TO THE EDITOR
The diabetes mellitus epidemic is rising worldwide with a significant impact on patients, healthcare and the community due to its microvascular and macrovascular complications. Diabetes mellitus affects all age groups with type 2 constituting 90%–95% of cases in adults, and type 1 diabetes being the most common metabolic abnormality in children[1].
Diabetes mellitus and cardiovascular disease are major determinants of morbidity and mortality. Poor glycemic control is the major cause of microvascular diseases including nephropathy, neuropathy and retinopathy. There is a close connection between the blood as a connective tissue and blood glucose, and red blood cell survival and other blood indices are disturbed by high blood glucose in diabetes mellitus[2]. Importantly, the hemoglobin is nonenzymatically glycated to form glycated hemoglobin (HbA1c), which reflects the blood glucose status throughout the red blood cell lifespan[3]. The American Diabetes Association approved HbA1c as a measure of glycemic control and recommended a HbA1c of < 7 to avoid diabetes microvascular complications. However, the test is inaccurate among patients with hemoglobinopathy and renal impairment[4,5]. In addition, HbA1c measures glycemic control over the red blood cell lifespan (4 mo). Therefore, a rapid test that reflects glycemic control over a shorter period is attractive. Platelet indices (PIs) are easy to measure, cheap and fulfill the above needs in diabetes mellitus.
Platelets that are important for homeostasis are produced in large amounts by megakaryocytes (100 billion/d), and their lifespan varies between 8 and 10 d. There is a direct association between increased platelet activity, diabetes, procoagulability and thrombosis risk, and evidence that patients with diabetes are at high risk of cardiac and cerebral disorders[6,7]. However, the clinical implications of mean platelet volume (MPV), plateletcrit (PLC), and platelet distribution width (PLDW) in diabetes management are lacking. Therefore, there is increasing interest in the role of PIs in glycemic control and homeostatic changes in diabetes[8].
We read with interest the article published by Regassa et al[9] who found that MPV, PLC, PLDW and platelet large cell ratio (PLCR) were associated with poor glycemic control and microvascular complications among patients with diabetes. Although previous studies were conducted on the same patients, they showed contradictory findings[10-13]. In addition, Regassa et al[9] adopted strict inclusion criteria to assess the PIs among healthy controls and diabetic patients with good/poor control. The importance of the Regassa et al[9] study is that it was conducted in a low-income country where measurement of HbA1c was not available or cost-effective. Therefore, their study is highly relevant.
Literature review and discussion
The subject of PIs is complex, and to address the important issues raised by Regassa et al[9], we searched PubMed/MEDLINE, Cochrane Library and Google Scholar for relevant articles from inception to August 17, 2024. The terms used were mean platelet volume, PLC, PLDW, PLCR, HbA1c, PIs, platelet activity, diabetes mellitus and HbA1c. Out of the 790 articles retrieved, 187 full texts were reviewed, and 49 were included (Figure 1).
Figure 1 Literature search showing the relationship between platelets indices and glycemic control.
PIs among patients with diabetes and their relation to glycemic control
Type 2 diabetes: The first retrieved article was published in Turkey[14]. The authors found a high MPV among patients with diabetes compared to healthy subjects. Additionally, MPV was higher among patients with poor glycemic control. MPV returned to normal following good glycemic control. In contrast, Singer et al[15] observed no improvement in platelet activity with improving HbA1c, which could be explained by the higher baseline HbA1c, long duration of diabetes, and statin and aspirin consumption. Wu et al[16] found a correlation between HbA1c and MPV, but no correlation was found with platelet count (PLT). A plausible explanation is that the PDW and PLT are influenced by the preceding 15–20 d (lifespan of platelets) rather than the glycemic period of 30–120 d [17]. Importantly, MPV is not an independent predictor of diabetes and glycemic control, because it is positively associated with treatment with angiotensin receptor blockers (ARBs), age, and hemoglobin levels, and negatively correlated with hypercholesterolemia[18].
