| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Department of Internal Medicine, and Center for Applied Biomedical Research, S. Orsola-Malpighi Hospital (A.G., L.P., R.D.I., U.P., R.P.), and Reproductive Endocrinology Center (B.C., G.E.C., M.F.), University of Bologna, 40138 Bologna, Italy; Department of Endocrinological and Metabolic Sciences and Center for Excellence for Biomedical Research, University of Genova (A.B.), 16132 Genova, Italy; and Endocrine Section, First Department of Internal Medicine, Athens University School of Medicine, Laiko General Hospital (E.D.-K.), 17562 Athens, Greece
Address all correspondence and requests for reprints to: Dr. Renato Pasquali, Division of Endocrinology, Department of Internal Medicine, S. Orsola-Malpighi Hospital, Via Massarenti 9, 40138 Bologna, Italy. E-mail: renato.pasquali{at}unibo.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The purpose of this study was to evaluate the effect of octreotide-LAR, a long-acting somatostatin analog, in anovulatory abdominal obese women with PCOS.
Design: A single-blind, placebo-controlled study was performed, lasting for 7 months.
Setting: The patients were ambulatory throughout the study.
Patients: Twenty PCOS subjects were enrolled. Eighteen completed the study.
Interventions: A low-calorie diet was given during the first month, a low-calorie diet plus octreotide-LAR (10 mg; n = 10 subjects) or placebo (n = 10 subjects) was then given, with one im injection every 28 d (for 6 months).
Main Outcome Measures: The main outcome measures were clinical features, computerized tomography measurement of fat distribution, androgens, GH, IGF-I, IGF-binding proteins (IGFBPs), fasting and glucose-stimulated insulin, and ovulation.
Results: Octreotide had no additional effect in reducing body fat or improving fat distribution than placebo. Conversely, octreotide produced an additional decrease in fasting (P = 0.018) and glucose-stimulated (P = 0.038) insulin levels, an increase in IGFBP-2 (P = 0.042) and IGFBP-3 (P = 0.047), and an improvement in hirsutism (P = 0.004). Moreover, a trend toward greater reductions in testosterone (P = 0.061) and androstenedione (P = 0.069) was observed in women treated with octreotide-LAR compared with those given placebo. All women treated with octreotide ovulated at the end of the study compared with only one of those receiving placebo (P < 0.001).
Conclusions: Octreotide-LAR may be usefully applied to hypocalorically dieting, abdominal obese PCOS women to improve hyperandrogenism and the insulin-IGF-I system. Restoration of ovulatory menstrual cycles appears to be another advantage of this treatment.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Hyperinsulinemia is considered one of the main factors involved in the pathogenesis of anovulation and infertility in PCOS women (4), as a consequence of an interaction with the IGF system (5). In particular, by suppressing hepatic and ovarian IGF-binding protein-1 (IGFBP-1) synthesis, insulin increases the bioavailability of the IGFs, therefore synergizing to stimulate ovarian steroidogenesis and to modulate ovarian follicle growth and apoptosis (5). The increased IGF availability observed in PCOS, however, seems to be differently determined in obese and normal weight subjects (6). In obese PCOS women, IGF-I alterations appear to be strictly related to the effects of hyperinsulinemia, whereas in normal weight PCOS women, they seem to be the consequence of increased pituitary secretion and relatively high circulating concentrations of GH (6). Therefore, the insulin-GH-IGF system appears to play a crucial role in the pathophysiology of PCOS.
