Cardiac Events in 735 Type 2 Diabetic Patients Who Underwent

Pathophysiology/Complications ORIGINAL ARTICLE

Cardiac Events in 735 Type 2 Diabetic Patients Who Underwent Screening for Unknown Asymptomatic Coronary Heart Disease 5-year follow-up report from the Milan Study on Atherosclerosis and Diabetes (MiSAD)

1 4 EZIO FAGLIA, MD NICOLETTA MUSACCHIO, MD raphy group may suggest, although it does not 2 5 FABRIZIO FAVALES, MD ARTURO MASTROPASQUA, MD prove, that the screening program followed by 2 6 PATRIZIA CALIA, MD GIANPAOLO TESTORI, MD appropriate management was effective for the 3 6 FELICE PALEARI, MD PIETRO RAMPINI, MD reduction of risk of major cardiac events. 3 7 GIOVANNI SEGALINI, MD FLAVIA MORATTI, MD 4 7 PIER LUIGI GAMBA, MD ANNA BRAGA, MD 4 8 ALBERTO ROCCA, MD ALBERTO MORABITO, PHD any studies have investigated the presence of unknown asymptom- M atic coronary heart disease (CHD) among diabetic subjects (1–7). Each of OBJECTIVE — To report the cardiac events in type 2 diabetic outpatients screened for un- these studies has shown, with different known asymptomatic coronary heart disease (CHD) and followed for 5 years. rates depending on the diagnostic meth- RESEARCH DESIGN AND METHODS — During 1993, 925 subjects aged 40–65 ods and selection criteria used, increased years underwent an exercise treadmill test (ETT). If it was abnormal, the subjects then underwent prevalence rates compared with the gen- an exercise scintigraphy. Of the 925 subjects, 735 were followed for 5 years and cardiac events eral population. However, to our knowl- were recorded. edge, none of these studies has followed screened patients prospectively. In 1993 RESULTS — At the entry of the study, 638 of the 735 followed subjects had normal ETT, 45 we created the Milan Study on Athero- had abnormal ETT with normal scintigraphy, and 52 had abnormal ETT and abnormal scintigraphy. The 52 subjects with abnormal scintigraphy and ETT underwent a cardiological and sclerosis and Diabetes (MiSAD), and un- diabetological follow-up; the subjects with just abnormal ETT had a diabetological follow-up dertook a screening procedure for CHD only. During the follow-ups, 42 cardiac events occurred: 1 fatal myocardial infarction (MI), 20 detection in type 2 middle-aged diabetic nonfatal MIs, and 10 cases of angina in the 638 subjects with normal ETT; 1 fatal MI in the 45 outpatients, free from known and asymp- subjects with normal scintigraphy; and 1 fatal MI and 9 cases of angina in the 52 subjects with tomatic CHD (8). The aim of this article abnormal scintigraphy. In these 52 subjects all cardiac events were signiﬁcantly more frequent is to report on cardiac events in 735 (␹2 ϭ 21.40, P Ͻ 0.0001) but the ratio of major (cardiac death and MI) to minor (angina) cardiac screened subjects over a 5-year follow-up events was signiﬁcantly lower (P ϭ 0.002). Scintigraphy abnormality (hazard ratio 5.47; P Ͻ period. 0.001; 95% CI 2.43–12.29), diabetes duration (1.06; P ϭ 0.021; 1.008–1.106), and diabetic retinopathy (2.371; P ϭ 0.036; 1.059–5.307) were independent predictors of cardiac events on multivariate analysis. RESEARCH DESIGN AND METHODS CONCLUSIONS — The low ratio of major to minor cardiac events in the positive scintig-

