Re: NFLD and diabetes

x-no-archive: yes



"Conclusions: The results of the present study support the view that unknown CS is not rare among patients with diabetes mellitus. This is the first demonstration that screening for CS may be feasible at the clinical onset of diabetes in an unselected cohort of patients. Therefore, early diagnosis and treatment of CS may provide the opportunity to improve the prognosis of diabetes."


Screening of Cushing's Syndrome in Adult Patients With Newly Diagnosed Diabetes Mellitus

Giuseppe Reimondo; Anna Pia; Barbara Allasino; Francesco Tassone; Silvia Bovio; Giorgio Borretta; Alberto Angeli; Massimo Terzolo

Clin Endocrinol. 2007;67(2):225-229. ©2007 Blackwell Publishing
Posted 08/30/2007
Summary and Introduction
Summary

Objective: Recent studies have shown that a relatively high number of diabetic patients may have unsuspected Cushing's syndrome (CS). The aim of the present study was to screen for CS in adult patients with newly diagnosed diabetes mellitus who were not selected for clinical characteristics, such as poor control and obesity, which may increase the pre-test probability of CS.
Design, patients and measurement: We prospectively evaluated 100 consecutive diabetic patients at diagnosis from 2003 to 2004. No patient had clear Cushingoid features. Screening was performed by using the overnight 1-mg dexamethasone suppression test (DST) after complete recovery from acute concomitant illnesses and attainment of satisfactory glycaemic control. The threshold of adequate suppression after DST was set at 110 nmol/l.
Results: Five patients failed to suppress cortisol after DST and underwent a repeated DST and a confirmatory standard 2-day, 2-mg DST after 3–6 months from the baseline evaluation. In one woman, a definitive diagnosis of CS was made by a surgically proven pituitary adenoma, and glycaemic control improved after cure of CS.


**************Conclusions: The results of the present study support the view that unknown CS is not rare among patients with diabetes mellitus. This is the first demonstration that screening for CS may be feasible at the clinical onset of diabetes in an unselected cohort of patients. Therefore, early diagnosis and treatment of CS may provide the opportunity to improve the prognosis of diabetes.*********************



Introduction

Cushing's syndrome (CS) is a chronic and insidious disease that is associated with reduced life expectancy; excess mortality is mostly due to cardiovascular disease.[1] There are striking similarities between Cushing's syndrome and the metabolic syndrome as both are characterized by central obesity, hypertension, insulin resistance, glucose intolerance, and dyslipidaemia.[2] Although glucose intolerance or type 2 diabetes occurs in approximately 80% of patients with CS,[3] there is still insufficient information about the prevalence of CS in patients with diabetes mellitus.

Three groups of authors have independently shown that a relatively high percentage of diabetic patients may harbour an unsuspected CS, although the studies had a different design and methodology. In a retrospective study on 90 diabetic patients with a BMI > 25 and glycosylated haemoglobin > 9%, Leibowitz et al.[4] found that three patients (3·3%) had a definitive diagnosis of CS.[4] In a prospective study on 200 overweight patients with poorly controlled type 2 diabetes, Catargi et al.[5] reported a 2% prevalence of previously unknown CS. More recently, in a prospective case-control study, Chiodini et al.[6] demonstrated a higher prevalence of subclinical CS in a cohort of 294 diabetic patients compared to 189 controls (9·4 vs. 2·1%).

Thus, a number of patients with CS may not be recognized while they are managed for diabetes, either because of a mild clinical presentation of hypercortisolism or because of insufficient awareness of the physicians who take care of such patients. Missing a diagnosis of CS may have detrimental consequences because hypercortisolism, although clinically unapparent, makes metabolic control more difficult to achieve and increases the probability of future cardiovascular events through induction/amplification of several risk factors (hypertension, central adiposity, thrombophilic state, etc.). Therefore, the results of such studies argue in favour of systematic screening of CS in the diabetic population, but consensus of the timing of such screening has not been established. It would be better to identify underlying CS when diabetes is first diagnosed than after many years of poor metabolic control and attending complications; however, hypercortisolism may be more subtle and difficult to recognize in the early stages. Further, the clinical onset of diabetes may be associated with stressful conditions that can trigger an aspecific activation of the hypothalamic–pituitary–adrenal (HPA) axis and this hinders the process of distinguishing CS.[7]

