
| 30–39 | 932,983 | 42.7 | 8261 | 8.9 |
| ≥40 | 169,195 | 7.8 | 2252 | 13.3 |
| Missing | 22,237 | 1.0 | 3580 | 161.0 |
| Preeclampsia | ||||
| Yes | 67,200 | 3.1 | 1242 | 18.5 |
| No | 2,115,556 | 96.9 | 21,512 | 10.2 |
| Total | 2,182,756 | 22,754 | 10.4 |
To calculate the absolute risk of fetal death (in percent) according to maternal age, the denominator was all women with a fetus still in utero at the gestational period being studied, and the numerator was women with stillbirths at that same period of gestation. For instance, in weeks 30-36, only pregnant women who delivered after week 30 were included.
Cox regression models were applied to estimate the relative risk of fetal death. The outcome variables was time (days) to fetal death. Births of live infants or live fetuses in utero at the end of a gestational length interval were treated as censored observations. In the adjusted analysis, the confounders were included as categorical variables with categorization as given in earlier text. Separate analyses were carried out for the different gestational length intervals. For term and postterm pregnancies, we also estimated the association of fetal death with maternal age in 2 different time periods: 1967-1986 and 1987-2006. We used the Statistical Package for the Social Sciences for statistical analyses (version 16.0; SPSS, Chicago, IL).
The study was approved by the Norwegian Data Inspectorate and the Publishing Committee for the Medical Birth Registry of Norway.
Results
A total of 22,754 fetal deaths occurred in 2,182,756 births during the study period, representing 1.04% of all births after 16 weeks of gestation. Of all pregnancies, 10.3% were in women 35 years old or older and 93.2% of the births were at the 37th gestational week or later.
The absolute risk of fetal death varied according to length of gestation ( Figure 1 ). Women aged 40 years or older had the highest risk of fetal death at all gestational ages, but their increased risk was most pronounced in early gestation and at term. After gestational week 36, the risk of fetal death increased rapidly in all age groups. However, the increase in risk seemed to be highest in women 40 years old or older.
The relative importance of high maternal age on fetal death risk was confirmed to be elevated at term and in early gestation by Cox regression analyses ( Table 2 ). The crude relative risk of fetal death in gestational weeks 16-22 was 3.45 (95% confidence interval [CI], 3.00–3.96) in women 40-44 years old as compared with women aged 20-24 years (reference), and it was 1.95 (95% CI, 1.66–2.28) after adjustment for period of delivery, parity, plurality, and paternal age.
