Regístrese
Buscar en
Endocrinología y Nutrición (English Edition)
Toda la web
Inicio Endocrinología y Nutrición (English Edition) Predictors of mean blood glucose control and its variability in diabetic hospita...
Journal Information
Vol. 62. Issue 6.
Pages 257-263 (June - July 2015)
Visits
1185
Vol. 62. Issue 6.
Pages 257-263 (June - July 2015)
Original article
DOI: 10.1016/j.endoen.2015.06.002
Full text access
Predictors of mean blood glucose control and its variability in diabetic hospitalized patients
Factores predictores del control glucémico promedio y de su variabilidad en pacientes diabéticos ingresados en el hospital
Visits
1185
Daniel Sáenz-Abada,
Corresponding author
danielsaenzabad@hotmail.com

Corresponding author.
, José Antonio Gimeno-Ornab, Beatriz Sierra-Berguaa, Juan Ignacio Pérez-Calvoc
a Servicio de Urgencias, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain
b Servicio de Endocrinología, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain
c Servicio de Medicina Interna, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain
This item has received
1185
Visits
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (1)
Tables (3)
Table 1. Patient characteristics and metabolic control at admission.
Table 2. Measurements of mean blood glucose (MBG), standard deviation (GV), and coefficient of variation (CV) depending on the baseline characteristics of patients.
Table 3. Univariate predictors of mean blood glucose (MBG), standard deviation (GV), and coefficient of variation (CV).
Show moreShow less
Abstract
Introduction

This study was intended to assess the effectiveness and predictors factors of inpatient blood glucose control in diabetic patients admitted to medical departments.

Material and methods

A retrospective, analytical cohort study was conducted on patients discharged from internal medicine with a diagnosis related to diabetes. Variables collected included demographic characteristics, clinical data and laboratory parameters related to blood glucose control (HbA1c, basal plasma glucose, point-of-care capillary glucose). The cumulative probability of receiving scheduled insulin regimens was evaluated using Kaplan–Meier analysis. Multivariate regression models were used to select predictors of mean inpatient glucose (MHG) and glucose variability (standard deviation [GV]).

Results

The study sample consisted of 228 patients (mean age 78.4 (SD 10.1) years, 51% women). Of these, 96 patients (42.1%) were treated with sliding-scale regular insulin only. Median time to start of scheduled insulin therapy was 4 (95% CI, 2–6) days. Blood glucose control measures were: MIG 181.4 (SD 41.7) mg/dL, GV 56.3 (SD 22.6).

The best model to predict MIG (R2: .376; p<.0001) included HbA1c (b=4.96; p=.011), baseline plasma glucose (b=.056; p=.084), mean capillary blood glucose in the first 24h (b=.154; p<.0001), home treatment (versus oral agents) with basal insulin only (b=13.1; p=.016) or more complex (pre-mixed insulin or basal-bolus) regimens (b=19.1; p=.004), corticoid therapy (b=14.9; p=.002), and fasting on admission (b=10.4; p=.098).

Conclusion

Predictors of inpatient blood glucose control which should be considered in the design of DM management protocols include home treatment, HbA1c, basal plasma glucose, mean blood glucose in the first 24h, fasting, and corticoid therapy.

Keywords:
Hyperglycemia
Diabetes
Insulin
Inpatient blood glucose control
Hospital management of hyperglycemia
Resumen
Introducción

Nuestros objetivos fueron evaluar el control glucémico intrahospitalario de pacientes con diabetes mellitus (DM) y determinar sus factores predictores.

Material y métodos

Estudio de cohortes retrospectivo analítico con inclusión de pacientes dados de alta de medicina interna con un diagnóstico relacionado con la DM. Se recogieron variables clínicas (demográficas y relacionadas con el manejo intrahospitalario del paciente) y analíticas relacionadas con el control glucémico (HbA1c, glucemia plasmática inicial, glucemias capilares durante el ingreso). Se evaluó la probabilidad de recibir insulina programada mediante curvas de Kaplan Meier y los factores predictores de la glucemia media (GM) y de su variabilidad (desviación estándar [VG]) mediante regresión múltiple.

