Interventions for Hirsutism Excluding Laser and Photoepilation Therapy Alone

Abridged Cochrane Systematic Review Including GRADE Assessments

E.J. van Zuuren; Z. Fedorowicz


The British Journal of Dermatology. 2016;175(1):45-61. 

In This Article


Search results are shown in Figure 1. We included 157 studies comprising 10 550 women.[21–177] The mean age varied between 15 years and 46 years, and the overall mean was 25·4 years.

Figure 1.

Flowchart showing allocation of search results.

Our primary outcome of participant-reported improvement was assessed in 26 of 157 studies, change in health-related quality of life in three of 157 studies and adverse events were reported in 63 of 157 studies. Our secondary outcomes were more frequently addressed.

Only four studies were considered to have a low risk of bias.[77,97,105,150] Most studies (123 of 157) were assessed as 'high risk' of bias (plausible bias that seriously weakens confidence in results), mainly owing to lack of blinding and attrition bias, and 30 studies were assessed as having an 'unclear risk' of bias (Fig. 2 and Table S1; see Supporting Information). One-third of the studies (48 of 157) reported data that could not be further analysed (Table S2; see Supporting Information).[130–177] The remaining 109 studies covered 133 comparisons (Table S3; see Supporting Information).[20–129] Although the minimal important difference for the FG score has not been established, a change score exceeding 7 was considered to be clinically important. We reported changes in androgen levels and BMI only when these changes were reflected in an associated clinically important difference in FG score (for full details, see Cochrane review).[15]

Figure 2.

Summary of risk of bias graph.

We highlight the interventions routinely prescribed for hirsutism. (For details of all interventions, see Table S3; Supporting Information and the Cochrane review).[15]

Lifestyle Interventions

Exercise alone, or exercise combined with a calorie-restricted diet did not result in statistically significant differences in hirsutism scores compared with the 'no exercise' or dietary intervention control groups.[74,75,82,126]

Topical Treatments

Eflornithine hydrochloride 13·9% cream compared with vehicle in 596 participants demonstrated a statistically significant effect rated on a visual analogue scale, assessed as 'overall bother' with facial hair, with an MD of 16·30 (95% CI: 11·72–20·88, P < 0·001). The physician assessments confirmed eflornithine to be more effective than vehicle (RR: 3·56, 95% CI: −2·24–5·66; P < 0·001) (NNT = 5, 95% CI: 4–7).[80,127]

There was no statistically significant difference in adverse events (RR: 1·16, 95% CI: 0·95–1·41) between groups and these were mainly acne and pseudofolliculitis barbae, with burning and stinging more frequently reported in the eflornithine group. Two studies comparing topical finasteride with placebo showed conflicting results.[79,81]

Oral Contraceptive Pills

Fourteen studies compared oral contraceptive pills (OCPs) with each other or with placebo.[33,38,39,75,86,89,95,96,104,108,109,114,118,120] Reductions in FG scores varied from minimal (2–3) to clinically important reductions (8–12). All OCPs showed increased SHBG accompanied by a reduction in free testosterone. Data from only three studies evaluating an OCP containing ethinyl oestradiol with 2 mg cyproterone acetate vs. ethinyl oestradiol combined with 0·15 mg desogestrel could be pooled.[38,95,96] Both treatments were effective for hirsutism, but the MD between the two OCPs was not statistically significant. The quality of the evidence for the outcomes varied from low to high (Table 1). No important changes in BMI were seen in the few studies that addressed this outcome and three studies showed a reduction in acne severity in the OCP groups, but there were no other significant differences between the various OCPs.[38,68,89]


Flutamide 250 mg twice daily was compared with placebo in two studies.[68,97] Study duration was 6 months[97] and 12 months[68] and although a greater reduction in FG score was reflected by the longer treatment duration, there was a clinically important difference in both studies compared with placebo ( Table 2 ). There was no difference in the number of adverse events between treatment groups, although gastrointestinal discomfort was reported in the flutamide group. The quality of the evidence for outcomes was rated low to very low.

Flutamide was compared with spironolactone in two studies, each with 20 participants.[59,97] Both treatments appeared to be effective for hirsutism (MD: −1·90, 95% CI: −5·01–1·21 and MD: 0·49, 95% CI: −1·99–2·97). Flutamide 250 mg twice daily yielded higher reductions in FG score than the single dose, as did the longer treatment duration (9 months vs. 6 months) (Table S3; see Supporting Information). There were no statistically significant differences in the proportion of participants who experienced adverse events. The most frequent adverse event with spironolactone was irregular bleeding. The quality of evidence for the outcomes was rated moderate to very low (Table S4; see Supporting Information).

Spironolactone 100 mg per day vs. placebo was assessed in one four-armed study with 40 participants.[97] The MD between treatments in change in FG score was −7·69 (95% CI: −10·12 to −5·26, P < 0·001), which is both statistically significant and a clinically important difference. The quality of evidence for the outcomes was low (Table S5; see Supporting Information).

