Medical Care in Diabetesd2018 Diabetes Care 2018;41(Suppl. 1):S55 S64 /dc18-s006

Transcription

1 Diabetes Care Volume 41, Supplement 1, January 2018 S55 6. Glyemi Targets: Standards of Medial Care in Diabetesd2018 Diabetes Care 2018;41(Suppl. 1):S55 S64 Amerian Diabetes Assoiation The Amerian Diabetes Assoiation (ADA) Standards of Medial Care in Diabetes inludes ADA s urrent linial pratie reommendations and is intended to provide the omponents of diabetes are, general treatment goals and guidelines, and tools to evaluate quality of are. Members of the ADA Professional Pratie Committee, a multidisiplinary expert ommittee, are responsible for updating the Standards of Care annually, or more frequently as warranted. For a detailed desription of ADA standards, statements, and reports, as well as the evidene-grading system for ADA s linial pratie reommendations, please refer to the Standards of Care Introdution. Readers who wish to omment on the Standards of Care are invited to do so at professional.diabetes.org/soc. 6. GLYCEMIC TARGETS ASSESSMENT OF GLYCEMIC CONTROL Patient self-monitoring of blood gluose (SMBG)andA1Careavailabletohealth are providers and patients to assess the effetiveness and safety of a management plan on glyemi ontrol. Continuous gluose monitoring (CGM) also has an important role in assessing the effetiveness and safety of treatment in subgroups of patients with type 1 diabetes and in seleted patients with type 2 diabetes. Data indiate similar A1C and safety with the use of CGM ompared with SMBG (1). Reommendations Most patients using intensive insulin regimens (multiple-dose insulin or insulin pump therapy) should perform self-monitoring of blood gluose (SMBG) prior to meals and snaks, at bedtime, oasionally postprandially, prior to exerise, when they suspet low blood gluose, after treating low blood gluose until they are normoglyemi, and prior to ritial tasks suh as driving. B When presribed as part of a broad eduational program, SMBG may help to guide treatment deisions and/or self-management for patients taking less frequent insulin injetions B or noninsulin therapies. E When presribing SMBG, ensure that patients reeive ongoing instrution and regular evaluation of SMBG tehnique, SMBG results, and their ability to use SMBG data to adjust therapy. E When used properly, ontinuous gluose monitoring (CGM) in onjuntion with intensive insulin regimens is a useful tool to lower A1C in adults with type 1 diabetes who are not meeting glyemi targets. A CGM may be a useful tool in those with hypoglyemia unawareness and/or frequent hypoglyemi episodes. C Given the variable adherene to CGM, assess individual readiness for ontinuing CGM use prior to presribing. E Suggested itation: Amerian Diabetes Assoiation. 6. Glyemi targets: Standards of Medial Care in Diabetesd2018. Diabetes Care 2018; 41(Suppl. 1):S55 S by the Amerian Diabetes Assoiation. Readers may use this artile as long as the work is properly ited, the use is eduational and not for profit, and the work is not altered. More information is available at

2 S56 Glyemi Targets Diabetes Care Volume 41, Supplement 1, January 2018 When presribing CGM, robust diabetes eduation, training, and support are required for optimal CGM implementation and ongoing use. E People who have been suessfully using CGM should have ontinued aess after they turn 65 years of age. E Self-monitoring of Blood Gluose Major linial trials of insulin-treated patients have inluded SMBG as part of multifatorial interventions to demonstrate the benefit of intensive glyemi ontrol on diabetes ompliations. SMBG is thus an integral omponent of effetive therapy (2). SMBG allows patients to evaluate their individual response to therapy and assess whether glyemi targets are being ahieved. Integrating SMBG results into diabetes management an be a useful tool for guiding medial nutrition therapy and physial ativity, preventing hypoglyemia, and adjusting mediations (partiularly prandial insulin doses). Among patients with type 1 diabetes, there is a orrelation between greater SMBG frequeny and lower A1C (3). The patient s speifi needs and goals should ditate SMBG frequeny and timing. Optimization SMBG auray is dependent on the instrument and user, so it is important to evaluate eah patient s monitoring tehnique, both initially and at regular intervals thereafter. Optimal use of SMBG requires proper review and interpretation of the data, by both the patient and the provider. Among patients who hek their blood gluose at least one daily, many report taking no ation when results are high or low. In a yearlong study of insulinnaive patients with suboptimal initial glyemi ontrol, a group trained in strutured SMBG (a paper tool was used at least quarterly to ollet and interpret 7-point SMBG profiles taken on 3 onseutive days) redued their A1C by 0.3 perentage points more than the ontrol group (4). Patients should be taught how to use SMBG data to adjust food intake, exerise, or pharmaologi therapy to ahieve speifi goals. The ongoing need for and frequeny of SMBG should be reevaluated at eah routine visit to avoid overuse (5 7). SMBG is espeially important for insulin-treated patients to monitor for and prevent asymptomati hypoglyemia and hyperglyemia. Patients should be advised against purhasing or reselling preowned or seondhand test strips, as these may give inorret results. Only unopened vials of gluose test strips should be used to ensure SMBG auray. For Patients on Intensive Insulin Regimens Most patients using intensive insulin regimens (multiple-dose insulin or insulin pump therapy) should perform SMBG prior to meals and snaks, at bedtime, oasionally postprandially, prior to exerise, when they suspet low blood gluose, after treating low blood gluose until they are normoglyemi, and prior to ritial tasks suh as driving. For many patients, this will require testing 6 10 (or more) times daily, although individual needs may vary. A database study of almost 27,000 hildren and adolesents with type 1 diabetes showed that, after adjustment for multiple onfounders, inreased daily frequeny of SMBG was signifiantly assoiated with lower A1C ( 0.2% per additional test per day) and with fewer aute ompliations (8). For Patients Using Basal Insulin and/or Oral Agents The evidene is insuffiient regarding when to presribe SMBG and how often testing is needed for patients who do not use intensive insulin regimens, suh as those with type 2 diabetes using oral agents and/or basal insulin. For patients using basal insulin, assessing fasting gluose with SMBG to inform dose adjustments to ahieve blood gluose targets results in lower A1Cs (9,10). For individuals with type 2 diabetes on less intensive insulin therapy, more frequent SMBG (e.g., fasting, before/after meals) may be helpful, as inreased frequeny is assoiated with meeting A1C targets (11). Several randomized trials have alled into question the linial utility and osteffetiveness of routine SMBG in noninsulin-treated patients (12 15). Meta-analyses have suggested that SMBG an redue A1C by % at 6 months (16,17), but the effet was attenuated at 12 months in one analysis (16). A key onsideration is that performing SMBG alone does not lower blood gluose levels. To be useful, the information must be integrated into linial and self-management plans. Continuous Gluose Monitoring CGM measures interstitial gluose (whih orrelates well with plasma gluose), and most CGM devies inlude alarms for hypoand hyperglyemi exursions. The intermittent or flash CGM devie, very reently approved for adult use only (18), differs from previous CGM devies. Speifially, it does not have alarms, does not require alibration with SMBG, and does not ommuniate ontinuously (only on demand). It is reported to have a lower ost than traditional systems. A study in adults with well-ontrolled type 1 diabetes found that flash CGM users spent less time in hypoglyemia than those using SMBG (19). However, due to signifiant differenes between flashcgmandother CGM devies, more disussion is needed