JAMDA xxx (2015) e1ee6
JAMDA journal homepage: www.jamda.com
Original Study
Severe Hypoglycemia Is Associated With Antidiabetic Oral Treatment Compared With Insulin Analogs in Nursing Home Patients With Type 2 Diabetes and Dementia: Results From the DIMORA Study Angela Marie Abbatecola MD, PhD a, *, Mario Bo MD b, Mario Barbagallo MD c, Raffaele Antonelli Incalzi MD d, Alberto Pilotto MD e, Giuseppe Bellelli MD f, g, h, Stefania Maggi MD i, Giuseppe Paolisso MD j on behalf of the Italian Society of Gerontology and Geriatrics (SIGG), Florence, Italy a
Italian Society of Gerontology and Geriatrics (SIGG), Florence, Italy Geriatric Section, Department of Medical Sciences, San Giovanni Battista Hospital, University of Turin, Turin, Italy c Department of Internal Medicine and Medical Specialties, University of Palermo, Palermo, Italy d Department of Geriatrics, Campus Bio-Medico University, Rome, Italy e Azienda ULSS, Geriatrics Unit, St. Antonio Hospital, Padua, Italy f Department of Health Sciences, University of Milano-Bicocca, Milan, Italy g Acute Geriatric Unit, S Gerardo Hospital, Monza, Italy h Milan Center for Neuroscicence (Neuro-Mi), Milan, Italy i National Research Council (CNR), Neuroscience Institute Aging Branch, Padua, Italy j Department of Medical, Surgical, Neurological, Metabolic and Geriatric Sciences, Second University of Naples, Naples, Italy b
a b s t r a c t Keywords: Severe hypoglycemia dementia insulin analogs antidiabetic oral agents aging
Objectives: Severe hypoglycemia is associated with cognitive decline and dementia in older persons with type 2 diabetes. The role of antidiabetic treatments on severe hypoglycemia is unknown in dementia. The aims were to determine the prevalence of severe hypoglycemic events and investigate associations among severe hypoglycemic and specific antidiabetic treatments (classes of oral agents and types of insulin analogs) in a large sample of nursing home patients with diabetes according to dementia status. Design: Cross-sectional observational study. Setting: A total of 150 nursing homes across Italy. Participants: A total of 2258 patients with type 2 diabetes (dementia ¼ 1138, no dementia ¼ 1120). Measurements: Diagnosis of dementia before nursing home admission. Data were collected regarding functional status, glycemic control, antidiabetic treatments, comorbidities, and biochemical and clinical measurements. Logistic regression models with severe hypoglycemia as the dependent variable were used to test associations with antidiabetic agents. Results: Patients had a mean age (SD) of 82 (8) years, body mass index (BMI) of 25.4 (4.8), fasting plasma glucose (FPG) of 7.5 (3.0) mmol/L, postprandial glucose (PPG) of 10.3 (3.6) mmol/L, HbA1c of 7.1% (54 mmol/L), and impairments in activities of daily living (ADLs) of 3.7 (2.1). Severe hypoglycemia was more prevalent in patients with dementia (18%) compared with patients without dementia (8%). Patients with dementia were older, showed greater ADL impairments, greater number of comorbidities, lower FPG, and higher PPG compared with those without dementia. Adjusted logistic regression models in patients with dementia showed that rapid- and long-acting insulin analogs were associated with reduced odds ratio (OR) (OR 0.333; 95% confidence interval [CI] 0.184e0.602; OR 0.248, 95% CI 0.070 e0.882, respectively), whereas sulphonylureas and combined metformin þ sulphonylurea were associated with increased ORs (OR 8.805, 95% CI 4.260e18.201; OR 6.639; 95% CI 3.273e14.710, respectively) of experiencing severe hypoglycemia. No correlations were found in patients without dementia.
