Is the current definition of diabetes relevant to the risk of all-cause, cardiovascular, and non-cardiovascular mortality? - Evidence review
The DECODE study found that, especially after a glucose load, higher values were associated with greater cardiovascular mortality without a clear threshold. Independent cohorts confirm the association and often the informative advantage of the post-load test, but inconsistent results prevent it from being considered a definitive predictive tool on its own.
- Published 13/11/2026
- Updated 13/11/2026
- Article history
ABSTRACT
The original study analyzed 29,714 European adults to assess whether fasting glycemia and glycemia measured two hours after an oral glucose load were associated with total, cardiovascular, and non-cardiovascular mortality. After approximately eleven years of follow-up, elevated glycemic values were associated with higher mortality. For cardiovascular mortality, the increase in risk in relation to post-load glycemia appeared progressive, without a sudden threshold.
In section 2.4.1 of “The Choice to Live in Health or Illness” by Dr. A. Colonnese and R. Panzironi, this research contributes to the narrative that post-load glycemia can identify individuals with increased cardiovascular risk who would not be fully identified by fasting glycemia alone. Independent cohorts support the association between elevated post-load glycemia, cardiovascular events, and mortality, but not all confirm a clear predictive superiority over fasting glycemia. The evidence is therefore consistent with a prognostic role for post-load glycemia, without demonstrating that it directly describes cardiovascular damage or that its treatment necessarily reduces events.
Analyzed study
Is the current definition for diabetes relevant to mortality risk from all causes and cardiovascular and noncardiovascular diseases?
PMID: 12610023
The research question was whether fasting glycemia and glycemia measured two hours after an oral glucose tolerance test were associated with the risk of death, and whether this relationship followed a clear threshold or a gradual trend. The oral glucose tolerance test, or OGTT (a test in which the body's management of a standard amount of glucose is measured), can reveal an increase in glycemia after a meal that is not always visible while fasting.
The relevance to paragraph 2.4.1 is specific: the publication examines whether post-load measurement identifies a risk of cardiovascular mortality across the entire glycemic range and not just beyond the diagnostic threshold for diabetes. In "The Choice to Live in Health or Illness" by Dr. A. Colonnese and R. Panzironi, the study is used to argue that post-load glycemia can signal at-risk individuals more effectively than fasting glycemia alone.
As a prospective observational study, the research can show that higher glycemic levels more frequently precede death in the observed groups. However, it cannot prove that an increase in post-load glycemia directly causes cardiovascular damage, nor can it establish that reducing that value prevents clinical events.
Materials and methods of the original study
DECODE was a prospective collaboration of 22 European cohorts. The analysis included 29,714 people aged between 30 and 89, with baseline glycemic measurements. The total observation period was approximately eleven years, for 329,050 person-years: this measure sums the participation time of all subjects followed.
The authors evaluated fasting plasma glucose and plasma glucose two hours after an oral glucose tolerance test. The outcomes were all-cause mortality, cardiovascular mortality, and non-cardiovascular mortality. To estimate risk differences, they used Cox regression models, an analysis that compares the rate at which an event occurs over time across the studied groups. The result is expressed as a hazard ratio or HR: an HR greater than 1 indicates a higher frequency of the event compared to the reference group.
The detailed inclusion and exclusion criteria for the individual cohorts, the exact enrollment period, the clinical characteristics of the participants, the distribution by sex, the glucose dosages used in the oral test, and the methods for ascertaining causes of death are not reported in the available material, which consists only of the abstract.
Results
The main finding is that both high and very low glycemic values were associated with increased mortality, overall forming a J-shaped relationship. In practice, the risk was not at its minimum at the extremes of the glycemic distribution. However, this trend does not equate to proof that low glucose levels cause mortality: they could reflect comorbidities, frailty, or other factors not reported in the abstract.
Compared to fasting glycemia between 4.50 and 6.09 mmol/L, fasting glycemia equal to or greater than 7.0 mmol/L was associated with an adjusted HR of 1.6 for all-cause mortality, 1.6 for cardiovascular mortality, and 1.6 for non-cardiovascular mortality. This corresponds to a higher rate of death in groups with fasting glycemia in the diagnostic range for diabetes compared to the reference group.
For post-load glycemia, the comparison with the group between 4.51 and 5.50 mmol/L showed that diabetes defined by glycemia two hours after the load equal to or greater than 11.1 mmol/L was associated with an HR of 2.0 for all-cause mortality, 1.9 for cardiovascular mortality, and 2.1 for non-cardiovascular mortality. These values indicate an approximately double risk compared to the reference. Very low post-load glycemia, between 3.01 and 4.50 mmol/L, was associated with an HR of 1.1 for all-cause mortality, 1.1 for cardiovascular mortality, and 1.1 for non-cardiovascular mortality; the estimates were small, and some confidence intervals included the value of no difference.