Platelets play an important role in homeostasis. The interplay between platelets and adhesive glycoproteins after endothelial damage initiates platelet aggregation, leading to thrombosis. Importantly, large platelets are more active and secrete more dense granules, because big platelets (high MPV) significantly contribute to platelet activation processes, and lead to thrombosis and acute arterial events[19,20]. Another plausible explanation of the association between platelet volume and microvascular complications could be their association with poor glycemic control (a predictor of diabetic microvascular complications)[14,15].
Large platelets in patients with diabetes contain dense granules, high enzymatic activity, and a high potential for thrombosis leading to thrombosis and diabetes microvascular complications[21]. A systemic review and meta-analysis[22] concluded the association between PIs and microvascular complications. The capillary non-perfusion, releases of platelet-driven growth factors, and endothelin-1 activation are proposed mechanisms.
PLC, mean platelet component, mean platelet dry mass, and MPV were associated with glycemic parameters in Serbia[2]. Further research confirmed the association between blood glucose, low-density lipoproteins, high-low-density lipoproteins, high cholesterol, and increased total white blood cells. In addition, MPV and PDW were increased. The authors noticed low platelets and high interferon (IFN) among patients with poor glycemic control[23]. The above changes resolved/decreased with hyperglycemia treatment. The quantitative and qualitative abnormalities in the blood are well-known among patients with diabetes[24]. However, thrombocytopenia could be a more beneficial parameter for hyperglycemia if combined with MPV and other PIs.
Mechanisms of hyperglycemic effects on platelets
The high IFN, and inflammatory markers in diabetes in particular poor glycemic control, are associated with increased platelet volume and thrombocytopenia[25]. The effects of hyperglycemia on platelet activity could be part of disturbed homeostasis in diabetes mediated by inflammation, increasing adhesion molecules, and circulating selectins[26]. The association of inflammatory markers, including C-reactive protein and platelet hyperactivity, was observed by Jabeen and colleagues in 2013[26]. In addition, protein kinase C activation, nonenzymatic glycation of the platelet surface proteins, and osmotic changes lead to platelet membrane rigidity. Activated platelets in diabetes form psuedopodia, and change their shape from the usual biconcave to spheres, affecting the RDW[9].
Platelet hyperactivity was evident in newly diagnosed and known patients with diabetes and correlated with HbA1c. Because of that, PIs could be a better alternative to HbA1c in areas where the facilities for HbA1c measurement are not available and among patients with hemoglobin disorders[27]. However, PIs, including MPV, are increased among patients with renal impairment and have an inverse relation with glomerular filtration rate[28]. Therefore, platelet activity has the same limitation as HbA1c.
Type 1 diabetes mellitus: Studies conducted on type 1 diabetes are scarce. Söbü et al[28] found that MPV was associated with poor glycemic control and duration of diabetes; however, no significant differences were evident in MPV between patients with good glycemic control and healthy controls. Venkatesh et al[29] assessed PLT, MPV, PLDW, PLCR and PLC and found higher MPV, PDW and PLCR, and lower PLC among children with poor glycemic control compared to control subjects and others with optimal HbA1c. Only PDW was high among newly diagnosed patients, pointing to the effect of diabetes duration. In another study published by Ma et al[30], MPV and PDW were significantly higher among type 1 diabetes with ketoacidosis.
Platelets indices and vascular disease: One of the major objectives in the treatment of diabetes is cardiovascular disease (microvascular and macrovascular complications, and cardiovascular risk factors) prevention and treatment[31]. Importantly, HbA1c can predict microvascular but not macrovascular and cardiovascular risk factors. Therefore, cardiovascular risk assessment at diagnosis is mandatory in diabetes care for the proper treatment and prescription of novel diabetic medications with cardiac and renal protection[32]. PIs and other blood parameters are easy to measure using impedance technology which is cost-effective, easy and accurate [33]. PIs can predict microvascular complications[34,35], coronary syndrome, glycemic control and thrombotic events in low-income countries where the facilities for HbA1c measurement are not available or not affordable[36,37].
Studies investigating calmodulin glycation in platelets as a measure of glycemic control showed significance only when correlated with serum fructosamine. However, the role of fructosamine as a glycemic parameter is controversial[38]. Because of that, platelet glycation adds no benefit compared to HbA1c and other glycemic control measures.
