Vitamin D deficiency during pregnancy and the neonatal period is a common global health concern (1, 2). Studies on pregnant women have reported a high prevalence of vitamin D deficiency in Colombia (23.9%) (3), France (41.5%) (4), and China (76.4%) (5). It has been confirmed that 25-hydroxyvitamin D [25(OH)D], the major circulating form of vitamin D, could readily traverse the hemochorial placenta (6). Fetal vitamin D concentrations are largely dependent on maternal vitamin D status (7). To date, guidelines concerning vitamin D screening in pregnant women have been conflicting (8–13). The Committee Opinion of the American College of Obstetricians and Gynecologists (8) and Australian Government antenatal care clinical practice guidelines (10) recommend screening only women at increased risk of vitamin D deficiency, including vegetarians, women with limited sun exposure, and women with darker skin (14). Hence, identifying pregnant women at increased risk of vitamin D deficiency is an important public health strategy.

The main source of vitamin D in most humans is the synthesis of 7-dehydrocholesterol in the skin upon exposure to ultraviolet B (UVB) solar radiation and its conversion to a measurable circulating metabolite, 25(OH)D, in the liver (15), which is used to determine individual vitamin D status (16). Thus, limited sunlight exposure may influence skin production of vitamin D, resulting in vitamin D deficiency. However, several factors influence the intensity and duration of exposure to UVB radiation, including geographic location, season, atmospheric conditions, and time spent outdoors (17). Notably, the ambient temperature is usually associated with both UVB radiation and its influencing factors. Although there is no direct causality between ambient temperature and vitamin D, given the access to ambient temperature, it may be a better practical predictor of individual vitamin D concentrations. However, to the best of our knowledge, the relationship between ambient temperature and individual vitamin D concentration has not been well established.

We hypothesized that maternal exposure to ambient temperature might be positively associated with fetal vitamin D concentrations and that cutaneous synthesis of previtamin D could be unavailable when the ambient temperature is greatly reduced (i.e., when the amount of solar ultraviolet radiation that reaches the earth’s surface is substantially reduced). This leaves the individual uniquely dependent on supplemental sources of vitamin D during periods when ambient temperature is below the threshold value. This study aimed to determine the implications of gestational ambient temperature on infant vitamin D status and to establish the necessity of maternal vitamin D supplementation when the ambient temperature is less than a threshold value. The significance of this study lies in its ability to reveal the effects of ambient temperature on vitamin D supplementation during pregnancy. If ambient temperature is shown to be a reliable predictor of vitamin D deficiency or supplementation needs, it could significantly impact public health strategies, especially in regions with seasonal temperature variations. The findings of this study provide new insights into how environmental factors, particularly temperature, should be considered when developing pregnancy-care guidelines. This could lead to more individualized recommendations for vitamin D supplementation based on the local climate, particularly during colder months when vitamin D synthesis is impaired owing to reduced UVB exposure. Furthermore, this study could lay the foundation for future research on the relationship between environmental factors and maternal vitamin D status, ultimately guiding more effective prenatal care policies to improve maternal and fetal health outcomes.

In this study, ambient temperature in the eighth GM was suggested as an effective predictor of vitamin D status in full-term neonates. The mean 25(OH)D concentration in cord blood significantly increased with increasing ambient temperature in the eighth GM. Maternal exposure to 10–24.4°C in the eighth GM had a significantly higher risk of vitamin D deficiency, and the risk was further substantially elevated when the mother experienced ambient temperature < 10°C in the eighth GM, even after controlling for potential covariates and vitamin D supplementation during pregnancy. Furthermore, an interesting finding was that the necessity of vitamin D supplementation during pregnancy appeared to be dependent on ambient temperature in the eighth GM. For maternal exposure to ≥24.5°C in the eighth GM, vitamin D supplementation during pregnancy failed to significantly enhance neonatal vitamin D concentrations. In contrast, for maternal exposure to <10°C in the eighth GM, maternal vitamin D supplementation may be a unique and effective way to increase the neonatal vitamin D concentration. To the best of our knowledge, this is the first study to focus on the relationship between maternal exposure to ambient temperature and fetal vitamin D concentrations.