An element directly relevant to the book is that subjects with diabetes not previously recognized via post-load glycemia had a risk of death not significantly different from that of people with already known diabetes. This risk was, however, significantly higher than that observed in diabetes undiagnosed by the fasting glycemia criterion alone. Furthermore, post-load values between 10.01 and 11.09 mmol/L were associated with mortality similar to that of subjects classified as diabetic via fasting glycemia equal to or greater than 7.0 mmol/L.
According to the authors, for cardiovascular mortality, the relationship with post-load glycemia was gradual and increasing, without an evident threshold. The abstract does not report detailed numerical values for each step of the relationship, nor formal statistics directly comparing the predictive accuracy of the two tests.
Interpretation
In simple terms, a person may have a fasting glycemia that is not clearly diabetic but show a significant increase in glycemia after a glucose load. The study suggests that this condition is not necessarily harmless from a prognostic standpoint. The segment of the book's narrative documented by the publication is therefore: higher post-load glycemia → greater observed cardiovascular mortality → possible identification of at-risk individuals whom fasting glycemia alone may not classify as diabetic.
The result is biologically plausible because post-load glycemia reflects the body's ability to control glucose when it rises after carbohydrate intake. A high value can be an indicator of reduced insulin sensitivity or an insufficient insulin response. However, these metabolic alterations were not measured directly in the available material and cannot be attributed as the cause of the deaths.
The statement in the book that cardiovascular damage may develop long before overt diabetes requires caution. The study demonstrates a prognostic association, does not directly measure early vascular lesions, and does not prove a causal sequence from the glycemic peak to cardiovascular damage. Furthermore, although the results indicate that the post-load test identified people at high risk, the abstract does not report a formal statistical comparison of the overall predictive capacity between OGTT and fasting glycemia.
Independent supporting evidence
A confirmation particularly close to the book's narrative comes from the Cardiovascular Health Study. In 4,014 U.S. adults aged 65 or older, without treated diabetes or a history of heart attack or stroke, 764 cardiovascular events were recorded over 8.5 years. Glycemia two hours after a glucose load showed a linear relationship with the risk of cardiovascular events: for every 10 mg/dL increase, the HR was 1.02. When fasting glycemia and post-load glycemia were considered together, only post-load glycemia remained associated with events. This study directly confirms the passage stating that the post-load measurement can offer additional prognostic information compared to the fasting value alone, at least in the elderly. The main limitation is the advanced age of the population, which does not allow for the automatic transfer of this estimate to younger people [1].
An additional cohort, obtained by combining three population studies in Mauritius, Fiji, and Nauru, followed 9,179 people for a period ranging from five to twelve years. In men with isolated post-load hyperglycemia—that is, post-load glycemia equal to or greater than 11.1 mmol/L but fasting glycemia less than 7.0 mmol/L—total mortality was higher compared to people without diabetes; in women, the HR was 2.0. Cardiovascular mortality was also higher, with an HR of 2.3 in men and 2.6 in women. This result supports the concept that an alteration detectable only with the post-load test can identify people with an unfavorable prognosis. It remains an observational study and does not establish which biological mechanism explains the increase in mortality [2].
In the U.S. national NHANES II sample, 3,092 adults between 30 and 74 years of age who underwent an OGTT were followed until 1992. People with isolated post-load hyperglycemia had higher total mortality compared to normoglycemic subjects, with a relative risk adjusted for age and sex of 1.6. In the fully adjusted analysis, both fasting glycemia and post-load glycemia were associated with mortality with very similar estimates. The data confirms that post-load hyperglycemia is a prognostic signal independent of various cardiovascular factors, but it tempers the idea of a constant superiority of the post-load test in every population [3].
Conflicting evidence
Evidence is not uniform regarding the greater efficacy of post-load glycemia compared to fasting glycemia. In the Chin-Shan Community Cardiovascular Cohort in Taiwan, 2,165 adults were followed for 10.5 years, during which 166 major cardiovascular events occurred. The highest quartile of fasting glycemia was associated with an increased cardiovascular risk, with a relative risk of 1.74 compared to the lowest quartile; for post-load glycemia, the increase was 2.05. However, the two measures had the same discriminative capacity reported by the authors, and in reciprocal models, both remained associated with cardiovascular events. This result does not negate the prognostic value of post-load glycemia, but it does not confirm a clear superiority over the fasting value [4].
The discrepancy may depend on age, geographic origin, the frequency of glycemic abnormalities, the number of events, diagnostic criteria, and the way risk prediction is assessed. No sufficiently relevant directly conflicting evidence has been identified that demonstrates an inverse association between elevated post-load glycemia and cardiovascular risk; this does not prove the absence of contrary results.