The aim of the present study was to screen for CS in adult patients admitted to a diabetic care unit with newly diagnosed diabetes mellitus who were not selected for any clinical characteristic that could have increased the probability of underlying hypercortisolism.
Patients and Methods
Patients

We prospectively evaluated 100 consecutive patients admitted to the diabetic care unit of the Santa Croce & Carle Hospital, Cuneo, from 1 June 2003 to 30 May 2004. They were 63 men and 37 women, aged 30–87 years (median age, 61 years) newly diagnosed as having diabetes mellitus. All patients were seen by the same physicians (AP, FT and GB) and underwent careful clinical examination to exclude the presence of specific signs of hypercortisolism, such as weakness associated with proximal muscle wasting, skin atrophy, ecchymoses, moon face, buffalo hump, or purple striae. One patient was excluded by further analysis because he was taking antiepileptic drugs. None of the remaining patients was receiving any drug known to affect the HPA axis. Furthermore, there were no patients with either alcohol abuse or current or previous history of major mood disorders that required psychiatric intervention. Any subject with body mass index (BMI) greater than 30 kg/m2 was categorized as obese.[8] Any subject with systolic blood pressure greater than 140 mmHg, diastolic blood pressure greater than 90 mmHg, or on antihypertensive treatment was categorized as hypertensive.[9] In the present series, 28 subjects were obese and 52 patients had hypertension, of whom 50 were on antihypertensive treatment. The prevalence of obesity and hypertension is highly comparable to that recently reported in Italian diabetic patients by the Istituto Superiore di Sanità.[10] None of the patients had severe nephropathy (creatinine clearance < 0·5 ml/s) or liver insufficiency; seven patients (7·1%) were diagnosed with type 1 diabetes because they had evidence of autoimmunity (presence of anti-GAD antibodies) with subsequent development of insulin deficiency.[11,12] This percentage is in keeping with population-based data obtained in our and in other geographical areas on adult patients referred for a new diagnosis of diabetes.[13,14]

Fifty-two patients (29 men and 23 women aged 30–87 years, median age, 63 years) were admitted to the diabetic unit ward from the emergency room because of newly detected severe hyperglycaemia with hyperosmolarity or ketoacidosis, or concomitant illnesses (i.e. respiratory or urinary tract infections, foot infections) (group A). The other 47 patients (34 men and 13 women, aged 33–86 years; median age, 56 years) were referred by general practitioners to the outpatient department of the diabetic unit for newly detected moderate to severe hyperglycaemia and/or presence of comorbidities, according to a locally validated disease management programme[15] (group B). All patients underwent screening within 30 days of the initial diagnosis of diabetes and after hypoglycaemic treatment (oral antidiabetic agents or insulin) had been commenced. None of the patients was taking thiazolidinedione agents. The main characteristics of the evaluated subjects are shown in Table 1 . The institutional review board approved the study and all patients provided written informed consent. The study was performed in accordance with the Declaration of Helsinki.
Protocol

All patients underwent a first screening step for CS by undergoing the overnight 1-mg dexamethasone suppression test (DST): 1 mg dexamethasone was administered orally at 23.00 hours, and blood samples were collected on the following morning at 8.00 hours for determination of plasma cortisol concentration. As the HPA axis could be influenced by either comorbidities or stressful conditions, screening for CS was performed after complete resolution of acute illness, as gauged by normalization of previously altered clinical and biochemical parameters, and when satisfactory glycaemic control had been attained (fasting glucose at the time of screening: median 6·9 mmol/l; range 5·6–11·9 mmol/l). The threshold of adequate suppression after DST was set at 110 nmol/l. This cut-off value was previously calculated by ROC analysis in 106 patients evaluated for suspected CS at our centre.[16] Patients who failed to suppress serum cortisol below the cut-off value underwent a repeated 1-mg DST and a confirmatory second-step hormonal evaluation by undergoing a standard 2-day 2-mg DST 3–6 months after baseline evaluation (0·5 mg dexamethasone was administered orally at 6.00, 12.00, 18.00 and 24.00 hours, and blood samples for cortisol determination were obtained at 08.00 hours on the morning after dexamethasone treatment). A cortisol concentration above 50 nmol/l was considered abnormal and prompted further endocrine evaluation to confirm the diagnosis of hypercortisolism and determine its cause.[7] Briefly, this evaluation included 24-h urine collection for urinary free cortisol measurement, blood drawing at 24.00 hours for determination of midnight plasma cortisol concentration, and ACTH measurement in baseline conditions and after the corticotropin-releasing hormone (CRH) stimulation test, as previously described.[17] Depending on ACTH values, pituitary magnetic resonance imaging (MRI) or adrenal computed tomography was also performed. The protocol flow-chart is outlined in Fig. 1. Reference ranges of the tests used to study the HPA axis were determined in large cohorts of healthy subjects, as previously described.[16,17]

Figure 1.

