Diagnosis and treatment of hypothyroidism in TSH deficiency compared to primary thyroid disease: pituitary patients are at risk of under-replacement with levothyroxine.

OBJECTIVE: Achieving optimal thyroid hormone replacement is more difficult in TSH deficiency compared to primary hypothyroidism because of the inability to be guided by serum TSH levels. A combination of clinical symptoms and free thyroxine levels (fT4) are typically used to make a diagnosis and monitor replacement. We investigated the diagnosis of TSH deficiency in patients with pituitary disease and the adequacy of levothyroxine replacement compared with primary thyroid disease. DESIGN: Using our department's clinical information system, we identified all patients with a diagnosis of any type of pituitary tumour who had been seen in clinic over a 2-year period. We divided the patients into those at high risk and low risk of TSH deficiency based on the presence of macroadenoma and/or intervention by surgery or radiotherapy. We compared fT4 values in these patients with values in patients with primary thyroid disease in our thyrotoxicosis shared-care scheme (TSC) and hypothyroid register within the same timescale, assessing only those samples considered euthyroid in which TSH was in the normal range. RESULTS: A database query identified 525 patients with a pituitary tumour of whom 344 were considered at high risk of TSH deficiency. A free T4 (fT4) value was found for 514 patients (97·9%). TSC and thyroid register databases revealed fT4 values for comparison with simultaneous normal TSH in patients on no treatment (n = 3777 samples) or on levothyroxine alone (n = 11,805). fT4 levels overall were lower in pituitary patients than in equivalent controls. Of the high risk group not taking levothyroxine 17% had a free T4 ≤ 11 pmol/l compared to only 8·4% of untreated controls. Furthermore, 38·9% of patients on levothyroxine had a free T4 ≤ 13 pmol/l compared to 9·5% of controls on levothyroxine with previous thyrotoxicosis and 13·4% of controls with primary hypothyroidism. Median fT4 in controls on levothyroxine was 16 pmol/l and 20-80th centile range was 14-19 pmol/l. CONCLUSION: Levothyroxine doses were generally under-replaced in pituitary patients compared to primary thyroid disease and the data imply that some untreated patients were actually TSH deficient. The distribution of fT4 in patients with primary thyroid disease on levothyroxine may guide optimum replacement levels in pituitary disease.

Severe endothelial dysfunction in young women with polycystic ovary syndrome is only partially explained by known cardiovascular risk factors.

OBJECTIVE: We aimed to assess whether metabolic abnormalities can explain endothelial dysfunction and associated cardiovascular disease risk (CVDr) in polycystic ovary syndrome (PCOS). Endothelial function, a recognized composite marker of CVDr, may be reduced in PCOS and can be precisely and noninvasively assessed by cardiovascular magnetic resonance (CMR). PATIENTS: Fourteen women with anovulatory PCOS (age [mean +/- SD] 33 +/- 4 years) and 13 controls (age: 33 +/- 6 years) with similar body mass index and regular menses. METHODS: Endothelium-dependent (flow-mediated dilatation - FMD) and -independent (glyceryl trinitrate - GTN) changes in the brachial artery area were measured using CMR in women with PCOS and controls. Arterial function was assessed twice, in the early follicular phase and mid cycle in controls and after an interval of 2 weeks in PCOS subjects. Fasting lipids, glucose, insulin and sex hormones were measured at the first visit. RESULTS: FMD was greatly reduced in women with PCOS compared to controls (-1%vs 5% and 2%vs 12%, P < 0.01) without differences in GTN responses. Risk factors were more prevalent in PCOS women and displayed significant linear relationships with FMD. PCOS status was the strongest predictor of FMD. Linear regression between PCOS and FMD remained significant after correction for all CVD risk markers linked to the metabolic syndrome. CONCLUSION: PCOS is associated with changes in CVD risk markers and pronounced endothelial dysfunction. However endothelial dysfunction with PCOS is only partly explained by recognized CVD risk markers.

Prevalence of polycystic ovaries in women with androgenic alopecia.

OBJECTIVE: Although androgenic alopecia is recognised to be a symptom of polycystic ovary syndrome (PCOS), it is not known whether polycystic ovaries (PCO) and associated endocrine abnormalities are present in patients who present with alopecia as a primary complaint. We therefore set out to determine the strength of the association between androgenic alopecia and PCO. We examined the prevalence of ultrasound-based polycystic ovarian morphology and associated clinical and biochemical features in a large multiethnic group of women whose presenting complaint was of alopecia, and in a control group. SUBJECTS AND METHODS: We studied 89 women of mixed ethnic origin with androgenic alopecia and compared them to 73 control women. A detailed history was taken, anthropometry was performed and assessment of body-hair distribution was made. The presence of PCO was established by pelvic ultrasound scan. Serum gonadotrophins, testosterone, androstenedione, dihydrotestosterone and sex hormone binding globulin concentrations were measured. RESULTS: Women with alopecia had a higher prevalence of PCO and hirsutism than the control population (PCO: 67% vs 27%, P<0.00001; hirsutism: 21% vs 4%, P=0.003). Women with alopecia (with or without PCO) had higher testosterone, androstenedione and free androgen index than controls, even though few had frankly abnormal androgens. CONCLUSIONS: These findings confirm an association between androgenic alopecia and PCO, and other symptoms of hyperandrogenaemia. Thus most women who present with androgenic alopecia as their primary complaint also have PCO and have indices of abnormal androgen production. Since PCO is a well known risk factor for development of type 2 diabetes, this association has important implications for long-term management.

Variation within the type 2 diabetes susceptibility gene calpain-10 and polycystic ovary syndrome.

Variation within the calpain-10 gene (CAPN10) has been proposed to account for linkage to type 2 diabetes on chromosome 2q in Mexican-Americans, and associations with diabetes have been reported in several other populations. Given the epidemiological, physiological, and genetic overlap between type 2 diabetes and polycystic ovary syndrome (PCOS), CAPN10 represents a strong candidate gene for a role in PCOS susceptibility. Using both family based and case-control association resources (146 parent-offspring trios; 185 additional PCOS cases; 525 control subjects, all of European ancestry), we sought association between CAPN10 variation and PCOS, focusing on four single nucleotide polymorphism (SNP) variants (SNP-44, SNP-43; SNP-19; SNP-63). On single-locus transmission disequilibrium analysis in the 146 trios, there was nominal evidence (P = 0.03) of excess transmission of the more common allele at SNP-63. This association was not, however, replicated in the case-control analysis. No other significant associations were observed at the single-locus or haplotype level in either the transmission-disequilibrium or case-control analyses. The relative risk for the high-risk diabetes susceptibility 112/121 genotype (SNPs 43-19-63) was 0.84 (95% confidence intervals, 0.40-1.71). No associations were seen with intermediate traits of relevance to diabetes and PCOS pathogenesis. We have found no evidence from these analyses that CAPN10 gene variation influences susceptibility to PCOS.