Our study provides important evidence that ambient temperature can serve as a useful predictor of fetal vitamin D status and suggests that maternal vitamin D supplementation should be tailored to seasonal changes in UVB exposure. Our study provides actionable recommendations for public health policies. First, temperature-based predictors are integrated into clinical guidelines for vitamin D supplementation, especially in regions where the ambient temperature falls below the critical threshold. Second, we created geographically and seasonally tailored policies for vitamin D supplementation with an emphasis on high-risk groups. Third, supporting public health campaigns and maternal health surveillance focused on the risk of seasonal vitamin D deficiency. Fourth, routine vitamin D screening is based on ambient temperature data, especially for pregnant women during winter months. Vitamin D supplementation is essential for maintaining maternal and fetal health (20), particularly to prevent complications, such as gestational diabetes mellitus (GDM) and preeclampsia (21, 22). Temperature-based adjustments to supplementation may be a practical strategy to ensure optimal vitamin D levels, particularly in regions with low sunlight exposure. Previous studies have shown that temperature changes increase the risk of GDM and preeclampsia (23, 24). However, the direct effects of temperature-based vitamin D supplementation on these outcomes have not yet been fully explored. Further studies are required to confirm the efficacy of temperature-guided supplementation for reducing pregnancy-related complications.

Vitamin D is synthesized in the skin through exposure to UVB radiation, and solar UVB radiation is the primary source of vitamin D for most humans. Therefore, the assessment of vitamin D status in populations must consider sunlight exposure (15). However, sunlight is a complex exposure to measure because ambient UVB concentrations vary up to 30-fold by latitude, season, ground surface, time of day, clothing, sunscreen, and cloud or tree cover (25–29). Thus, simply asking people how much time they spend outside may not correctly classify UVB exposures. Studies assessing the relationship between sunlight measures and serum 25(OH)D levels also showed low correlations (30, 31); sunlight measures only accounted for approximately 10% of the variation in serum 25(OH)D concentrations. The ambient temperature is easy for an individual to access for an individual. Although there was no direct causality, the predictive value of ambient temperature in this study was substantial because the ambient temperature in the eighth GM at ≥10°C could account for approximately 12% of the variation in infant 25(OH)D concentrations, which appears to be more important than the season (32) and is similar to sunlight. In addition, ambient temperature can serve as a useful indirect measure, because temperature and UVB radiation levels are often positively correlated. For example, UVB radiation tends to be higher during warmer months and at higher latitudes closer to the equator, which corresponds to higher ambient temperatures. Previous studies (33, 34) have demonstrated that the UVB radiation intensity is associated with seasonal temperature changes. We acknowledge that ambient temperature alone is an imperfect proxy but is often useful when combined with other contextual data (e.g., seasonality or geographic latitude).

It is important to note that there was a two-month lag effect of maternal exposure to ambient temperature on fetal vitamin D concentrations. The complex progression of the vitamin D endocrine system (35) and placental transmission of 25(OH)D (36) may explain this lag effect. The lag effect could be linked to the fact that maternal vitamin D synthesis is less effective during the winter months because of lower sunlight exposure, especially when ambient temperatures are low (e.g., <10°C). This reduced synthesis may take time to manifest as a deficiency or insufficiency in both the mother and the fetus. Another possible mechanism is placental transmission of 25(OH)D. Research suggests that the placenta is a key mediator of vitamin D transfer (37); however, the efficiency of this process can be affected by several factors, including maternal vitamin D status and ambient temperature. Placental transfer of vitamin D is not instantaneous and may involve a temporal delay, especially when maternal vitamin D levels are low. If the lag effect is further confirmed in the whole pregnancy through animal study, it will give an important “time window” to intervene ahead of time, which was supposed to start from 2 months before pregnancy. Furthermore, the findings of this study suggest that when pregnant women experience an ambient temperature of less than 10°C, vitamin D supplementation may be a unique and effective way to increase fetal vitamin D concentrations, as little detectable production of precholecalciferol occurs in the season of low ambient temperature when there are few ultraviolet photons with energies between 290 and 315 nm (38). This result was supported by a study from Australia showing that the contribution of sunlight to 25(OH)D concentrations is very important in the summer months but less important in the winter months (39). Given the adverse pregnancy outcomes associated with vitamin D deficiency, the potential public health implications of our findings are of immense importance.