●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● Patient population A total of 925 type 2 diabetic outpatients, From the 1Policlinico Multimedica, Internal Medicine Unit–Diabetology Center, Sesto San Giovanni, Milan, Italy; the 2Niguarda Hospital, Internal Medicine Unit, Milan, Italy; the 3S. Gerardo Hospital, Diabetology aged 40–65 years, who were free from Unit–Internal Medicine Departement, Monza, Milan, Italy; the 4E. Bassini Hospital, Diabetology Center, known and asymptomatic CHD were en- Cinisello Balsamo, Milan, Italy; the 5G. Salvini Hospital, Diabetology Center, Garbagnate Milanese, Milan, rolled in seven Diabetology Centers dur- Italy; the 6Fatebenefratelli Hospital, Diabetology Unit, Milan, Italy; the 7Legnano Hospital, Diabetology 8 ing 1993. These patients did not have Center–Internal Medicine Unit, Legnano, Milan, Italy; and S. Paolo Hospital, Milan University, Departe- proliferative retinopathy, their serum cre- ment of Medicine, Surgery and Dentistry, Milan, Italy. Ͻ Address correspondence and reprint requests to Ezio Faglia, Internal Medicine Unit, Diabetology Center, atinine was 1.5 mg/dl, peripheral or Policlinico MultiMedica, Via Milanese 300, 20099 Sesto San Giovanni (Milano), Italy. E-mail: cerebral arteriopathy was absent, hyper- [email protected]. tension requiring more than two drugs Received for publication 3 August 2001 and accepted in revised form 4 August 2002. was not present, and the patients were not Abbreviations: CHD, coronary heart disease; ECG, electrocardiogram; ETT, exercise treadmill test; MI, myocardial infarction; MiSAD, Milan Study on Atherosclerosis and Diabetes. poor-prognosis disease patients. CHD A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion was screened by exercise treadmill test factors for many substances. (ETT). When this test was abnormal, ex-

2032 DIABETES CARE, VOLUME 25, NUMBER 11, NOVEMBER 2002 Faglia and Associates

All 735 followed subjects underwent an average of four diabetological evalua- tions per year. Examination included de- termination of GHb (HbA1c, Hb%, high- pressure liquid chromatography; normal value 4–6.4%), albuminuria (nefelom- etry), arterial blood pressure, and BMI measurement. The presence of diabetic retinopathy was assessed once a year by fundus oculi examination performed by an ophthalmologist, as well as total cholesterol (colorimetry), HDL cholesterol Figure 1—Age distribution of patients completing the 5-year follow-up period (N ϭ 735). (Polyethylene Glycol 6000), triglycerides (colorimetry), and resting ECG. These data, together with drug therapy data, ercise thallium-201 scintigraphy was per- ological consultation. On the basis of the were recorded on a study form and sent to formed. The ETT was considered positive advice of a cardiologist, 22 subjects un- the coordinating center in charge of the in the presence of a horizontal or derwent coronary angiography. Stenoses central database. A special form was used downsloping ST-segment depression Ͼ50% of vessel lumen were found in all to record all events—cardiac and noncar- Ն0.1 mV, or an upsloping ST-segment cases. Five patients had three-vessel in- diac. Cardiac death, myocardial infarc- depression Ն0.2 mV, measured at 0.08 s volvement; four of these subjects under- tion, resting angina, and effort angina after the J point. Exercise thallium-201 went coronary bypass. The remaining