Regassa et al[9] found that PLDW, MPV, PLCR and PLC were higher among patients with type 2 diabetes, and suggested these as glucoregulators for follow-up in low-income countries. Regassa et al[9] applied strict inclusion and exclusion criteria. They excluded patients with cancer, infection and some drugs, and evaluated the controls for the same. Therefore, their results were robust because drugs[39], infection[40] and cancer[41] have been shown to affect PIs. However, the authors' strict measures are difficult to apply in the real world, because patients with diabetes are more prone to infection[42] and cancer[43]. Another limitation of the Regassa et al[9] study was that they did not screen the controls for undiscovered diabetes mellitus, and prediabetes, which are common and can alter PIs[44,45]. In addition, Regassa et al[9] used fasting blood glucose and mean blood glucose as a measure of glycemic control, but HbA1c measurement might not be available. However, fasting blood glucose and mean blood glucose do not reflect the glycemic status and continuous blood glucose measurements are more accurate. In the context of the short platelet lifespan, it is more accurate to use continuous blood glucose monitoring, because it has the advantage of recording the time in the range which is more beneficial than fasting blood glucose and mean blood glucose taken at different points[46].
DISCUSSION
Diabetes mellitus is a chronic lifelong disease with cardiovascular complications and a high economic burden worldwide, particularly in developing countries. As a result, affordable and readily available diagnostic methods are needed for diagnosis and monitoring of microvascular complications of diabetes[21].
The bidirectional relationship between PIs and diabetes mellitus in terms of endothelial dysfunction and microvascular complications has attracted researchers to the use of PIs as a cheap, easy-to-conduct method of diabetes follow-up. Previous studies showed the association between PIs, diabetes glycemic control and microvascular complications[14-30]. However, the correlation and the significance varied considerably between the studies, and the type of the measured PIs[17].
MPV, PLDW, P-LCR and PLC are sensitive measures of glycemic control (82%); however, they are less specific (54.5%)[33]. The PIs are not independent predictors of glycemic control and cardiac complications but are influenced by age, diabetes duration, and drugs, including ARBs, aspirin and lipid-lowering drugs, which are widely prescribed to patients with diabetes[24,47]. In addition, glucagon-like peptide-1 receptor agonists, and DPP-4 inhibitors affect platelet aggregation through nitric oxide modulation. Therefore, patients on GLP-1 receptor agonists and DPP-4 inhibitors might show misleading results for PIs[48]. Furthermore, the blood sample needs to be measured promptly without delay when blood is collected in EDTA to avoid misleading results[49].
PLC is measured from PLT and MPV (PLC = PLT × MPV/10 000); therefore, they are discussed together. Age, duration of diabetes, drugs including ARBs, statins and aspirin, hypercholesterolemia, and hemoglobin could affect MPV and PLC[16,18]. PIs including MPV are increased among patients with renal impairment and show an inverse relation with glomerular filtration rate. Therefore, platelet activity has the same limitation as HbA1c[28]. However, PLDW is influenced by the preceding 15–20 d (lifespan of platelets)[16] and is less likely influenced by duration of diabetes[31]. Therefore, it might be more accurate as a glucoregulator because it is useful among newly diagnosed patients.
PLCR is a new index that needs special equipment (Sysmex analyzer) that might not be available in remote areas, and because of that more studies are suggested[9].
PIs, although associated with glycemic control and microvascular complications, can predict macrovascular disease among patients with diabetes, but they are not specific because the evidence was based mostly on observational studies. Therefore, a cause and effect cannot be concluded. Additionally, PI results should be viewed in the face of many confounders including demographic factors, pregnancy, renal failure, medication, hemoglobin and duration of diabetes.
PIs are sensitive, rapid, readily available and affordable measures of glycemic control, and can predict microvascular complications among patients with type 1 and type 2 diabetes. PLDW is more specific if performed promptly; good patient selection is needed to rule out confounders such as components of the metabolic syndrome, demographic characteristics and drugs.
The study limitations were the reliance on observational studies that were prone to bias, and we could not control for various confounders, including demographic factors and comorbidities.
