Approaches to treatment 2: Comparison of American Association of Clinical Endocrinologists (AACE) and American Diabetes Association (ADA) type 2 diabetes treatment guidelines

Abstract

总的来说,2015年美国糖尿病协会(American Diabetes Association,ADA)1与美国临床内分泌学家协会(American Association of Clinical Endocrinologists,AACE)2推荐的2型糖尿病治疗指南都对当前控制糖尿病高血糖的方法与概念进行了深刻的分析。 ADA的方法是以选择血糖控制目标为起点,在控制血糖目标之前强调要控制HbA1c,建议通常的HbA1c目标应该设定为7%,虽然也指出了对于一名特定的患者来说更高或者更低水平的HbA1c目标也可能是合适的,这主要依赖于患者的医疗条件(疾病病程、预期寿命、合并疾病,特别是强调了血管并发症),并且还依赖于患者的态度、预期的努力程度以及可用的资源,这些因素都是可改变的,除此之外还有药物不良反应的风险,特别是低血糖。二甲双胍是推荐的首选初始用药,间隔3个月后可以增加到2种药物的联合使用,接着是3种药物的联合使用。另外还对6类治疗药物进行了陈述,其中胰岛素被认为具有“最高”效能;胰高血糖素样肽-1(GLP-1)受体激动剂(RA)、噻唑烷二酮类(TZD)以及磺脲类(SU)具有“高”效能;而二肽基肽酶(DPP)-4抑制剂与钠-葡萄糖共转运体(SGLT)-2抑制剂具有“中等”效能1。提到了将后面这两类药物引入推荐指南是“治疗方法的主要变化”1,但是没有推荐联合使用GLP-1 RAs进行治疗,这主要是因为缺乏临床试验证据。特别是明确提到了膀胱癌与吡格列酮并不相关,以及心脏衰竭在理论上与使用DPP-4抑制剂无关1;但假设根据一项新的来源于临床试验的证据这个结论将被修改3。在特定的治疗方案中,为了使治疗费用最小化需要将DPP-4抑制剂、SGLT2抑制剂、GLP-1 RAs以及胰岛素类似物排除在外;为了避免低血糖需要将SU与胰岛素排除在外;而为了避免体重增加还需将TZDs排除在外1。 如果HbA1c> 9%,推荐启用一种基础胰岛素进行治疗,剂量为10个单位或者0.1-0.2 U/kg体重,根据自我监测的空腹血糖每周调整剂量1到2次1。根据ADA临床实践标准定义的空腹血糖推荐目标为130 mg/dL(7.2 mmol/L),餐后峰值(假定为餐后1小时)血糖目标为180 mg/dL(10.0 mmol/L])4。对于使用基础胰岛素治疗后空腹血糖达标但是总体血糖控制却没有达标的患者来说,推荐的治疗方法有3种:(i)加用一种GLP-1 RA;(ii)在进食最多的餐前加用一针速效胰岛素;或者(iii)改成一种每日注射两次的预混胰岛素,而每餐前注射一次速效胰岛素是最终的选择1。 AACE推荐的(治疗策略)2纳入了所有已被批准的降糖药。4组药物的区别如下:(i)主要降低空腹血糖的药物(例如二甲双胍、TZDs与基础胰岛素);(ii)主要降低餐后血糖的药物(例如α-葡萄糖苷酶抑制剂[AGI]、胰淀素类似物普兰林肽、DPP-4抑制剂、格列奈类以及短效和/或速效胰岛素);(iii)具有两种作用的药物(例如GLP-1 RA、SU、SGLT2抑制剂以及混合性胰岛素制剂;(iv)适度有效的药物如胆汁酸螯合剂考来维仑以及一种多巴胺D2受体激动剂溴隐亭的快速吸收剂型(BCR)。AACE的治疗策略聚焦于安全性以及降糖获益:SU、格列奈类与胰岛素的低血糖问题;胰岛素、SU与TZDs的体重增加效应;以及GLP-1 RA与SGLT2抑制剂的体重减少效应2。中度肾功能受损导致二甲双胍的禁忌使用,而重度肾功能受损可导致艾塞那肽的禁忌使用。其他安全性问题包括使用SGLT2抑制剂治疗后的泌尿生殖道感染与血容量下降,以及使用GLP1 RAs治疗后的恶心、呕吐与腹泻。使用二甲双胍、AGI、考来维仑与BCR治疗后也可能出现胃肠道不良反应。心血管问题,特别是心脏衰竭,是引证过的TZDs不良效应;SU有可能会导致心血管不良效应(并且提到了SGLT2抑制剂可导致低密度脂蛋白胆固醇水平升高,虽然它的真实临床影响还尚未明确);而TZDs与SGLT2抑制剂还有潜在的导致骨量丢失的副作用2。 AACE在推荐治疗策略时有一个重要原则,就是要制定个体化的治疗计划,主要是根据基线HbA1c来确定。与ADA以及欧洲糖尿病研究协会(European Association for the Study of Diabetes,EASD)不同的是,AACE将药物选择按照以下方案进行了分级。如果HbA1c水平< 7.5%,推荐单药治疗,首选二甲双胍,其次是GLP-1 RAs,而SGLT2抑制剂是第三选择,接着是DPP-4抑制剂、AGI与TZD,最后是SU和/或格列奈类2。TZDs以及SU的降糖效能与排行在它们前面的药物相比有可能都更高,但是因为它们各自的副作用而导致排名靠后。对于当前HbA1c> 7.5%并且< 9.0%的患者或者是(与ADA推荐的一样)经过3个月的单药治疗之后血糖仍然没有达标的患者来说,推荐二联治疗法,接着是三联治疗法,再此给出了分层选择的方案,但是组织起来有点差别:为了使患者的风险更低,推荐的二联治疗法是在二甲双胍的基础上首选加用GLP-1 RA,次选SGLT2抑制剂,第三选择为DPP-4抑制剂,第四选择是TZD,第五选择是基础胰岛素,接着是考来维仑、BCR、AGI,SU与格列奈类再次成为最后的选择2。在三联治疗法中,DPP-4抑制剂的排名被下移到了TZDs与基础胰岛素之后。如果当前的HbA1c> 9.0%,推荐进行二联治疗或者三联治疗,除非患者出现了高血糖症状,对于这些病例来说需要使用胰岛素联用或不联用其他降糖药进行治疗2。 在AACE治疗方法中推荐的胰岛素使用策略是:HbA1c< 8%与> 8%的启始基础胰岛素剂量分别为0.1-0.2与0.2-0.3 U/kg。推荐每3日根据空腹血糖水平进行基础胰岛素剂量的滴定:>180 mg/dL(10.0 mmol/L)增加20%,140-180 mg/dL(7.8-10.0 mmol/L)增加10%,110-139 mg/dL(6.1-7.8 mmol/L)增加1个单位,而低血糖时如果血糖< 70 mg/dL(3.9 mmol/L)或< 40 mg/dL(2.2 mmol/L)则每日剂量分别要减少20%或40%。再次提到了SUs2,建议如果加用胰岛素治疗就要停用SUs或者减量,并且还认为使用基础胰岛素类似物治疗优于中效胰岛素(NPH)。