on outomes and regarding speifi reommendations. For most CGM systems, onfirmatory SMBG is required to make treatment deisions, though a randomized ontrolled trial of 226 adults suggested that an enhaned CGM devie ould be used safely and effetively without regular onfirmatory SMBG in patients with well-ontrolled type 1 diabetes at low risk of severe hypoglyemia (1). Two CGM devies are now approved by the U.S. Food and Drug Administration (FDA) for making treatment deisions without SMBG onfirmation (18,20), inluding the flash CGM devie. Although performed with older generation CGM devies, a 26-week randomized trial of 322 patients with type 1 diabetes showed that adults aged $25 years using intensive insulin therapy and CGM experiened a 0.5% redution in A1C (from ;7.6% to 7.1% [;60 mmol/mol to 54 mmol/mol]) ompared with those using intensive insulin therapy with SMBG (21). The greatest preditor of A1C lowering for all age-groups was frequeny of sensor use, whih was highest in those aged $25 years and lower in younger age-groups. Two linial trials in adults with type 1 diabetes not meeting A1C targets and using multiple daily injetions also found that the use of CGM ompared with usual are resulted in lower A1C levels than SMBG over weeks (22,23). Other small, short-term studies have demonstrated similar A1C redutions using CGM ompared with SMBG in adults with A1C levels $7% (53 mmol/mol) (24,25). A registry study of 17,317 partiipants onfirmed that more frequent CGM use is assoiated with lower A1C (26), whereas another study showed that hildren with.70% sensor use (i.e., $5 days per

3 are.diabetesjournals.org Glyemi Targets S57 week) missed fewer shool days (27). Small randomized ontrolled trials in adults and hildren with baseline A1C, % (53 58 mmol/mol) have onfirmed favorable outomes inluding a redued frequeny of hypoglyemia (defined as a blood gluose level,70 mg/dl [3.9 mmol/l]) and maintaining A1C,7% (53 mmol/mol) during the study period in groups using CGM, suggesting that CGM may provide further benefit for individuals with type 1 diabetes who already have good glyemi ontrol (28 30). A meta-analysis suggests that ompared with SMBG, CGM is assoiated with short-term A1C lowering of ;0.26% in insulin-treated patients (31). The longterm effetiveness of CGM needs to be determined. This tehnology may be partiularly useful in insulin-treated patients with hypoglyemia unawareness and/or frequent hypoglyemi episodes, although studies have not shown onsistent redutions in severe hypoglyemia (31 33). A CGM devie equipped with an automati low gluose suspend feature has been approved by the FDA. The Automation to Simulate Panreati Insulin Response (ASPIRE) trial of 247 patients with type 1 diabetes and doumented noturnal hypoglyemia showed that sensor-augmented insulin pump therapy with a low gluose suspend funtion signifiantly redued noturnal hypoglyemia over 3 months without inreasing A1C levels (34). These devies may offer the opportunity to redue hypoglyemia for those with a history of noturnal hypoglyemia. The FDA has also approved the first hybrid losed-loop system. The safety of hybrid losed-loop systems has been supported in the literature (35) and may have advantages over sensor-augmented pump therapy in speifi populations, suh as pregnant women with type 1 diabetes (36). Duetovariableadherene,optimal CGM use requires an assessment of individual readiness for the tehnology as well as initial and ongoing eduation and support (26,37). Additionally, providers need to provide robust diabetes eduation, training, and support for optimal CGM implementation and ongoing use. As people with type 1 or type 2 diabetes are living longer, healthier lives, individuals who have been suessfully using CGM should have ontinued aess to these devies after they turn 65 years of age (38). A1C TESTING Reommendations Perform the A1C test at least two times a year in patients who are meeting treatment goals (and who have stable glyemi ontrol). E Perform the A1C test quarterly in patients whose therapy has hanged or who are not meeting glyemi goals. E Point-of-are testing for A1C provides the opportunity for more timely treatment hanges. E A1C reflets average glyemia over approximately 3 months and has strong preditive value for diabetes ompliations (39,40). Thus, A1C testing should be performed routinely in all patients with diabetesdat initial assessment and as part of ontinuing are. Measurement approximately every 3 months determines whether patients glyemi targets have been reahed and maintained. The frequeny of A1C testing should depend on the linial situation, the treatment regimen, and the liniian s judgment. The use of point-of-are A1C testing may provide an opportunity for more timely treatment hanges during enounters between patients and providers. Patients with type 2 diabetes with stable glyemia well within target may do well with A1C testing only twie per year. Unstable or intensively managed patients (e.g., pregnant women with type 1 diabetes) may require testing more frequently than every 3 months (41). A1C Limitations The A1C test is an indiret measure of average glyemia and, as suh, is subjet to limitations. As with any laboratory test, there is variability in the measurement of A1C. Although suh variability is less on an intraindividual basis than that of blood gluose measurements, liniians should exerise judgment when using A1C as the sole basis for assessing glyemi ontrol, partiularly if the result is lose to the threshold that might prompt a hange in mediation therapy. Conditions that affet red blood ell turnover (hemolyti and other anemias, reent blood transfusion, use of drugs that stimulate erythropoesis, end-stage kidney disease, and pregnany) may result in disrepanies between the A1C result and the patient s true mean glyemia. Hemoglobin variants must be onsidered, partiularly when the A1C result does not orrelate with the patient s SMBG levels. Options for monitoring inlude more frequent and/ or different timing of SMBG or CGM use. Other measures of average glyemia suh as frutosamine and 1,5-anhydrogluitol are available, but their translation into average gluose levels and their prognosti signifiane are not as lear as for A1C. Though some variability exists among different individuals, generally the assoiation between mean gluose and A1C within an individual orrelates over time (42). A1C does not provide a measure of glyemi variability or hypoglyemia. For patients prone to glyemi variability, espeially patients with type 1 diabetes or type 2 diabetes with severe insulin defiieny, glyemi ontrol is best evaluated by the ombination of results from A1C and SMBG or CGM. A1C may also onfirm the auray of the patient s meter (or the patient s reported SMBG results) and the adequay of the SMBG testing shedule. A1C and Mean Gluose Table 6.1 shows the orrelation between A1C levels and mean gluose levels based on two studies: the international A1C- Derived Average Gluose (ADAG) study, whih assessed the orrelation between A1C and frequent SMBG and CGM in 507 adults (83% non-hispani whites) with type 1, type 2, and no diabetes (43), and an empirial study of the average blood gluose levels at premeal, postmeal, and bedtime assoiated with speified A1C levels using data from the ADAG trial (37). The Amerian Diabetes Assoiation (ADA) and the Amerian Assoiation for Clinial Chemistry have determined that the orrelation (r ) in the ADAG trial is strong enough to justify reporting both the A1C result and the estimated average gluose (eag) result when a liniian orders the A1C test. Cliniians should note that the mean plasma gluose numbers in the table are based on ;2,700 readings per A1C in the ADAG trial. In a reent report, mean gluose measured with CGM versus entral laboratory measured A1C in 387 partiipants in three randomized trials demonstrated that A1C may underestimate or overestimate mean gluose. Thus, as suggested, a patient s CGM profile has onsiderable potential for optimizing his or her glyemi management (42).