The authors declare no conflicts of interest. This study was funded by an unconditional grant from the research Foundation of Sanofi-Aventis, Italy, to the SIGG.
http://dx.doi.org/10.1016/j.jamda.2014.12.014 1525-8610/Ó 2015 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
* Address correspondence to Angela M. Abbatecola, MD, PhD, Italian National Research Center on Aging (INRCA), Scientific Direction, Via Santa Margherita 5, 60124 Ancona, Italy. E-mail address: [email protected] (A.M. Abbatecola).
e2
A.M. Abbatecola et al. / JAMDA xxx (2015) e1ee6
Conclusion: In older nursing home patients with type 2 diabetes, severe hypoglycemia was significantly higher in dementia. Our findings suggest that sulphonylureas should be used with caution, whereas rapid- and long-acting insulin analogs seem safer. Ó 2015 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
There is a rapid and growing interest regarding the complex role of severe hypoglycemia on cognitive impairment and dementia in older persons with type 2 diabetes.1e3 Glucose, the primary energetic substrate for neurons, cannot be synthesized or stored in brain tissue, thus glycemic control is essential for accurate cognitive functioning. Low blood glucose causes impaired cognitive functioning, which is reversible if recognized and quickly corrected. However, if not rapidly corrected, a state of prolonged severe hypoglycemia will provoke permanent neuronal damage.4 Older persons with dementia are less able to manage complex treatment regimens and recognize initial hypoglycemic symptoms. Therefore, these individuals have difficulty avoiding severe hypoglycemia, which in turn, facilitates accelerated neuronal deterioration5 and potential worsening of remaining cognitive abilities. One study found that a small subgroup of community-dwelling elders with type 2 diabetes and dementia were at a threefold risk for a severe hypoglycemic event.2 The authors also found that treatment with insulin was associated with an increased risk of hypoglycemia.2,3 However, this study did not test for different risks according to specific types of insulin or oral antidiabetic agents, probably because of the small study size (n ¼ 28).2,3 Older nursing home patients represent a specific population of very frail individuals who need assistance in performing daily activities. Nursing home patients are closely monitored, and medically verified severe hypoglycemia is a validated finding compared with self-reported measures often used in community-dwelling population studies. For this reason, older patients with type 2 diabetes in nursing homes are a particular group that can be easily monitored.6 Recent guidelines have underlined the importance of maintaining flexible glycemic control in older frail nursing home patients with the aim to avoid hypoglycemia.7e11 However, specific antidiabetic treatment regimens designed to reach and maintain “safe” glycemic control, including those with dementia, have not been identified. These recommendations are based on data extrapolation from younger adults and expert opinion citing reliable evidence. Considering the urgent need of a clinical trial basis for testing of antidiabetic agents in the rising population of older frail nursing home patients, this present study aimed at identifying the prevalence of severe hypoglycemic events in a large sample of nursing home patients with diabetes with and without dementia (n ¼ 2258). We also investigated for different associations among severe hypoglycemic and specific antidiabetic treatments (classes of oral agents and types of insulin analogs). Methods The DIMORA (“DIabete MellitO in RsA” [Diabetes Mellitus in the Nursing Home]) is an observational cohort study that was based on data from 150 nursing homes across different regions of Italy. A total of 2258 nursing home residents with type 2 diabetes aged 65 to 110 years (mean age: 83 7) were enrolled between 2011 and 2013. Inclusion criteria included age older than 65 years, diagnosis of type 2 diabetes mellitus of at least 1 year, and complete information regarding antidiabetic treatment and glycemic control. Information regarding diabetic health status was assessed using a questionnaire regarding information on antidiabetic treatment and laboratory assays on glycemic control from medical nursing
