Characterizing the Influence of Vitamin D Levels on IVF Outcomes

B. Rudick; S. Ingles; K. Chung; F. Stanczyk; R. Paulson; K. Bendikson


Hum Reprod. 2012;27(11):3321-3327. 

In This Article


Of 208 eligible patients, 18 did not have available serum for testing and 2 had missing outcome data, leaving 188 study participants in the final analysis. Of the 188 patients, 21% (39/188) were vitamin D deficient (25(OH)D <20 ng/ml), 37% (70/188) were vitamin D insufficient (20–30 ng/ml) and only 42% (79/188) were vitamin D replete (25(OH)D >30 ng/ml).

Race was categorized as non-Hispanic white (53%), Asian (26%), Hispanic white (14%) or other (7%), according to guidelines outlined in the 2010 USA Census Bureau ( census) Asian race included those patients whose origins were from Southeast Asia as well as from the Indian subcontinent. Serum 25(OH)D levels varied by race (P= 0.01), being highest in non-Hispanic whites (n= 100; mean: 30.6 ng/ml), intermediate in Asians (n= 49; mean: 27.1 ng/ml) and lowest in Hispanic whites (n= 26; mean: 25.6 ng/ml). Baseline patient characteristics including infertility diagnosis, age and BMI were compared among the different racial groups. Hispanic whites were on average younger with a higher BMI compared with other races. No other significant differences were noted.

Table I depicts patient and IVF cycle characteristics by vitamin D status. Vitamin D deficient women were, on average, younger (P= 0.03) and heavier (P= 0.03), and were less likely to have a diagnosis of diminished ovarian reserve (P= 0.01). Although vitamin D deficient women were heavier (P= 0.03), all patients had BMI values within the normal range. Vitamin D status was not associated with other infertility diagnoses, with parity, or previous IVF failure. Of all patients, 53% underwent long lupron protocol, 37% underwent microdose flare and 11% underwent an antagonist protocol. Women with vitamin D deficiency were more likely to have been treated with the long lupron protocol (P= 0.01). Vitamin D status was not associated with ovarian stimulation parameters (dose of medications required, peak estradiol levels, number of oocytes retrieved and number of mature oocytes) as noted in Table I. Although the number of embryos transferred differed significantly by vitamin D status, there was no trend (more embryos were transferred to women in the intermediate vitamin D group). Vitamin D status was not associated with fertilization rates or markers of embryo quality (the mean number of blastomeres or mean percent fragmentation).

Table II depicts the patient and IVF cycle characteristics by pregnancy outcome. Clinical pregnancy rates were 43% in non-Hispanic whites, 38% in Hispanic whites and 35% in Asians. A similar pattern was seen for live birth rates: 35% in non-Hispanic whites, 27% in Hispanic whites and 26% in Asians. Clinical pregnancy was associated with lower dose of medication needed (P= 0.05), higher peak of E2 (P= 0.03), higher number of oocytes retrieved (P= 0.05) and higher mean number of cells on Day 3 both among the entire cohort of embryos (P= 0.03) and among the embryos that were selected for transfer (P= 0.01). The number of embryos transferred was greater in the clinical pregnancy group (3.6 versus 3.3, P= 0.02). Only dose of medication (P= 0.02) and the mean percent fragmentation among transferred embryos (P= 0.05) were significantly associated with live birth.

As shown in Table III, race significantly modified the relationship between cycle outcome and vitamin D status (P < 0.01, for clinical pregnancy and live birth). Among non-Hispanic whites, clinical pregnancy rates progressively decreased with declining vitamin D status, from 51% in those who were vitamin D replete, to 44% in those who were insufficient, to 19% in those who were vitamin D deficient (P= 0.04). However, the opposite trend was seen among Asians, with pregnancy rates increasing with worsening vitamin D status (P= 0.01). Results were similar after fitting multivariable logistic models to adjust for maternal age, number of embryos transferred, embryo quality and diagnosis of diminished ovarian reserve (Table III). A similar relationship was observed between vitamin D status and the live birth rate, with worsening vitamin D status, live birth rates decreased among non-Hispanic whites (P= 0.03), but increased in Asians (P= 0.01).