OBSERVATIONS Mones in Order to Adapt the Treatment Pling and Frozen at Ϫ20°C Until the End of Dose

LETTERS

jects were monitored to measure sex hor- by centrifugation immediately after sam- OBSERVATIONS mones in order to adapt the treatment pling and frozen at Ϫ20°C until the end of dose. The study protocol was approved the trial, then all of the samples were by the Henri Mondor Hospital Ethics thawed and the analyses performed in a Androgen Therapy Committee. All of the included subjects single batch. All of the analysts were blind gave written informed consent. to the treatment allocation. Insulin was Improves Insulin Men with low levels of PTT (con- determined by the immunoradiometric Sensitivity and firmed by two measurements) were se- assay method (Medgenix Diagnostics, Decreases Leptin lected from a large occupation-based Fleurus, Belgique), leptin by a commer- population. The inclusion criteria were as cial radioimmunoassay (RIA) (Linco Re- Level in Healthy follows: 1) either PTT Յ3.4 ng/ml [5th search, St. Charles, MO), follicle- Adult Men With Low percentile value of PTT distribution in the stimulating and luteinizing hormone by Plasma Total 1,718 men of the TELECOM Study (7)] the Automated Chemiluminescence Sys- from 1985 to 1987 and Ͻ4.0 ng/ml (13th tem 180 (Ciba Corning), and androgens Testosterone percentile value) from 1992 to 1993 (3) and estradiol by RIA (7). The only missing or PTT Ͻ4.0 ng/ml from 1992 to 1993 datum was one 2-h plasma insulin mea- A 3-month randomized placebo- and Ͻ4.0 ng/ml a few days before inclu- surement at 3 months in a subject treated controlled trial sion; 2) no history of vascular thrombosis by DHT. or ischemic heart disease; 3) no treatment The primary end points to assess by androgens, anti-androgens, and an- insulin sensitivity were fasting plasma tidiabetic or antithrombotic drugs; 4) insulin–to–fasting plasma glucose ratio n men, an association between lower normal values of plasma prolactin, estra- and homeostasis model assessment plasma total testosterone (PTT) and in- diol, and thyroxin; 5) no current prostatic (HOMA) index. Plasma leptin, 2-h I sulin resistance has been found in disease and a normal PSA value. A total of plasma glucose and insulin, and blood cross-sectional studies (1,2) and in one 18 healthy men with stable low plasma pressure were taken as secondary criteria. nested case-control study (3) without any androgens (Table 1) and a range of PTT Treatment tolerance was assessed by in- possible conclusion in terms of causality from 1.4 to 3.7 ng/ml at baseline were terview, by prostatic examination, and by or direction of the relationship. Indeed, to included. PSA, as well as by weight, electrocardio- obtain such information, randomized The 18 selected men were randomly gram (ECG), lipid, hemoglobin, hemato- controlled trials are needed. Until now, assigned to one of three treatment groups: crit, fibrinolysis markers, and hepatic only one clinical trial has suggested that testosterone, DHT, or placebo. The ran- enzyme variations. testosterone therapy improves insulin domization code was known only to the A sample size of 36 subjects was sensitivity in obese men (4). Cross- study manager. Treatment was a gel ad- needed to detect a difference of 5 mg/dl sectional studies concerning leptin regu- ministered every morning by percutane- for the decrease of fasting plasma glucose, lation by androgens have provided no ous route. The daily dose during the first assuming SD ϭ 5 mg/dl, using a two- definitive conclusions as to whether the weeks was 125 mg for the testosterone tailed Student’s t test with ␣ϭ0.05 and negative association between androgens and 35 mg for the DHT treatment groups. ␤ϭ0.20. However, we could not reach and leptin level is independent (5) or de- The adaptation of treatment doses be- that number, and the recruitment was pendent (6). This randomized controlled tween days 10 and 20 aimed at obtaining closed after having included 18 subjects. trial was designed to assess the role of an- a trough level of PTT between 4 and 10 To evaluate the treatment effect, the dif- drogens on insulin sensitivity and leptin ng/ml for the testosterone group and a ference between the values at entry and at regulation in healthy adult men. trough level of plasma DHT between 4 the end of the treatment period was cal- This study was a randomized, dou- and 10 ng/ml for the DHT group. To culated for each subject, and then the ble-blind, unicentric, controlled, clinical maintain the double blinding, the study Kruskal-Wallis nonparametric test was trial. Three treatments (testosterone, di- manager also sometimes changed the used. When statistical significance (P Յ hydrotestosterone [DHT], and placebo) dose of placebo. The subjects were asked 0.05) was reached for any overall three- were compared in parallel groups during not to change their dietary and physical group comparison, two-by-two compari- a 3-month period. All of the examinations activity. Compliance to treatment was as- sons were performed using the Bonferroni were performed by only two physicians, sessed by interview and by measuring sex test to correct for multiple comparisons. using a standardized protocol. Blood was hormones and gonadotropins at the end At baseline, the three treatment drawn between 8:00 A.M. and 9:30 A.M. of the trial. groups were similar with respect to age, after an overnight fast to determine fasting Plasma glucose, total cholesterol, BMI, waist-to-hip ratio (WHR), blood plasma glucose, insulin, leptin, sex hor- HDL cholesterol, triglycerides, apoli- pressure, plasma glucose, lipids, insulin, mones, lipids, coagulation and fibrinoly- poprotein (apo)-A1, apoB, hepatic en- leptin, androgens, and sex hormone– sis parameters, hepatic enzymes, and zymes, and blood cell count were assayed binding globulin, as well as hemoglobin, prostate-specific antigen (PSA) and blood on the same day of venipuncture. PSA and hematocrit, coagulation, and fibrinolysis cell count. Then, a standard 75-g oral glu- fibrinolysis markers were measured parameters (data not shown). At the end cose tolerance test and a digital rectal ex- within 3 days after venipuncture. For hor- of the trial, a significant difference was amination were performed. In addition, mone measurements at baseline and at shown for the variation of fasting plasma between days 10 and 20, all of the sub- the end of the trial, plasma was separated insulin (P Ͻ 0.05), fasting plasma insu-

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2149 Letters

Table 1—Baseline characteristics and variations in the three treatment groups (after minus before)