如果加用基础胰岛素治疗后仍然不能达到血糖控制目标,有2种推荐的治疗方法:(i)加用非胰岛素类的治疗药物,建议加用GLP-1 RA或SGLT2抑制剂或DPP-4抑制剂;或者(ii)加用餐时胰岛素,开始时的每日胰岛素总剂量为0.3-0.5 U/kg/天,使用人胰岛素与预混胰岛素治疗是不太理想的治疗方法。如果餐后2小时的血糖水平超过180 mg/dL(10.0 mmol/L)就要对加用的治疗药物剂量进行滴定。 这些推荐的治疗方法都很复杂,并且ADA与AACE的方法具有许多相似性与差异性(表1)。两者都推荐了3个不同的HbA1c目标,通常来说AACE的方法都更为严格;两者都推荐改变生活方式与治疗药物一样重要;两者都强调了启动药物治疗后同时需要每3个月监测一次HbA1c目标,而这也是强化治疗所需。虽然两者都需要使用自我监测血糖(SMBG)水平来调整胰岛素剂量,但是AACE推荐的血糖治疗目标更低,并且AACE还推荐使用SMBG来判断一名特定患者的治疗方案是否最佳,包括空腹或者餐后血糖都要达标。虽然AACE与ADA/EASD都明确表明治疗必须以患者为中心并且要遵循个体化原则,但是ADA/EASD推荐的治疗方案一直都是使用二甲双胍作为初始治疗药物,然而AACE却是根据基线HbA1c水平进行分层治疗,允许一开始就是二联治疗或者三联治疗,更加强调要将那些不适合一开始就使用二甲双胍治疗的亚组患者识别出来(例如晚期肾病患者;在它的2015年更新部分中5,ADA/EASD提到了如果HbA1c超过9%,需要考虑一开始就使用二联治疗)。AACE推荐的治疗药物包括了所有经美国食品与药品管理局(FDA)批准的降糖药物并且将药物选择进行了排名,然而ADA/EASD指南仅仅罗列了主要的6类药物而没有推荐药物的排名顺序。AACE推荐指南明确聚焦于降低空腹与餐后血糖的差异,列出了更多种类的药物,并且对更多不同的潜在药物不良效应作出了评论;AACE与ADA的优化胰岛素使用方法在某种程度上有些差异。 最后,ADA与AACE的治疗方法代表的都是专家共识,因此具有其局限性,因为对于特定的患者来说其特征具有无数种可能性,缺乏来自随机对照试验的证据证实什么才是最佳的治疗方案。对于一名特定的患者来说,我们还缺乏患者水平的meta分析来帮助我们预测何种药物或者是哪一类药物最为安全或是最为有效,特别是在联合用药时。因此我们在治疗高血糖时应该谨慎用药,小心地根据HbA1c与SMBG来明确高血糖情况,逐步调整治疗药物的种类与剂量,并且还要聚焦于潜在的药物不良效应。有些治疗方法可能还具有额外获益(体重、血压以及血脂),因此对于一名特定的患者来说这些药物可能就会显得更加合适或者更加不合适。也许心血管结果数据将可以最终帮助我们提高药物分层选择的特异性6。ADA与AACE推荐的治疗方法都为我们提供了重要的见解。在未来的岁月里随着新型药物以及新研究结果的问世,我们期待会有进一步的变化。 Taken together, the American Diabetes Association (ADA)1 and the American Association of Clinical Endocrinologists (AACE)2 2015 treatment recommendations for type 2 diabetes provide fascinating insight into concepts of the current approach to control of hyperglycemia in the disease. The ADA approach begins with selection of the target of glycemic control, emphasizing HbA1c ahead of glucose targets, and suggesting a usual HbA1c goal of 7%, although pointing out that either higher or lower levels may be appropriate for a given individual depending primarily on the patient's medical condition (duration of disease, life expectancy, and comorbidities, with vascular complications particularly highlighted) but also on attitude, expected efforts, and available resources, which could be modifiable, as well as risks of adverse drug effects, in particular hypoglycemia. Metformin is recommended as the preferred initial agent, with 3-month intervals for advancing to two-drug combinations and then three-drug combinations. Six drug classes are mentioned for additional treatment, with insulin rated as having the “highest” efficacy; glucagon-like peptide-1 (GLP-1) receptor agonists (RA), thiazolidinedione (TZD), and sulfonylureas (SU) having “high” efficacy; and dipeptidyl peptidase (DPP)-4 inhibitors and sodium–glucose cotransporter (SGLT) 2 inhibitors being of “intermediate” efficacy.1 Introduction of the latter drug class to the recommendations is mentioned as a “major change in treatment options”,1 but their combined use with GLP-1 RAs is not recommended given the paucity of clinical trial evidence. Explicit