home staff (nurses and/or physicians). Data regarding physical and cognitive functional status were also collected. The data collection form was designed by the “Società Italiana di Gerontologia e Geriatria” (Italian Society of Gerontology and Geriatrics [SIGG]) scientific board and communicated to the research foundation Sanofi-Aventis (Italy). The Quintiles Company (Milan, Italy) was responsible for collecting all of the data collection forms, performing the data-entry process, and structuring and testing the homogeneity of the data over a 6-month period. A scientific SIGG board member, expert in medical statistics, used the provided database to perform all statistical analyses without any potential conflicts of interest relevant to this report. In this report, the total study population consisted of 2258 individuals who were categorized according to the presence (n ¼ 1138) or absence (n ¼ 1120) of diagnosed dementia12 before nursing home admittance. Antidiabetic treatment information was collected regarding the administration of antidiabetic oral drugs: sulphonylureas, metformin, glinides, a-glucosidase inhibitors, thiazolidinediones, Glucagon-Like Peptide-1 (GLP-1) analogs and Dipeptidyl peptidase-4 (DPP-4) inhibitors and in combination; and the type of insulin analogs: rapid, short, intermediate, or long acting, or premixed. At the time of data collection, physical functional status was measured using the basic activities of daily living (ADLs) and calculated from the questionnaire regarding physical independency levels ranging from 0 to 6 (maximum independency).13 All participants underwent an anthropometrical evaluation, including body mass index (BMI), calculated as weight in kilograms divided by the square of height in meters. Daily glycemic control and monitoring included data collection regarding information on fasting plasma glucose (FPG) and postprandial glucose (PPG) evaluations. Hemoglobin A1c (HbA1c) levels also were recorded during this observation period. Information was reported regarding the presence of specific comorbidities, including heart failure, cardiovascular disease, stroke, chronic obstructive pulmonary disease, cancer, osteoporosis, osteoarthritis, Parkinson disease, anemia, chronic respiratory failure, chronic renal failure, liver disease, arterial hypertension, gastrointestinal disease, bone fractures, and others (peripheral arterial disease, diabetic neuropathy, diabetic retinopathy, depression, hypothyroidism, prostate hypertrophy). The number of nondiabetic agents administered daily also was recorded. Severe hypoglycemic events were reported and defined as required documentation of a plasma glucose of 50 mg/dL (2.8 mmol/L) or lower and symptoms requiring assistance by a third party to administer oral carbohydrate, intravenous glucose, or parenteral glucagon14,15 during the stay in the nursing home. Episodes requiring hospitalization or care in an emergency department or from emergency personnel were recorded. The Scientific Review Board of the SIGG in Florence, Italy, approved the study protocol. All participating nursing homes in the Italian regions (Abruzzo, Campania, Lazio, Lombardia, Marche, Piemonte, Puglia, Sicilia, Veneto) were informed about the study design and data collection procedures. Patient informed consent was obtained at contributing nursing homes accordingly. All statistical analyses were performed using the statistical software SPSS, version 17.0 (SPSS, Chicago, IL). All data are presented as mean (SD) unless otherwise indicated. Analysis of variance (ANOVA)
A.M. Abbatecola et al. / JAMDA xxx (2015) e1ee6
e3
Table 1 Clinical Characteristics of Patients With Type 2 Diabetes in the Italian Nursing Home Study Population (n ¼ 2258) Characteristic
All, n ¼ 2258
Dementia, n ¼ 1138
No Dementia, n ¼ 1113
P
Age, y Sex (men/women) BMI, kg/m2 FPG, mmol/L PPG, mmol/L HbA1c mmol, % Diabetes duration, y Severe hypoglycemic events, n (%) Impairment in ADL Length of stay, y Comorbidities, n Nephropathy, n (%) Coronary heart disease, n (%) Cerebrovascular disease, n (%) Drugs, n Antidiabetic treatment AOD, n (%) Insulin, n (%) Insulin þ AOD, n (%) Diet, n (%)