Testosterone DHT Placebo P n 666 Age (years) 52.8 Ϯ 4.2 51.2 Ϯ 3.9 55.4 Ϯ 3.6 0.80 BMI (kg/m2) 29.9 Ϯ 0.9 27.8 Ϯ 0.9 28.0 Ϯ 1.1 0.84 WHR 0.95 Ϯ 0.02 0.96 Ϯ 0.02 0.96 Ϯ 0.03 0.99 Systolic blood pressure (mmHg) 152 Ϯ 5 143 Ϯ 7 126 Ϯ 8 0.08 Diastolic blood pressure (mmHg) 93 Ϯ 488Ϯ 280Ϯ 5 0.17 Fasting plasma glucose (mg/dl) 101 Ϯ 597Ϯ 299Ϯ 4 0.97 Total cholesterol (mg/dl) 212 Ϯ 14 228 Ϯ 11 221 Ϯ 14 0.70 HDL cholesterol (mg/dl) 45 Ϯ 444Ϯ 442Ϯ 6 0.81 Triglycerides (mg/dl) 126 Ϯ 20 142 Ϯ 18 123 Ϯ 18 0.64 Fasting plasma insulin (␮U/ml) 14 Ϯ 418Ϯ 413Ϯ 3 0.52 Leptin (ng/ml) 10.1 Ϯ 4.5 6.4 Ϯ 1.3 6.2 Ϯ 1.5 0.81 Plasma total testosterone (ng/ml) 2.4 Ϯ 0.1 2.9 Ϯ 0.3 2.7 Ϯ 0.3 0.20 Plasma bioavailable testosterone (ng/ml) 0.6 Ϯ 0.1 0.8 Ϯ 0.1 0.6 Ϯ 0.1 0.06 Plasma SHBG (nmol/l) 16.5 Ϯ 1.9 16.9 Ϯ 2.6 21.0 Ϯ 3.3 0.32 ⌬ Fasting plasma glucose (mg/dl) 4 Ϯ 3 Ϫ1 Ϯ 33Ϯ 4 0.42 ⌬ Fasting plasma insulin (␮U/ml) Ϫ0.8 Ϯ 2.0 Ϫ6.2 Ϯ 2.2 2.7 Ϯ 1.6 0.02* ⌬ Fasting plasma insulin/fasting plasma glucose Ϫ0.23 Ϯ 0.32 Ϫ1.09 Ϯ 0.29 0.43 Ϯ 0.35 0.003* ⌬ HOMA index Ϫ0.09 Ϯ 0.53 Ϫ1.54 Ϯ 0.69 0.73 Ϯ 0.39 0.012* ⌬ Leptin (ng/ml) Ϫ1.2 Ϯ 1.7 Ϫ1.8 Ϯ 0.6 0.4 Ϯ 0.4 0.05† ⌬ Total cholesterol (mg/dl) Ϫ7 Ϯ 7 Ϫ4 Ϯ 5 Ϫ12 Ϯ 5 0.66 ⌬ HDL cholesterol (mg/dl) Ϫ1 Ϯ 2 Ϫ5 Ϯ 11Ϯ 2 0.09 ⌬ Triglycerides (mg/dl) 6 Ϯ 18 11 Ϯ 12 19 Ϯ 24 0.78 ⌬ Systolic blood pressure (mmHg) 4 Ϯ 5 Ϫ3 Ϯ 521Ϯ 7 0.052† ⌬ Diastolic blood pressure (mmHg) 4 Ϯ 55Ϯ 514Ϯ 3 0.22 ⌬ Weight (kg) 3.3 Ϯ 1.1 1.4 Ϯ 1.0 Ϫ0.3 Ϯ 1.0 0.09 Data are means Ϯ SEM. *P Ͻ 0.01 for DHT vs. placebo; †P Ͻ 0.05 for DHT vs. placebo. lin–to–fasting plasma glucose ratio (P Ͻ three groups for lipids (Table 1), PSA, he- increased) the response to DHT treatment 0.01), and HOMA index (P Ͻ 0.05), patic enzymes, coagulation, and fibrino- and blood pressure, probably explaining which all decreased under androgens. lysis parameters, but hemoglobin and the nearly significant improvement of sys- The two-by-two comparisons showed a hematocrit increased under androgens tolic blood pressure under androgens by a significant improvement only for DHT (P Ͻ 0.05 and P Ͻ 0.01, respectively), regression to the mean phenomenon in compared with placebo (P Ͻ 0.01 for all mainly with testosterone (data not shown). the placebo group. On the contrary, the of these indexes of insulin sensitivity). No This randomized, controlled, double- parallel decrease in fasting plasma insulin significant differences were observed for blind trial provides evidence that in and leptin and the improvement in insu- 2-h plasma glucose and insulin among the healthy men, androgen treatment, partic- lin sensitivity under androgens appear three groups (data not shown), whereas ularly DHT, improves insulin sensitivity very consistent. Our study may appear plasma leptin significantly decreased un- and decreases plasma leptin level without limited because of the sample size (half of der androgen treatment (P Ͻ 0.05), notable side effects. The three treatment that planned), enjoining the use of con- mainly with DHT (P Ͻ 0.05 for DHT vs. groups were quite identical at baseline servative nonparametric tests, and caus- placebo). Systolic blood pressure in- concerning glucose tolerance status. In ing the final statistical analysis to be creased in the placebo group (P ϭ 0.052) the placebo group, one subject was dia- equivalent to a planned intermediary (Table 1). betic according to 2-h plasma glucose analysis. Indeed, to have confirmed the a The only serious event was the dis- (227 mg/dl, with fasting plasma glucose at priori hypotheses in these conditions of covery of a prostatic nodular hyperplasia, 85 mg/dl), and none had impaired glu- weak statistical power emphasizes the ef- benign at biopsy, in a subject treated by cose tolerance (IGT) or impaired fasting fect of androgens, mainly DHT, to im- testosterone. A trend for an increase in glucose (IFG). In the DHT group, one prove insulin sensitivity and to decrease weight was observed under androgen subject had IGT, and none had diabetes leptin concentrations in healthy men with treatment (P ϭ 0.09), mainly with testos- or IFG. In the testosterone group, all of low PTT. Very few side effects were ob- terone (Table 1), without any modifica- the subjects had normal glucose toler- served, including a tendency for weight tion of waist circumference and WHR ance. The primary differences at baseline increase and an increase in hemoglobin (data not shown). No change was ob- concerned bioavailable testosterone with and hematocrit, although these were re- served on the ECG recordings. No signif- a trend for a higher level in the DHT versible a few months later (data not icant difference was shown among the group, which should have blunted (not shown), thus indicating good patient

2150 DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 Letters compliance to the allocated treatment, onstrates that androgens improve insulin Eschwege E: Association between plasma also confirmed by hormone measurements. sensitivity and decrease leptin levels in total testosterone and cardiovascular risk From the literature, we can speculate adult men. We recruited healthy subjects factors in healthy adult men: the Telecom that the effect of androgens on insulin ranked in the lowest 10 percentiles of the Study. J Clin Endocrinol Metab 82:682– sensitivity could be caused by changes in PTT distribution from a large occupation- 685, 1997 4. Marin P, Holma¨ng S, Gustafsson C, Jo¨n- body composition and fat mass distribu- based population by systematically mea- sson L, Kvist H, Elander A, Eldh J, Sjo¨s- tion. Indeed, androgens are known to in- suring PTT. Therefore, these data can tro¨m L, Holm G, Bjo¨rntorp P: Androgen crease fat-free mass and muscle size and most likely be extrapolated to healthy treatment of abdominally obese men. to decrease visceral fat mass (8) by inhib- men in the first decile of the PTT distri- Obes Res 1:245–251, 1993 iting lipoprotein lipase activity, therefore bution. The pathways through which an- 5. Luukkaa V, Pesonen U, Huhtaniemi I, inhibiting triglyceride uptake and accel- drogens exert their inhibiting effects on Lehtonen A, Tilvis R, Tuomilehto J, Koulu erating triglyceride release from abdomi- insulin and leptin in humans deserve fur- M, Huupponen R: Inverse correlation be- nal adipose tissue (9). In turn, a decrease ther fundamental research. In parallel, as tween serum testosterone and leptin in in abdominal fat mass may induce an im- low levels of testosterone are predictive of men. J Clin Endocrinol Metab 83:3243– provement in insulin sensitivity via a re- the development of insulin resistance and 3246, 1998 6. Haffner SM, Miettinen H, Karhapaa P, duction in circulating free fatty acids (10). type 2 diabetes (14), and as type 2 dia- Mykkanen L, Laakso M: Leptin concen- In our study, no significant variation of betic patients are known to have a lower trations, sex hormones, and cortisol in waist circumference was found under an- level of PTT than nondiabetic men (15), nondiabetic men. J Clin Endocrinol Metab drogens, but this measurement was prob- larger studies on androgen treatment in 82:1807–1809, 1997 ably too imprecise to detect changes. In insulin-resistant nondiabetic subjects and 7. Simon D, Preziosi P, Barrett-Connor E, the only previous controlled clinical trial in type 2 diabetic patients are necessary. Roger M, Saint-Paul M, Nahoul K, Papoz having compared testosterone, DHT, and L: The influence of aging on plasma sex 1,2 placebo gel treatments, 31 abdominally DOMINIQUE SIMON, MD, PHD hormones in men: the Telecom Study. 1,2 Am J Epidemiol 135:783–791, 1992 obese subjects with a moderately low PTT MARIE-ALINE CHARLES, MD 3 8. Marin P, Holma¨ng S, Jo¨nsson L, Sjo¨stro¨m level (mean 4.5 ng/ml) were treated dur- NAJIBA LAHLOU, MD 4 L, Kvist H, Holm G, Lindstedt G, Bjo¨rn- ing 9 months (4). In that study, the tes- KHALIL NAHOUL, MD 5 torp P: The effects of testosterone treatment tosterone group had a significant decrease JEAN-MICHEL OPPERT, MD, PHD 6 on body composition and metabolism in in visceral fat mass seen by computerized MICHE`LE GOUAULT-HEILMANN, MD middle-aged obese men. Int J Obes Relat 7 tomography, and a marked augmentation NICOLE LEMORT, BSC Metab Disord 16:991–997, 1992 1 of glucose disposal rate was observed with NADINE THIBULT, BSC 9. Marin P, Oden B, Bjo¨rntorp P: Assimila- 8 euglycemic-hyperinsulinemic clamp, EVELYNE JOUBERT, MD tion and mobilization of triglycerides in 1 whereas plasma triglycerides and total BEVERLEY BALKAU, PHD subcutaneous abdominal and femoral ad- 1 cholesterol had decreased. Leptin was not EVELINE ESCHWEGE, MD ipose tissue in vivo in men: effects of androgens. J Clin Endocrinol Metab 80:239– measured. No significant improvement 1 was shown with DHT treatment. This From INSERM U 21/U 258, Paris XI University, 243, 1995 Villejuif, France; the 2Department of Endocrinol- 10. Boden G: Role of fatty acids in the patho- striking contrast with our study concern- ogy, Henri Mondor Hospital, Cre´teil, France; the genesis of insulin resistance and NIDDM. ing the respective effects of testosterone 3Department of Hormonal Biochemistry, Saint- Diabetes 46:3–10, 1997 and DHT could be explained by the Vincent-de-Paul Hospital, Paris, France; 4Claude 5 11. Sinha MK, Caro JF: Clinical aspects of lep- higher PTT level at baseline and mostly by Le´vy Laboratory, Ivry, France; the Department of tin. Vitam Horm 54:1–30, 1998 Nutrition, Hotel-Dieu Hospital, Paris, France; the an undertreatment in the DHT group in 6Department of Biological Hematology, Henri Mon- 12. Kosaki A, Yamada K, Kuzuya H: Reduced the trial by Marin et al. (4). We used larger dor Hospital, France; 7INSERM U 468, Paris XII expression of the leptin gene (ob) by cat- doses of DHT and adapted the doses of University, Cre´teil, France; and 8Besins-Iscovesco echolamine through a G(S) protein-cou- testosterone and DHT after 2 weeks, Laboratories, Paris, France. pled pathway in 3T3–L1 adipocytes. when necessary, according to the circulat- Address correspondence to Dr. Dominique Si- Diabetes 45:1744–1749, 1996 mon, INSERM U-258, 16 Ave. Paul Vaillant Coutu- ing androgen level, whereas in the trial by 13. Wabitsch M, Blum WF, Muche R, Braun rier, 94807 Villejuif Cedex, France. E-mail: M, Hube F, Rascher W, Heinze E, Teller Marin et al. (4) no monitoring of sex hor- [email protected]. W, Hauner H: Contribution of androgens mones level was performed. to the gender difference in leptin produc- The decrease in plasma leptin con- ●●●●●●●●●●●●●●●●●●●●●●● tion in obese children and adolescents. centration is also probably explained by References J Clin Invest 100:808–813, 1997 the supposed reduction in adipose tissue 1. Simon D, Preziosi P, Barrett-Connor E, 14. Stellato RK, Feldman HA, Hamdy O, Hor- mass (11), but the influence of androgens Roger M, Saint-Paul M, Nahoul K, Papoz ton ES, McKinlay JB: Testosterone, sex on leptin could also be mediated by a L: Interrelation between plasma testoster- hormone-binding globulin, and the de- stimulation of the splanchnic ␤-adreno- one and plasma insulin in healthy adult velopment of type 2 diabetes in middle- ceptors (12) or by a direct suppressive ef- men: the Telecom Study. Diabetologia 35: aged men: prospective results from the fect on ob gene expression (13). 173–177, 1992 Massachusetts Male Aging study. Diabetes 2. Haffner SM, Karhapaa P, Mykkanen L, Care 23:490–494, 2000 Nevertheless, our data clearly demon- Laakso M: Insulin resistance, body fat dis- 15. Goodman-Gruen D, Barrett-Connor E: strate the role of androgens to decrease tribution, and sex hormones in men. Di- Sex differences in the association of en- leptin levels in healthy men, as previously abetes 43:212–219, 1994 dogenous sex hormone levels and glucose suggested (5). 3. Simon D, Charles MA, Nahoul K, Orssaud tolerance status in older men and women. In conclusion, this clinical trial dem- G, Kremski J, Hully V, Joubert E, Papoz L, Diabetes Care 23:912–918, 2000