Summary table of evidence
The table distinguishes the study under review from the independent publications used to confirm or define the narrative of the book's paragraph.
| Reference | Study type | Population or model | Main result | Direction of evidence | Strength of evidence |
|---|---|---|---|---|---|
| Original study | Prospective cohort collaboration | 29,714 European adults, 22 cohorts, 11 years | Elevated post-load glycemia was associated with higher mortality; for cardiovascular mortality, the relationship was graded and increasing. | direct confirmation | moderate |
| [1] | Prospective observational | 4,014 US adults aged 65 or older | In the model including both measurements, only post-load glycemia remained predictive of cardiovascular events. | direct confirmation | moderate |
| [2] | Prospective observational | 9,179 adults in Mauritius, Fiji, and Nauru | Isolated post-load hyperglycemia was associated with higher total and cardiovascular mortality. | indirect confirmation | moderate |
| [3] | Prospective observational | 3,092 US adults from the NHANES II sample | Isolated post-load hyperglycemia was associated with higher mortality; the prognostic value was similar to that of fasting glycemia. | biologically consistent evidence | moderate |
| [4] | Prospective observational | 2,165 Taiwanese adults | Fasting and post-load glycemia were both associated with higher cardiovascular risk, with similar discriminative capacity. | discordant evidence | moderate |
Confirmation of the evidence
Claim in the book
In section 2.4.1 of “The Choice to Live in Health or Illness” by Dr. A. Colonnese and R. Panzironi, the relevant statement is that glycemia measured two hours after an oral glucose load shows a progressive increase in cardiovascular risk along the glycemic spectrum and can identify high-risk individuals more effectively than fasting glycemia alone.
Increase in post-load glycemia → identification of impaired glucose tolerance not always detected while fasting → greater observed risk of cardiovascular events or mortality → possible early detection of high-risk individuals
The publication subject to the Evidence Review was used in the book to support the following passage: elevated glycemia two hours after an oral glucose load is associated with higher cardiovascular mortality along a continuum, and can recognize subjects with high risk not fully identified by fasting glycemia alone.
The original publication directly documents the association between elevated post-load glycemia and increased cardiovascular mortality, as well as the absence of a clear threshold for this relationship. Its contribution to the narrative is prognostic: it does not demonstrate vascular damage, but shows that the information obtained after a glucose load is linked to unfavorable clinical outcomes.
Independent evidence supports this interpretation. In elderly individuals from the Cardiovascular Health Study, post-load glycemia, but not fasting glycemia, remained associated with cardiovascular events when both were analyzed in the same model [1]. Other cohorts have found higher mortality in people with hyperglycemia detectable only after the load, despite having fasting glycemia below the diagnostic threshold for diabetes [2,3]. These results make the statement from “The Choice to Live in Health or Illness” by Dr. A. Colonnese and R. Panzironi consistent, according to which the post-load test can identify a portion of risk not expressed by the fasting measurement alone.
The confirmation is not absolute. In a Taiwanese cohort, both glycemic measures maintained an independent association with cardiovascular events, and their discriminative capacity was similar [4]. Therefore, the idea of a prognostic value for post-load glycemia is supported, while its greater efficacy compared to fasting glycemia depends on the context, the population, and the outcome evaluated.
Overall level of confirmation
MODERATE CONFIRMATION
The original publication and several independent cohorts show a consistent relationship between elevated post-load glycemia, mortality, or cardiovascular events. However, the evidence is observational, does not demonstrate causality, and does not uniformly confirm a predictive superiority of the post-load test compared to fasting glycemia.
Conclusions
The study analyzed clearly supports that glycemia measured two hours after an oral glucose load is associated with mortality, including cardiovascular mortality, and that for the latter, the risk increases progressively without a clear, distinct threshold. Independent evidence confirms that post-load hyperglycemia can identify individuals with a less favorable cardiovascular prognosis even when fasting glycemia is not in the diabetic range. Its greater efficacy compared to fasting glycemia alone is moderately supported, but it is not consistent across all cohorts. The link with cardiovascular risk is therefore well-founded on an observational level, while the causal mechanism and the utility of targeted interventions require further validation.
Overall assessment
The narrative is moderately supported: elevated post-load glycemia shows consistent associations with increased mortality and cardiovascular events in various cohorts. The biological mechanism remains plausible but is not directly proven by observational studies. The greater prognostic capacity compared to fasting glycemia appears fairly well supported, but it depends on the population analyzed.
References
[1] Fasting and 2-Hour Postchallenge Serum Glucose Measures and Risk of Incident Cardiovascular Events in the Elderly
Editorial note
Artificial Intelligence was used solely as a tool to support editorial and documentary work. All content, interpretations, bibliographic citations and conclusions have been verified, reviewed and approved by the author, who assumes full scientific responsibility for them.
The manuscript was also subject to independent peer review prior to publication.
To ensure maximum transparency, the article's complete history — including the dates of submission, review, acceptance and publication, as well as information on the peer review process — is available on the dedicated Article History page.