Somatostatin is an endogenous hypothalamic peptide with a short half-life that, besides blunting the LH response to GnRH (7) and decreasing GH pituitary secretion (8), inhibits pancreatic insulin release (9). In addition, the recent discovery of somatostatin receptors at the adrenal (10) and ovary (11) levels suggests that these two organs may be under direct control of somatostatin. Preliminary studies by Prelevic and colleagues (12) showed that 7-d administration of octreotide, a synthetic somatostatin analog with a half-life of 80–110 min, improved pulsatile gonadotropin patterns, reduced androgen levels, and decreased fasting and glucose-stimulated insulin concentrations in PCOS women. More recently, a few other prospective, noncontrolled, short-term studies using octreotide expanded these findings (13, 14, 15, 16, 17, 18, 19, 20), showing that this compound not only reduced insulin, IGF-I, LH, and androgen levels (13, 14, 15, 16, 17, 18, 19, 20), but also increased spontaneous (13) and stimulated ovulation (14, 20) in PCOS patients. Unfortunately, the daily multiple sc injections required by the short life of octreotide makes this procedure inappropriate for a long-term treatment.
With this background, we carried out this placebo-controlled study to evaluate the effect of prolonged therapy with a long-acting somatostatin analog formulation (octreotide-LAR), injected im every 28 d, in a selected group of anovulatory PCOS women with abdominal obesity. Octreotide-LAR consists of octreotide acetate encapsulated with a biodegradable polymer (21). Its slow and constant drug release into the circulation makes it possible to improve patient compliance by avoiding repeated daily sc injections (21). An additional advantage of this formulation is a decreased occurrence of side effects, such as gallstone formation (21).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty consecutive overweight or obese women with PCOS attending the Division of Endocrinology, Department of Internal Medicine, S. Orsola-Malpighi Hospital of Bologna were invited to participate in the study. No additional women were screened, because all invited women consented to participate after clinical examination and baseline blood tests were performed, and diagnosis was made. Their body mass index (BMI) values were greater than 28 kg/m2, and the waist to hip ratios were greater than 0.80, consistent with an abdominal fat distribution phenotype (22). The diagnosis of PCOS was made according to the presence of chronic anovulation [supported by luteal progesterone (P) measurement] (23), oligomenorrhea/amenorrhea, hirsutism (Ferriman-Gallwey score, 
8) (24), or elevations in blood levels of total (T) and free (FAI) testosterone (25), and polycystic ovarian morphology at ultrasound (1). None had Cushing’s syndrome; late-onset congenital adrenal hyperplasia; hyperprolactinemia; thyroid dysfunction; diabetes; cardiovascular, renal, or liver disease; or gallstones. Moreover, no patient had taken any medication or significantly modified her body weight in the 3 months preceding the study, nor was any subject dieting.
Human chorionic gonadotropin was measured in each subject before the start of the study to exclude pregnancy. All women were also advised orally and in writing to use nonhormonal contraception throughout the study. The protocol was approved by the local ethics committee, and all women gave their informed and written consents to participate in the study.
Protocol study
At baseline, blood samples for hormonal and metabolic parameters were drawn between 0800 and 0830 h after overnight fasting, followed by an oral glucose tolerance test (OGTT; 75 g; Curvosio, Sclavo, Cinisello Balsamo, Italy). On the same day, an evaluation of anthropometric parameters (height, weight, and waist and hip circumferences), hirsutism, and acanthosis nigricans was performed. Hirsutism was estimated by the Ferriman-Gallwey score (24), whereas acanthosis nigricans was classified according to scale described in Ref. 26 . A computerized tomography (CT) scan measurement of body fat distribution was also performed at the level of L4–L5 to estimate total (TAT), visceral (VAT), and sc (SAT) adipose tissue areas (25). On the following day, all women were placed on a hypocaloric diet, ranging from 1200–1420 kcal/d, with 15% total proteins, 30% lipids, and 55% carbohydrates. All diets were prescribed by the same dietician, who calculated the dietary energy intake by subtracting 500 kcal from the usual individual energy intake. Habitual energy intake was calculated by means of the diet history method and a 3-d recall questionnaire. The women returned after 1 month for anthropometry and CT scan measurements. After this period and while continuing dietary treatment, PCOS women were scheduled to receive, in a random order and according to a single-blind design, octreotide-LAR (Novartis Farma S.p.A., Origgio-Varese, Italy; 10 mg; n = 10) or placebo (saline solution; n = 10), one im injection every 28 d for the following 6 months. The random allocation sequence to the two treatments was decided before the recruitment of the patients. Treatment with octreotide-LAR or placebo was preceded by 3 d of treatment with octreotide (Novartis Farma S.p.A.; 100 µg, sc, twice a day) or an equal amount of saline solution to test the individual tolerability to the drug. One investigator (L.P.) generated the random allocation sequence and administered the drugs (active or placebo) at the out-patient clinic for the entire period of the study, whereas another investigator (A.G.), blinded to group assignment, enrolled and regularly checked all patients at monthly intervals during the study to evaluate the appearance of side effects and tolerance to the treatments. The compliance with diet was evaluated by the same dietician at each visit, according to a previously defined method providing quantitative information on daily energy intake and macronutrient composition of the diet performed during the previous month. Patients failing to comply with dietary prescription (>30% excess kcal intake) were excluded from the study. Patients were also directed to maintain their usual physical activity, which was checked monthly by the self-administered questionnaire proposed by Baecke et al. (27). The frequency of menses during the study and in the 7 months previous to it were also evaluated and were expressed as menstrual status. At the end of the trial, a final examination was conducted by analyzing all measurements performed at baseline.