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2151 Letters

Lipid Lowering Does A total of 18 diabetic patients were the recent finding that glycated LDL from enrolled in the study. Five patients diabetic patients reduces endothelial cell Not Improve dropped out for logistical reasons, none nitric oxide synthesis and bioactivity (5). Endothelial Function because of adverse reactions. Of the 13 Therefore, in our study, we speculate that in Subjects With patients who completed the study, 7 were glycated LDL may have interfered with randomized to the simvastatin group (all endothelium-dependent vasorelaxation Poorly Controlled women; 5 African-Americans and 2 Lati- in both groups. Finally, it could not be Diabetes nos) and 6 to the placebo group (5 wom- excluded that the long duration of diabe- en; 5 African-Americans and 1 Latino). tes may have induced diabetic neuropa- The subjects were elderly (61 Ϯ 6 vs. thy, which may affect flow response to utaneous microangiopathy is sus- 60 Ϯ 5 years of age, simvastatin versus ACh (6). However, this possibility is less pected to play a role in the patho- placebo, respectively), obese (BMI: 33 Ϯ likely because we systematically excluded C genesis of diabetic foot ulcers (1). 5 vs. 32 Ϯ 5 kg/m2), and had a relatively patients with neuropathy on physical ex- Because microcirculatory flow is regu- long duration of diabetes (11 Ϯ 8.5 vs. amination. lated in part by the endothelium and prior 6.7 Ϯ 3.8 years). Diabetes was poorly In summary, LDL lowering does not Ϯ Ϯ studies of the coronary microcirculation controlled (HbA1c 9.3 1.7 vs. 9.1 appear to improve endothelial function showed significant improvement in endo- 2.5%) and LDL cholesterol levels were el- in the microcirculation without ade- thelium-dependent vasodilatory re- evated (4.6 Ϯ 0.5 vs. 4.2 Ϯ 0.6 mmol/dl) quate glycemic control. Further studies sponses with lipid-lowering therapy (2), in both groups. The two groups were not are needed to assess the benefit of opti- we examined the effect of lipid lowering significantly different in any of the above mal control of diabetes in conjunction on endothelial function in cutaneous parameters. Endothelium-dependent and with lipid lowering on endothelium- microcirculation in patients with type 2 -independent responses (ACh: 0.7 Ϯ 0.6 dependent microcirculation vasomotion. diabetes. vs. 0.8 Ϯ 0.3; SNP: 1.6 Ϯ 1.0 vs. 1.5 Ϯ 1 Patients aged 21–80 years with dia- 1.0) were also similar at baseline. LDL JACQUES MANSOURATI, MD 2 betes and LDL cholesterol Ͼ3.4 mmol/l cholesterol was significantly reduced at 3 LISA G. NEWMAN, MD 2 were randomized in a double-blind fash- months in the simvastatin group (from SHEILA H. ROMAN, MD, MPH 2 ion to treatment with either simvastatin 4.6 Ϯ 0.5 to 2.8 Ϯ 0.6 mmol/dl, P Ͻ ARLENE TRAVIS, RN 2 40 mg daily or placebo and followed for 3 0.01) but not in the placebo group (from MOHAMMAD RAFEY, MD 3 months after randomization. All patients 4.2 Ϯ 0.6 to 3.8 Ϯ 0.6 mmol/dl, P ϭ NS). ROBERT A. PHILLIPS, MD, PHD, FACC received dietary counseling with regard to There was no significant change in HbA1c From the 1Department of Cardiology, University lowering LDL, but no attempt was made over the course of the study. Endothelium- 2 to alter glycemic control. The study was Hospital of Brest, Brest, France; the Hypertension dependent vasodilatation remained un- and Cardiac Health Programs and the Endocrine Di- approved by the Institutional Review changed in both the simvastatin and the vision, Mount Sinai School of Medicine, New York, Board of the Mount Sinai School of Med- placebo groups (⌬ACh: Ϫ0.1 Ϯ 1 vs. New York; and the 3Department of Medicine, Lenox icine, and informed consent was obtained 0.3 Ϯ 1.1, P ϭ NS; ⌬SNP: Ϫ0.6 Ϯ 1.4 vs. Hill Hospital, New York, New York. before enrollment. 0.2 Ϯ 1.4, P ϭ NS). Address correspondence to Robert A. Phillips, MD, PhD, FACC, Director, Department of Medicine, Cutaneous microcirculatory flow The main result of this study is the Lenox Hill Hospital, 100 East 77th St., New York, from the same site of the dorsum of the lack of beneficial effect of lipid lowering NY 10021. E-mail: [email protected]. foot was measured with the Periflux Sys- on skin microcirculation vasomotion in L.G.N. holds stock in Merck. R.A.P. currently tem PF3 (Perimed, Ja¨rfa¨lla, Sweden) at ev- this population of poorly controlled dia- does limited consulting for and has a research grant ery visit in all patients. The flow response betic patients. Neither endothelium- with Merck. to heating was recorded at a skin temper- dependent nor -independent responses ature of 32°C and then at 44°C (the skin were significantly improved in the treated Acknowledgments— This work was sup- temperature at which maximal flow is group, although total and LDL cholesterol ported in part by a grant (5MO1-RR-00071) achieved) (3). Endothelium-dependent were significantly lowered by simvastatin. for the Mount Sinai General Clinical Research and -independent microcirculatory re- A similar lack of improvement in endo- Center from the National Center for Research sponses were recorded using acetylcho- thelial function in a large conduit artery Resources, National Institutes of Health, and line (ACh) and sodium nitroprusside (brachial artery) was recently reported in by a Medical School Grant from Merck. (SNP) iontophoresis, respectively (Peri- type 2 diabetic patients treated with sim- Iont Micropharmacology System PF380; vastatin (4). Our results extend this find- ●●●●●●●●●●●●●●●●●●●●●●● Perimed). All flow measurements were ing to the microcirculation. This negative References performed with subjects in a fasting state. result may be explained by several factors. 1. Flynn MD, Tooke JE: Aetiology of dia- Normalized values of flow (the ratio of First, a longer follow-up may be necessary betic foot ulceration: a role for the mi- flow in response to ACh or SNP to the to demonstrate significant improvement crocirculation? Diabet Med 8:320–329, flow reaction to heating to 44°C) were in cutaneous microcirculation in the pop- 1992 2. Anderson TJ, Meredith IT, Yeung AC, Frei used for comparisons between the ulation of patients with a long duration of B, Selwyn AP, Ganz P: The effect of cho- groups. When appropriate, Mann- diabetes. Second, glycemic control may lesterol-lowering and antioxidant therapy Whitney U and Wilcoxon tests were used be needed to achieve beneficial effects of onendothelium-dependentcoronaryvaso- for group comparisons. P Ͻ 0.05 was LDL lowering on endothelial function. motion. N Engl J Med 332:488–493, 1995 considered significant. This hypothesis is further supported by 3. Rendell M, Bamisedun O: Diabetic cuta-

