Introduction

Dystocia is defined as slow, abnormal progression of labor [1], [2], [3]. It is the result of problems related to either the passenger, such as abnormal fetal presentation; the passage, such as absolute or relative cephalopelvic disproportion; or the power, such as insufficient uterine activity [1].Known risk factors for dystocia include nulliparity, labor induction, high maternal age, short stature [1], [4], [5], postterm gestation, high birth weight and abnormal vertex presentation. [4], [5]. It is also debated whether epidural analgesia itself may cause dystocia [4], [5], [6], [7], [8]. Sub-fecundity, which has increased due to the higher age of nulliparous women, may also play a role, as failure to progress is frequently observed in pregnancies achieved by assisted reproductive technology. [5], [9]. Despite the liberal use of oxytocin augmentation, pregnancies complicated by dystocia often end with assisted deliveries, including forceps, vacuum or, commonly, caesarean section. It is estimated that dystocia accounts directly or indirectly for up to 30–60% of all caesarean deliveries in the United States [3], [10], [11].

Recent studies associated levels of vitamin D in pregnancy with complications such as gestational diabetes, preeclampsia, bacterial vaginosis, and preterm birth. They attempted to define the optimal levels of vitamin D in serum during pregnancy, and to recommend the optimal dose for supplementation [12], [13], [14].The prevalence of vitamin D deficiency during pregnancy varies from 18% to 84% [15], [16], [17] depending on the geographic area of residency and local clothing customs. An established symptom of vitamin D deficiency is proximal skeletal muscle weakness, caused by reduced stimulation of the cellular Vitamin D Receptor (VDR) [18], [19]. Several studies have described VDR expression in non-vascular smooth muscle [19] including the myometrium [20]. Thus, it is possible that vitamin D affects the contractility of the myometrium and may be involved in the risk of dystocia. In addition, calcium levels are affected by the concentration of vitamin D. Calcium itself also seems to play a part in myometrial contractility, by affecting the production of prostaglandins and the action of nitric oxidase [21]. Low concentration of extracellular calcium or inhibition of the entry of the ion into the myometrium cell reduces the contractility of the muscle fibers and their sensitivity to oxytocin [22].

If vitamin D deficiency causes dystocia, seasonal variation in the incidence of dystocia would be expected, particularly in areas with distinct seasonal variation in sunlight exposure. In northern latitude regions, such as Denmark, ultraviolet B (UVB) radiation is nonexistent from October until April [23]. The resulting lack of dermal vitamin D production during this period is reflected in a seasonal variation in plasma 25-hydroxyvitamin D (P-25OHD) [24], causing up to 90% of Danish women to be vitamin D deficient (<50 nmol/l) during winter. By contrast, this number is reduced to under one-fifth during summer [25]. Seasonal variation in the incidence of dystocia, if found, would necessitate further studies on actual levels of vitamin D and dystocia. If vitamin D depletion increases the risk of dystocia, it would imply that an inexpensive, safe, and easily accessible method of preventing dystocia and associated complications is readily available. This study therefore examined whether there is seasonal variation in dystocia in a Danish cohort of pregnant women.

Materials and Methods

Results

Of the 34,261 women included in the study, 751 (2.2%) delivered by acute cesarean section because of dystocia, 1,711 (5%) delivered by vacuum or forceps because of dystocia, and 9,929 (29%) received oxytocin augmentation due to dystocia after spontaneous dilatation of the orificium to 4 cm or more. Overall dystocia was observed in 10,655 women (31.1%); the remaining 23,606 women comprised the reference group.

All four outcomes showed non-monotonous time trends. The tendency over time was towards more cesarean deliveries (from 0.4% to 3.7%), fewer instrumental deliveries (from 8% to 2%) and more oxytocin augmentations (from 29% to 66%). Over time, a greater number of deliveries seemed to be characterized by dystocia (from 25% to 36%). During the time of the study, changes in diagnostic criteria and in criteria for intervention occurred. This may partially explain the observed time trends.