82 (8) 694/1564 25.4 (4.8) 7.5 (3.0) 10.2 (3.5) 54 (14), 7.1 (1.3) 12.6 (8.1) 316 (14) 3.7 (1.8) 3.1 6 3.4 (1.3) 293 (13) 644 (29) 494 (22) 5.1 (1.3)
83 (8) 305/833 24.4 (4.7) 7.3 (3.0) 10.5 (3.7) 53 (12), 7.0 (1.2) 13.3 (8.2) 202 (18) 4.4 (1.2) 3.3 (7.3) 3.9 (1.2) 145 (13) 317 (28) 271 (24) 5.2 (1.2)
81 (8) 389/724 26.3 (4.7) 7.6 (3.0) 9.9 (3.4) 54 (15), 7.1 (1.4) 11.9 (8) 87 (8) 2.9 (1.9) 2.9 (4.8) 2.8 (1.4) 148 (13) 327 (30) 223 (20) 5.0 (1.3)
JAMDA journal homepage: www.jamda.com
Original Study
Severe Hypoglycemia Is Associated With Antidiabetic Oral Treatment Compared With Insulin Analogs in Nursing Home Patients With Type 2 Diabetes and Dementia: Results From the DIMORA Study Angela Marie Abbatecola MD, PhD a, *, Mario Bo MD b, Mario Barbagallo MD c, Raffaele Antonelli Incalzi MD d, Alberto Pilotto MD e, Giuseppe Bellelli MD f, g, h, Stefania Maggi MD i, Giuseppe Paolisso MD j on behalf of the Italian Society of Gerontology and Geriatrics (SIGG), Florence, Italy a
Italian Society of Gerontology and Geriatrics (SIGG), Florence, Italy Geriatric Section, Department of Medical Sciences, San Giovanni Battista Hospital, University of Turin, Turin, Italy c Department of Internal Medicine and Medical Specialties, University of Palermo, Palermo, Italy d Department of Geriatrics, Campus Bio-Medico University, Rome, Italy e Azienda ULSS, Geriatrics Unit, St. Antonio Hospital, Padua, Italy f Department of Health Sciences, University of Milano-Bicocca, Milan, Italy g Acute Geriatric Unit, S Gerardo Hospital, Monza, Italy h Milan Center for Neuroscicence (Neuro-Mi), Milan, Italy i National Research Council (CNR), Neuroscience Institute Aging Branch, Padua, Italy j Department of Medical, Surgical, Neurological, Metabolic and Geriatric Sciences, Second University of Naples, Naples, Italy b
a b s t r a c t Keywords: Severe hypoglycemia dementia insulin analogs antidiabetic oral agents aging
Objectives: Severe hypoglycemia is associated with cognitive decline and dementia in older persons with type 2 diabetes. The role of antidiabetic treatments on severe hypoglycemia is unknown in dementia. The aims were to determine the prevalence of severe hypoglycemic events and investigate associations among severe hypoglycemic and specific antidiabetic treatments (classes of oral agents and types of insulin analogs) in a large sample of nursing home patients with diabetes according to dementia status. Design: Cross-sectional observational study. Setting: A total of 150 nursing homes across Italy. Participants: A total of 2258 patients with type 2 diabetes (dementia ¼ 1138, no dementia ¼ 1120). Measurements: Diagnosis of dementia before nursing home admission. Data were collected regarding functional status, glycemic control, antidiabetic treatments, comorbidities, and biochemical and clinical measurements. Logistic regression models with severe hypoglycemia as the dependent variable were used to test associations with antidiabetic agents. Results: Patients had a mean age (SD) of 82 (8) years, body mass index (BMI) of 25.4 (4.8), fasting plasma glucose (FPG) of 7.5 (3.0) mmol/L, postprandial glucose (PPG) of 10.3 (3.6) mmol/L, HbA1c of 7.1% (54 mmol/L), and impairments in activities of daily living (ADLs) of 3.7 (2.1). Severe hypoglycemia was more prevalent in patients with dementia (18%) compared with patients without dementia (8%). Patients with dementia were older, showed greater ADL impairments, greater number of comorbidities, lower FPG, and higher PPG compared with those without dementia. Adjusted logistic regression models in patients with dementia showed that rapid- and long-acting insulin analogs were associated with reduced odds ratio (OR) (OR 0.333; 95% confidence interval [CI] 0.184e0.602; OR 0.248, 95% CI 0.070 e0.882, respectively), whereas sulphonylureas and combined metformin þ sulphonylurea were associated with increased ORs (OR 8.805, 95% CI 4.260e18.201; OR 6.639; 95% CI 3.273e14.710, respectively) of experiencing severe hypoglycemia. No correlations were found in patients without dementia.
