Table 3.
Overview of articles concerning the effect of HC on brain function.
| Reference | Sample | Type of study and methods | Findings |
|---|---|---|---|
| Abler et al. (2013) | University students. N = 24, 12 oral HC subjects (4 using 0.03 mg EE with 0.13 mg levonorgestrel or 0.15 mg desogestrel and 8 using 0.03 mg EE combined with 2 mg chlormadinone, 3 mg drospirenone or 2 mg dienogest) and 12 NC subjects. Mean age for total sample: 24.0, age span: 20–29. | Observational RMWC fMRI-study using erotic stimuli. Whole brain analyses with uncorrected p-values (p < 0.001). | No significant difference in HC vs. NC group upon viewing of explicit erotic stimuli, activation in precentral gyrus increased in NC group during the follicular phase (positively correlated with estrogen levels) upon expectation of erotic stimuli. |
| Arnoni-Bauer et al. (2017) | N = 29, 11 oral HC subjects (Mean age: 26, SD 2) (All using 0.02 mg EE and 0.15 mg desogestrel) and 18 NC subjects (Mean age: 25, SD: 3), age span both groups 18–35. | Observational RMWC fMRI-study using food stimuli. fMRI group comparisons entirely ROI-based. | The HC group had similar fMRI activations in in all activated brain areas as the NC group in the luteal phase but differed from the NC group in the follicular phase. |
| Basu et al. (2016) | Recruitment through newspaper ads and from a family planning clinic. N = 8, subjects were not using HC for the 3 months prior to inclusion. One intramuscular dose of depot medroxyprogesterone acetate was administered to all subjects, age span 18–25. | Quasi-experimental pre-post fMRI study, subjects serving as own controls. fMRI measured before receiving an intramuscular dose of depot medroxyprogesterone acetate and 8 weeks after. Whole brain analyses corrected for FWE. | Increased BOLD response in frontal pole, superior frontal gyrus, orbitofrontal cortex, postcentral gyrus, cingulate cortex, paracingulate gyrus, postcentral gyrus, middle frontal gyrus, superior parietal lobule and lingual gyrus 8 weeks after injection of DMPA vs. baseline. |
| Bonenberger et al. (2013) | University students. N = 24, 12 oral HC subjects (4 using 0.03 mg EE with 0.13 mg levonorgestrel or 0.15 mg desogestrel and 8 using 0.03 mg ethinylestradiol combined with 2 mg chlormadinone, 3 mg drospirenone or 2 mg dienogest) and 12 NC subjects, mean age 24.0, age span 20–29 (same sample as Abler 2013). | Observational RMWC fMRI-study using a monetary incentive task. Whole brain analyses with uncorrected p-values (p < 0.001) and ROI-analyses. | In ROI based analyses, increased activation in the left anterior insula and inferior lateral prefrontal cortex upon expectation of monetary reward was seen in the HC group as compared to NC group in the follicular phase. No effects of HC in whole brain analyses. |
| Chung et al. (2016) | N = 46, 15 oral HC female subjects, [using estrogen and progestin (or estrogen and an antiandrogen progestin)] 15 NC female subjects and 16 male subjects, mean age 25.46, SD 3.2, age span 19–34. | Observational RMWC MRI-study with regard to effects of HC, but a placebo controlled experimental condition regarding the effects of androstadienone on psychosocial stress responses based on the Montreal Imaging Stress Task. Whole brain FWE-corrected analyses and dorsolateral prefrontal cortex ROI analysis. | No effects of HC. Increased activation of right premotor and supplementary motor areas as well as left somatosensory association cortex during stress in placebo condition, no significant differences during treatment with androstadienone. |
| De Bondt et al. (2015b) | N = 37, 19 oral HC female subjects (using 2nd or 3rd monophasic HC) 18 NC (38 NC and 27 HC before exclusion criteria). No information about age. | Observational RMWC resting state fMRI-study with data extraction using principal component and ICA. | No default mode network differences between any group in any phase were found. There was a positive correlation between functional connectivity in the posterior part of the default mode network and psychological premenstrual-like symptoms seen in inactive pill phase in HC group, not found in NC group. |
| Gingnell et al. (2013) | Women with previously reported negative affect with oral HC use, recruited through newspaper advertisements. N = 30, 15 subjects 30 mcg EE and 0.15 mg levonorgestrel, and 15 placebo, mean age 25.5 +−5 in HC group, 24.5 +−3.3 in placebo group, age span 18–45. | Double-blinded, randomized parallel group clinical trial fMRI-study with pre and post treatment measurement. fMRI paradigm using an emotional face matching task. fMRI group comparisons were entirely ROI-based. | HC women more depressed mood after treatment. Reduced emotion-induced activity in left insula, left middle frontal gyrus and bilateral inferior frontal gyri compared with the NC group and reduced emotion-induced activity in bilateral inferior frontal gyri post HC treatment vs. pre-HC treatment. The placebo group only showed reduced BOLD activity in the amygdala in the last scans. |