mention is given of bladder cancer, not being a concern with pioglitazone and of heart failure as a rationale for not using DPP-4 inhibitors;1 one assumes this will be revised based on new evidence from clinical trials.3 Specific approaches are given to minimize cost by excluding DPP-4 inhibitors, SGLT2 inhibitors, GLP-1 RAs, and insulin analogs; to avoid hypoglycemia by excluding SU and insulin; and to avoid weight gain by also excluding TZDs.1 For HbA1c >9%, initiation of a basal daily insulin dose at 10 units, or 0.1–0.2 units/kg body weight, is recommended, with once to twice weekly adjustment based on self-monitored fasting glucose.1 Recommended targets of fasting glucose (130 mg/dL [7.2 mmol/L]) and peak postprandial (assume 1 h) glucose (180 mg/dL [10.0 mmol/L]) were defined by the ADA standard of practice.4 For individuals reaching the fasting glucose target with basal insulin but not achieving overall glycemic control, three options are suggested: (i) adding a GLP-1 RA; (ii) adding a rapid-acting insulin before the largest meal; or (iii) changing to a twice-daily premixed insulin, with rapid-acting insulin before each meal an eventual option.1 The AACE recommendations2 (algorithm) incorporate all approved hypoglycemic agents. Four groups of drugs are distinguished: (i) those that primarily lower fasting glucose (e.g. metformin, TZDs, and basal insulin); (ii) those with postprandial effects (e.g. α-glucosidase inhibitors [AGI], the amylin analogue pramlintide, DPP-4 inhibitors, glinides, and short- and/or rapid-acting insulin); (iii) those with both actions (e.g. GLP-1 RA, SU, SGLT2 inhibitors, and mixed insulin preparations); and (iv) the modestly effective bile acid sequestrant colesevelam and a rapidly absorbed formulation of the dopamine D2 receptor agonist bromocriptine (BCR). The AACE algorithm focuses on safety as well as glycemic benefits: hypoglycemia issues with SU, glinides, and insulin; weight gain as effects of insulin, SU, and TZDs; and weight loss as effects of GLP-1 RA and SGLT2 inhibitors.2 Moderate renal function impairment precludes the use of metformin, whereas severe kidney function precludes the use of exenatide. Other safety issues include genitourinary infection and reduced intravascular volume with SGLT2 inhibitors, and nausea, vomiting, and diarrhea with GLP1 RAs. The potential for adverse gastrointestinal effects is also seen with metformin, AGI, colesevelam, and BCR. Cardiovascular issues, in particular heart failure, are cited as an issue with TZDs; the possibilities of adverse cardiovascular effects are raised with SU (and the increase in low-density lipoprotein cholesterol with SGLT2 inhibitors is mentioned, although its real clinical impact is not clear); and bone loss is given as a potential side effect of TZDs and SGLT2 inhibitors.2 An important