Inpatient Glycemic Control in the Chinese Population

To the Editor: To our knowledge, the status of inpatient glycemic control has not been reviewed extensively in Chinese populations and has never been investigated in different care units within one hospital. Therefore, we used an electronic informatics system to extract data on inpatient point-of-care bedside glucose (POC-BG) tests for patients admitted from April to June 2012. Mean POC-BG values and hyperglycemia (>180 mg/dL) and hypoglycemia (<70 mg/dL) rates were calculated for intensive care unit (ICU) and general wards, and patient-day-weighted mean POC-BG values were calculated for general wards. A total of 3002 patients with 70 303 POC-BG measurements were analyzed. Mean POC-BG was 193.2 ± 87.3 mg/dL for patients in ICUs and 192.7 ± 89.3 mg/dL for patients in general wards. The prevalence of hospital hyperglycemia and hypoglycemia was 47.0% and 1.9%, respectively, for the ICUs and 46.4% and 1.8%, respectively, for the general wards. The mean POC-BG values did not differ between medical and surgical ICUs (193.5 ± 94.5 mg/dL vs 192.9 ± 77.7 mg/dL, P = .588); however, there were more hypoglycemic events at a value lower than 70 mg/dL in the medical ICU than in the surgical ICU (2.4% vs 1.3%, P < .01). The ICUs with lower mean POC-BG values were associated with a higher prevalence of hospital hypoglycemia. In the general wards, patients were admitted for 7.8 ± 4.6 days. The mean patient-dayweighted POC-BG value varied from 208.3 ± 99.6 mg/dL on admission day to 176.8 ± 65.6 mg/dL at the seventh day, and to 179.2 ± 63.5 mg/dL at the day before discharge. There was no difference between medical and surgical wards analyzed day by day. Analysis for patientday-weighted mean POC-BG values based on ward characteristics among 7 medical wards was carried out: 3 wards with the subspecialty of (a) chest medicine, (b) hematology, and (c) rheumatology, cardiology, and endocrine and metabolism had significantly higher mean patient-day-weighted POC-BG values than the other 4 wards during the first week of admission. These observations were not found after the first week of admission nor at the day before discharge. The patient-day-weighted mean POC-BG values of the latter 4 medical wards did not differ during admission. We could not find any difference in the mean patient-day-weighted POC-BG values among surgical wards. American Association of Clinical Endocrinologists/ American Diabetes Association consensus statements have proposed that glucose values should be maintained in the range of 140 to 180 mg/dL for ICU glycemic control. Some meta-analyses have indicated that tight glycemic control is associated with more severe hypoglycemic events, which are associated with inpatient mortality. To achieve a tight glycemic target, adequate manpower to perform frequent measurements of glucose levels and subsequently to adjust insulin dosage in time is important. Therefore, some institutions might choose a loose glycemic target if hypoglycemia risk cannot be easily reduced. In our study, 3 medical wards with a certain subspecialty had significantly higher mean patient-day-weighted POC-BG values than the others during the first week of admission. One possibility is that we admitted patients with a specific disease to the same ward and steroids are often prescribed in these wards for patients with respiratory disease, rheumatology problems, and oncology problems. Practitioners often lack the management techniques of insulin therapy to maintain glycemic control in these patients. Without a timely adjustment of the insulin dose along with changes in the steroid dosage, these patients often remain in the hyperglycemic range or at high risk of hypoglycemia. A couple of ideal insulin protocols that support consistently reaching and maintaining blood glucose within a specified optimal target range while minimizing hypoglycemia attacks should be investigated and endorsed in different settings of the hospital.