4 S58 Glyemi Targets Diabetes Care Volume 41, Supplement 1, January 2018 Table 6.1 Mean gluose levels for speified A1C levels (37,43) A1C Mean plasma gluose* Mean fasting gluose Mean premeal gluose Mean postmeal gluose Mean bedtime gluose % (mmol/mol) mg/dl mmol/l mg/dl mmol/l mg/dl mmol/l mg/dl mmol/l mg/dl mmol/l 6 (42) 126 ( ) 7.0 ( ) (37 47) 122 ( ) 6.8 ( ) 118 ( ) 6.5 ( ) 144 ( ) 8.0 ( ) 136 ( ) 7.5 ( ) (47 53) 142 ( ) 7.9 ( ) 139 ( ) 7.7 ( ) 164 ( ) 9.1 ( ) 153 ( ) 8.5 ( ) 7 (53) 154 ( ) 8.6 ( ) (53 58) 152 ( ) 8.4 ( ) 152 ( ) 8.4 ( ) 176 ( ) 9.8 ( ) 177 ( ) 9.8 ( ) (58 64) 167 ( ) 9.3 ( ) 155 ( ) 8.6 ( ) 189 ( ) 10.5 ( ) 175 ( ) 9.7 ( ) 8 (64) 183 ( ) 10.2 ( ) (64 69) 178 ( ) 9.9 ( ) 179 ( ) 9.9 ( ) 206 ( ) 11.4 ( ) 222 ( ) 12.3 ( ) 9 (75) 212 ( ) 11.8 ( ) 10 (86) 240 ( ) 13.4 ( ) 11 (97) 269 ( ) 14.9 ( ) 12 (108) 298 ( ) 16.5 ( ) Data in parentheses represent 95% CI, unless otherwise noted. A alulator for onverting A1C results into eag, in either mg/dl or mmol/l, is available at *These estimates are basedonadagdataof;2,700 gluose measurements over 3 months per A1C measurement in 507 adults with type 1, type 2, and no diabetes. The orrelation between A1C and average gluose was 0.92 (43). A1C Differenes in Ethni Populations and Children In the ADAG study, there were no signifiant differenes among raial and ethni groups in the regression lines between A1C and mean gluose, although the study was underpowered to detet a differene and there was a trend toward a differene between the Afrian/Afrian Amerian and non-hispani white ohorts, with higher A1C values observed in Afrians/Afrian Amerians ompared with non-hispani whites for a given mean gluose. Other studies have also demonstrated higher A1C levels in Afrian Amerians than in whites at a given mean gluose onentration (44,45). Moreover, Afrian Amerians heterozygous for the ommon hemoglobin variant HbS may have, for any level of mean glyemia, lower A1C by about 0.3 perentage points than those without the trait (46). Another geneti variant, X-linked gluose-6-phosphate dehydrogenase G202A, arried by 11% of Afrian Amerians, was assoiated with a derease in A1C of about 0.8% in hemizygous men and 0.7% in homozygous women ompared to those without the trait (47). A small study omparing A1C to CGM data in hildren with type 1 diabetes found a highly statistially signifiant orrelation between A1C and mean blood gluose, although the orrelation (r 5 0.7) was signifiantly lower than in the ADAG trial (48). Whether there are linially meaningful differenes in how A1C relates to average gluose in hildren or in different ethniities is an area for further study (44,49,50). Until further evidene is available, it seems prudent to establish A1C goals in these populations with onsideration of both individualized SMBG and A1C results. A1C GOALS For glyemi goals in hildren, please refer to Setion 12 Children and Adolesents. For glyemi goals in pregnant women, please refer to Setion 13 Management of Diabetes in Pregnany. Reommendations A reasonable A1C goal for many nonpregnant adults is,7% (53 mmol/mol). A Providers might reasonably suggest more stringent A1C goals (suh as,6.5% [48 mmol/mol]) for seleted individual patients if this

5 are.diabetesjournals.org Glyemi Targets S59 an be ahieved without signifiant hypoglyemia or other adverse effets of treatment (i.e., polypharmay). Appropriate patients might inlude those with short duration of diabetes, type 2 diabetes treated with lifestyle or metformin only, long life expetany, or no signifiant ardiovasular disease. C Less stringent A1C goals (suh as,8% [64 mmol/mol]) may be appropriate for patients with a history of severe hypoglyemia, limited life expetany, advaned mirovasular or marovasular ompliations, extensive omorbid onditions, or long-standing diabetes in whom the goal is diffiult to ahieve despite diabetes self-management eduation, appropriate gluose monitoring, and effetive doses of multiple gluose-lowering agents inluding insulin. B A1C and Mirovasular Compliations Hyperglyemia defines diabetes, and glyemi ontrol is fundamental to diabetes management. The Diabetes Control and Compliations Trial (DCCT) (2), a prospetive randomized ontrolled trial of intensive versus standard glyemi ontrol in patients with type 1 diabetes, showed definitively that better glyemi ontrol is assoiated with signifiantly dereased rates of development and progression of mirovasular (retinopathy [51], neuropathy, and diabeti kidney disease) ompliations. Follow-up of the DCCT ohorts in the Epidemiology of Diabetes Interventions and Compliations (EDIC) study (52) demonstrated persistene of these mirovasular benefits despite the fat that the glyemi separation between the treatment groups diminished and disappeared during follow-up. The Kumamoto Study (53) and UK Prospetive Diabetes Study (UKPDS) (54,55) onfirmed that intensive glyemi ontrol signifiantly dereased rates of mirovasular ompliations in patients with type 2 diabetes. Long-term follow-up of the UKPDS ohorts showed enduring effets of early glyemi ontrol on most mirovasular ompliations (56). Therefore, ahieving A1C targets of,7% (53 mmol/mol) has been shown to redue mirovasular ompliations of diabetes. Epidemiologial analyses of the DCCT (2) and UKPDS (57) demonstrate a urvilinear relationship between A1C and mirovasular ompliations. Suh analyses suggest that, on a population level, the greatest number of ompliations will beavertedbytakingpatientsfromvery poor ontrol to fair/good ontrol. These analyses also suggest that further lowering of A1C from 7% to 6% [53 mmol/mol to 42 mmol/mol] is assoiated with further redution in the risk of mirovasular ompliations, although the absolute risk redutions beome muh smaller. Given the substantially inreased risk of hypoglyemia in type 1 diabetes trials and with polypharmay in type 2 diabetes, the risks of lower glyemi targets outweigh the potential benefits on mirovasular