Finally, the occurrence of ovulation in each patient was monitored during both the first and the last menstrual cycle occurring during treatment with octreotide-LAR or placebo. Ovulation was assessed by performing pelvic ultrasound and by measuring serum levels of LH, FSH, estradiol (E2), and P at 7, 14, 21, and 28 d after the start of menstrual bleeding. Ovulation was assumed to have occurred when a dominant follicle with a diameter higher than 12 mm or a corpus luteum was seen on pelvic ultrasound scanning, and P and E2 concentrations were higher than 8 ng/ml and 100 pg/ml, respectively (23).
During the treatment period, two women in the placebo group were excluded because of noncompliance with the diet, one in the first and the other in the third month of treatment. The cohort available for final statistical analysis, therefore, included 18 PCOS women, 10 treated with octreotide-LAR and eight given placebo. A schematic representation of the study design is shown in Fig. 1
.
|
To assess the tolerance and the appearance of side effects, a self-administered questionnaire was given, and a blood sample was obtained at each monthly visit. The questionnaire evaluated the presence of gastrointestinal disturbances, such as abdominal pain, abdominal swelling, flatulence, diarrhea, nausea, and vomiting, and the appearance of pain, swelling, and blushing at the site of the im injection. Blood tests included liver enzymes, bilirubin, and complete blood count. A gallbladder ultrasound was also performed at the start and the end of the study to detect any gallstone formation.
Assays
Blood samples were centrifuged immediately, and serum was stored at –20 C; plasma was kept at –80 C until assayed. Plasma glucose levels were determined by the glucose oxidase method. Insulin, LH, FSH, T, dehydroepiandrosterone sulfate (DHEA-S), E2, P, androstenedione (A), 17-hydroxyprogesterone (17-OHP), 17-hydroxypregnenolone (17-OHPreg), cortisol, SHBG, IGFBP-1, IGFBP-2, IGFBP-3, and total plasma ghrelin were measured as previously described (28, 29, 30). GH was measured by the immunochemiluminescence assay, and IGF-I was determined by immunoradiometric assay. FAI was calculated as the ratio between total T and SHBG (31). To investigate insulin sensitivity, the homeostasis model assessment (HOMA), the quantitative insulin-sensitivity check index (QUICKI), and the insulin sensitivity index during the OGTT [ISI(composite)] were calculated (32, 33, 34).
The intraassay coefficients of variation in our laboratory were 3.0% for insulin, 4.8% for LH, 1.9% for FSH, 7.0% for T, 5.9% for DHEA-S, 5.6% for E2, 4.1% for P, 6.0% for A, 13% for 17-OHP, 8% for 17-OHPreg, 10% for cortisol, 6.5% for SHBG, 2.5% for IGFBP-1, 6% for IGFBP-2, 3.25% for IGFBP-3, 5.3% for ghrelin, 5% for GH, and 6% for IGF-I.