were considered cardiac events. Major scintigraphy was classified as abnormal if subject was considered not to be an ap- events were defined as cardiac death and perfusion defects were present. In the propriate candidate for surgical interven- MI; minor events were defined as resting MiSAD article (8), patients with abnormal tion. Six patients had two-vessel involve- and effort angina. Each cardiac event was scintigraphy were defined as affected by ment, and four of these subjects under- reported in this work by means of hospital asymptomatic unknown CHD. went coronary angioplasty. Eleven had admission chart or the cardiologist medi- One center was unable to participate involvement that was limited to a single cal report. in the follow-up study, and the 150 pa- vessel, and none underwent a revascular- tients screened in this center did not par- ization procedure. These 52 patients fol- Statistical analysis ticipate in the follow-up. Of the patients lowed the cardiologist’s follow-up protocol. Cardiac events were the primary end referring to the remaining six centers par- All of the subjects with normal ETT points. Differences in event frequencies ticipating in the follow-up, 40 moved to and abnormal ETT, but with normal scin- were evaluated by a ␹2 test or an Fisher’s another city before completing the 5-year tigraphy, did not undergo any further car- exact test. follow-up period and thus were com- diologic work-up (except for the annual An incidence rate estimation with the pletely removed from the follow-up anal- resting electrocardiogram [ECG]) in the CI of cardiac events was performed. The ysis. At the time of removal from the absence of any cardiac symptoms. Cox model with the Stata statistical soft- study, these subjects did not show any events. recorded in 1993 and values at (735 ؍ The study population of 735 patients Table 1—Clinical parameters of study population (N (267 females and 468 males) were fol- the end of the study of the variables monitored during follow-up lowed for 5 years. At the time of the screening, 638 of 735 presented normal Variables 1993 1998 ETT, 45 presented abnormal ETT but normal scintigraphy, and 52 had abnor- Family history for CHD (n) 198 — mal ETT and abnormal scintigraphy. Fasting C-peptide (nmol/l) 0.90 Ϯ 0.40 — Postprandial C-peptide (nmol/l) 2.10 Ϯ 0.98 — Follow-up studies GHb (% of Hb) 7.3 Ϯ 1.5 7.7 Ϯ 1.6 At the end of the follow-up the mean age BMI (kg/m2) 27.4 Ϯ 3.9 27.4 Ϯ 4.0 was 59.6 years and the mean diabetes du- Total cholesterol (mmol/l) 5.56 Ϯ 0.99 5.58 Ϯ 1.0 ration was 12.4 Ϯ 6.1 years. Figure 1 HDL cholesterol (mmol/l) 1.20 Ϯ 0.3 1.23 Ϯ 0.3 shows the age distribution. Table 1 shows Triglycerides (mmol/l) 2.22 Ϯ 1.26 2.11 Ϯ 1.26 the variables recorded in the study popu- Systolic blood pressure (mmHg) 142.4 Ϯ 18.7 141.6 Ϯ 17.7 lation in 1993 and the values at the end of Diastolic blood pressure (mmHg) 84.3 Ϯ 8.8 81.9 Ϯ 7.9 the study of the variables monitored dur- Smokers 198 (26.9) 126 (17.1) ing the follow-up (Table 1). Autonomic neuropathy 162 (22.0) — On the basis of screening results in Retinopathy 39 (5.3) 39 (5.3) 1993, all 52 subjects with abnormal scin- Microalbuminuria (Ͼ20 and Ͻ200 ␮g/min) 162 (22.0) 166 (22.6) tigraphy were placed on 100 mg daily of Proteinuria (Ն200 ␮g/min) 10 (1.4) 10 (1.4) acetylsalicylic acid and underwent cardi- Data are n (%) or means Ϯ SD.