2152 DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 Letters

neous microangiopathy. Am J Med 93: 611–618, 1992 4. Sheu WH-H, Juang B-L, Chen Y-T, Lee W-J: Endothelial dysfunction is not re- versed by simvastatin treatment in type 2 diabetic patients with hypercholes- terolemia (Letter). Diabetes Care 22: 1224–1225, 1999 5. Posch K, Simecek S, Wascher TC, Ju¨ rgens G, Baumgartner-Parzer S, Kostner GM, Graier WF: Glycated low-density lipoprotein attenuates shear stress–induced nitric oxide synthesis by inhibition of shear stress–activated L-arginine uptake in endothelial cells. Diabetes 48:1331–1337, 1999 6. Arora S, Smakowski P, Frykberg RG, Simeone LR, Freeman R, LoGerfo FW, Veves A: Differences in foot and forearm skin microcirculation in diabetic patients with and without neuropathy. Diabetes Care 21:1339–1344, 1998

Effects of Felted Foam on Plantar Figure 1—Mean peak plantar pressures of 28 diabetic patients with neuropathic foot ulcers up to Pressures in the a Wagner grade 2. D, day; FF, application of felted foam. Treatment of Neuropathic Diabetic tients with neuroischemic or ischemic di- mean of all pressures measured in each Foot Ulcers abetic foot ulcers were excluded from the walk (three steps per walk and two walks study. per test) was calculated for each patient. The felted foam (i.e., a combination of Differences between the plantar pressures t is generally accepted that (besides in- 0.635-cm thick rubber foam with a at each day were compared by analysis of fection control and revascularization, 0.158-cm layer of felt adhered, fixed by variance; P Ͻ 0.05 was considered signif- I when necessary) pressure relief is the rubber glue) was measured exactly to fit icant. most important measure in the treatment the plantar aspect of the foot. Using a scal- The mean ulcer area in the patients of diabetic foot ulcers. The use of felted pel, an aperture was cut from the felted studied was 159.9 Ϯ 102.6 mm2.Bythe foam dressings is a promising but not yet foam at the exact location of the ulcer, application of the felted foam dressing, well-standardized technique for the treat- allowing clear visualization of the ulcer. the mean peak plantar pressures at the ment of neuropathic diabetic foot ulcers Gauze was then wrapped around the foot ulceration site was significantly reduced, and may have some advantage over total and the felted foam pad to secure the pad. from 297.3 Ϯ 120.0 kPa before to 90.3 Ϯ contact casting (1–4). We aimed to assess The wound was covered with a saline- 38.2 kPa immediately after the applica- the effects of felted foam on plantar pres- soaked sponge, which was cut according tion (P Ͻ 0.0001). In the following pe- sure reduction during the therapy of neu- to the size of the ulcer and changed every riod, over at least 4 days with the ropathic foot ulcers and to define the day. The felted foam was kept dry at all mounted felted foam dressing, the plantar optimal time course for renewal of the times. A compress was placed over the load in the area of interest significantly felted foam according to the plantar pres- wet sponge and fixed with Peha-haft. The increased, from 93.6 Ϯ 39.6 kPa the day sure. Using felted foam dressings, plantar felted foam dressing was exclusively used after the application to 222.6 Ϯ 97.8 kPa pressure reduction and wound healing for plantar ulcerations. Foot pressure was at day 4 (P Ͻ 0.0001, Fig. 1). On days 2 was determined in 9 type 1 and 19 type 2 measured using the FastScan system, as and 3, the plantar pressures at the ulcer- diabetic patients (15 men and 13 women, described elsewhere (5,6). Recordings of ation site varied, from 113.8 Ϯ 47.6 to aged 61.0 Ϯ 13.6 years) with neuropathic the plantar pressures were first done with- 137.5 Ϯ 63.9 kPa, without significant foot ulcers up to a Wagner grade 2. Phys- out any dressings and then with the at- day-to-day changes. However, from day 3 ical examination included the inspection tached felted foam dressing every day to day 4 there was a clear-cut increase of of the foot and the palpation of the pe- from the beginning of the study (day 0) to the plantar pressure in the area of interest, ripheral pulses. Peripheral diabetic neu- day 4. The plantar pressures were mea- from 137.5 Ϯ 63.9 to 222.6 Ϯ 97.8 kPa ropathy was evaluated by measuring the sured by selecting the area of interest un- (P ϭ 0.0001). vibration perception threshold with the der the foot, within an area of 25 ϫ 25 Because the relief of the plantar load calibrated Rydell-Seiffer tuning fork. Pa- mm2 in the center of the ulceration. A at the ulceration site is one of the most