5α-reductase Inhibitors

Finasteride 5–7·5 mg per day was compared with placebo in three studies.[47,87,97] Pooled data showed a difference in change in FG score of −5·73 (95% CI: −6·87 to −4·58, P < 0·001; I2 = 0%); however, this was confirmed by the participant assessments in only one study[87] and not in the other.[97] There was no statistically significant difference in adverse events (RR: 1·13, 95% CI: 0·48–2·67; I2 = 18%). There was a statistically significant MD in reduction of dihydrotestosterone levels in the finasteride groups because finasteride reduces the conversion of testosterone into dihydrotestosterone. The quality of evidence was mainly very low for the different outcomes ( Table 3 ). Two further studies comparing dosages of finasteride 2·5 mg, 5 mg and 7·5 mg per day showed that a longer treatment duration resulted in more substantial reductions in the FG scores, while treatment with a higher dose had less impact (Table S3; see Supporting Information).[23,35]

Insulin Sensitizers

Several studies evaluated metformin, rosiglitazone, troglitazone and pioglitazone for hyperinsulinaemia and improvement of the menstrual cycle in PCOS and also reported outcomes for hirsutism. However, owing to serious adverse events, rosiglitazone and troglitazone have been withdrawn from the U.K. market, as has pioglitazone elsewhere, but these are still available in some countries. Metformin was compared with placebo in eight studies, but showed no effect on hirsutism.[56,68,74,75,94,98,103,105] It did have a beneficial effect on the menstrual cycle and ovulation in several studies, although this did not reach statistical significance (Table S3; see Supporting Information). The quality of evidence was moderate to very low (Table S6; see Supporting Information). No other insulin sensitizers had a positive effect on hirsutism.

Combined Treatments

In 64 comparisons, combinations of treatments were evaluated against a single treatment or other combined treatments. For further details, see Table S3 (see Supporting Information) and the Cochrane review.[15]

The following combinations did not result in a meaningful reduction in FG score:

  1. Adding metformin to lifestyle interventions, OCP or flutamide[58,67,68,74,75,76]

  2. Adding simvastatin to OCPs or metformin[29,30]

  3. Adding sibutramine to OCPs[113]

  4. Any combinations with clomiphene.[112,129]

Oral Contraceptive Pills + Cyproterone Acetate. Although clinical and methodological heterogeneity did not allow data-pooling from studies that assessed OCPs in combination with cyproterone acetate (20–100 mg), this combination resulted in greater reductions of FG score than OCPs alone.[31,37] However, only one comparison demonstrated a statistically significant difference.[37]

Oral Contraceptive Pills + Flutamide. One four-armed study compared triphasic OCP alone vs. OCP plus three different dosages of flutamide (125–375 mg).[41] Adding flutamide to an OCP provided better improvements in hirsutism, but the higher dosage of flutamide was no more effective than the 125 mg dosage (MD: 1·45, 95% CI: −0·31–3·21). Adverse events related to flutamide included dry skin, nausea, vomiting and diarrhoea.

Oral Contraceptive Pills + Finasteride. Adding 5 mg finasteride to OCP provided a small difference in reduction of FG score compared with OCP alone (MD: −2·00, 95% CI: −3·83 to −0·17; P = 0·03), which although statistically significant is not clinically important.[121]

Oral Contraceptive Pills + Cyproterone Acetate vs. Oral Contraceptive Pills + Spironolactone. Differences in compositions of OCPs and dosages of cyproterone acetate and varying study durations, did not permit data-pooling.[55,60,84,92,102] No studies demonstrated a statistically significant difference in reduction of FG score between treatment groups, suggesting that both combination therapies are similarly effective.

Combinations With Gonadotropin-releasing Hormone Agonists. Gonadotropin-releasing hormone agonists (GnRH-As) combined with OCPs or hormone replacement therapy (HRT) were assessed in 11 studies vs. GnRH-A alone or vs. OCP.[26,42,44,53,57,62–64,106,123,124] There was a small additional effect with the combination therapy when compared with GnRH-A or OCP alone in only three of the studies,[42,44,123] but no difference could be demonstrated for the other eight studies.

Adverse events associated with GnRH-As included hot flushes, depression, sweating, vaginal dryness, headache, decreased libido and breast tenderness. Addition of an OCP or HRT reduced those complaints.

Other Treatment Comparisons

The following treatments appeared to be ineffective for hirsutism:[48–50,70,82,83,90,100,107,112,113,116]

  1. sibutramine

  2. clomiphene

  3. acarbose

  4. myo-inositol

  5. D-chiro-inositol

  6. spearmint tea, camomile tea

  7. low-frequency electroacupuncture

  8. statins

  9. cimetidine

  10. bromocriptine.

Oral Contraceptive Pills vs. Finasteride. One study of OCP vs. finasteride with a high risk of bias showed no statistically significant difference in reduction of FG score (MD: −0·48, 95% CI: −2·42–1·46).[115]

Finasteride vs. Spironolactone. Two studies of 6 months' duration addressed these interventions.[97,128] Both treatments were equally effective in improving hirsutism based on participant and physician assessments. The quality of evidence was moderate to low (Table S7; see Supporting Information).

Finasteride vs. Flutamide. Four studies examined these treatments.[65,66,97,99] Two of the four,[65,99] showed a statistically significant difference in reduction of the FG score in favour of flutamide, while the other two did not.[66,97] Adverse events included dry skin, decreased libido, headache and gastrointestinal discomfort, with no difference between groups. The quality of evidence was rated high to very low (Table S8; see Supporting Information).