Hair Loss
Alopecia has been reported as an adverse effect of hormonal contraceptives over the past several decades.[60] Although the reports and drug labels often do not specify which type of alopecia it can trigger, the most commonly discussed subtypes include androgenetic alopecia (AGA) and telogen effluvium (TE).
AGA, or patterned hair loss, is the most common form of hair loss in both men and women. It is characterized by a miniaturization of hairs in a symmetric distribution over the crown, vertex, and frontal areas of the scalp. As the name suggests, androgens play a clear role in the pathogenesis, with various genetic loci encoding androgen and estrogen receptors implicated, emphasizing the function of hormones in this condition.[61–63]
AGA can be triggered or worsened in relation to hormonal contraceptives, particularly early generation COCs with androgenic effects, such as levonorgestrel (Table 3).[64–66] One of the earliest reports of alopecia caused by COCs was by Cormia in 1967, who presented two cases of women who developed male-patterned hair loss after starting first-generation COCs as early as 5 months after initiation.[67] To explain this effect, investigations on the specific changes occurring within the hair cycle were carried out. In a study of 50 females taking COCs, half of the participants exhibited a transient increase in the proportion of resting follicles, with the original pretreatment ratio re-established at 6 months. Conversely, 11 patients demonstrated an increase in the proportion of actively growing follicles. Of note, these patients had relatively high levels of resting follicles at treatment initiation.[68,69] Present day, there are less reports of alopecia related to COCs in the literature, possibly due to the increased use of third- and fourth-generation pills.
Like acne and hirsutism, AGA is a possible adverse effect of levonorgestrel-releasing implants due to the androgenic effects. Alopecia has been reported with progestin implants, with incidence rates of 1.1–11.5%, and is the third most common reason for requesting removal.[70–73] This adverse effect appears to be more common during the initial months. For example, the annual rates of hair loss in the US Norplant trial were 5.2 and 3.5 per 100 woman-years in the first and subsequent years of use, respectively.[74] Although the levonorgestrel IUD generally causes fewer adverse effects than the implant or pills due to lower systemic progestogen levels, the adverse effect of alopecia may still occur. A New Zealand study estimated the cumulative incidence of alopecia associated with levonorgestrel IUD placement to be 0.33% over 1 year.[75] In a survey-based study by Lullo et al. 20% of women reported hair loss after levonorgestrel IUD insertion, however this finding was not statistically significant.[22]
TE is the most common type of drug-induced hair loss and is characterized by diffuse hair shedding with or without visible hair thinning. In the majority of cases, drugs trigger TE by advancing the follicle into premature rest. However, TE may also be secondary to discontinuation of medications that prolong the anagen phase, such as COCs. Several case series have reported an association between TE and COC cessation, most commonly mestranol–norethisterone and EE–norethisterone acetate.[67,76] Unlike the male-patterned hair loss seen in some women while using COCs, the hair loss seen after discontinuation is more diffuse. This is postulated to be due to a large number of follicles transitioning into the telogen stage after interruption. Like AGA, there are few recent reports of TE related to COCs. This may be due to the fact that the estrogen implicated in earlier reports was mestranol, which was used in amounts of up to 150 μg. Currently, available COCs contain 20–35 μg of EE. As the newer COCs have less estrogenic activity, their cessation would theoretically be less likely to trigger TE. Furthermore, given that this form of hair loss is likely related to estrogen withdrawal, there are unsurprisingly no reports of TE related to progestin implants.