DIABETES CARE, VOLUME 25, NUMBER 11, NOVEMBER 2002 2033 5-Year follow-up of the MiSAD

Table 2—Univariate analysis results of the association with the variables considered and of the six subjects who presented resting -angina underwent myocardial revascular (735 ؍ in study population (N (42 ؍ cardiac events (n ization in 1993, by means of coronary by- Hazard pass in one and coronary angioplasty in Variables ratio P 95% CI the other. No cardiac events occurred in the remaining patients having undergone Age (years) 1.06 0.03 1.01–1.12 a myocardial revascularization. Women 0.39 0.02 0.18–0.85 The incidence of cardiac events is Family history for CHD (no/yes) 2.99 0.11 0.77–11.56 3.85/100 persons/year (95% CI1.84– Diabetes duration (years) 1.10 0.02 1.01–1.18 7.07) in subjects with abnormal scintigra- GHb (% of Hb) 1.16 0.18 0.93–1.45 phy, 0.44/100 persons/year (0.01–2.47) BMI (kg/m2) 1.05 0.24 0.96–1.15 in subjects with abnormal ETT but nor- Fasting C-peptide (nmol/l) 0.99 0.83 0.96–1.04 mal scintigraphy, and 0.97/100 persons/ Postprandial C-peptide (nmol/l) 0.99 0.73 0.98–1.01 year (0.66–1.38) in subjects with normal Total cholesterol (mmol/l) 1.00 0.79 0.99–1.01 ETT. The number of cardiac events was HDL cholesterol (mmol/l) 0.98 0.29 0.95–1.02 significantly higher in patients with ab- Triglycerides (mmol/l) 1.00 0.67 0.99–1.00 normal scintigraphy: ␹2 ϭ 21.40, P Ͻ Systolic blood pressure (mmHg) 0.99 0.99 0.98–1.02 0.0001. The proportion between major Diastolic blood pressure (mmHg) 1.00 0.88 0.95–1.06 and minor events is significantly lower in Smokers (no/yes) 1.08 0.83 0.52–2.28 subjects with abnormal scintigraphy Autonomic neuropathy (no/yes) 0.98 0.77 0.84–1.14 compared with subjects who had normal Retinopathy (no/yes) 2.07 0.02 1.14–3.78 ETT or abnormal ETT but normal scintig- Microalbuminuria (Ͻ20 and Ͼ200 ␮g/min) (no/yes) 1.06 0.20 0.97–1.17 raphy at the screening in 1993 (P ϭ Proteinuria (Ն 200 ␮g/min) (no/yes) 1.00 1.00 — 0.002). The incidence of all cardiac events Abnormal scintigraphy 4.57 0.00 2.00–9.52 between subjects with normal ETT and subjects with abnormal ETT but with normal scintigraphy was not statistically sig- ware (Statistics/Data Analysis; Stata, Col- went coronary angiography and one un- nificant (P ϭ 0.715). lege Station, TX) was used to accomplish derwent coronary bypass. Of the five Univariate analysis results of the asso- the survival analysis, to estimate relative patients with effort angina, two were ad- ciation with the variables considered and risk and 95% CI limit, and to assess the mitted to the hospital and one underwent cardiac events are summarized in Table 2 relationship between clinical features of coronary bypass; three patients were di- . the 735 followed subjects and time at on- agnosed in an outpatient setting by means Of the variables found to be associ- set of all cardiac events. For each contin- of ETT and scintigraphy. ated in the univariate analysis, the follow- uous variable considered, mean and SDs In the group of 45 patients with ab- ing maintained independent prognostic are reported. normal ETT but normal scintigraphy 1 fa- significance for cardiac events on multi- tal MI occurred in the hospital. variate analysis: scintigraphy abnormality RESULTS — During follow-up, nine In the group of 52 patients with ab- (hazard ratio 5.47; P Ͻ 0.0001; CI 2.43– patients died of noncardiac events: seven normal ETT and scintigraphy, 1 case of 12.27), diabetes duration (1.06; P ϭ from malignancies, one from a stroke, fatal MI, 6 cases of resting angina, and 3 0.021; 1.01–1.12), presence of diabetic and one from gastrointestinal bleeding. A cases of effort angina occurred. The pa- retinopathy (2.37; P ϭ 0.036; 1.06– total of 42 cardiac events occurred; 3 tient with inoperable three-vessel disease 5.31). were fatal. had the fatal MI in the hospital. All six In the group of 638 subjects with nor- subjects with resting angina were admit- CONCLUSIONS — There is a signif- mal ETT, there was 1 fatal MI, 20 nonfatal ted to the hospital; three underwent cor- icantly different rate of all cardiac events MIs (3 of which were silent), 5 cases of onary angiography and one underwent in subjects with abnormal scintigraphy resting angina, and 5 cases of effort an- a coronary bypass. The three subjects with respect to subjects with normal ETT gina. The fatal MI occurred in a hospital who underwent coronary angiography at or abnormal ETT but normal scintigra- setting. Of the 20 subjects with nonfatal the moment of the event were subjects phy. These data suggest that our screen- MIs, 18 were admitted to the hospital and who did not undergo coronary angiogra- ing algorithm may effectively identify underwent coronary angiography. Two of phy in 1993, and the three subjects who middle-aged type 2 diabetic subjects with the three patients with silent nonfatal MI, did not undergo coronary angiography at preclinical CHD (9). For this purpose, which was discovered during a routine the moment of the event were subjects thallium scintigraphy is the most accurate annual ECG and confirmed by an echo- who underwent coronary angiography in diagnostic test (10). However, it is costly cardiogram and scintigraphy, were not 1993. Of the three patients with effort an- and not widely available and therefore not admitted. Of the 18 subjects who under- gina, only one was hospitalized, and none suitable as a first-line screening proce- went coronary angiography, 5 underwent had coronary angiography in 1993 or at dure. Thus, a more realistic approach is coronary angioplasty and 3 coronary by- the time of the event. In all three patients, the use of an initial test such as ETT, pass. All five subjects with resting angina the diagnosis was made by means of ETT, which may not be very specific but is in- were admitted to the hospital; four under- echocardiogram, and scintigraphy. Two expensive and widely available. Patients