ompliations. ACCORD, ADVANCE, and VADT Three landmark trials (Ation to Control Cardiovasular Risk in Diabetes [ACCORD], Ation in Diabetes and Vasular Disease: Preterax and Diamiron MR Controlled Evaluation [ADVANCE], and Veterans Affairs Diabetes Trial [VADT]) showed that lower A1C levels were assoiated with redued onset or progression of some mirovasular ompliations (58 60). The onerning mortality findings in the ACCORD trial (61), disussed below, and the relatively intense efforts required to ahieve near-euglyemia should also be onsidered when setting glyemi targets. However, on the basis of physiian judgment and patient preferenes, selet patients, espeially those with little omorbidity and long life expetany, may benefit from adopting more intensive glyemi targets (e.g., A1C target,6.5% [48 mmol/mol]) as long as signifiant hypoglyemia does not beome a barrier. A1C and Cardiovasular Disease Outomes Cardiovasular Disease and Type 1 Diabetes Cardiovasular disease (CVD) is a more ommon ause of death than mirovasular ompliations in populations with diabetes. There is evidene for a ardiovasular benefit of intensive glyemi ontrol after longterm follow-up of ohorts treated early in theourseoftype1diabetes.inthedcct, there was a trend toward lower risk of CVD events with intensive ontrol. In the 9-year post-dcct follow-up of the EDIC ohort, partiipants previously randomized to the intensive arm had a signifiant 57% redution in the risk of nonfatal myoardial infartion (MI), stroke, or ardiovasular death ompared with those previously randomized to the standard arm (62). The benefit of intensive glyemi ontrol in this ohort with type 1 diabetes has been shown to persist for several deades (63) and to be assoiated with a modest redution in all-ause mortality (64). Cardiovasular Disease and Type 2 Diabetes In type 2 diabetes, there is evidene that more intensive treatment of glyemia in newly diagnosed patients may redue long-term CVD rates. During the UKPDS, there was a 16% redution in CVD events (ombined fatal or nonfatal MI and sudden death) in the intensive glyemi ontrol arm that did not reah statistial signifiane (P ), and there was no suggestion of benefit on other CVD outomes (e.g., stroke). However, after 10 years of observational follow-up, those originally randomized to intensive glyemi ontrol had signifiant long-term redutions in MI (15% with sulfonylurea or insulin as initial pharmaotherapy, 33% with metformin as initial pharmaotherapy) and in all-ause mortality (13% and 27%, respetively) (56). ACCORD, ADVANCE, and VADT suggested no signifiant redution in CVD outomes with intensive glyemi ontrol in partiipants followed for years who had more advaned type 2 diabetes than UKPDS partiipants. All three trials were onduted in relatively older partiipants with longer known duration of diabetes (mean duration 8 11 years) and either CVD or multiple ardiovasular risk fators. The target A1C among intensive ontrol subjets was,6% (42 mmol/mol) in ACCORD,,6.5% (48 mmol/mol) in ADVANCE, and a 1.5% redution in A1C ompared with ontrol subjets in VADT, with ahieved A1C of 6.4% vs. 7.5% (46 mmol/mol vs. 58 mmol/mol) in ACCORD, 6.5% vs. 7.3% (48 mmol/mol vs. 56 mmol/mol) in ADVANCE, and 6.9% vs. 8.4% (52 mmol/mol vs. 68 mmol/mol) in VADT. Details of these studies are reviewed extensively in Intensive Glyemi Control and the Prevention of Cardiovasular Events: Impliations of the ACCORD, ADVANCE, and VA Diabetes Trials (65). The glyemi ontrol omparison in ACCORD was halted early due to an inreased mortality rate in the intensive ompared with the standard treatment arm (1.41% vs. 1.14% per year; hazard ratio 1.22 [95% CI ]), with a similar inrease in ardiovasular deaths. Analysis

6 S60 Glyemi Targets Diabetes Care Volume 41, Supplement 1, January 2018 Table 6.2 Summary of glyemi reommendations for many nonpregnant adults with diabetes A1C <7.0% (53 mmol/mol)* Preprandial apillary plasma gluose mg/dl* ( mmol/l) Peak postprandial apillary plasma gluose <180 mg/dl* (10.0 mmol/l) *More or less stringent glyemi goals may be appropriate for individual patients. Goals should be individualized based on duration of diabetes, age/life expetany, omorbid onditions, known CVD or advaned mirovasular ompliations, hypoglyemia unawareness, and individual patient onsiderations. Postprandial gluose may be targeted if A1C goals are not met despite reahing preprandial gluose goals. Postprandial gluose measurements should be made 1 2 h after the beginning of the meal, generally peak levels in patients with diabetes. of the ACCORD data did not identify a lear explanation for the exess mortality in the intensive treatment arm (61). Longer-term follow-up has shown no evidene of ardiovasular benefit or harm in the ADVANCE trial (66). The end-stage renal disease rate was lower in the intensive treatment group over follow-up. However, 10-year follow-up of the VADT ohort (67) showed a redution in the risk of ardiovasular events (52.7 [ontrol group] vs [intervention group] events per 1,000 person-years) with no benefit in ardiovasular or overall mortality. Heterogeneity of mortality effets aross studies was noted, whih may reflet differenes in glyemi targets, therapeuti approahes, and population harateristis (68). Mortality findings in ACCORD (61) and subgroup analyses of VADT (69) suggest that the potential risks of intensive glyemi ontrol may outweigh its benefits in higher-risk patients. In all three trials, severe hypoglyemia was signifiantly more likely in partiipants who were randomly assigned to the intensive glyemi ontrol arm. Those patients with long duration of diabetes, a known history of hypoglyemia, advaned atheroslerosis, or advaned age/frailty may benefit from less aggressive targets (70,71). Providers should be vigilant in preventing hypoglyemia and should not aggressively attempt to ahieve near-normal A1C levels in patients in whom suh targets annot be safely and