Protocol flowchart.


Assays

Hormones were measured in-house with commercially available reagents. All samples for an individual subject were batched and run in duplicate using the same assay. Serum and urinary cortisol were measured by RIA (Sorin Biomedica, Saluggia, Italy), and plasma ACTH by immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Sera were immediately separated and stored at –20 °C until assayed (samples for ACTH measurement were collected into prechilled tubes and centrifuged at –20 °C). Intra- and interassay coefficients of variation for all of the above-mentioned variables were below 8 and 12%, respectively.
Statistical Analysis

The sample size was statistically estimated a priori. Seventy-seven patients were the minimum sample size needed to confirm the hypothesis of a prevalence rate for CS of at least 1% (α = 0·05, β = 0·80).

Rates and proportions were calculated for categorical data, and means and standard deviations for continuous data. Normality of data was assessed by the Kolmogorov-Smirnov test. For continuous variables, differences were analysed by means of the two-tailed Student's t-test when data were normally distributed and by the Mann–Whitney U-test for nonparametric data. For categorical variables, differences were analysed by means of the χ2-test and Fisher's exact test. Levels of statistical significance were set at P < 0·05. All analyses were performed using the Statistica® software package (Microsoft Corp, Tucla, OK).
Results

The patients of group A presented either higher glycaemia or higher glycosylated haemoglobin than group B patients; the c-peptide levels were conversely lower in group A than in group B ( Table 1 ). Moreover, group A patients had higher cortisol levels after 1-mg DST (64 ± 44 vs. 39 ± 14 nmol/l, P = 0·001). A satisfactory metabolic control was attained in both groups of patients at the time when the 1-mg DST was performed (fasting glucose of 7·6 ± 1·3 mmol/l for group A vs. 7·1 ± 1·3 mmol/l for group B, P = NS).

Five patients (three women and two men, aged 31–74 years) failed to suppress cortisol levels to less than 110 nmol/l after 1-mg DST. Application of the recently proposed cut-off point of 50 nmol/l7 would have resulted in poor specificity, yielding a rate of nonsuppression as high as 32·3% in our series. All of the five patients who failed to suppress cortisol levels were in group A; four had hypertension and only one subject was obese. Neither fasting glucose nor glycosylated haemoglobin levels were different between patients who suppressed and those who did not ( Table 2 ). Patients showing post-DST cortisol ≥ 110 nmol/l had higher cortisol levels, either at 8.00 hours or at midnight, than the other patients ( Table 2 ). None of the patients had plasma ACTH levels < 2·2 pmol/l. Comparison of clinical and biochemical parameters between DST suppressors and nonsuppressors are reported in Table 2 .

As established by the protocol, the DST nonsuppressor patients underwent a repeated 1-mg DST and a confirmatory test by the standard 2-day 2-mg DST at least 3 months after the baseline evaluation. Four patients suppressed cortisol levels below 110 nmol/l after the repeated 1-mg DST and below 50 nmol/l after 2-mg DST. A 70-year-old obese and hypertensive woman, who was admitted for frank hyperglycaemia and foot infection, failed to suppress cortisol levels either after the repeated 1-mg DST (607 nmol/l) or after the 2-day 2-mg DST (179 nmol/l). Further endocrine work-up demonstrated both increased UFC (1390 nmol/24 h; normal values 138–552 nmol/24 h) and midnight serum cortisol levels (497 nmol/l; normal values < 229 nmol/l), whereas ACTH responded to CRH (baseline ACTH, 12·9 pmol/l; peak ACTH, 27·3 pmol/l). MRI was negative for any pituitary mass lesion and bilateral inferior petrosal sinus sampling was performed. The centre-to-periphery ACTH ratio after CRH stimulation was 10·3, thus establishing the diagnosis of pituitary-dependent CS. On February 2006, the patient underwent surgery and a 4-mm pituitary microadenoma, cells of which stained positive for ACTH, was found and removed. Three months after surgery, the patient is hypoadrenal on glucocorticoid replacement therapy. No other anterior pituitary hormone deficiency has been evident and better glycaemic control has been attained, with a daily insulin requirement reduced from 18 to 9 IU. Therefore, the incidence of confirmed CS in our series of 99 patients newly diagnosed with diabetes mellitus was 1% (95% confidence interval; 0·9–2·9%).
Discussion

In a series of 99 consecutive patients newly diagnosed with diabetes mellitus we identified one patient with definitive CS, sustained by a pituitary microadenoma. An incidence of CS of 1% (95% confidence interval, 0·9–2·9%) could occur by chance but is consistent with the findings of previous studies,[4-6] thus supporting the view that undiagnosed CS is not rare among diabetic patients.

