In this study on a large monocentric cohort of women with genetically defined CAIS syndrome, sexual health impairment and body uneasiness resulted worryingly common in adult patients.

Sexual dysfunction has already been reported in CAIS, although findings vary across highly heterogeneous cohorts. In particular, Minto et al. [10] investigated sexual function in a relatively large group of patients with CAIS through the Golombok-Rust Inventory of Sexual Satisfaction scale and found a high prevalence of sexual difficulties, especially sexual infrequency and vaginal penetration difficulty. These results contrast with earlier studies, mostly reporting satisfactory sexual outcomes in CAIS, yet in small series often employing non-validated methods of sexual function assessment [7]–[9]. More recently, Fliegner et al. [11] examined sexual health in 11 patients with CAIS, all gonadectomized, and reported a lack of sexual confidence and satisfaction, excluding clinically relevant depression as a confounding factor. FSFI results were available in only 6 patients, as the rest were not sexually active [11]. The dsd-LIFE multicentre study included 69 individuals with CAIS, analyzed together with complete XY gonadal dysgenesis, XX gonadal dysgenesis and unknown steroid synthesis defects (XY-DSD + XX-GD-F–na group) and found this group to be least satisfied with sex life compared to individuals with other DSD conditions, with over 50% reporting lack of sexual desire when asked [12]. Conversely, in the study by Engberg et al. 20 women with CAIS were assessed showing normal sexual functioning using the Profile of Female Sexual Function and Personal Distress Scale scales, comparable to population-derived controls and women with POI [13]. Thus, available data on sexual function in patients with CAIS remain inconclusive, with significant variability in both the cohorts studied and the methods of assessment used. Moreover, the ongoing HRT is not specified in previous literature on the subject.

Beyond sexual dysfunction, we believe that our result on the highly reported distress related to sexual life registered trough the FSDS-R scale is also concerning, since it clearly influences well-being.

Interestingly, in the present study sexual impairment resulted highly prevalent also in the subgroup of patients with gonads in situ. Despite this subset of patients being very small, fully reflecting clinical practice, we believe it is an interesting preliminary result, also considering that this is the first study to investigate this issue. In fact, avoiding gonadectomy and maintaining the gonads has been suggested by some authors to help maintaining sexual health: in our experience, the presence of functional gonads did not seem to preserve sexual well-being.

We also systematically assess the aspect of body uneasiness, on which very limited data are available. Wisniewski et al. [7] previously assessed body image perception in a sample of 14 subjects with CAIS, using a questionnaire with three response options (mainly satisfied, somewhat dissatisfied, mainly dissatisfied) and also reported varying degree of body dissatisfaction in 43% of the cohort, related to physical appearance and linked to the presence of obesity. In the dsd-LIFE study [20], participants with 46,XY conditions showed different results concerning body image, as expected given the heterogeneous group composition, with female-identifying participants registering more positive scores than participants with partial androgen action or hypospadias, but no specific data on CAIS are available.

In our study, the patients who were not taking HRT performed worse in all questionnaires, which could be expected and possible biased by a worse attitude toward the psychological and medical issues. We did not register significant differences among different HRT regimes and questionnaires’ results, but this could have been influenced by the small number of patients included in each subgroup, and further multicentric studies are needed to better assess this aspect. As mentioned, very limited data are available in CAIS; however, similar results have been observed in young women with spontaneous POI, where FSFI scores resulted lower in comparison to age-matched controls despite taking HRT with optimal compliance, suggesting other factors playing a role in sexuality outcomes [21].

The FSFI questionnaire was also used in the recent study by Birnbau et al. [6] to explore the differences in sexual function between testosterone and oestradiol transdermal therapy: at the end of the study, the difference of FSFI total score between the two groups was not significant, and testosterone resulted superior to oestradiol only in improving the domain of sexual desire [6]. In the present study we could not properly explore this point, as at recruitment only 2 of our patients were receiving testosterone, yet we still noticed that their sexual and arousal domains’ results were in the high-normal range. It will be interesting to further explore the different sub-domains of sexual health in larger cohorts. Considering the high prevalence of sexual dysfunction in our patients, a trial with testosterone therapy aiming to improve sexual desire could be worth a try in the future, even though the unsatisfactory results registered with respect to the overall sexual satisfaction in the mentioned trial [6] seems to suggest it might not be enough.

Another issue that probably needs to be better explored in the future is target oestradiol levels and the goals of HRT. Since no specific guidelines on the optimal management are available for patients with CAIS, we derived target oestradiol levels during HRT from the most recent European guidelines for POI, suggesting to aim to obtain physiological oestradiol levels as found in the serum of women with normal menstrual cycles, average 50–100 pg/ml [22]. The recent guidelines on Turner syndrome propose even higher goals of oestradiol concentrations of 100–150 pg/ml [23]. Our study underlines how in clinical practice these levels seem to be hard to achieve: as shown, most of our cohort had insufficient serum oestradiol concentration < 50 pg/ml, and none of them levels above 100, therefore we could not explore if higher oestrogen levels could have resulted in improved sexual health. The type of formulation employed and the distance from the last administration could also impact the oestradiol levels registered.

Finally, in our study the prevalence of sexual dysfunction resulted high regardless of the presence of vaginal hypoplasia and of the dilation treatment performed. Again, this observation could be affected by the small cohort, and it is consistent with previous limited literature. In the study by Minto et al. mentioned above [10], women with shorter vaginal length had a higher incidence of vaginal penetration problems, yet without reaching statistical significance. Previous dilator treatment for vaginal hypoplasia did not affect sexual function scores nor self-perceptions of vaginal normality [10]. Further multicentric studies are needed to better explore the impact of vaginal hypoplasia and the efficacy of treatment on the incidence of sexual difficulty. In this regard, it is worth noticing how, as underlined in the review by Callens et al. [24], the definition itself of functional success after treatments for vaginal hypoplasia is quite variable and mostly defined as “satisfactory sexual intercourse” without relying on standardized methods of assessment. Only a few studies have utilized reliable questionnaires (i.e. FSFI), and none focusing specifically on CAIS [24].

In patients with CAIS, the burden of sexual distress is complex and multifaceted and could be referred not only to the effect of the physical and hormonal changes, but also to the emotional burden of the diagnosis and its consequent infertility, which can have consequences on self-perception and relationships. Moreover, age at evaluation and cultural and economic conditions can also play a role.

The main limitation of the present study is the small number of patients included. However, the rarity of the condition and the lack of centralized adult care pose challenges in collecting data on patients with CAIS. Moreover, the cross-sectional design of the work prevents us from drawing conclusions on causal effects. Still, we believe that the present study represents a promising starting point for future multicentre prospective studies to focus on this often-neglected issue. It also presents the strength of stratifying the sexual outcomes analysed according to gonadal status and type of HRT, posing new challenges in the management of the condition. Another limitation is that steroid hormone determination was based on radioimmunoassay and not tandem mass spectrometry. Finally, we did not register the relationship status of the patients included, which also has an impact on sexual results.

In conclusion, our study demonstrates how preservation of gonads and standard-dose HRT does not guarantee adequate sexual function nor optimal hormone levels. Psychological and sexual aspects must be taken into consideration in the management and therapeutic choices for these patients, as they highly impact the quality of life despite being still largely ignored in routine clinical management.