reasonably ahieved. Severe or frequent hypoglyemia is an absolute indiation for the modifiation of treatment regimens, inluding setting higher glyemi goals. Many fators, inluding patient preferenes, should be taken into aount when developing a patient s individualized goals (Table 6.2). proposes optimal targets, but eah target must be individualized to the needs of eah patient and his or her disease fators. When possible, suh deisions should be made with the patient, refleting his or her preferenes, needs, and values. Fig. 6.1 is not designed to be applied rigidly but to be used as a broad onstrut to guide linial deision-making (72), in both type 1 and type 2 diabetes. Reommended glyemi targets for many nonpregnant adults are shown in Table 6.2. The reommendations inlude blood gluose levels that appear to orrelate with ahievement of an A1C of,7% (53 mmol/mol). The issue of preprandial versus postprandial SMBG targets is omplex (73). Elevated posthallenge (2-h oral gluose tolerane test) gluose values have been assoiated with inreased ardiovasular risk independent of fasting plasma gluose in some epidemiologial studies, but intervention trials have not shown postprandial gluose to be a ardiovasular risk fator independent of A1C. In subjets with diabetes, surrogate measures of vasular pathology, suh as endothelial dysfuntion, are negatively affeted by postprandial hyperglyemia. It is lear that postprandial hyperglyemia, like preprandial hyperglyemia, ontributes to elevated A1C levels, with its relative ontribution being greater at A1C levels that are loser to 7% (53 mmol/mol). However, outome studies have learly shown A1C to be the primary preditor of ompliations, and landmark trials of glyemi ontrol suh as the DCCT and UKPDS relied overwhelmingly on preprandial SMBG. Additionally, a randomized ontrolled trial in patients with known CVD found no CVD benefit of insulin regimens targeting postprandial gluose ompared with those targeting preprandial gluose (74). Therefore, it is reasonable for postprandial testing to be reommended for individuals who have premeal gluose values within target but have A1C values above target. Measuring postprandial plasma gluose 1 2 h after the start of a meal and using treatments aimed at A1C and Glyemi Targets Numerous aspets must be onsidered when setting glyemi targets. The ADA Figure 6.1 Depited are patient and disease fators used to determine optimal A1C targets. Charateristis and prediaments toward the left justify more stringent efforts to lower A1C; those toward the right suggest less stringent efforts. Adapted with permission from Inzuhi et al. (72).

7 are.diabetesjournals.org Glyemi Targets S61 Table 6.3 Classifiation of hypoglyemia* Level Glyemi riteria Desription Hypoglyemia alert value (level 1) #70 mg/dl (3.9 mmol/l) Suffiiently low for treatment with fast-ating arbohydrate and dose adjustment of gluose-lowering therapy Clinially signifiant hypoglyemia (level 2),54 mg/dl (3.0 mmol/l) Suffiiently low to indiate serious, linially important hypoglyemia Severe hypoglyemia (level 3) No speifi gluose threshold Hypoglyemia assoiated with severe ognitive impairment requiring external assistane for reovery *Adapted from ref. 75. reduing postprandial plasma gluose values to,180 mg/dl (10.0 mmol/l) may help to lower A1C. An analysis of data from 470 partiipants in the ADAG study (237 with type 1 diabetes and 147 with type 2 diabetes) found that atual average gluose levels assoiated with onventional A1C targets were higher than older DCCT and ADA targets (Table 6.1) (37,39). These findings support that premeal gluose targets may be relaxed without undermining overall glyemi ontrol as measured by A1C. These data prompted the revision in the ADA-reommended premeal gluose target to mg/dl ( mmol/l) but did not affet the definition of hypoglyemia. HYPOGLYCEMIA Reommendations Individuals at risk for hypoglyemia should be asked about symptomati and asymptomati hypoglyemia at eah enounter. C Gluose (15 20 g) is the preferred treatment for the onsious individual with blood gluose #70 mg/dl [3.9 mmol/l]), although any form of arbohydrate that ontains gluose may be used. Fifteen minutes after treatment, if SMBG shows ontinued hypoglyemia, the treatment should be repeated. One SMBG returns to normal, the individual should onsume a meal or snak to prevent reurrene of hypoglyemia. E Gluagon should be presribed for all individuals at inreased risk of linially signifiant hypoglyemia, defined as blood gluose,54 mg/dl (3.0 mmol/l), so it is available should it be needed. Caregivers, shool personnel, or family members of these individuals should know where it is and when and how to administer it. Gluagon administration is not limited to health are professionals. E Hypoglyemia unawareness or one or more episodes of severe hypoglyemia should trigger reevaluation of the treatment regimen. E Insulin-treated patients with hypoglyemia unawareness or an episode of linially signifiant hypoglyemia should be advised to raise their glyemi targets to stritly avoid hypoglyemia for at least several weeks in order to partially reverse hypoglyemia unawareness and redue risk of future episodes. A Ongoing assessment of ognitive funtion is suggested with inreased vigilane for hypoglyemia by the liniian, patient, and aregivers if low ognition or delining ognition is found. B Hypoglyemia is the major limiting fator in the glyemi management of type 1 and type 2 diabetes. Reommendations from the International Hypoglyemia Study Group regarding the lassifiation of hypoglyemia in linial trials are outlined in Table 6.3 (75). Of note, this lassifiation sheme onsiders a blood gluose,54 mg/dl (3.0 mmol/l) deteted by SMBG, CGM (for at least 20 min), or laboratory measurement of plasma gluose as suffiiently low to indiate linially signifiant hypoglyemia that should be inluded in reports of linial trials of gluose-lowering drugs for the treatment