Resultados

Se incluyeron 228 pacientes (edad media 78,4 [DE 10,1] años, 51% mujeres); 96 (42,1%) recibieron solo pauta correctora. La mediana del tiempo hasta el inicio de insulina programada fue 4 días (IC 95%: 2–6). Las medidas de control fueron: GM 181,4 (DE 41,7) mg/dl, VG 56,3 (DE 22,6) mg/dl.

El mejor modelo predictor de la GM (R2: 0,376; p<0,0001) incluyó HbA1c (b=4,96; p=0,011), glucemia plasmática inicial (b=0,056; p=0,084), glucemia media de las primeras 24h (b=0,154; p<0,0001), tratamiento domiciliario (versus antidiabéticos orales) con insulina basal (b=13,1; p=0,016) o mezclas o basal-bolo (b=19,1; p=0,004), tratamiento con corticoides (b=14,9; p=0,002) y ayuno al ingreso (b=10,4; p=0,098).

Conclusión

Los determinantes del control glucémico intrahospitalario, que deberían considerarse en protocolos de actuación, son el tratamiento previo, la HbA1c, la glucemia inicial y media de las primeras 24h de ingreso, el ayuno y la utilización de corticoides.

Palabras clave:
Hiperglucemia
Diabetes
Insulina
Control glucémico hospitalario
Tratamiento hospitalario de la hiperglucemia
Full Text
Introduction

Diabetes mellitus (DM) is a chronic disease of carbohydrate metabolism affecting 371 million people worldwide and with an increasing prevalence.1 According to a recent study, DM prevalence in Spain is 13.8%.2 The hospital management of patients with DM is a common problem, as they are estimated to account for 30–40% of patients attending the emergency room and for up 25% of admissions to both medical and surgical hospital wards.3 When these figures are interpreted, however, one should also bear in mind that this is still an underdiagnosed disease, because this diagnosis is not reported at discharge in one third of patients with hyperglycemia upon hospital admission.4

DM is not only a common problem, but also a relevant one, because hyperglycemia upon admission is associated with greater morbidity and mortality5 and resource utilization, with a resultant increase in costs.6 Despite this epidemiological association, few interventional studies intended to show the efficacy of intensive blood glucose control in inpatients not admitted to intensive care units (ICUs) are available; a recent meta-analysis7 showed that intensive hospital treatment of hyperglycemia outside the ICU decreased infections (RR=0.41; 95% CI: 0.21–0.77), but not mortality, with a borderline significant increase in the risk of hypoglycemia (RR=1.58; 95% CI: 0.97–2.57).

Based on the available evidence on the risks and benefits of metabolic control in non-ICU hospitalized patients, medical bodies8 now recommend preprandial and random blood glucose levels of <140mg/dL and <180mg/dL respectively (<200mg/dL in patients with limited life expectation or a high risk of hypoglycemia). These goals may be difficult to achieve9 due to the sometimes unpredictable nutritional intake, the stress from disease, with an increase in counterregulatory hormones, and the use of hyperglycemic drugs, especially corticosteroids.

An additional factor that impairs hyperglycemic control in hospitals is clinical inertia, due to the difficulty in persuading staff responsible for patients to change their beliefs and behavior.3 To overcome this inertia, educational programs have been suggested. These would be reinforced by the distribution of hospital management protocols, followed by regular evaluations of their implementation and efficacy at the hospital.10 The basic rules for preparing a protocol for hyperglycemia include the establishment of glycemic goals, the calculation of total insulin dose (TID) with its basal and prandial components (avoiding the use of corrective rapid insulin regimens [RIRs] only), and guidance on insulin adjustment during hospital stay and when treatment on discharge is being planned.11

This study was undertaken to assess whether the treatment of patients with DM at our hospital complied with the recommendations made by national and international medical associations.3,8 The specific objectives included:

  • (1)

    To assess the use of scheduled insulin versus RIR alone in patients with DM.

  • (2)

    To assess glycemic control in patients with DM.

  • (3)

    To find out the predictors of inpatient glycemic control, which would help us to design specific medical protocols for improving the management of patients with DM.

Patients and methodsDesign

This was a retrospective descriptive and analytical cohort study.