The authors declare no conflicts of interest. This study was funded by an unconditional grant from the research Foundation of Sanofi-Aventis, Italy, to the SIGG.
http://dx.doi.org/10.1016/j.jamda.2014.12.014 1525-8610/Ó 2015 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
* Address correspondence to Angela M. Abbatecola, MD, PhD, Italian National Research Center on Aging (INRCA), Scientific Direction, Via Santa Margherita 5, 60124 Ancona, Italy. E-mail address: [email protected] (A.M. Abbatecola).
e2
A.M. Abbatecola et al. / JAMDA xxx (2015) e1ee6
Conclusion: In older nursing home patients with type 2 diabetes, severe hypoglycemia was significantly higher in dementia. Our findings suggest that sulphonylureas should be used with caution, whereas rapid- and long-acting insulin analogs seem safer. Ó 2015 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
There is a rapid and growing interest regarding the complex role of severe hypoglycemia on cognitive impairment and dementia in older persons with type 2 diabetes.1e3 Glucose, the primary energetic substrate for neurons, cannot be synthesized or stored in brain tissue, thus glycemic control is essential for accurate cognitive functioning. Low blood glucose causes impaired cognitive functioning, which is reversible if recognized and quickly corrected. However, if not rapidly corrected, a state of prolonged severe hypoglycemia will provoke permanent neuronal damage.4 Older persons with dementia are less able to manage complex treatment regimens and recognize initial hypoglycemic symptoms. Therefore, these individuals have difficulty avoiding severe hypoglycemia, which in turn, facilitates accelerated neuronal deterioration5 and potential worsening of remaining cognitive abilities. One study found that a small subgroup of community-dwelling elders with type 2 diabetes and dementia were at a threefold risk for a severe hypoglycemic event.2 The authors also found that treatment with insulin was associated with an increased risk of hypoglycemia.2,3 However, this study did not test for different risks according to specific types of insulin or oral antidiabetic agents, probably because of the small study size (n ¼ 28).2,3 Older nursing home patients represent a specific population of very frail individuals who need assistance in performing daily activities. Nursing home patients are closely monitored, and medically verified severe hypoglycemia is a validated finding compared with self-reported measures often used in community-dwelling population studies. For this reason, older patients with type 2 diabetes in nursing homes are a particular group that can be easily monitored.6 Recent guidelines have underlined the importance of maintaining flexible glycemic control in older frail nursing home patients with the aim to avoid hypoglycemia.7e11 However, specific antidiabetic treatment regimens designed to reach and maintain “safe” glycemic control, including those with dementia, have not been identified. These recommendations are based on data extrapolation from younger adults and expert opinion citing reliable evidence. Considering the urgent need of a clinical trial basis for testing of antidiabetic agents in the rising population of older frail nursing home patients, this present study aimed at identifying the prevalence of severe hypoglycemic events in a large sample of nursing home patients with diabetes with and without dementia (n ¼ 2258). We also investigated for different associations among severe hypoglycemic and specific antidiabetic treatments (classes of oral agents and types of insulin analogs). Methods The DIMORA (“DIabete MellitO in RsA” [Diabetes Mellitus in the Nursing Home]) is an observational cohort study that was based on data from 150 nursing homes across different regions of Italy. A total of 2258 nursing home residents with type 2 diabetes aged 65 to 110 years (mean age: 83 7) were enrolled between 2011 and 2013. Inclusion criteria included age older than 65 years, diagnosis of type 2 diabetes mellitus of at least 1 year, and complete information regarding antidiabetic treatment and glycemic control. Information regarding diabetic health status was assessed using a questionnaire regarding information on antidiabetic treatment and laboratory assays on glycemic control from medical nursing