of diabetes (75). However, a hypoglyemia alert value of #70 mg/dl (3.9 mmol/l) an be important for therapeuti dose adjustment of gluose-lowering drugs in linial are and is often related to symptomati hypoglyemia. Severe hypoglyemia is defined as severe ognitive impairment requiring assistane from another person for reovery (76). Symptoms of hypoglyemia inlude, but are not limited to, shakiness, irritability, onfusion, tahyardia, and hunger. Hypoglyemia may be inonvenient or frightening to patients with diabetes. Severe hypoglyemia may be reognized or unreognized and an progress to loss of onsiousness, seizure, oma, or death. It is reversed by administration of rapid-ating gluose or gluagon. Clinially signifiant hypoglyemia an ause aute harm to thepersonwithdiabetesorothers,espeially if it auses falls, motor vehile aidents, or other injury. A large ohort study suggested that among older adults with type 2 diabetes, a history of severe hypoglyemia was assoiated with greater risk of dementia (77). Conversely, in a substudy of the ACCORD trial, ognitive impairment at baseline or deline in ognitive funtion during the trial was signifiantly assoiated with subsequent episodes of severe hypoglyemia (78). Evidene from DCCT/EDIC, whih involved adolesents and younger adults with type 1 diabetes, found no assoiation between frequeny of severe hypoglyemia and ognitive deline (79), as disussed in Setion 12 Children and Adolesents. Severe hypoglyemia was assoiated with mortality in partiipants in both the standard and the intensive glyemia arms of the ACCORD trial, but the relationships between hypoglyemia, ahieved A1C, and treatment intensity were notstraightforward. An assoiation of severe hypoglyemia with mortality was also found in the ADVANCE trial (80). An assoiation between self-reported severe hypoglyemia and 5-year mortality has also been reported in linial pratie (81). Younghildrenwithtype1diabetesand the elderly, inluding those with type 1 and type 2 diabetes (77,82), are noted as partiularly vulnerable to linially signifiant hypoglyemia beause of their redued ability to reognize hypoglyemi symptoms and effetively ommuniate their needs. Individualized gluose targets, patient eduation, dietary intervention (e.g., bedtime snak to prevent overnight hypoglyemia when speifially needed to treat

8 S62 Glyemi Targets Diabetes Care Volume 41, Supplement 1, January 2018 low blood gluose), exerise management, mediation adjustment, gluose monitoring, and routine linial surveillane may improve patient outomes (76). CGM with automated low gluose suspend has been showntobeeffetiveinreduinghypoglyemia in type 1 diabetes (34). For patients with type 1 diabetes with severe hypoglyemia and hypoglyemia unawareness that persists despite medial treatment, human islet transplantation may be an option, but the approah remains experimental (83,84). In 2015, the ADA hanged its preprandial glyemi target from mg/dl ( mmol/l) to mg/dl ( mmol/l). This hange reflets the results of the ADAG study, whih demonstrated that higher glyemi targets orresponded to A1C goals (37). An additional goal of raising the lower range of the glyemi target was to limit overtreatment and provide a safety margin in patients titrating gluose-lowering drugs suh as insulin to glyemi targets. Hypoglyemia Treatment Providers should ontinue to ounsel patients to treat hypoglyemia with fastating arbohydrates at the hypoglyemia alert value of 70 mg/dl (3.9 mmol/l) or less. Hypoglyemia treatment requires ingestion of gluose- or arbohydrateontaining foods. The aute glyemi response orrelates better with the gluose ontent of food than with the arbohydrate ontent of food. Pure gluose is the preferred treatment, but any form of arbohydrate that ontains gluose will raise blood gluose. Added fat may retard and then prolong the aute glyemi response. In type 2 diabetes, ingested protein may inrease insulin response without inreasing plasma gluose onentrations (85). Therefore, arbohydrate soures high in protein should not be used to treat or prevent hypoglyemia. Ongoing insulin ativity or insulin seretagogues may lead to reurrent hypoglyemia unless further food is ingested after reovery. One the gluose returns to normal, the individual should be ounseled to eat a meal or snak to prevent reurrent hypoglyemia. Gluagon The use of gluagon is indiated for the treatment of hypoglyemia in people unable or unwilling to onsume arbohydrates by mouth. Those in lose ontat with, or having ustodial are of, people with hypoglyemia-prone diabetes (family members, roommates, shool personnel, hild are providers, orretional institution staff, or oworkers) should be instruted on the use of gluagon kits inluding where the kit is and when and how to administer gluagon. An individual does not need to be a health are professional to safely administer gluagon. Care should be taken to ensure that gluagon kits are not expired. Hypoglyemia Prevention Hypoglyemia prevention is a ritial omponent of diabetes management. SMBG and, for some patients, CGM are essential tools to assess therapy and detet inipient hypoglyemia. Patients should understand situations that inrease their risk of hypoglyemia, suh as fasting for tests or proedures, delayed meals, during or after intense exerise, and during sleep. Hypoglyemia may inrease the risk of harm to self or others, suh as with driving. Teahing people with diabetes to balane insulin use and arbohydrate intake and exerise are neessary, but these strategies are not always suffiient for prevention. In type 1 diabetes and severely insulindefiient type 2 diabetes, hypoglyemia unawareness (or hypoglyemia-assoiated autonomi failure) an severely ompromise stringent diabetes ontrol and quality of life. This syndrome is haraterized by defiient ounterregulatory hormone release, espeially in older adults, and a diminished autonomi response, whih both are risk fators for, and aused by, hypoglyemia. A