Polycystic ovarian syndrome (PCOS) is one of the most common endocrine disorders affecting women of reproductive age with a global prevalence ranging from 6% to 21% depending on the diagnostic criteria used⁽¹ . In the Indian context studies have reported a prevalence of 9.13% to 22.5% reflecting both genetic and environmental influences ^(2,3).

The diagnosis of PCOS is based on the Rotterdam criteria 2003 which requires 2 out of 3 following features: oligoovulation/anovulation, clinical or biochemical hyperandrogenism and polycystic ovarian morphology (PCOM) on ultrasound. Based on these criteria, 4 phenotypes of PCOS have been identified:

1. Phenotype A (Hyperandrogenism + ovarian dysfunction + PCOM)
2. Phenotype B (Hyperandrogenism + ovarian dysfunction)
3. Phenotype C (Hyperandrogenism + PCOM)
4. Phenotype D (Ovarian dysfunction + PCOM)

Each phenotype exhibits distinct profiles influencing long term outcomes such as infertility, insulin resistance, metabolic syndrome and cardiovascular risk.^(4,5)

Therefore, this cross-sectional study was designed to evaluate the phenotypic distribution of PCOS among women presenting to a tertiary care centre and to assess the clinical, metabolic and hormonal implication of each phenotype.

MATERIALS AND METHODS

 This is a hospital based cross-sectional study conducted in the department of Obstetrics and Gynaecology at Kolkata Medical College, Kolkata, West Bengal, over a period of 18 months from 1^st January 2023 to 30^th June 2024 after taking approval from institute research ethics committee, vide Ref. No. MC/KOL/IEC/NON-SPON/1669/11/2022, Dated 30/11/2022. Study population included women aged 15- 40 years attending the Gynecology outpatient department and diagnosed with PCOD based on the Rotterdam criteria 2003. Patients with other causes of hyperandrogenism such as Cushing’s syndrome, androgen secreting tumours, congenital adrenal hyperplasia, thyroid disorders and hyperprolactinemia were excluded. Sample size was calculated using appropriate statistical formula for cross-sectional study [n=4PQ/d², n=180] and a total of 200 women were enrolled using convenience sampling. After obtaining written informed consent, detailed clinical history and physical examination was conducted. A pre-specified proforma was used for obtaining demographic and clinical examination findings. Detailed clinical assessment was done on hirsutism (assessed by modified Ferriman-Gallway score), acne, acanthosis nigricans. Metabolic data included fasting blood glucose and lipid profile, hormonal profiles on LH, FSH, LH/FSH ratio, total testosterone, TSH and prolactin level. Pelvic ultrasonography was performed to detect polycystic ovarian morphology. All data of 200 women were analysed and they were categorised into four phenotypes A, B, C, and D based on Rotterdam criteria 2003.

STATISTICAL ANALYSIS PLAN:  Data was analysed using SPSS software. Descriptive statistics was used for demographic variable. Comparison between phenotypes with respect to clinical, metabolic and hormonal parameters was done using chi square test for categorical variable and ANOVA or Kruskal Wallis test with multiple comparison test for continuous variable. A p value of <0.05 was considered statistically significant.

RESULT

A total 200 women diagnosed with PCOD based on the Rotterdam criteria were included in this study. Based on the presence of Hyperandrogenism (HA), oilgo/anovulation (OA), and polycystic ovarian morphology (PCOM) they were classified into 4 phenotypic categories in which phenotype A was the most prevalent 89 (44.5%) and phenotype B was least prevalent.

Table 1: Distribution of PCOD phenotypes

The clinical features varied among different phenotypes. Most women presenting with PCOD were 20-27 years of age and there was no significant age variation among different phenotypes. Hirsutism and menstrual irregularities were more common in phenotype A and B while acanthosis nigricans was noted more in phenotype A and C. Appearance of acne was relatively more in phenotype

TABLE 2: Clinical features among different phenotypes

No significant variations were found in fasting blood sugar level and HDL level among all phenotypes of PCOS. However, triglycerides were higher in phenotype A than type C and D (162.89±7.76 Vs 154.86±9.77). In LDL study, phenotype A had more LDL level than others (phenotype A 101.75±6.93 Vs Phenotype D 95.84±8.08). Results of both harmful lipids are statistically significant (p<0.0001 and p=0.0001)

Table 3: Metabolic parameters among different phenotypes

More alteration of LH/FSH ratio was seen in phenotype A than other phenotypes. Total Testosterone level in blood of phenotype A was more than type C and D. Results of LH/FSH and mean testosterone level was statistically significant (p<0.0001 and p=0.0019)

Table 4: hormonal profile among different phenotypes

No significant difference in number of ovarian follicles and volume of ovaries were found in all phenotypes of PCOD.