Statistical analysis
Results were reported as the mean ± SD unless otherwise indicated. The responses of glucose and insulin to the OGTT were analyzed by calculating the area under the curve (AUC) by the trapezoidal method. A one-way ANOVA was applied to compare values between groups at baseline, whereas the repeated measure ANOVA was performed to estimate the within- and between-group modifications. The occurrence of ovulation and side effects and the tolerance to the treatments were analyzed by means of Fisher’s exact test (baseline values between groups), the McNemar test (modifications within groups), and the hierarchical log-linear models (modifications between groups). Statistical analyses were performed by running the SPSS/PC+ (SPSS, Inc., Chicago, IL) software package (35). Two-tailed P < 0.05 was used to define statistical significance.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients completing the study had excellent compliance with both dietary and pharmacological treatments, without any significant difference in energy intake or physical activity score between the two groups. Treatments were well tolerated by all of the women. In particular, none reported the appearance of vomiting or of pain, swelling, and blushing at the site of the im injection for the entire period of the study. Three women treated with octreotide reported slight abdominal pain and nausea once during the study compared with none of those receiving placebo (P = 0.216 for abdominal pain and P = 0.147 for nausea in the between-group analysis). Three octreotide-treated patients developed slight diarrhea and complained about flatulence at least once compared with one in the placebo-treated group, but no between-group differences were present (P = 0.588 for both parameters). Moreover, three women in each group reported the appearance of moderate abdominal swelling once during the study (P = 0.563 in the between-group analysis). None developed side effects, including gallstones.
Anthropometry and fat distribution
No significant difference in baseline parameters existed between the two groups (Table 1). Body weight, BMI, and TAT values decreased significantly and in a comparable manner from baseline to month 1 in the two groups, without any significant change in the other parameters. Octreotide-LAR-treated, but not placebo-treated, patients also significantly reduced body weight, BMI, waist circumference, and SAT values during the 6-month treatment. However, we found no significant difference in the changes in these parameters after each treatment between the two groups (Table 1). In addition, no treatment had a significant effect on waist to hip ratio, TAT, VAT, or the VAT/SAT ratio (Table 1).
|
Steroid hormones and SHBG concentrations at baseline were comparable in the two groups. No treatment significantly modified serum levels of LH, FSH, FAI, 17-OHP, 17-OHPreg, P, DHEA-S, E2, SHBG, or cortisol. T and A levels significantly decreased in the octreotide-LAR-treated, but not in the placebo-treated, group (Table 2). There were no significant differences in the hormonal changes after each treatment between the two groups, although a trend toward greater reductions in T (P = 0.061) and A (P = 0.069) levels was observed in PCOS women treated with octreotide-LAR (Table 2).
|
Fasting and glucose-stimulated values of glucose and insulin (Fig. 2) and the insulin resistance indices (Fig. 3) at baseline were not different in the two groups. Fasting and 60-min stimulated glucose values increased with a borderline significance only in the octreotide-LAR group, whereas glucose values at 120 min from the glucose load significantly decreased only in the placebo group. However, no significant changes occurred in the glucose AUC in either group. Fasting insulin significantly decreased in both groups, but a greater reduction was observed in the octreotide-LAR group than in the placebo group (P = 0.018). The 60-min glucose-stimulated insulin values significantly decreased only in the octreotide-LAR group, but no between-group differences were present (P = 0.332). The insulin AUC significantly decreased in the octreotide-LAR group, whereas only a trend toward a reduction was observed in the placebo-treated group (Fig. 2). Thus, when changes in insulin AUC were compared between the two groups, a significant difference was found (P = 0.038).
|
|
). There were no significant differences in the changes in insulin resistance indices after each treatment between the two groups, apart from a trend toward a greater increment in ISI(composite), with a borderline significance (P = 0.059), after octreotide-LAR administration. GH, IGF, IGFBPs, and ghrelin
No significant between-group differences in baseline GH, IGF, IGFBPs, or ghrelin levels were observed. GH, IGF-I, and total ghrelin levels significantly decreased, and IGFBP-1, IGFBP-2, and IGFBP-3 significantly increased in the group treated with octreotide-LAR, whereas no significant modifications occurred in the placebo group (Table 3). However, compared with placebo, only changes in IGFBP-2, IGFBP-3, and total ghrelin were significantly higher after octreotide-LAR administration (P = 0.042 for IGFBP-2; P = 0.047 for IGFBP-3; P = 0.002 for total ghrelin).