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2153 Letters important factors in the outcome of neu- 979–982, 1996 tivates DNA gyrase. As the antibiotic is ropathic foot ulcerations, the application 4. Stess RM, Jensen SR, Mirmiran R: The role chelated with 99 Tc, the area of bacterial of the felted foam appears to be useful to of dynamic plantar pressures in diabetic infection should be identifiable during reduce the peak plantar pressures at the foot ulcers. Diabetes Care 20:855–858, imaging (4). site of ulceration. We have shown that the 1997 A total of 16 diabetic patients with a 5. Ahroni JH, Boyko EJ, Forsberg R: Reliabil- pressure relief from attaching the felted ity of F-scan in-shoe measurements of hot swollen foot were studied prospec- foam dressing at the ulceration site lasts plantar pressure. Foot Ankle Int 19:668– tively using plain radiographs, MRI, Ga- up to 3 days after its application. Taking 673, 1998 lium-67, and Tc99 m Infecton. The MRI into account the distinct increase in plan- 6. Pitei DL, Lord M, Foster A, Wilson S, and plain radiographs were reported in- tar pressure on the fourth day, we recom- Watkins PJ, Edmonds ME: Plantar pres- dependently, blinded from the radionu- mend changing the felted foam each 3rd sures are elevated in the neuroischemic clide imaging, and vice versa. The to 4th day. Interestingly, we did not ob- and the neuropathic diabetic foot. Diabe- definitive diagnosis was established by serve the development of callosity at the tes Care 22:1966–1970, 1999 findings at surgery, microbiological re- ulceration site, which underlines the effi- 7. Ritz G, Kushner D, Friedman S: A suc- sults, or definitive imaging (e.g., plain ra- cacy of the felted foam technique for pres- cessful technique for the treatment of di- diograph to detect fractures). sure relief. In contrast to other methods abetic neurotrophic ulcers. J Am Podiatr In our prospective study, four (25%) Med Assoc 82:479–481, 1992 for pressure relief, such as total contact 8. Armstrong DG, Nguyen HC, Lavery LA, patients had osteomyelitis, three (19%) cast, felted foam also enables daily dress- van Schie CH, Boulton AJ, Harkless LB: had neuropathic fractures, and nine ing changes and can be used in patients Off-loading the diabetic foot wound: a (56%) had soft tissue swelling. MRI accu- with smaller infections (7,8). randomized clinical trial. Diabetes Care rately diagnosed all of the four cases with We conclude that in diabetic patients 24:1019–1022, 2001 osteomyelitis. Tc99 m Infecton was only with neuropathic foot ulcerations, the able to localize infection to bone in one of felted foam technique effectively reduces the four cases with osteomyelitis. In the the pressure load at the ulceration site. rest of the cases, Infecton could not dif- This pressure relief persists for 3 days, ferentiate whether infection was confined and we therefore recommend renewing Diagnosing to soft tissue or bone. Plain radiographs the felted foam after each 3–4 days of Osteomyelitis in were able to diagnose two of the four cases treatment. Patients With with osteomyelitis. MRI correctly diagnosed fractures in all of the three patients Diabetic Neuropathic who had evidence of fractures on plain STEFAN ZIMNY, MD Osteoarthropathy radiograph. Infecton and Gallium scans BERNADETTE REINSCH, MD reported bone or soft tissue as infected in HELMUT SCHATZ, MD all of the three cases. Therefore, the nu- MARTIN PFOHL, MD pproximately 15% of diabetic peo- clear medicine scans can falsely indicate ple will develop foot ulcers during infection or inflammation in the presence From Berufsgenossenschaftliche Kliniken Berg- A their lifetime, and early detection of of fractures. mannsheil Universita¨tsklinik, Ruhr-Universita¨t Bo- osteomyelitis is crucial to the manage- Radionuclide imaging is not reliable chum, Medizinische Klinik und Poliklinik, Bochum, ment of diabetic foot ulcers (1). Differen- to differentiate among infection, inflam- Germany. tiating osteomyelitis from neuropathic Address correspondence to Stefan Zimny, Beruf- mation around fractures, or Charcot joint, sgenossenschaftliche Kliniken Bergmannsheil Uni- osteoarthropathy is clinically difficult, as even when infection is correctly identi- versita¨tsklinik, Ruhr-Universita¨t Bochum, Medizin- the symptoms and signs are nonspecific. fied. The limited spatial resolution in the ische Klinik und Poliklinik, Buerkle-de-la-Camp- These patients all present with hot and forefoot does not allow accurate discrim- Platz 1, D-44789 Bochum, Germany. E-mail: erythematous feet. At presentation, there [email protected]. ination between soft tissue infection and is often no change on plain radiographs osteomyelitis. Plain radiograph was es- (2). Many of the imaging findings are also sential in the initial work-up, as hot spots ●●●●●●●●●●●●●●●●●●●●●●● similar, especially in rapidly progressing, on Infecton scans and Gallium 67 scans References noninfected neuro-osteoarthropathy. The can indicate fracture rather than infection. 1. Caputo GM, Cavanagh PR, Ulbrecht JS, most reliable method of establishing in- MRI is the imaging of choice to distinguish Gibbons GW, Karchmer AW: Assessment fection is to analyze microbiological sam- osteomyelitis from other conditions, such and management of foot disease in pa- ples of the lesion. However, this is not as cellulitis and neuropathic osteoarthropa- tients with diabetes. N Engl J Med 331: always practical and may lead to seeding thy in diabetic patients with a hot swollen 854–860, 1994 of the infection or damage to the area bi- foot. Infecton scans are helpful when used 2. Lavery LA, Vela SA, Lavery DC, Quebe- opsied. Magnetic resonance imaging in conjunction with MRI to localize an in- deaux TL: Reducing dynamic foot pres- (MRI) is a useful method of tissue local- fected area before surgery but cannot be sures in high-risk diabetic subjects with ization and is currently the most sensitive used independently as a diagnostic tool in foot ulcerations: a comparison of treatments. Diabetes Care 19:818–821, 1996 method to detect osteomyelitis (3). the assessment of a hot swollen diabetic 3. Murray HJ, Young MJ, Hollis S, Boulton Technetium (Tc)-99 m Infecton con- foot. AJ: The association between callus forma- sists of ciprofloxacin linked to Tc99 m. 1 tion, high pressures and neuropathy in di- The antibiotic is taken up and bound spe- DEVASENAN DEVENDRA, MB, CHB, MRCP 2 abetic foot ulceration. Diabet Med 13: cifically by living bacteria, where it inac- KIM FARMER, MB, CHB

2154 DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 Letters

2 GRAHAM BRUCE, BSC mitted for acute pancreatitis of unknown Abdominal tomodensitometry was 2 PHILIP HUGHES, FRCR etiology associated with diabetes. His normal, and there was no alteration of 2 GILL VIVIAN, MRCP, FRCR mother had type 2 diabetes. The patient’s exocrine pancreatic function. There were 2 BEVERLEY ANNE MILLWARD, MSC, MD, FRCP3 BMI was 31 kg/m . After the onset of no islet cell antibodies, GAD autoantibod- acute pancreatitis, the patient discontin- ies, or anti-insulin and anti-insulin– From 1University Medicine, University of Plymouth, 2 ued his antidiabetic treatment. Five years receptor antibodies. Insulin secretion was Plymouth, U.K.; the Departments of Podiatry and later, he was admitted in a hyperosmolar preserved (basal C-peptide before gluca- Radiology, Derriford Hospital, Plymouth, U.K.; and the 3Molecular Medicine Research Group, Univer- state (fasting blood glucose of 1,000 mg/ gon: 5 ng/ml; after glucagon: 10 ng/ml). sity of Plymouth, Plymouth, U.K. dl) without ketoacidosis. His BMI was 28 Levels of glucagon, vasoactive intestinal Address correspondence to Dr. D. Devendra, kg/m2. Subcutaneous insulin was re- peptide, and somatostatin were normal. University Medicine, Level 7 Derriford Hospital, sumed, using multi-injections, followed The euglycemic clamp study showed a Plymouth PL6 8DH, U.K. E-mail: senan.devendra@ ϭ phnt.swest.nhs.uk. by subcutaneous treatment with an insu- decrease in both sensitivity (ED50 1.13 lin pump. Despite a dramatic increase in vs. 0.62 mU kg–1 min–1 in control sub- the insulin dose (up to 1,800 units/day), jects) and responsiveness (mean maximal ●●●●●●●●●●●●●●●●●●●●●●● –1 –1 metabolic control was bad (HbA1c 11%; glucose disposal: 10.2 mg kg min , References high-performance liquid chromatogra- or 73% of the control subjects) to intrave- 1. Palumbo PJ, Melton LJ: Peripheral vascu- phy Ͻ6% [normal]). Intravenous insulin nous insulin. lar disease and diabetes. In Diabetes in infusion was then attempted through an Insulin binding was normal on the America. Harris MI, Hamman RF, Eds. implanted venous access port. Continu- patient’s erythrocytes. Insulin receptor Washington, DC, U.S. Government Print- ous insulin was given with an external in- number and phosphorylation, studied on ing Office, 1985, p. XV1–21 (NIH publ. sulin pump through a Port-A-Cath skin-cultured fibroblasts, were not al- no. 85-1468) (Braun, Boulogne Billancourt, France), tered. We did not find any coding muta- 2. Edmonds ME, Watkins PJ: The Diabetic foot. In International Textbook of Diabetes whereby the catheter was connected to a tion in the insulin receptor or the lamin Mellitus. KGMM Alberti, RA DeFronzo, H subcutaneous reservoir. After initial suc- A/C genes (2). Finally, plasma insulin keen, P Zimmet, Eds. Chicester, West cess, this treatment was withdrawn be- measurements showed subcutaneous Sussex, John Wiley and Sons, 1995 cause the metabolic state was deteriorated malabsorption of insulin; plasma free in- 3. Cook TA, Rahim N, Simpson HC, Galland in a septic context (catheter infection). sulin was 26 mU/l when 3,600 units/day RB: Magnetic resonance imaging in the Continuous subcutaneous infusion was insulin was subcutaneously infused and 7 management of diabetic foot infections. resumed with freeze-dried insulin (endo- mU/l when 30 units/day was intraperito- Br J Surg 83:245–248, 1996 pancrine; Organon Eragny/Epte, France), neally infused. 4. Oyen WJG, Boerman OC, Van der Laken up to 4,000 units/day, and then with lis- We have described a case of severe CJ, Claessens RA, van der Meer JW, Cor- pro insulin (Lilly), up to 420 units/day. subcutaneous insulin resistance that was stens FH: The uptake mechanisms of inflammation and infection localizing agents. Then, a protease inhibitor, aprotinin (Ini- treated by intraperitoneal insulin therapy. Eur J Nucl Med :459–465, 1996 prol 8 million units/day for 4 days; Labo- Such cases have already been described. ratoire Choay, Gentilly, France) was For some of them, an insulin-degrading mixed with insulin, but Ͼ1,500 units/day activity has been found in muscle or adi- insulin was needed to achieve euglyce- pose tissue (1,3). Therefore, some pa- Extreme mia. Corticosteroid therapy (dexametha- tients were successfully treated with a sone [Soludecadron; MSD Chibret, protease inhibitor, aprotinin (1,3–5). Subcutaneous Clermond-Ferrand, France] 40 mg for 3 Other patients did not respond to this Insulin Resistance days, then prednisolone [Solupred, treatment (6). Intramuscular insulin ther- Successfully Treated Houde, Paris-Defense, France] 60 mg/day apy was successfully attempted in some for 3 days) and intravenous immunoglob- cases (7,8). All of these treatments were by an Implantable ulin therapy were also ineffective. For 8 ineffective in our patient. Intraperitoneal Pump years, blood glucose remained very high, insulin therapy, through a subcutaneous between 300 and 400 mg/dl and, con- access device, was proposed in some cases sequently, diabetic complications ap- (9–11) with good results. We tried the xtreme subcutaneous insulin resis- peared: retinopathy (macular edema), same route of insulin therapy using an tance is a very rare syndrome char- painful neuropathy with neuropathic ul- implantable pump (MIP 2001). As a re- E acterized by severe resistance to cer, nephropathy, and sepsis. Then, an sult, glycemic control dramatically im- subcutaneous insulin together with nor- intraperitoneal insulin infusion with an proved with good efficiency for 1 year. mal or near normal intravenous insulin implantable pump (MIP 2001; Minimed, Therefore, extreme subcutaneous in- sensitivity. The pathophysiology is un- Sylman, CA) was started. After implanta- sulin resistance is a new indication of im- known. An increased insulin-degrading tion, the metabolic state dramatically implantable pump use. The pathophysiology activity has been reported in the subcuta- proved; weight increased from 65 to 70 of this syndrome remains unexplained. neous adipose tissue fraction (1). The kg (BMI from 21.2 to 22.8 kg/m2), insulin 1 proposed treatments are disappointing. requirement decreased to 40 units/day, JEAN-PIERRE RIVELINE, MD 2 We report such a case that was success- and HbA1c dropped from 10 to 6%. Six- JACQUELINE CAPEAU, MD, PHD 3 fully treated by intraperitoneal insulin. teen months later, the good metabolic JEAN-JACQUES ROBERT, MD, PHD 1 A 36-year-old male patient was ad- state was maintained. MICHEL VARROUD-VIAL, MD