| Gingnell et al. (2016) | Same sample as Gingnell et al. (2013). | Double-blinded, randomized parallel group clinical trial fMRI-study with pre and post treatment measurement. fMRI paradigm using a go/no-go inhibition task. Whole brain analyses, uncorrected for FWE (p < 0.001). | Reduced BOLD response in right middle frontal gyrus in HC group, vs. placebo group, no change in behavioral go/no-go performance. |
| Hornung et al. (2019) | University students. N = 67, 29 oral HC female subjects, 20 NC female subjects in luteal phase, 28 male subjects. Mean age 24.07 years, SD 3.64. | Observational fMRI-study using an emotional dot-probe attention-modulation paradigm. Whole brain analyses corrected for FWE and ROI based analyses. | No evidence of differences in neural attentional bias processing between HC and NC group, neither behaviorally nor with regard to brain activity. |
| Hwang et al. (2015) | N = 85, 16 oral HC subjects, 32 NC subjects (median split into 16 with high estradiol and 16 with low estradiol levels) and 37 male subjects, mean age 23.3, SD 2.4 in female group and 29.8, SD 8.8 in male group. | Observational fMRI study using a fear-conditioning and extinction paradigm. FWE-corrected ROI analyses only. | Reduced activation in posterior insula, middle cingulate cortex, amygdala and hypothalamus during a fear conditioning paradigm HC vs. high estrogen NC group. The main findings revolved around the high estrogen group driving most effects. |
| Lisofsky et al. (2016) | Women planning to start using HCs recruited through flyers at gynecologists' offices. N = 56 (28 HC, mean age 21.25 years, age span 16–33 years, 28 NC, mean age 21.5 years, age span 16–28 years. | Quasi-experimental pre-post start use of HC MRI and resting state fMRI-study (no randomization), matched NC controls. Whole brain FWE corrected analyses. | Functional connectivity between left amygdala/anterior parahippocampal gyrus and the right dorsolateral prefrontal cortex changed from positive to negative in the HC group, but findings uncertain due to lack of FWE correction. |
| Mareckova et al. (2014) | Experiment I: University students N = 20, 10 oral HC subjects using 30 mcg EE and 150 mcg levonorgestrel/75 mcg gestodene and 10 NC subjects, mean age 22.0 SD 3.26 in HC group and 20.44 SD 2.69 in NC group, age span 18–29. Experiment II: Adolescents recruited for The IMAGEN European multi-site study through school visits. N = 110, 55 oral HC female subjects, 55 NC female subjects, no information about type of oral HC, age span 13.5–15.5. | Observational fMRI-study using movies of either ambiguous or angry faces as stimuli. Whole brain FWE corrected analyses and ROI analyses. | Experiment I: Increased BOLD response to faces in right FFA in HC vs. NC group. Experiment II: Findings replicated. In experiment I, also positive correlation between duration of HC use and the BOLD response in left FFA. |
| Merz et al. (2012) | N = 122, 30 oral HC subjects (using 0.02–0.035 mg EE with a gestagen component, 30 NC subjects in luteal phase, 30 NC subjects in follicular phase, 32 male subjects. Age span 18–35, mean age 21.3–24.8. | Observational fMRI-study using a fear learning paradigm with an experimental condition where cortisol was administered to half the sample. Whole brain FWE corrected analyses and ROI analyses. | Cortisol enhanced fear-learning in HC vs. NC women and there was increased BOLD responses in the left hippocampus and parahippocampal gyrus. |
| Merz et al. (2013) | University students. N = 50, 15 oral HC subjects, 15 NC subjects in the luteal phase, 20 male subjects. Age span 18–35, mean age 23.60, SD 2.13 in HC group, 25.27, SD 3.69 in NC group, 24.15, SD 3.08 in male group. | Observational fMRI study using a fear conditioning paradigm. Whole brain FWE corrected analyses and ROI analyses. | Endogenous cortisol levels positively associated with amygdala BOLD contrasts between CS+ and CS- in men and HC-using women, but not in NC women in the luteal phase. |
| Miedl et al. (2018) | N = 53, 23 oral HC subjects (using EE and a progestin) 30 NC subjects, mean age 22.28, SD 3.73, age span 18–34. | Observational fMRI-study using “traumatic” film viewing as stimuli, with neutral films as reference. All analyses ROI based with corrections for FWE. | Increased BOLD activations in insula and dorsal anterior cingulate cortex during traumatic film viewing in HC groups as compared to NC group. |
| Monciunskaite et al. (2019) | N = 70, 33 combined monophasic anti-androgenic HC users and 37 NC women. Total sample age range 19–38, mean age 23.6, SD 4. No information about recruitment. | Observational EEG/event-related potential study using negative emotional images as stimuli. | HC users showed smaller global field power to visual stimuli than NC women in late latencies (>351 milliseconds). HC users showed blunted late positive potentials to unpleasant stimuli. |