orollary to this viious yle is that several weeks of avoidane of hypoglyemia has been demonstrated to improve ounterregulation and hypoglyemia awareness in many patients (86). Hene, patients with one or more episodes of linially signifiant hypoglyemia may benefit fromatleast short-term relaxation of glyemi targets. INTERCURRENT ILLNESS For further information on management of patients with hyperglyemia in the hospital, please refer to Setion 14 Diabetes Care in the Hospital. Stressful events (e.g., illness, trauma, surgery, et.) may worsen glyemi ontrol and preipitate diabeti ketoaidosis or nonketoti hyperosmolar state, life-threatening onditions that require immediate medial are to prevent ompliations and death. Any ondition leading to deterioration in glyemi ontrol neessitates more frequent monitoring of blood gluose; ketosisprone patients also require urine or blood ketone monitoring. If aompanied by ketosis, vomiting, or alteration in the level of onsiousness, marked hyperglyemia requires temporary adjustment of the treatment regimen and immediate interation with the diabetes are team. The patient treated with noninsulin therapies or medial nutrition therapy alone may temporarily require insulin. Adequate fluid and alori intake must be ensured. Infetion or dehydration is more likely to neessitate hospitalization of the person with diabetes than the person without diabetes. A physiian with expertise in diabetes management should treat the hospitalized patient. For further information on the management of diabeti ketoaidosis and the hyperglyemi nonketoti hyperosmolar state, please refer to the ADA onsensus report Hyperglyemi Crises in Adult Patients With Diabetes (87). Referenes 1. Aleppo G, Ruedy KJ, Riddlesworth TD, et al.; REPLACE-BG Study Group. REPLACE-BG: a randomized trial omparing ontinuous gluose monitoring with and without routine blood gluose monitoring in adults with well-ontrolled type 1 diabetes. Diabetes Care 2017;40: The Diabetes Control and Compliations Trial Researh Group. The effet of intensive treatment of diabetes on the development and progression of long-term ompliations in insulin-dependent diabetes mellitus. N Engl J Med 1993;329: Miller KM, Bek RW, Bergenstal RM, et al.; T1D Exhange Clini Network. Evidene of a strong assoiation between frequeny of self-monitoring of blood gluose and hemoglobin A 1 levels in T1D Exhange lini registry partiipants. Diabetes Care 2013;36: Polonsky WH, Fisher L, Shikman CH, et al. Strutured self-monitoring of blood gluose signifiantly redues A1C levels in poorly ontrolled, noninsulin-treated type 2 diabetes: results from the Strutured Testing Program study. Diabetes Care 2011;34: Gellad WF, Zhao X, Thorpe CT, Mor MK, Good CB, Fine MJ. Dual use of Department of Veterans Affairs and Mediare benefits and use of test strips in veterans with type 2 diabetes mellitus. JAMA Intern Med 2015;175: Grant RW, Huang ES, Wexler DJ, et al. Patients who self-monitor blood gluose and their unused testing results. Am J Manag Care 2015;21:e119 e Endorine Soiety. Choosing wisely [Internet], Available from Aessed 18 August Ziegler R, Heidtmann B, Hilgard D, Hofer S, Rosenbauer J, Holl R; DPV-Wiss-Initiative. Frequeny of SMBG orrelates with HbA1 and aute ompliations in hildren and adolesents with type 1 diabetes. Pediatr Diabetes 2011;12: Rosenstok J, Davies M, Home PD, Larsen J, Koenen C, Shernthaner G. A randomised, 52-week,

9 are.diabetesjournals.org Glyemi Targets S63 treat-to-target trial omparing insulin detemir with insulin glargine when administered as add-on to gluose-lowering drugs in insulin-naive people with type 2 diabetes. Diabetologia 2008;51: Garber AJ. Treat-to-target trials: uses, interpretation and review of onepts. Diabetes Obes Metab 2014;16: Elgart JF, González L, Prestes M, Rui E, Gagliardino JJ. Frequeny of self-monitoring blood gluose and attainment of HbA1 target values. Ata Diabetol 2016;53: Farmer A, Wade A, Goyder E, et al. Impat of self monitoring of blood gluose in the management of patients with non-insulin treated diabetes: open parallel group randomised trial. BMJ 2007;335: O Kane MJ, Bunting B, Copeland M, Coates VE; ESMON study group. Effiay of self monitoring of blood gluose in patients with newly diagnosed type 2 diabetes (ESMON study): randomised ontrolled trial. BMJ 2008;336: Simon J, Gray A, Clarke P, Wade A, Neil A, Farmer A; Diabetes Glyaemi Eduation and Monitoring Trial Group. Cost effetiveness of self monitoring of blood gluose in patients with non-insulin treated type 2 diabetes: eonomi evaluation of data from the DiGEM trial. BMJ 2008;336: Young LA, Buse JB, Weaver MA, et al.; Monitor Trial Group. Gluose self-monitoring in noninsulin-treated patients with type 2 diabetes in primary are settings: a randomized trial. JAMA Intern Med 2017;177: Malanda UL, Welshen LMC, Riphagen II, Dekker JM, Nijpels G, Bot SDM. Self-monitoring of blood gluose in patients with type 2 diabetes mellitus who are not using insulin. Cohrane Database Syst Rev 2012;1:CD Willett LR. ACP Journal Club. Meta-analysis: self-monitoring in non-insulin-treated type 2 diabetes improved HbA1 by 0.25%. Ann Intern Med 2012;156:JC6 JC U.S. Food and Drug Administration. FDA approves first ontinuous gluose monitoring system for adults not requiring blood sample alibration [Internet]. Available from NewsEvents/Newsroom/PressAnnounements/ um htm. Aessed 2 Otober Bolinder J, Antuna R, Geelhoed-Duijvestijn P, Kröger J, Weitgasser R. Novel gluose-sensing tehnology and hypoglyaemia in type 1 diabetes: a multientre, non-masked, randomised ontrolled trial. Lanet 2016;388: U.S. Food and Drug Administration. FDA expands indiation for ontinuous gluose monitoring system, first to replae fingerstik testing for diabetes treatment deisions [Internet]. Available from newsroom/pressannounements/um htm. Aessed 14 September Tamborlane WV, Bek RW, Bode BW, et al.; Juvenile Diabetes Researh Foundation Continuous