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2155 Letters

1 ISABELLE CERF-BARON, MD patients nonresponsive to subcutaneous C-peptide excretion rate was 4 ␮g/day, 1 ANNE DEBURGE, MD insulin. Diabetes 31:470–473, 1982 and the increment of serum C-peptide in 1 GUILLAUME CHARPENTIER, MD 10. Campbell IW, Kritz H, Najemnik C, Hag- response to intravenous administration of mueller G, Irsigler K: Treatment of a type 1mgofglucagonwasnotdetectable,indicat- 1 I diabetic with subcutaneous insulin resis- From the Service d’Endocrino-Diabe´tologie, Cen- ing severe impairment of insulin secre- tance by a totally implantable insulin in- tre Hospitalier Sud-Francilien, Corbeil, France; the tion. Islet cell antibody, anti-GAD antibody, 2Institut National de la Sante´ et de la Recherche fusion device (“Infusaid”). Diabetes Res Me´dicale, Faculte´ de Me´decine, Saint-Antoine, 1:83–88, 1984 and anti–IA-2 (tyrosine phosphatase–like France; and the 3Hoˆpital Necker-Enfants Malades, 11. Dandona P, Fonseca V, Fernando O, Me- protein) antibody were negative. Dy- Paris, France. non RK, Weerakoon J, Kurtz A, Stephen namic computed tomography showed Address correspondence to Jean-Pierre Riveline, R: Control of diabetes through a subcuta- swelling of the pancreas, with enhance- MD, 59 Bd Henri Dunand, 91100 Corbeil-Essonnes, neous peritoneal access device (SPAD) in ment on the early phase. Magnetic reso- France. E-mail: [email protected]. patients with resistance to subcutane- nance cholangiography and pancrea- ously injected insulin. Diabetes Res 5:47– tography examinations were normal. Vi- 49, 1987 ●●●●●●●●●●●●●●●●●●●●●●● rus antibodies showing acute infection with cytomegalovirus, Epstein-Barr virus, References 1. Paulsen EP, Courtney JW, Duckworth rubella virus, mumps virus, herpes sim- WC: Insulin resistance caused by massive Association of Rapid- plex 1 and 2, herpes zoster virus, cox- degradation of subcutaneous insulin. Di- sackie B viruses, and rotavirus were all abetes 28:640–645, 1979 Onset Type 1 negative. The patient possessed HLA 2. Vigouroux C, Magre J, Vantyghem MC, Diabetes and Clinical DQA1*0102 and DQB1*0602, the HLA Bourut C, Lascols O, Shackleton S, Lloyd types resistant to type 1 diabetes in Japa- DJ, Guerci B, Padova G, Valensi P, Acute Pancreatitis nese individuals (1). Other HLA types in Grimaldi A, Piquemal R, Touraine P, Positive for this case were A24, A33, B7, B44, Cw7, Trembath RC, Capeau J: Lamin A/C gene: Autoantibodies to DRB1* 0101 and 1501, DQA1* 0101, sex-determined expression of mutations and DQB1* 0501. The bentiromide test in Dunnigan-type familial partial lipodys- the Exocrine trophy and absence of coding mutations value on day 28 was 63.7% (73.1–90.1), in congenital and acquired generalized Pancreas showing mild exocrine dysfunction. Even lipoatrophy. Diabetes 49:1958–1962, 2000 after acute pancreatitis was remedied with 3. Maberly GF, Wait GA, Kilpatrick JA, Lo- ulinastatin, the subject remained insulin- ten EG, Gain KR, Stewart RDH, Eastman 24-year-old woman presented with dependent. In this case, it is unlikely that CJ: Evidence for insulin degradation by epigastralgia on day 0. About 2 diabetes is secondary to classical pancre- muscle and fat tissue in an insulin resis- A weeks before day 0, she had a low- atitis, considering the severity of insulin tant diabetic patient. Diabetologia 23:333– grade fever for a few days. On day 3, she deficiency with only mild exocrine dys- 336, 1982 consulted a clinic, where hyperamy- function. Rather, this patient was diag- 4. Freidenberg GR, White N, Cataland S, lasemia (2.8 multiples of the upper nor- nosed as an association of type 1B (idio- O’Dorisio TM, Sotos JF, Santiago JV: Dia- betes responsive to intravenous but not mal limit) and swelling of the pancreas on pathic) diabetes and clinical acute pancre- subcutaneous insulin: effectiveness of ultrasonography were detected. The sub- atitis. This association suggests a common aprotinin. N Engl J Med 305:363–368, 1981 ject’s fasting plasma glucose (FPG) level etiopathogenesis in both exocrine and en- 5. Muller WA, Taillens C, Lereret S, Berger was 85 mg/dl. Her serum insulin and C- docrine dysfunction, although the mech- M, Philippe J, Halban PA, Offord RE: Re- peptide levels, as measured later with the anism remains to be elucidated. sistance against subcutaneous insulin suc- frozen plasma, were 20.5 ␮U/ml and 7.7 Simultaneous involvement of exo- cessfully managed by aprotinin. Lancet ng/ml, respectively. On day 5, the sub- crine and endocrine pancreas has been re- 7:1245–1246, 1980 ject’s serum amylase was 4.58 multiples ported in type 1 diabetic patients (2,3). 6. Schade DS, Duckworth WC: In search of of the upper normal limit. On day 6, her Recently, a novel subtype of type 1 diabe- the subcutaneous-insulin-resistance syn- FPG level was elevated to 370 mg/dl and tes, characterized by a rapid onset, an ab- drome. N Engl J Med 315:147–153, 1986 ϩϩϩ 7. Brossard JH, Havrankova J, Rioux D, Ber- her urinary ketone showed . After sence of diabetes-related antibodies, and trand S, D’Amour P: Long-term use of in- treatment with intravenous administra- hyperamylasemia with lymphocytic intramuscular insulin therapy in a type I tion of glucose, insulin, and ulinastatin filtration in the exocrine pancreas, was diabetic patient with subcutaneous insu- (serine protease inhibitor) at the clinic, postulated to be “nonautoimmune” ful- lin resistance. Diabet Med 10:174–176, she was referred and admitted to Ohtsu minant type 1 diabetes (4). Our case sub- 1993 Red Cross Hospital. The subject had no ject showed characteristics, i.e., hyper- 8. Pickup JC, Home PD, Bilous RW, Keen H, history of diabetes, pancreatitis, or alco- glycemia with low HbA1c, an absence of Alberti KGMM: Management of severely hol consumption. Her BMI was 17.2 kg/ autoantibodies to the endocrine pancreas, brittle diabetes by continuous subcutane- m2, her serum amylase was 3.22 and hyperamylasemia, that seem compat- ous and intramuscular insulin infusions: multiples of the upper normal limit, and ible with this entitiy. The relatively high evidence for a defect in subcutaneous insulin absorption. Br Med J 282:347–350, her serum elastase 1 was 1,400 ng/dl levels of serum insulin and C-peptide for 1981 (range 100–400). The subject’s plasma normoglycemia on day 3 appear to reflect 9. Schade DS, Eaton RP, Warhol RM, Greg- glucose level was 329 mg/dl, and her the ongoing destruction of ␤-cells. ory JA, Doberneck RC: Subcutaneous HbA1c was 4.9% (3.0–6.0). The serum Serum levels of autoantibodies peritoneal access device for type I diabetic C-peptide level was 0.2 ng/ml, the urinary against human carbonic anhydrase II

