Re: Association of elevated fasting glucose with lower patency and increased major adverse limb events among patients with diabetes undergoing infrapopliteal balloon angioplasty

In this issue of Vascular Medicine, Singh et al. describe the association of pre-procedural fasting blood glucose (FBG) level with 1-year outcomes for infrapopliteal angioplasty in patients with critical limb ischemia (CLI). ¹ Assessment of long-term patency was made by duplex ultrasonography, but not invasive imaging. The authors found that elevated FBG (median >144 mg/dL), even when corrected for insulin use and other lesion-specific anatomic parameters, was associated with a twofold lesser primary patency. More striking are the quartiles of FBG and the overall very low primary patency (roughly 10%) associated with a FBG of more than 200 mg/dL. It is unclear how many of these failed procedures went on to further invasive procedures or amputation. Nonetheless, the correlation of this single FBG measurement to interventional outcomes is potentially important. One could envision using the FBG level in practice by either modifying the intervention, not performing the intervention, or using this in discussion with the patient and family in terms of success of the procedure.

While FBG may be a useful biomarker when considering endoluminal therapy for those with CLI, several important questions are not answered. First, it is unclear whether using FBG as a biomarker correlates with outcomes in open bypass or in more extensive arterial endoluminal interventions than just infrapopliteal procedures. The anatomic differences and therapeutic ways to approach the disease may be quite different, including different biomarker prognostication. ² Second, it is not detailed whether the group of patients with higher FBG simply had more severe tissue loss or uncontrolled infection than others, as the Rutherford scale may be somewhat subjective. This is a potential confounder which would render these data less important. For example, an active, poorly controlled infection may render blood glucose harder to control than a small ulcer with chronic osteomyelitis, and thus FBG may simply be a surrogate for more advanced tissue loss; even with similar Rutherford scores, as shown in this study. Thus, the severity of tissue loss and degree of CLI itself may portend worsened long-term angioplasty success. Although the glycated hemoglobin was higher in the group with a higher FBG, this difference was not clinically very great, suggesting that underlying poorly controlled diabetes was not to blame for the patency differences found. Third, there are no data on the patient’s periprocedural treatment of hyperglycemia; was this acted upon, or simply measured? Last, true acute limb ischemia usually does not manifest with the need for infrapopliteal interventions but rather larger vessel thrombectomy. It is not clear why these patients were included in this cohort given the stress response associated with acute limb ischemia, and whether they were evenly distributed.

The pathophysiology of diabetes-related peripheral artery disease (PAD) is not well understood. Certainly, the distribution of anatomical lesions is different to that in non-diabetics, with infrapopliteal disease more common. The histopathology varies, and may be less calcific and more fibrotic, particularly in the younger patient. Basic investigations into this process have shown that hyperglycemia and decreased nitric oxide may drive the pathology, at the endothelial cellular level. ³ More work is needed, and reasonably good rodent models now exist for type I and type II diabetes, revealing the interesting effects of insulin on post-injury vascular responses. ⁴ It would have been interesting to see if the FBG correlated with C-reactive protein or brain natriuretic peptide levels, or other markers of inflammation such as interleukin-6. This could be pursued in a prospective design, or with a registry.

The paper by Singh and colleagues also underscores the idea of managing PAD in a comprehensive, multi-therapeutic manner – similar to type I diabetes ⁵ – even when diabetes is not present. Single medication or procedural interventions on patients with PAD is short sighted as there are multiple factors that affect the overall outcomes of these patients, as well as the procedural outcomes. Aggressive medical therapy empirically makes sense for these patients to have well-controlled hypertension (often neglected), tight blood sugar control, hyperlipidemic management, as well as antiplatelet therapy. While no effect on outcomes was reported in this study, only 66–71% of patients were on a statin, and only 56–59% of patients were on an angiotensin-converting enzyme (ACE) inhibitor. ¹ Control of these risk factors is even more important in those patients who have very advanced atherosclerosis, of which the infrapopliteal tibial vascular bed is reflective.