Patients

All patients discharged from the internal medicine department of Hospital Clínico Universitario Lozano Blesa in Saragossa for approximately four consecutive months (from January 3 to April 25, 2010) with some diagnosis (primary or secondary) included in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) related to DM (DM, secondary DM, abnormal glucose, and postoperative hypoinsulinemia) were enrolled into the study. Patients in whom management upon admission required treatment at the ICU and patients with a primary diagnosis of diabetic ketoacidosis or hyperosmolar hyperglycemic nonketotic decompensation requiring the use of initial intravenous insulin were excluded.

Variables

The following variables were collected from each patient.

  • (1)

    Clinical variables: age at admission, sex, reason for admission, Charlson comorbidity index, type of DM, time since DM onset, type of outpatient treatment for DM, length of hospital stay, intake or fasting state upon admission, use of corticosteroids during admission, and type of treatment at hospital discharge.

  • (2)

    Variables related to glycemic control: use or otherwise (and day of start) of scheduled insulin, plasma blood glucose at admission, HbA1c measured during hospital stay or in the previous three months, glucose values obtained by capillary glucose monitoring, and frequency and severity of hypoglycemia (hypoglycemia was defined as blood glucose values <70mg/dL, and considered severe if glucose values were <40mg/dL or concomitant loss of consciousness occurred).

  • (3)

    Generated variables: the Charlson score was codified using three categories <3 (reference category), 3 and >3. Outpatient treatment before admission was recodified using three categories (diet and/or oral antidiabetic drugs as reference category, basal insulin with or without oral antidiabetic drugs and mixtures or basal-bolus) because of the small number of patients on basal-bolus treatment. To assess inpatient glycemic control, the following were estimated from capillary blood glucose values: maximum blood glucose level (the highest capillary blood glucose value during hospital stay), mean blood glucose (MBG) in the first 24h, MBG during hospital stay, defined as the mean of all capillary blood glucose values, corrected MBG (the mean of the daily means of capillary blood glucose values), standard deviation of all capillary blood glucose values of the patient (glucose variability, GV) and coefficient of variation (CV), defined as the ratio between standard deviation and mean blood glucose (SD/MBG).

Laboratory methods

Plasma blood glucose was measured using an enzymatic method with hexokinase, HbA1c by high-performance chromatography, and capillary blood glucose using an Optium Xceed® meter with a precision of 3–3.6% (CV), an accuracy r=0.98 for plasma blood glucose, and a 99% compliance with the ISO standard.

Statistical methods

Quantitative variables are presented with the mean and standard deviation (SD), while qualitative variables are given as frequency distribution. Quantitative variables were compared using a Student's t test for independent samples, ANOVA, or non-parametric Mann–Whitney or Kruskal–Wallis tests. Qualitative variables were compared using a Chi-square test or Fisher's exact test. The correlation between quantitative variables (mean blood glucose and corrected mean blood glucose) was evaluated using the Pearson correlation coefficient.

The probability of receiving scheduled insulin during the hospital stay was determined using Kaplan–Meier survival curves. A log-rank test was used to compare the curves depending on the outpatient treatment received.

The predictors of inpatient glycemic control were determined using univariate and multivariate linear regression. The dependent variables taken into account included MBG, GV, and CV. To determine the best predicting models, variables were included in multivariate models based on statistical criteria (p<0.1 in univariate analysis) and on clinical meaning. The measure of variability selected was the standard deviation of all capillary blood glucose values of the patient (GV) instead of CV, because the R2 of the predictive model was higher.

An association with a value of p<0.05 was considered significant. SPSS version 17.0 software was used for the statistical analysis.

Results

A total of 255 patients who met the inclusion criteria were initially selected. However, the clinical histories of 27 patients (10.6%) could not be reviewed, which left 228 valid patients for data analysis. Diagnoses related to DM included type 2 diabetes (n=198), secondary diabetes (n=14), and stress hypoglycemia (n=16).

The mean age was 78.4 (SD 10.1) years, with a similar proportion of males (49%) and females (51%). Table 1 shows the clinical characteristics of the patients at admission. The main reasons for admission were dyspnea (42%) and febrile syndrome (12.3%).

Table 1.

Patient characteristics and metabolic control at admission.