home staff (nurses and/or physicians). Data regarding physical and cognitive functional status were also collected. The data collection form was designed by the “Società Italiana di Gerontologia e Geriatria” (Italian Society of Gerontology and Geriatrics [SIGG]) scientific board and communicated to the research foundation Sanofi-Aventis (Italy). The Quintiles Company (Milan, Italy) was responsible for collecting all of the data collection forms, performing the data-entry process, and structuring and testing the homogeneity of the data over a 6-month period. A scientific SIGG board member, expert in medical statistics, used the provided database to perform all statistical analyses without any potential conflicts of interest relevant to this report. In this report, the total study population consisted of 2258 individuals who were categorized according to the presence (n ¼ 1138) or absence (n ¼ 1120) of diagnosed dementia12 before nursing home admittance. Antidiabetic treatment information was collected regarding the administration of antidiabetic oral drugs: sulphonylureas, metformin, glinides, a-glucosidase inhibitors, thiazolidinediones, Glucagon-Like Peptide-1 (GLP-1) analogs and Dipeptidyl peptidase-4 (DPP-4) inhibitors and in combination; and the type of insulin analogs: rapid, short, intermediate, or long acting, or premixed. At the time of data collection, physical functional status was measured using the basic activities of daily living (ADLs) and calculated from the questionnaire regarding physical independency levels ranging from 0 to 6 (maximum independency).13 All participants underwent an anthropometrical evaluation, including body mass index (BMI), calculated as weight in kilograms divided by the square of height in meters. Daily glycemic control and monitoring included data collection regarding information on fasting plasma glucose (FPG) and postprandial glucose (PPG) evaluations. Hemoglobin A1c (HbA1c) levels also were recorded during this observation period. Information was reported regarding the presence of specific comorbidities, including heart failure, cardiovascular disease, stroke, chronic obstructive pulmonary disease, cancer, osteoporosis, osteoarthritis, Parkinson disease, anemia, chronic respiratory failure, chronic renal failure, liver disease, arterial hypertension, gastrointestinal disease, bone fractures, and others (peripheral arterial disease, diabetic neuropathy, diabetic retinopathy, depression, hypothyroidism, prostate hypertrophy). The number of nondiabetic agents administered daily also was recorded. Severe hypoglycemic events were reported and defined as required documentation of a plasma glucose of 50 mg/dL (2.8 mmol/L) or lower and symptoms requiring assistance by a third party to administer oral carbohydrate, intravenous glucose, or parenteral glucagon14,15 during the stay in the nursing home. Episodes requiring hospitalization or care in an emergency department or from emergency personnel were recorded. The Scientific Review Board of the SIGG in Florence, Italy, approved the study protocol. All participating nursing homes in the Italian regions (Abruzzo, Campania, Lazio, Lombardia, Marche, Piemonte, Puglia, Sicilia, Veneto) were informed about the study design and data collection procedures. Patient informed consent was obtained at contributing nursing homes accordingly. All statistical analyses were performed using the statistical software SPSS, version 17.0 (SPSS, Chicago, IL). All data are presented as mean (SD) unless otherwise indicated. Analysis of variance (ANOVA)
A.M. Abbatecola et al. / JAMDA xxx (2015) e1ee6
e3
Table 1 Clinical Characteristics of Patients With Type 2 Diabetes in the Italian Nursing Home Study Population (n ¼ 2258) Characteristic
All, n ¼ 2258
Dementia, n ¼ 1138
No Dementia, n ¼ 1113
P
Age, y Sex (men/women) BMI, kg/m2 FPG, mmol/L PPG, mmol/L HbA1c mmol, % Diabetes duration, y Severe hypoglycemic events, n (%) Impairment in ADL Length of stay, y Comorbidities, n Nephropathy, n (%) Coronary heart disease, n (%) Cerebrovascular disease, n (%) Drugs, n Antidiabetic treatment AOD, n (%) Insulin, n (%) Insulin þ AOD, n (%) Diet, n (%)
82 (8) 694/1564 25.4 (4.8) 7.5 (3.0) 10.2 (3.5) 54 (14), 7.1 (1.3) 12.6 (8.1) 316 (14) 3.7 (1.8) 3.1 6 3.4 (1.3) 293 (13) 644 (29) 494 (22) 5.1 (1.3)
83 (8) 305/833 24.4 (4.7) 7.3 (3.0) 10.5 (3.7) 53 (12), 7.0 (1.2) 13.3 (8.2) 202 (18) 4.4 (1.2) 3.3 (7.3) 3.9 (1.2) 145 (13) 317 (28) 271 (24) 5.2 (1.2)
81 (8) 389/724 26.3 (4.7) 7.6 (3.0) 9.9 (3.4) 54 (15), 7.1 (1.4) 11.9 (8) 87 (8) 2.9 (1.9) 2.9 (4.8) 2.8 (1.4) 148 (13) 327 (30) 223 (20) 5.0 (1.3)