The background characteristics of women in each group are shown in Table 1. Women who received an intervention due to dystocia tended to have higher levels of education, to be older, to be of shorter stature, to have higher pre-pregnancy BMI, and to give birth to larger infants. This was most pronounced for women who delivered by emergency caesarian section due to dystocia.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Background characteristics of nulliparous women with dystocia, the Aarhus Birth Cohort, Denmark (1992–2010).

https://doi.org/10.1371/journal.pone.0094432.t001

Figure 1 shows the seasonal variation in emergency cesarean delivery, instrumental vaginal delivery, oxytocin augmentation due to dystocia, and overall dystocia according to date of delivery. The figure presents the seasonal variation from the spline method, with 95% confidence intervals. The monthly relative risk estimates without smoothing technique, but after adjustment for time trend, can also be seen in Figure 1. The relative risk estimates (RR) for the four outcomes at different periods during the year (with 95% confidence intervals shown in brackets) varied from 0.7 [0.48–1.00] to 1.35 [1.06–1.72]. The women who gave birth by instrumental vaginal delivery because of dystocia had the largest variation in relative risk estimates, ranging from the lowest risk estimate of 0.94 [0.72–1.22] for women giving birth in November to the highest risk estimates of 1.35 [1.06–1.72] for women giving birth in May. Dystocia overall had the smallest variation in risk estimates, ranging from 0.95 [0.84–1.06] for women giving birth in November to 1.09 [0.98–1.22] for women giving birth in August. The less frequent outcomes, such as emergency cesarean delivery because of dystocia, showed a greater variation in risk estimates during the year, while the more frequent outcomes, such as dystocia overall, presented minor variation in risk estimates during the year.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Seasonal variation of dystocia.

The solid line represents the estimated seasonal variation of dystocia after adjusting for time-trend, with 95% CI (dashed line). Dots represent the relative risk (RR) obtained by grouping the season into months. Some monthly point estimates (dots) are outside the CIs of the seasonal variation curve, and one is significant, (marked with a triangle).

https://doi.org/10.1371/journal.pone.0094432.g001

No evidence for seasonal variation was found for any of the outcomes: emergency cesarean delivery due to dystocia (p = 0.44); instrumental vaginal delivery due to dystocia (p = 0.69); oxytocin augmentation due to dystocia (p = 0.46); and overall dystocia (p = 0.91). Similar results were yielded when we investigated seasonal variation in the incidence of dystocia according to date of conception (results not shown).

There were no changes in these results based on the secondary analysis, in which the estimated relative risk for dystocia for each month of delivery was adjusted for pre-pregnancy BMI, birth weight of the child, maternal ethnicity or educational level (results not shown). Moreover, we found no seasonal variation in the overall incidence of emergency cesarean delivery (p = 0.9) regardless of indication, nor of overall instrumental vaginal delivery (p = 0.7), overall emergency cesarean delivery and instrumental vaginal delivery together (p = 0.6), or women registered as having dystocia irrespective of interventions due to dystocia (p = 0.60). Furthermore, we found no seasonal variation in the incidence of oxytocin augmentation due to dystocia after changing the parameter regarding cervical dilation from 4 to 6 cm (p = 0.69). Finally, grouping the year into quarters did not change the result.

Discussion

Our results show no seasonal variation in the incidence of dystocia, defined as emergency cesarean delivery, instrumental vaginal delivery, or oxytocin augmentation all indicated by dystocia, as well as overall dystocia, in a Danish cohort. Some tendency towards more dystocia-indicated cesarean deliveries during the spring and fall was observed, but the results were not statistically significant.