Gluose Monitoring Study Group. Continuous gluose monitoring and intensive treatment of type 1 diabetes. N Engl J Med 2008;359: Lind M, Polonsky W, Hirsh IB, et al. Continuous gluose monitoring vs onventional therapy for glyemi ontrol in adults with type 1 diabetes treated with multiple daily insulin injetions: the GOLD randomized linial trial. JAMA 2017;317: Bek RW, Riddlesworth T, Ruedy K, et al. Effet of ontinuous gluose monitoring on glyemi ontrol in adults with type 1 diabetes using insulin injetions: the DIAMOND randomized linial trial. JAMA 2017;317: Deiss D, Bolinder J, Riveline J-P, et al. Improved glyemi ontrol in poorly ontrolled patients with type 1 diabetes using real-time ontinuous gluose monitoring. Diabetes Care 2006;29: O Connell MA, Donath S, O Neal DN, et al. Glyaemi impat of patient-led use of sensorguided pump therapy in type 1 diabetes: a randomised ontrolled trial. Diabetologia 2009;52: Wong JC, Foster NC, Maahs DM, et al.; T1D Exhange Clini Network. Real-time ontinuous gluose monitoring among partiipants in the T1D Exhange lini registry. Diabetes Care 2014;37: Hommel E, Olsen B, Battelino T, et al.; SWITCH Study Group. Impat of ontinuous gluose monitoring on quality of life, treatment satisfation, and use of medial are resoures: analyses from the SWITCH study. Ata Diabetol 2014;51: Battelino T, Phillip M, Bratina N, Nimri R, Oskarsson P, Bolinder J. Effet of ontinuous gluose monitoring on hypoglyemia in type 1 diabetes. Diabetes Care 2011;34: Bek RW, Hirsh IB, Laffel L, et al.; Juvenile Diabetes Researh Foundation Continuous Gluose Monitoring Study Group. The effet of ontinuous gluose monitoring in well-ontrolled type 1 diabetes. Diabetes Care 2009;32: Bode B, Bek RW, Xing D, et al.; Juvenile Diabetes Researh Foundation Continuous Gluose Monitoring Study Group. Sustained benefit of ontinuous gluose monitoring on A1C, gluose profiles, and hypoglyemia in adults with type 1 diabetes. Diabetes Care 2009;32: Yeh H-C, Brown TT, Maruthur N, et al. Comparative effetiveness and safety of methods of insulin delivery and gluose monitoring for diabetes mellitus: a systemati review and meta-analysis. Ann Intern Med 2012;157: Choudhary P, Ramasamy S, Green L, et al. Real-time ontinuous gluose monitoring signifiantlyredues severehypoglyemiain hypoglyemiaunaware patients with type 1 diabetes. Diabetes Care 2013;36: Choudhary P, Rikels MR, Senior PA, et al. Evidene-informed linial pratie reommendations for treatment of type 1 diabetes ompliated by problemati hypoglyemia. Diabetes Care 2015;38: Bergenstal RM, Klonoff DC, Garg SK, et al.; ASPIRE In-Home Study Group. Threshold-based insulin-pump interruption for redution of hypoglyemia. N Engl J Med 2013;369: Bergenstal RM, Garg S, Weinzimer SA, et al. Safety of a hybrid losed-loop insulin delivery system in patients with type 1 diabetes. JAMA 2016; 316: Stewart ZA, Wilinska ME, Hartnell S, et al. Closed-loop insulin delivery during pregnany in women with type 1 diabetes. N Engl J Med 2016; 375: Wei N, Zheng H, Nathan DM. Empirially establishing blood gluose targets to ahieve HbA 1 goals. Diabetes Care 2014;37: Herman WH, Ilag LL, Johnson SL, et al. A linial trial of ontinuous subutaneous insulin infusion versus multiple daily injetions in older adults with type2 diabetes. Diabetes Care 2005;28: Albers JW, Herman WH, Pop-Busui R, et al.; Diabetes Control and Compliations Trial/Epidemiology of Diabetes Interventions and Compliations Researh Group. Effet of prior intensive insulin treatment during the Diabetes Control and Compliations Trial (DCCT) on peripheral neuropathy in type 1 diabetes during the Epidemiology of Diabetes Interventions and Compliations (EDIC) Study. Diabetes Care 2010;33: Stratton IM, Adler AI, Neil HAW, et al. Assoiation of glyaemia with marovasular and mirovasular ompliations of type 2 diabetes (UKPDS 35): prospetive observational study. BMJ 2000;321: Jovanovič L, Savas H, Mehta M, Trujillo A, Pettitt DJ. Frequent monitoring of A1C during pregnany as a treatment tool to guide therapy. Diabetes Care 2011;34: Bek RW, Connor CG, Mullen DM, Wesley DM, Bergenstal RM. The fallay of average: how using HbA 1 alone to assess glyemi ontrol an be misleading. Diabetes Care 2017;40: Nathan DM, Kuenen J, Borg R, Zheng H, Shoenfeld D, Heine RJ; A1-Derived Average Gluose Study Group. Translating the A1C assay into estimated average gluose values. Diabetes Care 2008;31: Selvin E. Are there linial impliations of raial differenes in HbA 1? A differene, to be a differene, must make a differene. Diabetes Care 2016;39: Bergenstal RM, Gal RL, Connor CG, et al.; T1D Exhange Raial Differenes Study Group. Raial differenes in the relationship of gluose onentrations and hemoglobin A1 levels. Ann Intern Med 2017;167: Lay ME, Wellenius GA, Sumner AE, et al. Assoiation of sikle ell trait with hemoglobin A1 in Afrian Amerians. JAMA 2017;317: Wheeler E, Leong A, Liu C-T, et al.; EPIC-CVD Consortium; EPIC-InterAt Consortium; Lifelines Cohort Study. Impat of ommon geneti determinants of hemoglobin A1 on type 2 diabetes risk and diagnosis in anestrally diverse populations: a transethni genome-wide meta-analysis. PLoS Med 2017;14:e Wilson DM, Kollman; Diabetes Researh in Children Network (DireNet) Study Group. Relationship of A1C to gluose onentrations in hildren with type 1 diabetes: assessments by high-frequeny gluose determinations by sensors. Diabetes Care 2008;31: Buse JB, Kaufman FR, Linder B, Hirst K, El Ghormli L, Willi S; HEALTHY Study Group. Diabetes sreening with hemoglobin A 1 versus fasting plasma gluose in a multiethni middle-shool ohort. Diabetes Care 2013;36: Kamps JL, Hempe JM, Chalew SA. Raial disparity in A1C independent of mean blood gluose in hildren with type 1 diabetes. Diabetes Care 2010;33: Diabetes Control and Compliations Trial (DCCT)/Epidemiology of Diabetes Interventions and Compliations (EDIC) Researh Group. Effet of intensive diabetes therapy on the progression