The novelty of the present study is the demonstration that previously unsuspected CS may be detected at the time of diagnosis of diabetes in unselected patients. Previous studies have indeed included patients with clinical characteristics that may increase the pre-test probability of CS, such as central obesity and long-standing uncontrolled diabetes, whereas our cohort included unselected patients with newly detected diabetes mellitus (in 7·1% of them type 1 diabetes was eventually diagnosed). The clinical setting we investigated was particularly challenging because in about half of our patients the clinical presentation was severe enough to require emergency admission and it has been shown that the HPA axis may be activated among patients with poor glycaemic control and/or diabetic complications.[18-22] Furthermore, our cohort included a higher percentage of men, which could reduce the likelihood of CS, as the female : male ratio ranges from 3 to 8.[3] We tried to circumvent the problem of stress-induced activation of the HPA axis by attaining a satisfactory metabolic control before the 1-mg DST was performed, as it is known that only minor functional abnormalities of the HPA axis may be found in well-controlled patients,[23,24] and by using a more stringent criterion for the 1-mg DST. The cut-off point at 110 nmol/l was previously validated in our centre and proved to have the best sensitivity and specificity to distinguish patients with CS from patients with the metabolic syndrome by ROC analysis.[16]

We cannot exclude the possibility that some patients with a milder degree of CS were missed; however, because of the particular characteristics of our cohort, we were particularly concerned by the issue of inadequate specificity associated with the use of lower thresholds for 1-mg DST. The pivotal study of Catargi et al.[5] demonstrated this problem well, as 63·8% of the patients who failed to suppress below 58 nmol/l were categorized as patients with functional activation of the HPA axis (false positive results). In our series, application of the recently proposed cut-off value of 50 nmol/l7 would have generated a 32·3% rate of nonsuppression. Conversely, using our validated criterion of 110 nmol/l for post-DST cortisol, only a limited number of patients (5%) needed to be re-evaluated. We obtained concordant results in the nonsuppressor patients by either a repeated 1-mg DST or the 2-day 2-mg DST. Thus, the 2-day 2-mg DST, which is a classic confirmatory test, could be safely omitted in this multistep diagnostic strategy.

Taking into account the differences between the previous series and the present one, the prevalence of occult CS among diabetic patients is consistent and comparable across the various studies, ranging between 1 and 3%, with the only exception being the study of Chiodini et al.[6] However, they focused on subclinical CS using low thresholds for any test employed to evaluate the function of the HPA axis, in order to maximize sensitivity. Indeed, they categorized 2·1% of control subjects as subclinical CS.

Moreover, a very mild hypercortisolaemic state may have a less important metabolic impact, as the critical issue is whether the recognition and cure of unsuspected CS may reduce the mortality associated with diabetes. In the few diabetic patients with occult CS who were submitted to surgical treatment, either an improvement in metabolic control or a relative weight reduction was observed.[4-6] Also, our patient with a definitive diagnosis of CS attained better metabolic control following cure of hypercortisolism.

To summarize, in our clinical setting organic hypercortisolism may be less frequent than in patients selected for long-standing and poorly controlled diabetes, and also more difficult to demonstrate because of a possible nonspecific, stress-induced activation of the HPA axis. However, our results appear to be representative of a very significant incidence of undiagnosed CS in diabetes mellitus that should not be considered as negligible in light of the incidence of the metabolic syndrome and type 2 diabetes in western countries. Further, it should be considered that type 2 diabetes is a progressive disease with a certain steady deterioration of metabolic control, notwithstanding intensive treatment, which reflects ongoing failure of beta-cell function.[25] In this respect, early detection of hypercortisolism at the time when diabetes is first diagnosed and treated may provide the opportunity to change the prognosis of the disease, and the present study provides the first demonstration that such a screening is feasible.