| Petersen et al. (2014) | Participants recruited from University of California and surrounding community. N = 91, 46 oral HC subjects (24 active pill phase, 22 inactive pill phase, HC type consisted of several different types of combined oral contraceptives with different amounts of EE and different types and amount of progestins), 45 NC subjects (20 follicular phase, 25 luteal phase), age-span 18–40. | Observational fMRI resting state study, analyzed using ICA whole brain FWE corrected analyses. | Greater connectivity with the aDMN in left angular gyrus in the follicular NC vs. active phase HC group, increased connectivity with the aDMN in the right caudate nucleus in the follicular group compared to inactive phase HC users. Increased connectivity with ECN in the left anterior cingulate cortex and left middle frontal gyrus seen in follicular group vs. active phase HC group. |
| Petersen and Cahill (2015) | Same recruitment as Petersen et al. (2014), but included N = 83, 40 oral HC subjects (20 active phase, 20 inactive pill phase), 43 NC subjects (20 follicular phase, 23 luteal phase). | Observational fMRI study using 72 negatively valenced, arousing images as stimuli vs. 72 neutral images as reference. Narrow ROI of parts of amygdala. | Reduced bilateral amygdala reactivity in response to negative emotional stimuli seen in HC vs. NC group. |
| Pletzer et al. (2014) | N = 46, 14 oral HC subjects using a combined oral contraceptive, no information about type of estrogen, progestin component consisted of levonorgestrel, other type of 2nd generation progestin or drospirenone, mean age 23.22, SD 3.51, 16 NC subjects, mean age 26.57, SD 6.01, 16 male subjects, mean age 25.14, SD 4.35. | Observational fMRI-study, using number comparison, number bisection. Whole brain FWE corrected analyses. | Different BOLD-activation patterns in HC women as compared to NC women during both numerical tasks, but without any behavioral performance differences. Authors propose that HC women's activation patterns resemble that of men. |
| Rumberg et al. (2010) | N = 36, 12 female subjects using combined oral HC, 12 NC female subjects, 12 male subjects, mean age 31, age span 18–45 in male and NC groups, mean age 23, age span 18–37 in HC group. | Observational fMRI study using a verb-generation task, backwards number counting as reference task. The NC group was scanned twice (ovulatory vs. menstrual period), the other groups were scanned once. Whole brain FWE corrected analyses. | Increased activation during verb generation in right superior temporal cortex in the HC vs. NC group in menstrual phase, and in right inferior frontal cortex in the HC vs. NC group in luteal phase. |
| Scheele et al. (2016) | N = 40, 21 oral HC female subjects, 19 NC female subjects tested in luteal or follicular phase, mean age 24.38, SD 3.26. | Observational fMRI study with regard to HC usage, but randomized, placebo-controlled, double-blind within-group design with regard to effects of oxytocin. Using photographs of partner face, unknown men, familiar women or unfamiliar women. Whole brain FWE corrected analyses. | Oxytocin increased behavioral evaluation of partner attractiveness and BOLD responses in nucleus accumbens and ventral tegmentum in NC group upon viewing partner's face but not in HC group. |
| Smith et al. (2018) | Recruitment through advertisements. Relevant subset for the post-hoc HC analyses: N = 28, 17 oral HC subjects, 11 NC subjects. | [18F] fallypride PET-study (targeting dopamine D2/D3 receptor availability) using a dexamphetamine dose as stimuli. Whole brain FWE corrected analyses and ROI analyses. | Plasma estradiol did not correlate with change in dopamine D2/D3 receptor availability and did not differ between HC and NC groups. |
| Vincent et al. (2013) | Recruitment by advertisement and “word of mouth.” N = 24, 12 HC subjects (monophasic combined oral HC) 12 NC subjects, no information about age. | Observational RMWC fMRI-study using noxious thermal stimuli. Whole brain analyses, corrected for FWE and ROI based analyses. | No omnibus effects were found, but in a subgroup of HC users with low testosterone levels, activity in rostral ventromedial medulla was ↓ in HC vs. NC group and this was also associated with lowered pain thresholds. |
RMWC, Repeated measures within one cycle; HC, hormonal contraceptives; NC, naturally cycling; EE, ethinyl estradiol; fMRI, functional magnetic resonance imaging; BOLD, blood-oxygen-level dependent; FFA, fusiform face area, SD: standard deviation, ROI: Region of interest FWE: Family-wise error; DMPA, depomedroxyprogesterone acetate; D2-receptor, Dopamine 2-receptor; D3-receptor, Dopamine 3-receptor; ICA, independent component analysis; aDMN, anterior default mode network; ECN, executive control network; CS, conditioned stimulus.
Information about recruitment is only described if described in the article.