Quantitative variables  Mean  SD 
Age (years)  78.4  10.1 
Time since DM onset (years)  11.7  9.5 
Length of stay (days)  12.6  8.7 
Plasma blood glucose at admission (mg/dL)  195.4  94.6 
HbA1c (%)  7.3  1.5 
Blood glucose measurements during hospital stay  26.6  12.4 
Blood glucose measurements by day of hospital stay  2.4  0.74 
Blood glucose at breakfast (mg/dL)  158.1  43.3 
Blood glucose at lunch (mg/dL)  195.3  44.5 
Blood glucose at dinner (mg/dL)  189.2  53.4 
Maximum blood glucose during hospital stay (mg/dL)  305.0  81.7 
Mean blood glucose during hospital stay (mg/dL)  181.4  41.7 
Standard deviation (GV) (mg/dL)  56.3  22.6 
Coefficient of variation (%)  31  10.5 
Mean blood glucose in first 24h (mg/(dL)  212.1  75.2 
Qualitative variables  No. 
Home treatment
Diet and/or oral antidiabetic drugs  135  59.2 
Basal insulin  57  25 
Mixtures  29  12.7 
Basal-bolus  1.3 
Charlson index
<3  101  44.3 
62  27.2 
>3  65  28.5 
Fasting at admission  39  17.1 
Hypoglycemia
None  161  70.6 
Mild (40–70mg/dL)  30  13.2 
Severe (<40mg/dL or loss of consciousness)  34  14.9 
Corticosteroid treatment during hospital stay  85  37.3 
Treatment at discharge
Diet and/or oral antidiabetic drugs  126  55.2 
Basal insulin  72  31.6 
Mixtures  26  11.4 
Basal-bolus  1.8 
Place of discharge
Home  203  89 
Death  19  8.3 
Another hospital  1.3 
Voluntary discharge  1.3 

Forty-one patients (18%) were treated with oral antidiabetic drugs during their hospital stay. Scheduled insulin regimens were received by 132 patients (57.9%) during their hospital stay, while 96 patients (42.1%) received RIR alone. The median time to the start of scheduled insulin regimens (the day of the hospital stay when ≥50% of patients with DM were receiving such regimens) was four days (95% CI: 2–6).

The proportion of patients who received scheduled insulin regimens was statistically higher (p<0.001) in patients on home treatment with insulin (93%) as compared to those treated with diet or oral antidiabetic drugs (35%). The difference in survival curves, which reflected the rate of administration of scheduled insulin which was itself dependent on the initial treatment of patients, was also statistically significant (p<0.0001; Fig. 1). In a follow-up restricted to three weeks of hospitalization, the probability of receiving scheduled insulin was 49% for patients taking oral antidiabetic drugs at home and virtually 100% for patients on insulin therapy before admission.

Figure 1.

Cumulative probability of remaining free of treatment with scheduled insulin depending on the type of home treatment for diabetes mellitus received by the patient.

(0.1MB).

MBG was 181.4mg/dL (SD 41.7), GV 56.3mg/dL (SD 22.6), and CV 31% (SD 10.5). Table 2 shows blood glucose values depending on the baseline characteristics of the patients. Corrected mean blood glucose was 182.4mg/dL (SD 41.6), with a correlation coefficient of 0.992 (p<0.0001) as compared to MBG. MBG was therefore used for all calculations of regression analyses.

Table 2.

Measurements of mean blood glucose (MBG), standard deviation (GV), and coefficient of variation (CV) depending on the baseline characteristics of patients.

