News & Perspective

Stringent glucose targets may not be beneficial to all diabetic patients

1 year ago, OP Editor

The J-shaped curvilinear association between hemoglobin A1c (HbA1c) and the risk of cardiovascular disease (CVD) remains a controversial concept, considering the lack of consensus regarding the lower threshold of HbA1c.1 To elucidate this issue, Mr. Eric Wan, Dr. Esther Yu, and their team from the University of Hong Kong (HKU) conducted a territory-wide retrospective cohort study to evaluate the association between HbA1c and incidence of CVD.1 The results of this study were recently published in Diabetes & Metabolism.1

“Lower HbA1c is not always better and may even be potentially harmful,” referring to the J-shaped curvilinear association between HbA1c and CVD risk,1 is a concept that has been identified in earlier studies but less well-understood.2 Action to Control Cardiovascular Risk in Diabetes (ACCORD), a randomized study that involved 10,251 type 2 diabetes patients and investigated a target HbA1c level below 6.0%, was terminated early due to the increased mortality.2 When compared to the standard therapy targeting a level of 7.0% to 7.9%, the intensive glucose lowering did not significantly reduce major cardiovascular events as well.2

Given the stringent design of clinical trials, a territory-wide retrospective cohort study based on real-world primary care outpatient setting was conducted by the researchers from the HKU.1 A total of 115,782 ethnic Chinese patients with type 2 diabetes, with clinical data that can be retrieved from the administrative database of the Hospital Authority (HA), was included in the analysis.1 Over a median follow-up period of 5.8 years, the cumulative mean measurement of HbA1c was used to evaluate the relationship between HbA1c and the incidence of CVD.1

The results suggested that the optimal goal of HbA1c depends on the duration of diabetes. “A key finding of this study was that no minimum threshold for HbA1c level was identified in patients with recently diagnosed [diabetes] of <2-year duration. This suggests that more stringent HbA1c targets (<6.5%) may be used in newly diagnosed patients.”1

However, the authors suggested that the more stringent HbA1c targets may not be beneficial to patients with long-standing diabetes. “For others, [the lowest risks for CVD were] observed at HbA1c levels of approximately 7.0% with a J-shaped curvilinear association between HbA1c and the risk of CVD.”1

In the analysis, age (<65 years or ≥65 years) and comorbidities (Charlson’s Index <5 or ≥5) did not appear to impact the J-shaped pattern, but there was a shift to right in patients with ≥10-year duration of diabetes or those taking ≥2 classes of anti-diabetic drugs.1

A potential explanation for these results came from the United Kingdom Prospective Diabetes Study (UKPDS).1,3 Known as the legacy effect, this described the long-lasting benefits that patients with newly or recently diagnosed diabetes might gain due to an intensive initial glucose control.3 Several pathophysiological mechanisms for this legacy effect have been proposed, including the increased intracellular formation of advanced glycation end products.3

Acknowledging that a convincing high-level evidence from a real-world study might be difficult to obtain, the authors suggested that “these findings supported the importance of early control of glucose at an earlier disease stage and clinicians should be vigilant against overtreatment of other diabetic patients in glycemic control for the primary prevention of CVD.”1


1. Wan EYF, Yu EYT, Fung CSC, et al. Relation between HbA1c and incident cardiovascular disease over a period of 6 years in the Hong Kong population. Diabetes Metab. 2018 [Epub ahead of print].

2. Action to Control Cardiovascular Risk in Diabetes Study Group, Gerstein HC, Miller ME, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358(24):2545-59.

3. Holman RR, Paul SK, Bethel MA, et al. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359(15):1577-89.


Menu Section