In our study, we did not find a significant association between pre-pregnancy BMI or parity and fetal vitamin D levels (measured using cord blood 25(OH)D concentrations). This result contrasts with previous studies that have suggested that these maternal factors may influence vitamin D status (40, 41) during pregnancy. This may be because of several reasons. First, it was conducted in Hefei, China, a region that may have different seasonal patterns of UVB exposure and lifestyle factors. Second, in our cohort, maternal vitamin D supplementation was likely more widespread, especially in regions with lower UVB radiation, potentially reducing the impact of maternal BMI or parity on fetal vitamin D levels.

Our study has two major strengths. First, this is the first ecological study on the relationship between maternal exposure to gestational ambient temperature and fetal vitamin D status. Second, in contrast to previous ecological studies, adjustment for individual information (including GWG, pregnancy complications, paternal alcohol consumption, and infant sex) strengthened the conclusions of this study. Third, we considered the timing of gestational months when analyzing the effects of ambient temperature on fetal vitamin D status. This stratification allowed us to capture more nuanced relationships, which may have been overlooked in studies that did not account for such temporal factors.

However, this study has several limitations that should also be acknowledged. The missing data from October to December is an important limitation that may affect the ability of the study to fully capture the seasonal variations in ambient temperature and fetal vitamin D levels. The absence of this key period could lead to underestimation or overestimation of seasonal effects, especially because these months typically represent the transition from fall to winter, when temperature and sunlight levels fluctuate more dramatically. In addition, our study had a short duration, and the results were preliminary. Another limitation was that the duration of supplementation, but not the dose, was used to estimate maternal vitamin D supplementation in this study. Although the dose of commercial vitamin D3 that most pregnant women consumed ranged from 300 to 600 IU/d, the absence of an exact dose might have confounded the contribution of vitamin D supplementation in our study. Without knowing the exact dosage, it is challenging to determine the precise contribution of supplementation to vitamin D levels observed in cord blood. Additionally, we acknowledge that relying on self-reported data on vitamin D supplementation, including the duration of supplementation, introduces the possibility of recall bias. This bias occurs when participants may have forgotten, misremembered, or exaggerated the extent of their supplementation. Recall bias is a common challenge in retrospective studies, especially when assessing dietary or health behaviors over long periods. Although we did not capture the exact dosages, the range of supplementation (300–600 IU/day) was relatively narrow, and this range is consistent with the general guidelines for pregnant women. Overall exposure to vitamin D supplementation was within a reasonable range for most participants. In addition, a previous study (5) successfully used self-reported supplementation duration as a proxy for the exact dose of vitamin D supplementation in pregnant women, particularly when the exact doses are difficult to capture. Third, our study was based on a single cohort from Hefei, China, which may limit the generalizability of our findings to other regions with differing latitudes, UVB exposures, and socioeconomic conditions. Therefore, the results are geographically specific, and further multicenter or multi-region studies are needed to validate the relationship between ambient temperature and neonatal vitamin D levels. This would help identify whether ambient temperature during gestation is a universal predictor of fetal vitamin D status or whether it varies according to regional factors such as UVB exposure, diet, and cultural practices regarding sunlight exposure. Fourth, data collection from 2008 may not fully reflect current conditions such as climate change, advancements in supplementation practices, or shifts in maternal behaviors. Rising temperatures and shifting seasonal patterns may affect the amount of UVB exposure and, in turn, the vitamin D status of the pregnant women. Additionally, advancements in vitamin D supplementation practices (42), which have become more standardized over the past decade, could have influenced our results. Public health campaigns promoting prenatal vitamin D supplementation may increase compliance among pregnant women, potentially altering the relationship between ambient temperature and vitamin D levels. Despite the potential limitations of using historical data, these findings can provide valuable insights into the fundamental relationship between ambient temperature and vitamin D levels during pregnancy. Fifth, while the R² value is significant, it does not explain a large proportion of the variance in neonatal vitamin D levels. This is an important point, as it suggests that there may be additional factors contributing to the variability in the data that are not captured by the ambient temperature alone. We suggest that the inclusion of other uncollected covariates could improve the model’s predictive power.

Maternal exposure to ambient temperature below 10°C during pregnancy is a critical predictor of vitamin D deficiency in newborns. These findings suggest that ambient temperature can be used as an indicator to guide vitamin D supplementation. Further studies are needed to confirm the predictive value of temperature and to optimize supplementation protocols based on environmental conditions.