We were interested in the recent article from Pantalone et al. (1), who reported that 23.7% of 6,973 subjects with poorly controlled type 2 diabetes (HbA1c >9%) attained HbA1c below 8% within 1 year, based on the electronic health record at Cleveland Clinic. Although the 6,973 participants represented only 6.7% of the 103,969 encountered patients in the database, their fate is a matter of concern, as high rates of vascular and infectious complications have been related to HbA1c above 9%. The follow-up of HbA1c trajectories has shown that high HbA1c usually remains stable or ascending for years, and long-term frank reductions are uncommon, only occurring for 3–4% of the patients (2). Real improvement of HbA1c has however been reported 3 months after hospitalizing patients with HbA1c >9% (3), but we do not know whether it may persist in the long term.
A total of 364 subjects were hospitalized in our diabetologic ward for uncontrolled type 2 diabetes (HbA1c >9%: mean ± SD 10.6 ± 1.5%) from 2009 to 2017. The diagnosis of type 2 diabetes was confirmed on admission, and the intensification of the treatment was planned, using lifestyle modification and all the available pharmacologic treatments as indicated for individual patients: oral antidiabetes drugs (all, except thiazolidinediones and sodium–glucose cotransporter 2 inhibitors that were not available in France) and injectable drugs such as glucagon-like peptide 1 receptor agonists and insulin analogs when necessary and accepted by the patient. Capillary glucose levels were monitored before and after meals, and often at night, and doses were adapted twice a day with the active participation of the patient, owing to individual glycemic objectives. Based on the initial educational diagnosis, continuous individual education was performed by the whole team during the hospital stay.
We obtained an HbA1c result 1 year later: N = 92, HbA1c 8.4 ± 1.7% (P < 0.001 vs. initial value); 43.0% were below 8%. For 109 subjects, later HbA1c was available 4.9 ± 2.2 years after the hospitalization: 8.6 ± 1.7% (P < 0.001 vs. initial value); 40.4% were below 8%. Accordingly, in the multicentric IDAHO 2 study, Raccah et al. (4) reported that HbA1c decreased from 10.0 ± 2.2% to 7.8% after 6 months and 7.9 ± 1.4% 1 year after hospitalization for uncontrolled type 2 diabetes.
Depending on the organization of health systems, the rates of hospitalizations for therapeutic intensification for type 2 diabetes varies among countries. The financial cost of these hospitalizations, and the overbooking of hospitals, have led to reduced rates of hospitalization, but we should keep in mind that the admission in a specialized ward can durably improve the HbA1c in poorly controlled type 2 diabetes. Subjects with diabetes are indeed twice as frequently hospitalized as those without diabetes. Diabetes itself is an important primary reason for these hospitalizations, as it is the 2nd reason after angina pectoris in men (14.5%) and the first reason in women (12.5%) (5). We hypothesize that some of the 6,973 subjects studied by Pantalone et al. were hospitalized during the year between their two HbA1c determinations. If this was registered in the electronic health record system, it would be interesting to analyze whether these hospitalisations contributed to some subjects’ improvement of HbA1c.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.