The crucial point is to estimate the risk–benefit balance of a widespread screening of patients with a common disease for a rare one. Neither the present nor previous studies have addressed this issue, and this warrants further investigation on a population-based scale. Taking into account the costs and the potentially harmful consequences of a widespread screening, we suggest the employment of robust thresholds for tests used for this purpose. This would limit the number of subjects submitted to further testing because of false positive results and make screening more readily applicable.


CLICK HERE for subscription information about this journal.

Table 1. Main Characteristics of the Evaluated Subjects: Inpatients (Group A) and Outpatients (Group B)


Table 1: Main Characteristics of the Evaluated Subjects: Inpatients (Group A) and Outpatients (Group B)


Table 2. Comparison of Clinical and Biochemical Parameters Between the Patients Who Suppressed After the 1-mg DST and Patients Who Did Not


Table 2: Comparison of Clinical and Biochemical Parameters Between the Patients Who Suppressed After the 1-mg DST and Patients Who Did Not




References

1. Pivonello, R., Faggiano, A., Lombardi, G. & Colao, A. (2005) The metabolic syndrome and cardiovascular risk in Cushing's syndrome. Endocrinology and Metabolism Clinics of North America, 34, 327–339
2. Walker, B.R. (1996) Abnormal glucocorticoid activity in subjects with risk factors for cardiovascular disease. Endocrine Research, 22, 701–708
3. Boscaro, M., Barzon, L., Fallo, F. & Sonino, N. (2001) Cushing's syndrome. Lancet, 357, 783–791
4. Leibowitz, G., Tsur, A. & Chayen, S.D. (1996) Pre-clinical Cushing's syndrome: an unexpected frequent cause of poor glycaemic control in obese diabetic patients. Clinical Endocrinology, 44, 717–722
5. Catargi, B., Rigalleau, V., Poussin, A., Ronci-Chaix, N., Bex, V., Vergnot, V., Gin, H., Roger, P. & Tabarin, A. (2003) Occult Cushing's syndrome in type-2 diabetes. Journal of Clinical Endocrinology and Metabolism, 88, 5808–5813
6. Chiodini, I., Torlontano, M., Scillitani, A., Arosio, M., Bacci, S., Di Lembo, S., Epaminonda, P., Augello, G., Enrini, R., Ambrosi, B., Adda, G. & Trischitta, V. (2005) Association of subclinical hypercortisolism with type 2 diabetes mellitus: a case-control study in hospitalized patients. European Journal of Endocrinology, 153, 837–844
7. Arnaldi, G., Angeli, A., Atkinson, A.B., Bertagna, X., Cavagnini, F., Chrousos, G.P., Fava, G.A., Findling, J.W., Gaillard, R.C., Grossmann, A.B., Kola, B., Lacroix, A., Mancini, T., Mantero, F., Newell-Price, J., Nieman, L.K., Sonino, N., Vance, M.L., Giustina, A. & Boscaro, M. (2003) Diagnosis and complications of Cushing's syndrome: a consensus statement. Journal of Clinical Endocrinology and Metabolism, 88, 5593–5602
8. Bray, G.A. (2001) An approach to the classification and evaluation of obesity. In: P. Bjorntorp, B.N. Brodoff eds. Obesity. Lippincott, Philadelphia, 294–308
9. Guidelines Committee European Society of Hypertension (2003) European Society of Cardiology guidelines for the management of arterial hypertension. Journal of Hypertension, 21, 1011–1053
10. Studio Quadri (2004) Studio nazionale sulla qualità dell'assistenza alle persone diabetiche nelle Regioni italiane. http://www.epicentro.iss.it/quadri/pdf/SOCO%20quadri.pdf (accessed 15 September 2006)
11. Tuomi, T., Groop, L.C., Zimmet, P.Z., Rowley, M.J., Knowles, W. & Mackay, I.R. (1993) Antibodies to glutamic acid decarboxylase reveal latent autoimmune diabetes mellitus in adults with a non-insulin-dependent onset of disease. Diabetes, 42, 359–362
12. Groop, L.C., Bottazzo, G.F. & Doniach, D. (1986) Islet cell antibodies identify latent type 1 diabetes in patients aged 35–75 years at diagnosis. Diabetes, 35, 237–244