To our knowledge, no previous study has tested the association between season of birth and the risk of dystocia. Two previous studies examined the association between serum vitamin D and acute cesarean delivery, but they showed conflicting results. The first study demonstrated no association between a low level (<30 nmol/l) of vitamin D and the risk of an acute cesarean delivery in a Pakistani population [32]. However, the women included in that study were characterized by malnourishment in addition to their vitamin D deficiency; moreover, the clothing customarily worn by Pakistani women may predispose them to vitamin D deficiency. The second study, from Boston, demonstrated that nulliparous women who delivered by cesarean section had lower serum vitamin D compared to controls who delivered vaginally (P-25OHD median of 45 nmol/liter and 63 nmol/liter, respectively). However, in this study dystocia was the indication for cesarean delivery in only 17 of the 43 cases, while the rest had a cesarean delivery indicated by non-reassuring fetal tracing, fetal malpresentation, or other indications [33]. Both studies have a number of limitations, which were addressed in the present study. Unlike previous studies, we used the spline technique with a day-by-day analysis and tested for seasonal variation using four degrees of freedom after adjusting for time trends. Furthermore, this study involved a large cohort of healthy women who were relatively homogeneous in terms of socioeconomic and nutritional status. All background information used for the secondary analyses was collected by self-administered questionnaires at gestational week 12, and is therefore less prone to recall biases. Generally, dystocia is difficult to define by objective measures and to date there is no consensus for the diagnostic criteria [34]. This lack of specificity in the outcome definition, which is unrelated to the season of birth, would lead to non-differential misclassification and thus cause bias towards no association. We tried to increase the validity of the outcome assessment by using well-registered interventions as proxies, and still no significant seasonal variation was found. Nonetheless it is possible that the lack of seasonal variation in our study is partially explained by some misclassification of dystocia, especially in the oxytocin augmentation group.

According to our primary hypothesis, plasma 25OHD variations due to variation in exposure to sunlight according to the date of delivery should cause a seasonal variation in the risk of dystocia. However, plasma 25OHD concentrations among women may vary as a consequence of differences in eating habits as well as sun exposure, and the fact that we did not directly measure plasma 25OHD is a limitation of the present study. Additionally, plasma 25OHD concentration at other gestational ages may also be important for delivery, e.g. at the time of conception and early in pregnancy, where the myometrium starts to grow and develop. Such potential vitamin D-mediated effects may counteract each other and lead to negative results, even though we restricted our study to term deliveries in order to reduce some of the variation in vitamin D levels in earlier stages of pregnancy and repeated the analysis according to conception dates. Furthermore, restricting our study to term deliveries excludes women with preeclampsia and preterm deliveries, both associated with vitamin D deficiency. This may have caused selection of women with less variation in their plasma 25OHD, reducing the ability to find an association with dystocia.

Seasonal variation according to month of birth has been seen in some immune system-related diseases of pregnancy, such as preeclampsia [35], [36]. The immune system is not known to be involved in the pathogenesis of dystocia. However, if any other factors subject to seasonal change, such as eating habits or exposure to viral infections, or meteorological factors, such as humidity, temperature or exposure to daylight [28], affect myometrial function, we would expect to see a seasonal variation in dystocia. With the present study design, it could be difficult to isolate the association between vitamin D and dystocia. If these factors pull in different directions, the net association without adjustment for those in our study would be no association. Finally, dystocia is not only caused by insufficient contractility of the myometrium, but it is also affected by the size and position of the fetus, the pelvic conditions, and the cervical collagen concentration [37]. In our study we included infants with vertex presentation and adjusted in a secondary analysis for birth weight, but, clearly, risk factors for dystocia that were not taken into account could cause bias towards no association. Our results therefore do not preclude the possibility that vitamin D is of importance in the pathogenesis of dystocia, but indicate the need for further studies with strict definition of dystocia, measurements of actual levels of vitamin D in different stages of pregnancy, and exclusion of causes of dystocia other than myometrial function.

The external validity of the study should most likely be restricted to geographical areas with sunlight exposure, dietary and clothing habits similar to those in Denmark. The Danish National Board of Health recently increased the amount of recommended daily vitamin D intake for pregnant women to 10 µg of supplement during the entire pregnancy, but the exact amount of vitamin D supplementation during pregnancy that is necessary in order to achieve a beneficial effect remains unknown. Our study did not supply evidence for an association between vitamin D and dystocia, but more direct measures of vitamin D status in pregnancy should be studied in relation to pregnancy and delivery complications in order to draw more certain conclusions.

Acknowledgments

We would like to thank The Biostatistic Section of the Department of Public Health for providing us statistic consultancy. Furthermore, thank you to Kate Thulin for proofreading the manuscript.