Prevalence of Diabetes, Prediabetes, & Stress Hyperglycemia
Prevalence of Diabetes, Prediabetes, & Stress Hyperglycemia
The study included 605 patients. The characteristics of the patients and the TPN infused are shown in Table 1. Of the 605 patients, 106 (17.5%) received corticosteroids, 57 (9.4%) received octreotide or somatostatin, and 13 (2.1%) received tacrolimus or cyclosporine.
Figure 1 shows the percentages of patients with carbohydrate metabolism disorders, and Table 2 shows the clinical and TPN characteristics in the patients according to the particular carbohydrate metabolism disorder. Significant differences were found between the groups with respect to age, BMI, the prior presence of comorbidity, grams of carbohydrates infused, and the values of venous blood glucose and HbA1c.
(Enlarge Image)
Figure 1.
Prevalence of diabetes and stress hyperglycemia in hospitalized patients receiving total parenteral total parenteral nutrition (TPN). Data are presented as percentages. Known diabetes: documented history of diabetes. Unknown diabetes: no record of having diabetes mellitus and glycated hemoglobin <6.5%. Prediabetes: glycated hemoglobin ≥5.7% but <6.5%. Stress hyperglycemia: glycated hemoglobin <5.7% plus blood glucose ≥126 mg/dL prior to TPN infusion.
Table 3 shows the logistic regression data for the risk of having a capillary blood glucose concentration >180 mg/dL during TPN infusion. After adjustment, an increased risk was significantly associated with the CRP level (>116 mg/L, third tertile), grams of carbohydrates in the TPN, age >65 years, HbA1c concentration >5.7%, the presence of diabetes, infectious complications, or the concomitant use of glucose-elevating drugs. Although the venous blood glucose level prior to starting TPN was not quite significant (P = .056), after taking HbA1c out of the model, those persons with baseline venous blood glucose levels >140 mg/dL had a relative risk that was 2.2-times greater (95% confidence interval, 1.279 to 3.811; P = .004) than those whose levels were lower. All the other variables mentioned above remained significant.
A total of 433 (71.6%) patients received insulin at some time during their TPN infusion (55.4% only subcutaneously and 44.6% with intravenous insulin: 35.8% in the bag and subcutaneous, and 8.8% with insulin perfusion independent of the TPN), with or without subcutaneous correctional insulin dosing. Table 4 shows the type and dose of the insulin therapy used and the degree of metabolic control achieved during the TPN infusion in the various patient groups, according to their prior metabolic status. Significant differences were seen in all the variables related to the type of insulin treatment used as well as with the metabolic control achieved.
Results
The study included 605 patients. The characteristics of the patients and the TPN infused are shown in Table 1. Of the 605 patients, 106 (17.5%) received corticosteroids, 57 (9.4%) received octreotide or somatostatin, and 13 (2.1%) received tacrolimus or cyclosporine.
Figure 1 shows the percentages of patients with carbohydrate metabolism disorders, and Table 2 shows the clinical and TPN characteristics in the patients according to the particular carbohydrate metabolism disorder. Significant differences were found between the groups with respect to age, BMI, the prior presence of comorbidity, grams of carbohydrates infused, and the values of venous blood glucose and HbA1c.
(Enlarge Image)
Figure 1.
Prevalence of diabetes and stress hyperglycemia in hospitalized patients receiving total parenteral total parenteral nutrition (TPN). Data are presented as percentages. Known diabetes: documented history of diabetes. Unknown diabetes: no record of having diabetes mellitus and glycated hemoglobin <6.5%. Prediabetes: glycated hemoglobin ≥5.7% but <6.5%. Stress hyperglycemia: glycated hemoglobin <5.7% plus blood glucose ≥126 mg/dL prior to TPN infusion.
Table 3 shows the logistic regression data for the risk of having a capillary blood glucose concentration >180 mg/dL during TPN infusion. After adjustment, an increased risk was significantly associated with the CRP level (>116 mg/L, third tertile), grams of carbohydrates in the TPN, age >65 years, HbA1c concentration >5.7%, the presence of diabetes, infectious complications, or the concomitant use of glucose-elevating drugs. Although the venous blood glucose level prior to starting TPN was not quite significant (P = .056), after taking HbA1c out of the model, those persons with baseline venous blood glucose levels >140 mg/dL had a relative risk that was 2.2-times greater (95% confidence interval, 1.279 to 3.811; P = .004) than those whose levels were lower. All the other variables mentioned above remained significant.
A total of 433 (71.6%) patients received insulin at some time during their TPN infusion (55.4% only subcutaneously and 44.6% with intravenous insulin: 35.8% in the bag and subcutaneous, and 8.8% with insulin perfusion independent of the TPN), with or without subcutaneous correctional insulin dosing. Table 4 shows the type and dose of the insulin therapy used and the degree of metabolic control achieved during the TPN infusion in the various patient groups, according to their prior metabolic status. Significant differences were seen in all the variables related to the type of insulin treatment used as well as with the metabolic control achieved.