13. Bruno, G., De Salvia, A., Arcari, R., Borra, M., Grosso, N., Carta, Q., Trovati, M., Veglio, M. & Pagano, G. (1999) Clinical, immunological, and genetic heterogeneity of diabetes in an Italian population-based cohort of lean newly diagnosed patients aged 30–54 years. Piedmont Study Group for Diabetes Epidemiology. Diabetes Care, 22, 50–55
14. Romkens, T.E., Kusters, G.C., Netea, M.G. & Netten, P.M. (2006) Prevalence and clinical characteristics of insulin-treated, anti-GAD-positive, type 2 diabetic subjects in an outpatient clinical department of a Dutch teaching hospital. Netherlands Journal of Medicine, 64, 114–118
15. A.M.D., Associazione Medici, Diabetologi; S.I.M.G., Società Italiana di Medicina Generale; S.I.D., Società Italiana di Diabetologia (2000) L'assistenza al paziente diabetico: raccomandazioni cliniche ed organizzative di AMD-SID-SIMG. Dall'assistenza integrata al team diabetologico e al Disease Management della malattia diabetica. http://www.siditalia.it/Pubblicazioni/Protocollodintesa.pdf (accessed 15 September 2006)
16. Reimondo, G., Allasino, B., Bovio, S., Paccotti, P., Angeli, A. & Terzolo, M. (2005) Evaluation of the effectiveness of midnight serum cortisol in the diagnostic procedures for Cushing's syndrome. European Journal of Endocrinology, 153, 803–809
17. Reimondo, G., Paccotti, P., Minetto, M., Termine, A., Stura, G., Bergui, M., Angeli, A. & Terzolo, M. (2003) The corticotrophin-releasing hormone test is the most reliable noninvasive method to differentiate pituitary from ectopic ACTH secretion in Cushing's syndrome. Clinical Endocrinology, 58, 718–724
18. Lentle, B.C. & Thomas, J.P. (1964) Adrenal function and the complications of diabetes mellitus. Lancet, 14, 545–549
19. Tsigos, C., Young, R.J. & White, A. (1993) Diabetic neuropathy is associated with increased activity of the hypothalamic pituitary adrenal axis. Journal of Clinical Endocrinology and Metabolism, 76, 554–558
20. Roy, M.S., Roy, A., Gallucci, W.T., Collier, B., Young, K., Kamilaris, T.C. & Chrousos, G.P. (1993) The ovine corticotrophin-releasing hormone stimulation test in type 1 diabetic patients and controls: suggestion of mild hypercortisolism. Metabolism, 42, 696–700
21. Faiman, C. & Moorhouse, J.A. (1967) Diurnal variation in the levels of glucose and related substances in healthy and diabetic subjects during starvation. Clinical Science, 32, 111–126
22. Dacou-Voutetakis, C., Peppa-Patrikiou, M. & Dracopoulou, M. (1998) Urinary free cortisol and its nyctohemeral variation in adolescents and young adults with IDDM; relation to endothelin 1 and indices of diabetic angiopathy. Journal of Pediatric Endocrinology and Metabolism, 11, 437–445
23. Couch, R.M. (1992) Dissociation of cortisol and adrenal androgen secretion in poorly controlled insulin-dependent diabetes mellitus. Acta Endocrinologica (Copenhagen), 127, 115–117
24. Asfeldt, V.H. (1972) Hypophyseo-adrenocortical function in diabetes mellitus. Acta Medica Scandinavica, 191, 349–354
25. UK Prospective Diabetes Study (UKPDS) Group. (1998) Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet, 352, 837–853

Acknowledgements

We thank the nurses of the diabetic care unit of the Santa Croce & Carle Hospital, Cuneo, for their assistance with patient care. We thank Mrs A. Termine for her skilful technical assistance.
Funding Information

This work was partially supported by a grant from the Università degli Studi di Torino (ex-60% funds).
Reprint Address

Giuseppe Reimondo, Dipartimento di Scienze Cliniche e Biologiche, Università di Torino, A.S.O. San Luigi, Regione Gonzole, 10,10043 Orbassano, Italy. Tel.: +39 011 9026292; Fax: +39 011 9038655; E-mail: g.reimondo@xxxxxxxxxxx

Giuseppe Reimondo*, Anna Pia†, Barbara Allasino*, Francesco Tassone†, Silvia Bovio*, Giorgio Borretta†, Alberto Angeli*, and Massimo Terzolo*

*Dipartimento di Scienze Cliniche e Biologiche, Medicina Interna I, A.S.O. San Luigi, Università di Torino
†Endocrinologia, A.S.O. S. Croce e Carle, Cuneo, Italy
.