Some authors suggest that hormonal contraceptives can affect the immune system and possibly be linked to the development of autoimmune diseases, such as alopecia areata.[77] Alopecia areata presents as single or multiple patches of hair loss and is not believed to be influenced by hormones. However, Vallings reported three patients who developed alopecia areata shortly after starting COCs.[78] Another author reported a 30-year-old woman who developed a patch of alopecia 4 months after contraceptive implant insertion. She had no prior or family history of hair loss, and her hair regrew 3 weeks after removal.[79] Despite these reports, experts agree that the prevalence of alopecia areata in women of child-bearing age is common and these cases can be coincidental.[80,81]
Despite the previously mentioned reports that hormonal contraceptives may incite hair loss, they have become an important therapeutic option for specific subtypes of alopecia. While the androgenic effects of the progesterone component of COCs may contribute to hair loss, the estrogen component may support hair growth by lengthening the anagen phase of the hair cycle via follicular estrogen receptors.[82] In addition, estrogen may have an indirect effect on hair growth by increasing SHBG, which reduces the levels of free androgens in circulation. The increase in SHBG is greatest when COCs incorporate third-generation or newer progestins, as they do not inhibit the estrogenic effect as much as earlier generations. The use of COCs in treating AGA is becoming increasingly common in dermatology.[83] It was reported as the second most common therapy for female AGA after minoxidil (81% vs. 98%) by dermatologists in Spain.[84] Several cohort studies and case series have shown that women taking CPA and EE for AGA have significantly decreased hair loss and thinning, particularly in frontocranial regions of the scalp.[85,86] When comparing this combination with megestrol acetate and EE, both regimens appear to have similar efficacies in treating AGA.[87] COCs have also been used in combination with other treatment modalities, such as finasteride. In a case series of 37 women with female pattern hair loss taking oral finasteride (2.5 mg/day) and a COC containing drospirenone and EE, 62% of patients demonstrated improvement at 12-month follow-up.[88] Furthermore, a cohort study of 40 women with AGA demonstrated a significant increase in total hair density in those treated with CPA 50 mg and EE 20 mg, but only if serum ferritin levels were above 40 μg/L, suggesting a more optimal treatment response at specific ferritin concentrations.[89] Overall, the efficacy of COCs in treating female AGA is not as high as with acne and hirsutism.[90] In addition, an RCT of three therapeutic options for premenopausal hyperandrogenic alopecia revealed that the COC option (CPA 50 mg with EE) had no statistically significant treatment effect.[91]
While COCs have been reported to cause TE, one study demonstrated the efficacy of COCs in treating postpartum alopecia, a subtype of TE.[92] In this investigation, 34 women with telogen hair loss were included—13 with postpartum hair loss and 21 with hair loss of unknown etiology. After a placebo washout phase, 2 of the postpartum women experienced hair growth, while the remaining 11 had no change. Of the 21 non-postpartum women, 8 experienced hair growth, 12 demonstrated no change, and 1 developed exaggerated hair loss. The 11 postpartum women who did not experience improvement were then started on a COC containing lynestrenol and mestranol, resulting in 9 women demonstrating improved hair growth. Of the 13 nonpostpartum women treated, only 3 improved after starting COCs. These results highlight a difference in pathogenesis for postpartum alopecia and TE not associated with pregnancy, and emphasize the efficacy of COCs in managing the former.
Frontal fibrosing alopecia (FFA) is a variant of lichen planopilaris, a cicatricial alopecia characterized by a lymphocytic infiltrate, follicular destruction, and permanent hair loss. FFA presents with the loss of terminal and vellus hairs in the frontotemporal regions in a band-like distribution. Its pathogenesis likely involves an interplay of autoimmune, environmental, and hormonal factors. In a survey study completed by 105 women with FFA and 100 controls, a history of COC use was significantly greater in the control group, suggesting a protective role of COCs.[93] This protective role may also carry over to IUDs. In a case–control study of 104 females with FFA and 208 controls, the only protective factor identified was the previous use of an IUD (odds ratio 0.22).[94] Of note, the authors did not classify IUD use as copper versus hormonal. Ultimately, this may suggest a role for hormonal therapies as treatment options for FFA. However, a recent genome-wide association study identified several susceptibility loci for FFA, including 2p22.2, likely due to a missense variant in the gene that encodes cytochrome P450 1B1, an important sex hormone-processing enzyme.[95] This suggests that alterations in hormone metabolism may be a potential mechanism for predisposition to FFA and may explain the appearance of the disease after the creation of COCs. However, further investigations on gene–environment interactions are needed to support this.
Am J Clin Dermatol. 2021;22(1):69-80. © 2021 Adis Springer International Publishing AG