principle of the AACE recommendations is to develop a personalized management plan determined primarily by baseline HbA1c. The AACE, unlike the ADA and European Association for the Study of Diabetes (EASD), ranks the drug choice as follows. At HbA1c levels <7.5%, monotherapy is recommended, with metformin as the first choice, GLP-1 RAs the second choice, and SGLT2 inhibitors the third choice, followed by DPP-4 inhibitors, AGI, TZD, and finally SU and/or glinides.2 Both TZDs and SU have greater glycemic efficacy than some of the drugs ranked above them, but are ranked lowest because of their respective side effects. For patients who present with an HbA1c >7.5% and <9.0%, or (as with the ADA recommendations) after 3 months without reaching goal from monotherapy, dual therapy and then triple therapy approaches are recommended, again giving a hierarchy of choices, but somewhat differently organized; because there is a better tolerance to risk, in dual therapy the recommended additions to metformin are GLP-1 RA first, SGLT2 inhibitors second, DPP-4 inhibitors third, TZD fourth, basal insulin fifth, and then colesevelam, BCR, AGI, and, again last, SU and glinides.2 In triple therapy, DPP-4 inhibitors are moved down in ranking below TZDs and basal insulin. At presentation with HbA1c >9.0%, initial dual or triple combinations are recommended unless the patient is experiencing hyperglycemic symptoms, in which case insulin with or without other agents is imperative.2 The insulin algorithm recommended in the AACE approach suggests that the initial basal insulin dose be 0.1–0.2 and 0.2–0.3 units/kg for HbA1c <8% and >8%, respectively. Dose titration of basal insulin is recommended every 3 days based on fasting glucose: >180 mg/dL (10.0 mmol/L) increasing by 20%, 140–180 mg/dL (7.8–10.0 mmol/L) increasing by 10%, and 110–139 mg/dL (6.1–7.8 mmol/L) increasing by 1 unit, with a reduction in daily dose by 20% and 40% for hypoglycemia with glucose <70 mg/dL (3.9 mmol/L) and <40 mg/dL (2.2 mmol/L), respectively. The SUs are again mentioned,2 suggesting that these agents be discontinued or reduced when insulin is added, and the use of basal insulin analogs is suggested as being preferred to neutral protamine Hagedorn (NPH) insulin. When addition of basal insulin fails to achieve glycemic goal, two recommendations are made: (i) the addition of non-insulin treatment, suggesting GLP-1 RA or SGLT2 inhibitors or DPP-4 inhibitors; or (ii) the addition of prandial insulin, starting with a total daily insulin dose of 0.3–0.5 units/kg per day, with human insulin and premixed insulin less desirable as approaches. These additional treatments should be titrated if 2-h postprandial glucose levels exceed 180 mg/dL (10.0 mmol/L). These are complex recommendations, and the ADA and AACE approaches have a number of similarities and