2156 DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 Letters

(ACA) and autoantibodies against lacto- ●●●●●●●●●●●●●●●●●●●●●●● One Touch glucose meter and continued ferrin (ALF), which are distributed in the References to monitor her glucose frequently for the pancreatic duct cells and the acinar cells, 1. Ikegami H, Ogihara T: Genetics of insu- next few months. The glucose readings lin-dependent diabetes mellitus. Endo- respectively, were measured using the were consistently 30–40 mg/dl. She often solid-phase ELISA method, as previously crine J 43:605–613, 1996 2. Nakanishi K, Kobayashi T, Sugimoto T, reported symptoms of lightheadedness, described (5). ACA and ALF were positive Murase T, Itou T, Kosaka K: Does pan- fatigue, and sweating, but in retrospect, in the present case subject. pancreatic involvement occur in IDDM? the relationship between these symptoms It was reported that ACA and ALF (Letter) Diabetes Care 11:100–101, 1988 and the glucose readings was inconsis- were detected in patients with autoim- 3. Moriai T, Morita Y, Matsui T, Okada M: tent. mune pancreatitis, whereas they were Type 1 diabetes mellitus associated with A second endocrinologist repeated clinical acute pancreatitis in an adult. Pan- not detected in any of the patients with the glucose and C-peptide tests, and these alcoholic or gall stone–related pancreati- creas 20:415–420, 2000 were again normal. A urine screen for sul- tis (5). Although our present case was un- 4. Imagawa A, Hanafusa T, Miyagawa J, Mat- fonylurea agents was negative. characteristic of autoimmune chronic suzawa Y: A novel subtype of type 1 dia- pancreatitis after completion of imaging betes mellitus characterized by a rapid The patient was then referred for a onset and an absence of diabetes-related prolonged fast. On greeting the patient, studies and the clinical course (6,7), the antibodies. N Engl J Med 342:301–307, presence of these antibodies suggests the her hands were white and cold. Over the 2000 subsequent 2 h, using a Freestyle meter involvement of autoimmunity against the 5. Okazaki K, Uchida K, Ohana M, Nakase exocrine pancreas. H, Uose S, Inai M, Matsushima Y, Kata- (FM) and a Precision QID meter (PM), We have recently demonstrated the kura K, Ohmori K, Chiba T: Autoim- glucose measurements were obtained presence of ACA and ALF in type 1 dia- mune-related pancreatitis is associated from the patient’s fingertips and forearms betic patients and proposed the concept with autoantibodies and a Th1/Th2-type and by venipuncture. of autoimmune exocrinopathy and endo- cellular immune response. Gastroenterol- At 9:00 A.M., glucose levels from the crinopathy of the pancreas (8). ogy 118:573–581, 2000 fingertips were 53 and 56 (FM) and 49 In conclusion, we reported the first 6. Yoshida K, Toki F, Takeuchi T, Wata- nabe S, Shiratori K, Hayashi N: Chronic and 38 (PM) mg/dl, and the forearm level case of an association of rapid-onset type pancreatitis caused by an autoimmune was 83 mg/dl (FM). At 10:00 A.M., glucose 1 diabetes and clinical acute pancreatitis abnormality: proposal of the concept of levels from the fingertips were 50 (FM) that tested positive for autoantibodies to autoimmune pancreatitis. Dig Dis Sci 40: and 48 (PM) mg/dl, and the forearm level the exocrine pancreas. Although known 1561–1568, 1995 was 73 (FM) mg/dl. At 11:00 A.M., glucose antibodies against the endocrine pancreas 7. Taniguchi T, Seko S, Azuma K, Tamegai levels from the fingertips were 42 (FM) were not detected, autoimmunity, at least M, Nishida O, Inoue F, Okamoto M, Mi- and 53 (PM) mg/dl, and the forearm level against the exocrine pancreas, was sug- zumoto T, Kobayashi H: Association of was 78 (FM) mg/dl. Also, at 11:00 A.M., a gested. autoimmune pancreatitis and type 1 diabetes: autoimmune exocrinopathy and venous blood sample was drawn. This endocrinopathy of the pancreas. Diabetes sample was used for a glucose check on 1 TAKAO TANIGUCHI, MD,PHD Care 23:1592–1594, 2000 each meter and was sent to the clinical 1 JUNNYA TANAKA, MD 8. Taniguchi T, Okazaki K, Okamoto M, laboratory (LAB). The 11:00 A.M. glucose 1 SHUJI SEKO, MD,PHD Seko S, Uchida K, Seino Y: Presence of 2 levels from the venous sample were 88 KAZUICHI OKAZAKI, MD,PHD autoantibodies to carbonic anhydrase II 1 (FM), 93 (PM), and 86 (LAB) mg/dl. MOTOZUMI OKAMOTO, MD,PHD and lactoferrin in type 1 diabetes: proposal of the concept of autoimmune exo- Fingertip capillary glucose levels were consistently lower than simulta- 1 crinopathy and endocrinopathy of the From the Department of Internal Medicine, Ohtsu pancreas (Letter). Diabetes Care 24:1695– neous forearm capillary glucose levels Red Cross Hospital, Shiga, Japan; and the 2Depart- ment of Endoscopic Medicine and Gastroenterol- 1696, 2001 and/or venous glucose levels. We believe ogy, Kyoto University Faculty of Medicine, Kyoto, the patient’s false low fingertip glucose Japan. readings were secondary to the circula- Address correspondence to Takao Taniguchi, MD, PhD, Department of Internal Medicine, Ohtsu tory change from Raynaud’s phenome- Red Cross Hospital, 1-1-35, Nagara, Ohtsu, Shiga, Case of non. Similar pseudohypoglycemia has Japan. Pseudohypoglycemia been reported in patients with altered cir- culation from shock (1).

Acknowledgments— This work was sup- he patient is a 44-year-old white 1 ported by grant-in-Aid for Scientiﬁc Research woman with a history of myasthenia ROBERT J. RUSHAKOFF, MD (C) from the Ministry of Culture and Science gravis, hypothyroidism, epilepsy, STEPHEN B. LEWIS, MD of Japan (11670495), Grant-in-Aid for Re- T and Raynaud’s phenomenon. In May search for the Future Program from the Japan 2000, while hospitalized for gastroenter- Society for Promotion of Science (JSPS- 1 itis, a ﬁngerstick glucose reading reported From the University of California, San Francisco, RFTF97I00201), Supporting Research Funds San Francisco, California. from the Japanese Foundation for Research as “low.” Subsequent outpatient testing Address correspondence to Robert J. Rushakoff, and Promotion of Endoscopy (JFE-1997), and showed a normal random venous glucose MD, University of California, San Francisco, P.O. Supporting Research Funds from the Shimid- level with concurrent normal C-peptide Box 1616, San Francisco, CA 94143. E-mail: zu Foundation for Immunology. and insulin levels. The patient obtained a [email protected].