|
Because eight of the 11 PCOS patients who were hirsute at baseline were treated with octreotide-LAR and three were given placebo, the pretreatment hirsutism score was significantly higher in the former than in the latter (P = 0.01). All women had acanthosis nigricans, and both groups had a similar score at baseline. After treatment, the hirsutism and acanthosis nigricans scores significantly decreased only in the octreotide-LAR group (Table 4). The greater effect of octreotide-LAR with respect to placebo was confirmed when the changes that occurred in these two parameters were compared between the two groups (P = 0.004 for hirsutism score; P = 0.008 for acanthosis nigricans score).
|
). Anovulation was present during the first posttreatment menstrual cycle in all women. Conversely, during the last treatment cycle, ovulation occurred in all patients treated with octreotide-LAR, but in only one of those given placebo (Table 4). Overall ovulatory rates were therefore significantly higher during octreotide-LAR than placebo treatment (P < 0.001). The hormonal profile assessed in the first and last treatment cycles of each group is shown in Fig. 4. Both groups had a similar hormonal profile at the first cycle, apart from FSH levels on d 28, which were significantly higher in the octreotide-LAR group than in the placebo group. However, significant differences appeared between the two groups at the last treatment cycle. In particular, P was significantly higher in the octreotide-LAR group than in the placebo group on both d 21 and 28 of the menstrual cycle. Moreover, the octreotide-LAR-treated group showed higher levels of LH and E2 and lower levels of FSH than the placebo group on d 28. The overall profiles of LH, FSH, E2, and P did not significantly change during the first treatment cycle for either group. Conversely, significant increases in E2 (P = 0.049) and P (P = 0.012) and a significant decrease in FSH (P = 0.023) were observed only in the octreotide-LAR group in the last treatment cycle. No significant overall modifications occurred in the LH profile in either group. Compared with the effects of placebo, only changes in E2 and P were persistently significantly higher after octreotide-LAR administration (P = 0.044 for E2; P = 0.009 for P).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As expected, octreotide-LAR reduced circulating GH levels, and this can be explained by its well-known ability to inhibit pituitary GH secretion and GHRH hypothalamic release (38). In contrast, the significant and selective decrease in circulating ghrelin, the most important endogenous GH secretagogue, may represent an additional mechanism to explain the inhibitory effect of octreotide-LAR on the GH/IGF-I system, as reported in previous studies (39). Octreotide-LAR also significantly increased IGFBP-1, IGFBP-2, and IGFBP-3, thus favoring a decrease in free IGF-I levels and, therefore, in IGF-I availability to the target tissues. We speculated that combined reduction of insulin and IGF-I levels may represent the key mechanism responsible for the decrement in androgen levels we found in PCOS patients treated with octreotide-LAR, because no direct effects of somatostatin or its analogs on ovarian steroidogenesis have been demonstrated to date (40). Our data also indirectly support the concept that octreotide-LAR did not affect adrenal steroidogenesis, because no significant changes in cortisol and DHEA-S concentrations occurred during the study. Unexpectedly, we found that octreotide-LAR had no effect on LH levels, a finding that contradicts other clinical studies in PCOS women (12, 13, 15, 16, 18, 19). However, previous studies examined integrated LH concentrations after frequent blood sampling (12, 13) or LH responsiveness to GnRH agonist (16), whereas we performed only single fasting LH measurements; this methodological approach may have limited our ability to assess variations in LH secretion. Furthermore, in contrast with all other published studies, in which normal weight or moderately obese PCOS women were examined, we mainly assessed frankly obese PCOS women with an abdominal phenotype, a condition characterized by lower and sometimes virtually normal LH levels (2).