Table 5: ovarian morphology study

DISCUSSION

PHENOTYPIC VARIETY

In the current study, phenotype A emerged as the most common (44.5%) phenotypes of PCOD followed by phenotype D (26.0%) and least common variety was phenotype B (12.5%). Multiple studies across various regions indicate that phenotype A is the most common particularly in hospital-based cohorts. For example, a study by Joshi et al in India reported that phenotype A accounted for 52.6% followed by phenotype B (18.6%), phenotype C (16.3%) and phenotype D (12.5%) ⁽⁶⁾. In contrast studies from USA and Europe show a higher frequency of phenotype C and D possibly reflecting differences in genetics, environment, lifestyle and health seeking behaviour. ^(7,8)

Recognising phenotypic subtype is crucial for both diagnostic clarity and tailored treatment. While phenotypes A and B are more likely to seek medical help due to overt symptoms, phenotype C and D remain undetected unless actively screened. Hence phenotype-based stratification enables more precise risk assessment and management specially in population with diverse clinical presentations.

CLINICAL PROFILE

The present cross-sectional study provides insight into the clinical spectrum of PCOD with particular emphasis on the phenotype distribution and their associated clinical presentations. According to the Rotterdam criteria, PCOD is classified into four phenotypes (A-D) which differ not only in metabolic parameters but also significantly in clinical symptomatology impacting diagnosis and management.

In the present study, menstrual irregularities were noted across all Phenotypic subtypes of PCOD although the frequency and type of abnormalities were varied significantly. We observed 89.88% of phenotype A, 84.0% of phenotype B and 70.5% of phenotype C presented with some form of menstrual irregularities.

In the study of Lizneva D et al, almost all women with phenotype A presented with oligomenorrhoea or amenorrhoea indicating chronic anovulation and menstrual dysfunction which is usually long lasting. ⁽⁴⁾

We observed prevalence of acne in PCOD patients were not prominent presentation in any phenotype. Among phenotype B 20.5% and phenotype C 17.6% had complaints of acne. On the contrary Diamanti kandarakis et al 2012 study observed acne was the main or only clinical presentation of phenotype C and they mainly attended in dermatology clinic rather than gynaecology department. ⁽⁹⁾

In the current study hirsutism was one of the most common presentations of phenotype A (67.4%) and phenotype B (56%). Lizneva et al 2016⁽⁴⁾ and Balen AH et al 1995⁽¹⁰⁾ study concluded hirsutism was the most prominent and often severe, seen in up to 70-80% of PCOD women.

Another clinical presentation, acanthosis nigricans, a clinical marker of insulin resistance was observed in 20.6% women of phenotype C and 20.2% women of phenotype A. Gambineri A et al 2002⁽¹¹⁾ study noted it to be frequent in phenotype A women and Wijeyaratne CN et al 2002⁽¹²⁾ study showed moderate to high prevalence of acanthosis nigricans in phenotype B.

Overall, the clinical presentation of PCOD varies significantly across phenotypes. Phenotype A and B exhibit the most severe hyperandrogenic and metabolic features including hirsutism, acne, acanthosis nigricans along with profound menstrual irregularities, though phenotype C and D women are not excluded totally from menstrual abnormalities.

METABOLIC PROFILE

From metabolic perspective, phenotype A and phenotype B showed significantly higher dyslipidaemia when compared to phenotype C and phenotype D. Although there was no significant fasting blood glucose level variation in all types of PCOD. These findings support the view that hyperandrogenism particularly when combined with anovulation contributes to adverse metabolic profiles as reported by Dunaif et al.⁽¹³⁾ These metabolic abnormalities are associated with increased risk of type 2 diabetes mellitus and cardiovascular diseases. ⁽¹⁴⁾

HORMONAL PROFILE

Hormonal profiles study further delineated phenotypic differences. Phenotype A demonstrated highest LH/FSH ratio (p<0.0001) and elevated testosterone levels (p<0.0019) indicating more profound ovarian dysfunction which has been previously linked to the severity of PCOS shown by Dewailly et al 2011⁽¹⁵⁾ and another study by Sahmay S et al. 2013.⁽¹⁶⁾

These hormonal observations may contribute to the ovulatory dysfunction and subfertility associated with this phenotype.