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2157 Letters

●●●●●●●●●●●●●●●●●●●●●●● deed, at the onset of the disease, a state of Ricerche Farmacologiche Mario Negri, Consorzio References hypertrophy and hyperfunction (with hy- Mario Negri Sud, 66030 Santa Maria Imbaro, Italy. 1. Atkin SH, Dasmahapatra A, Jaker MA, perfiltration) of the kidneys is often ob- E-mail: [email protected]. Chorost MI, Reddy S: Fingersick glucose served. GFR may be as much as 40% determination in shock. Ann Intern Med above normal. The next stage is the ap- 114:1020–1024, 1991 Acknowledgments— S.B. is the recipient of a pearance of microalbuminuria, a phe- Marie Curie Training Site Fellowship. nomenon that is thought to be an early marker of diabetic nephropathy. This is of ●●●●●●●●●●●●●●●●●●●●●●● COMMENTS AND special interest because GFR is the stron- References gest independent predictor of plasma 1. Pavia C, Ferrer I, Valls C, Artuch R, Co- RESPONSES tHcy concentration. The study performed lome C, Vilaseca MA: Total homocysteine by Wollesen et al. (5) clearly showed that in patients with type 1 diabetes. Diabetes in diabetic patients with a normal serum Care 23:84–87, 2000 creatinine concentration (Ͻ115 ␮mol/l) 2. Cotellessa M, Minniti G, Cerone R, Pri- Homocysteinemia in gione F, Calevo MG, Lorini R: Low total Patients With Type 1 and GFRs in the lower range, the concen- plasma homocysteine concentrations in tration of tHcy was higher than in subjects patients with type 1 diabetes (Letter). Di- Diabetes in Relation who had GFRs in the upper range. Hyper- abetes Care 24:969–970, 2001 to Renal Function filtration or GFRs above the normal values 3. van Guldener C, Stam F, Stehouwer CD: for their age and sex were found in Ͼ80% Homocysteine metabolism in renal fail- of patients. The authors found a strong ure. Kidney Int 59 (Suppl. 78):S234– n the study by Pavia et al. (1) and the inverse linear correlation between plasma S237, 2001 subsequent commentary by Cotellessa tHcy and GFR in the range of 47–165 ml 4. Kemperman FAW, Silberbush J, Slaats et al. (2), the problem of homocysteine min–1 1.73 m–2. In this study, GFR deter- EH, van Zanten AO, Weber JA, Krediet I RT, Arisz L: Glomerular filtration rate es- levels in patients with type 1 diabetes was mined plasma levels of tHcy independently timation from plasma creatinine after in- raised. The authors found no differences of age, serum folic acid and B-group vita- hibition of tubular secretion: relevance of in plasma total homocysteine (tHcy) con- min concentrations, serum creatinine the creatinine assay. Nephrol Dial Trans- centration between diabetic children concentration, and urine-albumin excre- plant 14:1247–1251, 1999 and/or adolescents and age-matched con- tion rate. 5. Wollesen F, Brattstrom L, Refsum H, trol subjects. They also did not observe Thus, because no patients suffered Ueland PM, Berglund L, Berne C: Plasma any association between tHcy levels and from overt nephropathy in the population total homocysteine and cysteine in rela- either duration or metabolic control of the of patients studied by Pavia et al. (1), a tion to glomerular filtration rate in diabe- disease or its complications. The 91 pa- relative hyperfiltration is the most plausi- tes mellitus. Kidney Int 55:1028–1035, tients analyzed had a duration of type 1 ble cause of their results, which showed 1999 diabetes ranging from 1 to 15 years, and low tHcy concentrations in this group of in ϳ50% of them, the duration of the dis- diabetic patients. Although they found a Glitazones, ease was Ͼ5 years. Patients did not have correlation between tHcy and creatinine microalbuminuria and had serum creati- concentration, measuring only creatinine Glycemia, and nine within the normal range. concentration is not sufficient for assess- Global Health Status Pavia et al. (1) pointed out that hyper- ing GFR (with special regard to hyperfil- homocysteinemia is already present in the tration) in this specific population. We early stages of renal failure (3). However, agree that hyperhomocysteinemia is not he recent study by Raskin et al. (1) no change in renal function, measured as the cause of vascular complications in di- failed to emphasize three important the serum creatinine concentration, was abetic patients, at least in those without T points. In the study (1), the mean found in their patients. overt nephropathy. weight gain was 4.0 and 5.3 kg for the In our opinion, there are two points study groups given rosiglitazone 4 and 8 1,2 that should be emphasized in this con- SZYMON BRZOSKO, MD mg/day, respectively. No range of weight 2 text. First, serum creatinine concentra- MICHAL MYSLIWIEC, MD, PHD change or standard deviation was given, 1 tion is not an accurate measure of MARIA BENEDETTA DONATI, MD, PHD making it impossible to appreciate the 1 glomerular filtration rate (GFR), espe- LICIA IACOVIELLO, MD, PHD maximum and minimum weight changes cially in the range of 50–140% of normal. of this 26-week trial. It would be interest- Therefore, mildly disturbed renal func- From the 1Angela Valenti Laboratory of Genetic and ing to see if any factors predicted weight tion is underestimated when assessed Environmental Risk Factors for Thrombotic Disease, change for combination therapy. Further- only on the basis of serum creatinine con- Department of Vascular Medicine and Pharmacol- more, the authors failed to comment on ogy, Istituto di Ricerche Farmacologiche Mario Ne- centration. Although not perfect, the gri, Consorzio Mario Negri Sud, Santa Maria the adverse health consequences of fur- Cockroft-Gault formula should at least be Imbaro, Italy; and the 2Department of Nephrology ther weight gain in these obese study pa- used to measure creatinine clearance as a and Internal Medicine, Bialystok Medical Univer- tients. Despite improved glycemia and surrogate of GFR (4). The second and sity, Bialystok, Poland. other potential cardiovascular benefits of Address correspondence to Szymon Brzosko, An- probably more important point is that re- gela Valenti Laboratory of Genetic and Environmen- thiazolidinediones (2), at some point, nal abnormalities may already be present tal Risk Factors for Thrombotic Disease, Department weight gain, which is common, signifi- at very early stages of type 1 diabetes. In- of Vascular Medicine and Pharmacology, Istituto di cant, and seen with all agents in this drug

2158 DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 Letters class when given with insulin, will prob- some pause in their enthusiasm for these ●●●●●●●●●●●●●●●●●●●●●●● ably outweigh any putative positive ben- drugs. Perhaps exercise testing would be References eﬁts. Also, because these drugs cause more conﬁrmatory of the cardiovascular 1. Raskin P, Rendell M, Riddle MC, Dole JF, differentiation of preadipoctyes into adi- safety of these agents in the short term. Freed MI, Rosenstock J, the Rosiglitazone pocytes (3), it is possible that weight gain Finally, despite a therapeutic effect, Clinical Trials Study Group: A random- may be progressive with time in some pa- the mean achieved HbA (8.5 and 7.9% ized trial of rosiglitazone therapy in pa- 1c tients with inadequately controlled insulin- tients. We do not consider weight in- for 4 and 8 mg/day rosiglitazone, respec- Ն treated type 2 diabetes. Diabetes Care 24: creases of 10 kg unusual when tively) is still much greater than the gly- 1226–1232, 2001 thiazolidinediones with insulin are being cemic goals mandated by the American 2. Parulkar AA, Pendergrass ML, Granda- used; in two of our patients, we saw Diabetes Association (5). Although the Ayala R, Lee TR, Fonseca VA: Nonhypo- weight increases of Ͼ40 kg. Intensive di- combination of a thiazolidinedione and glycemic effects of thiazolidinediones. etary intervention may be necessary to insulin is useful for the control of glyce- Ann Intern Med 134:61–71, 2001 preclude this development in patients on mia, it will nonetheless fail to achieve ad- 3. Toseland CDN, Campbell S, Francis I, Bu- insulin treatment concomitantly with a equate glycemic control in many patients. gelski PJ, Mehdi N: Comparison of adi- thiazolidinedione. pose tissue changes following adminis- Edema was noted in 17 of 103 pa- WILLIAM L. ISLEY, MD tration of rosiglitazone in the dog and rat. tients on insulin plus 8 mg rosiglitazone. Diabetes Obes Metab 3:163–170, 2001 Although edema is not considered to be a From Saint Luke’s Lipid and Diabetes Research Cen- 4. Food and Drug Administration, Center serious adverse event, it can be troubling ter, University of Missouri-Kansas City, Kansas City, for Drug Evaluation and Research: 73rd Missouri. Meeting of the Endocrinologic and Metabolic for some patients. Expanded extracellular Address correspondence to William L. Isley, MD, Drugs Advisory Committee, Bethesda, MD, water is commonly seen with thiazo- Saint Luke’s Lipid and Diabetes Research Center, April 22-23, 1999. Bethesda, MD, FDA, lidinediones. Although echocardio- MPI, Suite 128, 4320 Wornall Rd., Kansas City, MO 1999. graphic studies in humans have failed to 64111. E-mail: [email protected]. W.L.I. has received honorarium from Smith- 5. American Diabetes Association: Stan- detect deleterious effects in the short Kline Beecham, who provides grant support to Saint dards of medical care for patients with di- term, the adverse cardiac effects seen in Luke’s Lipid and Diabetes Research Center for the abetes mellitus. Diabetes Care 24 (Suppl. animal models (4) should give physicians study of diabetes and hypertension drugs. 1):S33–S43, 2001

DIABETES CARE, VOLUME 24, NUMBER 12, DECEMBER 2001 2159