Octreotide-LAR also had evident clinical effects, because it significantly improved hirsutism and acanthosis nigricans, probably related to the efficacy of octreotide in reducing hyperandrogenemia and in improving insulin resistance and hyperinsulinemia. A novel feature of this study was the strong effect exerted by octreotide-LAR on the ovulatory status of PCOS women, which appeared to be relatively independent from changes in menstrual cyclicity. Our finding that menstrual cycle patterns improved similarly in both octreotide-LAR and placebo groups suggests that weight loss and energy restriction were probably the major factors responsible for this effect. The lack of an increase in P in the last menstrual cycle of the study, found in all but one woman in the placebo group, indicates that anovulation persisted in these subjects. Conversely, all women receiving octreotide-LAR ovulated at the end of treatment despite limited changes in serum gonadotropin levels. The presence of somatostatin receptors in human granulosa-luteal cells and human follicles makes it conceivable that octreotide-LAR affects the ovulatory process through a direct action at the ovarian level (11). This process may particularly involve all mechanisms dependent on regulation of the insulin-IGF system, including IGFBP action. In particular, although the action of IGFBPs in the modulation of ovarian function, such as follicular growth, recruitment, and atresia, and hormone biosynthesis is not well known (5), it is likely that changes in their circulating concentrations and intraovarian functions are important to explain the good performance of octreotide-LAR to increase ovulation in abdominal obese PCOS.
Potential limitations of this study include the single-blind design, which is only partially compensated by the careful criteria of selection and randomization of the patients and by the fact that the investigator assessing the outcomes was blinded to treatment assignment. Therefore, additional, larger, double-blinded trials are needed to confirm the clinical relevance of our findings.
In conclusion, this study shows that long-term administration of octreotide-LAR had several benefits on the clinical, metabolic, and hormonal parameters of dieting abdominal obese PCOS women, and that it was able to restore ovulatory status in otherwise anovulatory patients. This treatment seems to be safe and well tolerated, thus providing the basis for a novel application of octreotide-LAR in the treatment of hormonal dysfunction and infertility in PCOS women.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Deceased, September 2004, in Genova, Italy. 
Abbreviations: A, Androstenedione; AUC, area under the curve; BMI, body mass index; CT, computerized tomography; DHEA-S, dehydroepiandrosterone sulfate; E2, estradiol; FAI, free testosterone; HOMA, homeostasis model assessment; IGFBP, IGF-binding protein; ISI(composite), insulin sensitivity index during the OGTT; OGTT, oral glucose tolerance test; 17-OHP, 17-hydroxyprogesterone; 17-OHPreg, 17-hydroxypregnenolone; P, progesterone; PCOS, polycystic ovary syndrome; QUICKI, quantitative insulin-sensitivity check index; SAT, sc adipose tissue; T, testosterone; TAT, total adipose tissue; VAT, visceral adipose tissue.
Received December 17, 2004.
Accepted April 6, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Codner and H. F. Escobar-Morreale Hyperandrogenism and Polycystic Ovary Syndrome in Women with Type 1 Diabetes Mellitus J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1209 - 1216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gambineri, L. Patton, A. Vaccina, M. Cacciari, A. M. Morselli-Labate, C. Cavazza, U. Pagotto, and R. Pasquali Treatment with Flutamide, Metformin, and Their Combination Added to a Hypocaloric Diet in Overweight-Obese Women with Polycystic Ovary Syndrome: A Randomized, 12-Month, Placebo-Controlled Study J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3970 - 3980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Pasquali and A. Gambineri Insulin-sensitizing agents in polycystic ovary syndrome. Eur. J. Endocrinol., June 1, 2006; 154(6): 763 - 775. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and after combined treatment with hypocaloric diet plus octreotide-LAR or placebo (
). P values refer to the comparison between values at baseline and at the end of the treatment in each group. To convert glucose to millimoles per liter, multiply by 0.056; to convert insulin to picomoles per liter, multiply by 7.17.
