First, in full disclosure, as a Ph.D. who does not directly interact with patients, I perhaps lay myself open to some flack here. But, I’ve been wondering for some time now if the initial diabetes or pre-diabetes diagnosis of a patient were a little more detailed, couldn’t the tailoring of an individual’s therapeutic strategy be more effectively planned? Currently, the diagnosis information is a fasting plasma glucose (prediabetes 100-125mg/dl; diabetes ≥126mg/dl), a plasma glucose measurement at a 2h time point of an OGTT (prediabetes 140-199mg/dl; diabetes ≥200mg/dl), an HbA1C level (prediabetes 5.7-6.4%; diabetes ≥ 6.5%), and, of course, all of these measures considered relative to a BMI (overweight ≥ 25 kg/m2; obese ≥ 30 kg/m2), other physical symptoms, and family history. This is all very good information. But does it really tell us what kind of diabetes—type 1 or 2—or how functional and/or numerous a patient’s beta cells are at the time of diagnosis? Maybe a measure of plasma insulin (and/or C-peptide) level at diagnosis might help?
Our current approach to diagnosis often leaves the situation cloudy in obesity-linked type 2 diabetes where there is much heterogeneity, compounded by the pathogenesis of the disease: early attempts by the functional beta cell mass to compensate for insulin resistance but then followed by eventual beta cell failure. At the time of diabetes diagnosis, using only measures of glucose and HbA1C levels as we currently do, it is close to impossible to pinpoint where a patient might be in the pathogenesis of the disease. Here, extra measurements of plasma insulin/C-peptide after overnight fasting and at a 2h time point of an OGTT could well assist in the diagnosis and help decide on effective treatment strategies (including diet and exercise, of course).
This approach makes sense, especially with a mandate to hold onto what functional beta cells remain. For example, measures of high glucose and insulin/C-peptide levels in plasma are a good indicator that the patient still has some functional beta cells left working against significant insulin resistance. Insulin sensitizers to alleviate insulin resistance (as well as to give the beta cells some respite), could be applied here. In contrast, if an obese patient has high glucose but low insulin/C-peptide levels, then the diagnosis could indicate loss of functional beta cell mass later in the pathogenesis of the disease. Then perhaps insulin sensitizers and an incretin or insulin replacement therapy (to help out the few remaining beta cells) would be appropriate.
Early measurement of insulin levels could also be helpful in the diagnosis of patients with type 1 diabetes, although here additional tests may be required. Slim individuals with high fasting plasma glucose and low insulin/C-peptide levels (especially along with symptoms of weight loss, polyuria, fatigue, etc.), for example, would likely have type 1 diabetes, but it is also possible that they could have lean type 2 or a rare genetic form of diabetes. Here an islet autoantibody (IAA) test should nail a diagnosis of autoimmune type 1 diabetes and then the initiation of insulin replacement therapy. But a negative IAA result could suggest an insulin secretory defect, as often found in lean type 2 diabetes, in which case sulfonylurea and/or incretin-based therapies could be considered to bolster beta cell function. Alternatively, a follow-up genetic screen to identify a known form of monogenic diabetes could be considered.
So, is adding insulin and/or C-peptide measurements to the diagnosis of diabetes a good idea? Is it practical? I think maybe, but thoughts, especially from my MD colleagues, are welcome. Please understand that the scenarios above are merely fictional examples and should not be taken as guidelines. I include them simply as illustrations that some additional information of insulin/C-peptide levels at the diagnosis of diabetes could be valuable. It is likely that academic medical centers do this already, but much diabetes is diagnosed in places where measurements of plasma insulin/C-peptide levels are not necessarily convenient and, given their cost, restrictive if done routinely. But then again, if diabetes is better diagnosed in the first place, treatment can be better tailored, and, in the long run, avoiding diabetes complications makes sense economically.