2034 DIABETES CARE, VOLUME 25, NUMBER 11, NOVEMBER 2002 Faglia and Associates with abnormal ETT can then be more ac- literature concerning diabetic people isty C, Ward JD: Asymptomatic myocar- curately assessed by a more specific test (12,17–19). This discrepancy may be ex- dial ischemia in diabetes and its that will highlight false positives (11). In plained on the basis of the CHD risk cat- relationship to diabetes neuropathy: an patients with a positive ETT but a normal egory of our study population. In fact, the exercise electrocardiography study in scintigraphy the incidence of cardiac MiSAD excluded patients with advanced middle-aged diabetic man. Diabetes Care 9:384–388, 1986 events is not significantly different from age, known peripheral or cerebral arteri- 4. Koistinen MJ: Prevalence of asymptom- that of patients with a normal ETT. None- opathy, renal insufficiency, and severe atic myocardial ischaemia in diabetic sub- theless, we have to note that the CI in the hypertension. These selection criteria re- jects. BMJ 301:92–95, 1990 group with positive ETT but normal scin- sulted in the inclusion of a diabetic subset 5. Langer A, Freeman MR, Josse RG, Steiner tigraphy is very large. with low-risk features. In addition, in the G, Amstrong PW: Detection of silent myo- We believe that the most relevant data MiSAD the most relevant modifiable car- cardial ischemia in diabetes mellitus. Am J emerging from the study is the low inci- diovascular risk factors, such as total and Cardiol 67:1073–1078, 1991 dence of fatal cardiac events in subjects HDL cholesterol, triglycerides, and hy- 6. Naka M, Hiramatsu K, Aizawa T, Momose with abnormal scintigraphy (0.38/100 pertension, were optimal at enrollment A, Yoshizawa K, Shigematzu S, Ishihara F, persons/year). The only death that oc- and remained such throughout follow-up Niwa A, Yamada T: Silent myocardial ischemia in patients with non-insulin- curred in this group was in a patient with (20). We do not believe that variations, dependent diabetes mellitus as judged by inoperable three-vessel disease. The inci- even if statistically significant, of the clin- treadmill exercise test and coronary an- dence of cardiac deaths among patients ical modifiable variables monitored dur- giography. Am Heart J 123:46–52, 1992 identified as coronaropathic in the MiSAD ing the course of the follow-up are 7. Janand-Delenne B, Savin B, Habib G, Bory screening is superimposable to that of clinically relevant (21). We believe that M, Vague P, Lassmann-Vague V: Silent nondiabetic, noncoronaropathic subjects the exclusion criteria used in our study myocardial ischemia in patients with dia- reported in the Kuopio Study, which fol- population and the optimal status of betes: who to screen. Diabetes Care 22: lowed over a 7-year period both type 2 CHD-modifiable major risk factors re- 1396–1400, 1999 diabetic and nondiabetic subjects aged sulted in the low frequency of cardiac 8. Milan Study on Atherosclerosis and Dia- 45–64 years with or without a history of events in our study. betes (MiSAD) Group: Prevalence of un- recognized silent myocardial ischemia MI (12). In conclusion, the low ratio of major and its association with atherosclerotic The proportion between major and to minor cardiac events in patients iden- risk factors in non insulin-dependent di- minor cardiac events is significantly less tified as coronaropathic in the MiSAD abetes mellitus. Am J Cardiol 79:134–139, in subjects with an abnormal scintigra- study suggests but does not prove that the 1997 phy, and only two patients who under- screening program was effective probably 9. Inzucchi SE: Noninvasive assessment of went a myocardial revascularization after because interventions were possible and, the diabetic patient for coronary artery screening have presented cardiac events, thus, were carried out. However, it also disease. Diabetes Care 24:1519–1521, 2001 both being minor, in the5-year follow-up. suggests that ETT may be inadequate as 10. Nesto RW, Phillips RT, Kett KG, Hill T, This ratio may indicate that preclinical di- an initial screening test because in the Perper E, Young E, Leland OS Jr: Angina agnosis of CHD and the consequent early larger group, patients with coronary ar- and exertional myocardial ischemia in diabetic and nondiabetic patients: assess- institution of appropriate management tery disease were not pinpointed; subse- ment by exercise thallium scintigraphy. might be a relevant factor in the preven- quently, no intervention was performed, Ann Intern Med 108:170–175, 1988 tion of major cardiac events in the subset and the risk of a major event was therefore 11. Pollock SG, Abbott RD, Boucher CA, of patients with CHD detected at screen- higher. So our data can be interpreted to Beller GA, Kaul S: Independent and incre- ing (13). These patients were treated with understand that routine scintigraphy is mental prognostic value of tests per- acetylsalicylic acid (14), and in some in- probably best but, if for financial reasons formed in hierarchical order to evaluate stances revascularization procedures this is not possible, methods should be patients with suspected coronary artery were performed and appropriate cardio- found to improve accuracy of low-cost disease: validation of models besed on logic therapy and follow-up were insti- screening procedures such as the ETT. these tests. Circulation 85:237–248, 1992 tuted (15). It is probable that these However, the MiSAD follow-up data sug- 12. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M: Mortality from cor- interventions have had an influence on gest that in middle-aged type 2 diabetic onary heart disease in subjects with type 2 the low rate of major cardiac events. subjects the preclinical diagnosis of CHD, diabetes and in nondiabetic subjects with These data are enchanting, but without a followed by appropriate management, and without prior myocardial infarction. control study in our population, it is not may be advantageous. N Engl J Med 339:229–234, 1998 possible to assess the impact of these in- 13. Nesto RW: Screening for asymptomatic terventions on the rate of major cardiac coronary artery disease in diabetes. Dia- events. Nevertheless, we could not ob- References betes Care 22:1393–1395, 1999 serve any danger of the screening (16). 1. Bellet S, Roman L: The exercise test in di- 14. Rolka DB, Fagot-Campagna A, Narayan Another interesting finding from our abetic patients as studied by radioelectro- V: Aspirin use among adults with diabe- entire patient population is that the inci- cardiography. Circulation 36:245–254, tes: estimates from the Third National 1967 Health and Nutrition Examination Sur- dence of all cardiac events, 1.14/100 per- 2. Abenavoli T, Rubler S, Fisher VJ, Axelrod vey. Diabetes Care 24:197–201, 2001 sons/year, and the incidence of fatal HI, Zuckerman KP: Exercise testing with 15. Grundy SM, Benjamin IJ, Burke GL, Chait cardiac events, 0.08/100 persons/year, myocardial scintigraphy in asymptomatic A, Eckel RH, Howard BV, Mitch W, Smith are low. This low incidence is in contrast diabetic males. Circulation 63:54–64, 1981 SC Jr, Sowers JR: Diabetes and cardiovas- with the data generally reported in the 3. Hume L, Oakley GD, Boulton AJM, Hard- cular disease: a statement for healthcare