So, despite the increasing awareness, the phenotypic variability of PCOS remains under recognized in clinical practice, especially in low resource settings. Understanding the distribution of PCOS phenotypes and their associated clinical, metabolic and hormonal characteristics is essential for tailoring patient specific management, strategies and anticipating long term complications.

CONCLUSION

The phenotypic distribution of PCOD reflects the complex and heterogeneous nature of the disorder, emphasizing the need for individualized evaluation and management. Among the various  phenotypes those fulfilling all three Rotterdam criteria tend to exhibit more pronounced clinical, metabolic and hormonal derangements, including higher prevalence of hyperandrogenism, menstrual irregularities, dyslipidemia and increased cardiometabolic risk. In contrast, ovulatory or normo-androgenic phenotypes often demonstrate milder metabolic burden but remain clinically significant, particularly in reproductive health.

Understanding these phenotype-specific patterns is essential for inproving diagnostic accuracy, tailoring therapeutic strategies and predicting long term outcomes. Early identification of high -risk phenotypes can guide timely interventions targeting weight optimization, insulin sentization, androgen control and reproductive planning. Overall, a phenotype-based approach enhances the precision of PCOD management and contributes to better long term reproductive, metabolic and hormonal health in affected women.

LIMITATIONS: A larger sample size could have increased the power of the study. The effect of socio demographic, environmental and psychological determinants of health were not taken into account in this study. More extensive laboratory tests to rule out androgen excess would have strengthened the study.

ACKNOWLEDGEMENT: Authors would like to express their gratitude to their junior postgraduate trainee doctors for their valuable contributions throughout the development of this work. Authors are thankful to all women who participated in this study. Lastly, authors also thank their institution for necessary facilities and support.

CONFLICT OF INTEREST: No.

Source of Funding: There was no financial support concerning the work.

REFERENCES

1. Azziz R, Carmina E, Chen Z, et al. (2016). Polycystic ovary syndrome. Nature reviews disease primers,2,16057.
2. Nidhi R, Padmalatha V, Nagarathna R, Ram A (2011) Prevalence of polycystic ovarian syndrome in Indian adolescents. Journal of paediatric and adolescent gynaecology, 24 (4),223-227.
3. Teede HJ, Misso ML, Costello MF, et al. (2010) Recommendation from the international evidence-based management of PCOS. Human reproduction 25(4),900-912
4. Lizneva D, Suturina L, Walker W, Brakta S, Gavrilova-Jordan L, Azziz R. (2016) Criteria, prevalence, and phenotypes of polycystic ovary syndrome. Fertility and sterility, 106 (1), 6-15.
5. Dunaif A (2016). Polycystic ovary syndrome mechanisms and new treatment approaches. Endocrinology, 157 (12), 4599-4615.
6. Joshi B, Mukherjee S, Patil A et al. (2014) A cross sectional study of the phenotypic and biochemical profile of PCOS in Indian women. Indian J, Endocrinol Metab, 18(3), 317-324.
7. Carmina E, Dewailly D, Escobar HF, et al. (2012). PCOS phenotypes and prevalence: a global perspective. Fertil steril 93(3), 591-598.
8. March WA, Moore VM, Wilson, KJ et al. (2010). The prevalence of polycystic ovary syndrome in a community sample. Hum Reprod, 25 (2), 544-551.
9. Diamanti-kandarkis E, Dunaif A. Insulin resistance and the polycystic ovary syndrome. Reprod Biol Endocrinol 2012; 10-22.
10. Balen AH et al. Obesity and the polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum Reprod.1995; 1098): 2107-2111.
11. Gambineri A et al. Obesity and the polycystic ovary syndrome. Int J Obes (lond) 2002; 26(7): 883-896.
12. Wijeyaratne CN et al. Clinical manifestation and insulin resistance in PCOS among south Asians. Clin Endocrinol (oxf). 2002: 57(3): 343-350.
13. Dunaif A. Insulin resistance and the polycystic ovary syndrome: Mechanism and implications for pathogenesis. Endocr Rev. 1997; 18(6): 774-800.
14. Moran LJ, Misso ML, Wild RA, Norman RJ. Impaired glucose tolerance, type 2 diabetes and metabolic syndrome in PCOS: A systemic review and meta-analysis. Hum Reprod update. 2010; 16(4): 347-363.
15. Dewailly D, et al (2011). The physiology and clinical utility of anti mullerian hormone in women. Human Reproduction update,17(6), 770-785.
16. Sahmay S, et al. (2013). Serum anti mullerian hormone levels in the main phenotypes of polycystic ovary syndrome. European journal of obstetric & gynaecology and reproductive biology, 170 (1) 157-161.