Measurements of blood glucose  MBG  p  GV  p  CV  p 
Age
<75 years  186.7  NS  57.3  NS  0.31  NS 
≥75 years  178.7    57.6    0.32   
Home treatment
Diet and/or oral antidiabetic drugs  171.9  0.001  52.1  <0.001  0.30  0.01 
Basal insulin  188.0    63.0    0.33   
Mixtures or basal-bolus  197.4    66.2    0.34   
Charlson index
<3  176.1  NS  53.8  0.036  0.31  NS 
188.6    59.1    0.31   
>3  182.1    61.5    0.34   
Fasting at admission
No  180.0  0.064  56.9  NS  0.31  NS 
Yes  187.3    60.5    0.33   
HbA1c
<7%  168.7  <0.001  47.9  <0.001  0.28  <0.001 
7–<9%  185.5    64.5    0.35   
≥9%  211.9    71.8    0.35   
Administration of scheduled insulin
No  168.3  <0.001  46.7  <0.001  0.28  <0.001 
Yes  189.3    64.3    0.34   
Treatment with corticosteroids during hospital stay
No  175.5  0.009  53.1  0.002  0.30  0.039 
Yes  192.1    65.6    0.34   
Blood glucose at admission
≤140mg/dL  159.5  <0.001  47.1  <0.001  0.29  0.013 
>140 and ≤180mg/dL  182.1    54.4    0.30   
>180mg/dL  194.1    65.9    0.34   
Blood glucose in first 24h
≤140mg/dL  152.0  <0.001  42.8  <0.001  0.28  0.008 
>140 and ≤180mg/dL  167.4    50.0    0.30   
>180mg/dL  195.5    64.9    0.34   

Table 3 shows the univariate predictors of blood glucose measurements. Variables consistently affecting all blood glucose measures during the hospital stay (mean blood glucose and variability) were those related to the metabolic state of the patient before admission (treatment for DM and HbA1c), the initial degree of decompensation (basal plasma blood glucose and capillary blood glucose in the first 24h after admission), and the need for corticosteroid treatment. Patients in whom scheduled insulin regimens were started also had higher mean blood glucose and glucose variability during hospitalization. In this case, the scheduled insulin regimen was probably the consequence of the greater degree of metabolic decompensation, rather than the cause. Therefore, since our primary objective was to identify factors that should be taken into consideration when designing hospital management protocols, this variable was not included in multiple regression models with predictive purposes.

Table 3.

Univariate predictors of mean blood glucose (MBG), standard deviation (GV), and coefficient of variation (CV).

Variables  MBG (mg/dL)GV (mg/dL)CV
  b  p  b  p  b  p 
HbA1c (1%)  10.46  <0.001  6.25  <0.001  0.018  <0.001 
Basal blood glucose (1mg/dL)  0.18  <0.001  0.095  <0.001  0.001  <0.001 
Glucose in first 24h (1mg/dL)  0.27  <0.001  0.17  <0.001  0.001  <0.001 
Corticosteroid treatment  15.2  0.008  10.21  0.001  0.029  0.042 
Home treatment
Basal insulin  17.6  0.006  11.9  0.001  0.038  0.024 
Mixtures or basal-bolus  24.7  0.002  15.3  <0.001  0.048  0.019 
Fasting at admission  20.7  0.004         
Charlson index (1 point)      2.91  0.001  0.012  0.005 
Scheduled insulin  17.1  0.002  18.5  <0.001  0.076  <0.001 

In the multivariate analysis, the best predictive model of MBG (R2 of the model 0.376; overall p<0.0001) consisted of HbA1c (b=4.96; 95% CI: 1.17–8.76; p=0.011), baseline plasma blood glucose (b=0.056; 95% CI: 0–0.12; p=0.084), mean blood glucose in the first 24h (b=0.154; 95% CI: 0.08–0.23; p<0.0001), home treatment with basal insulin (b=13.1; 95% CI: 2.5–23.8; p=0.016) or basal-bolus mixtures (b=19.1; 95% CI: 6.2–32; p=0.004), the need for corticosteroid treatment (b=14.9; 95% CI: 5.4–24.4; p=0.002), and fasting at admission (b=10.4; 95% CI: 1.9–22.7; p=0.098). The best predictive model for GV (R2 of the model 0.488; overall p<0.0001) consisted of age (b=0.28; 95% CI: 0.04–0.52; p=0.023), HbA1c (b=3.59; 95% CI: 1.78–5.4; p<0.0001), mean blood glucose in the first 24h (b=0.136; 95% CI: 0.1–0.17; p<0.0001), home treatment with basal insulin (b=8.35; 95% CI: 2.8–13.9; p=0.003) or basal-bolus mixtures (b=9.44; 95% CI: 2.75–16.1; p=0.006), and the need for corticosteroid treatment (b=12.2; 95% CI: 7.3–17.2; p<0.0001).