differences (Table 1). Both recommend three different HbA1c goals, with the AACE approach in general being more stringent; both recommend lifestyle modification as critical to therapy; both recommend initiating medications concomitantly and stress the need to monitor the HbA1c goal every 3 months and intensify treatment as needed. Although both use self-monitoring of blood glucose (SMBG) levels for insulin dose adjustment, the AACE approach suggests lower targets and the AACE recommendations suggest using SMBG to decide whether a given individual's treatment would optimally involve targeting fasting or postprandial glycemia. Although both AACE and the ADA/EASD explicitly state that management must be patient centric and individually based, the ADA/EASD recommendation is to always start with metformin, whereas the AACE stratifies treatment by baseline HbA1c, allowing initial dual and triple therapy and putting greater emphasis on recognizing the subsets of patients for whom initiation of treatment with metformin is not appropriate (e.g. those with advanced kidney disease; in its 2015 update,5 the ADA/EASD mentions consideration of initial dual therapy when HbA1c exceeds 9%). The AACE recommendations incorporate all the glucose-lowering agents approved by the US Food and Drug Administration (FDA) and ranks the drug choices, whereas the ADA/EASD guidelines list only the main six classes and do not recommend an order of ranking. The AACE recommendations explicitly focus on differences in fasting and postprandial glucose lowering, list a greater variety of agents, and review a larger number of different potential adverse effects; the AACE and ADA offer somewhat different approaches in optimizing the use of insulin. Ultimately, both the ADA and AACE approaches represent a consensus of experts and thus are limited by the lack of evidence from randomized controlled trials to support optimal treatment for the myriad of possible specific patient characteristics. We lack patient-level meta-analyses to aid in the prediction of which agent or groups of agents will be safest and most effective, especially in combinations, for a given individual. We therefore need to use caution in the management of hyperglycemia, making progressive adjustments in treatment agent and dose based on careful ascertainment of hyperglycemia using HbA1c and SMBG, and focusing on potential adverse effects. There may be additional benefits of various approaches (on weight, blood pressure, and lipids) that make them more or less applicable to a given patient. Perhaps cardiovascular outcome data will ultimately help us to add specificity to the hierarchy of medication choice.6 Both the ADA and AACE approaches offer important insights. As new agents and new studies become available in the coming years, we should expect further changes.