DIABETES CARE, VOLUME 25, NUMBER 11, NOVEMBER 2002 2035 5-Year follow-up of the MiSAD

professionals from the American Heart hain B, Stiegler H, Hormann A, Holle R: 5-year all-cause mortality. Diabetologia Association. Circulation 100:1134–1146, Predictors of 10-year macrovascular and 38:318–325, 1995 1999 overall mortality in patients with NIDDM: 20. UK Prospective Diabetes Study (UKPDS) 16. Sayer JW, Simmis AD: Investigation of the Munich General Practitioner Project. Group: Intensive blood-glucose control coronary artery disease in diabetes: is Diabetologia 39:1540–1545, 1996 with sulphonylureas or insulin compared screening of asymptomatic patients nec- 19. Muggeo M, Verlato G, Bonora E, Bressan with conventional treatment and risk of essary? Heart 78:525–526, 1997 E, Girotto S, Corbellini M, Gemma ML, complications in patients with type 2 di- 17. Geiss LS: In Diabetes in America. Washing- Moghetti P, Zenere M, Cacciatori V, Zop- abetes (UKPDS 33). Lancet 352:837–853, ton, DC, U.S. Govt. Printing Ofﬁce, 1995, pini G, de Marco R: The Verona Diabetes 1998 p. 233–257 (NIH publ. no. 95-1468) Study: a population-based survey on 21. Jolley D: The glitter of the t table. Lancet 18. Standl E, Balletshofer B, Dahl B, Weichen- known diabetes mellitus prevalence and 342:27–29, 1993

2036 DIABETES CARE, VOLUME 25, NUMBER 11, NOVEMBER 2002