Discussion

A review of the clinical histories of patients discharged from our hospital with a diagnosis related to the presence of DM showed that the management of these patients did not completely follow the recommendations of the medical bodies. Eighteen percent of patients continued to take oral antidiabetic drugs after hospital discharge, and 42.1% received rapid insulin regimens (RIRs) for correction alone, without scheduled insulin. The preprandial MBG of patients was approximately 180mg/dL, above the recommended goal of 140mg/dL. After identifying where the management of these patients could be improved, the main factors predicting for MBG and glucose variability were found to be those related to prior outpatient management (HbA1c and type of treatment) and to the degree of initial metabolic decompensation (plasma blood glucose at baseline and in the first 24h after admission), and the need for corticosteroid treatment.

It is currently agreed that the best treatment for patients with DM is the replacement of oral antidiabetic drugs by insulin. Insulin should ideally be administered as a scheduled regimen of basal insulin (fasting condition) or as basal insulin plus prandial insulin (preserved intake condition). Except in special cases, the use of insulin as RIR only is not recommended.3,8,11 Despite these recommendations, we found at our hospital that more than 40% of patients, especially those treated on an outpatient basis with oral antidiabetic drugs, were not receiving scheduled insulin. A study conducted in 44 US hospitals12 showed that 16% of patients with type 1 DM and 35% of patients with type 2 DM only received insulin as RIR. The lack of use of scheduled insulin has been attributed to unfamiliarity with insulin therapy on the part of the professionals responsible for patient care, the fear of hypoglycemia, clinical inertia, and the belief that intensive treatment is not a priority in the context of the disease that causes admission.11 We therefore think that there is room for improvement in the management of patients with DM admitted to the hospital ward, and agree with Pérez et al. that the clinical inertia that prevents the administration of scheduled insulin regimens should be overcome from the time of patient admission,3 i.e. in the emergency room.

The patients in our hospital showed deficient metabolic control according to the recommended control criteria,8 but this should be put into context, bearing in mind the advanced age and high comorbidity of the patients. In a recent study13 of 620 non-ICU hospitalized patients, glycemic control (MBG: 182mg/dL; SD: 57.9mg/dL; CV: 31.9%) was virtually identical to that seen in our patients, which confirms the difficulties involved in DM management in hospitalized patients.

Although intensive blood glucose treatment has not been shown to decrease mortality in hospitalized patients,14 there are epidemiological studies showing that various glycemic control parameters may impact on patient prognosis. The most consistent of these parameters is MBG: in the Kosiborod et al. study15 patients admitted to hospital with acute myocardial infarction and preexistent DM had an increased inpatient mortality risk (OR=4.1; 95% CI: 1.8–9.3) when MBG was >200mg/dL. Therefore, and because of its simple calculation, MBG is the control parameter that could become the reference for assessing the quality of the hospital care of patients with DM, facilitating best-practices benchmarking16 among hospitals. However, the importance of inpatient glycemic variability both for the mid-term (one year) development of cardiovascular events17 and 90-day mortality has recently been recognized13. We therefore decided to assess the factors associated with measures not only of MBG, but also of variability.

The variables selected in the three models were those related to prior outpatient management (HbA1c as a marker of long-term glycemic control and type of treatment as a marker of time since onset and, probably, of the insulin reserve of patients), to the degree of decompensation shortly after admission (plasma blood glucose at baseline and in the first 24h after admission) and the need for corticosteroid treatment. Such variables should be taken into account in the design of the treatment algorithm included in hospital management protocols. It is generally agreed that baseline blood glucose, diabetes type and/or prior treatment, and the possibility of oral intake condition the initial insulin therapy of patients.3 Special mention should be made of corticosteroid treatment, because its marked impact on glycemic control requires specific protocols adapted to this situation.10,18 The high power of mean blood glucose in the first 24h for predicting MBG throughout the hospital stay emphasizes the importance of an adequate treatment approach from the time of hospital admission.

It should also be borne in mind that the dissemination of any protocol should be combined with an education program. In one survey,19 69% of surgeons admitted the need to improve their training in the management of hyperglycemia. The Institute for Safe Medication Practices20 has recognized that insulin is a drug associated with risks in the hospital, and in order to improve patient safety, specifically recommends the implementation of protocols which lay down the rules for insulin use as well as courses for all hospital professionals using insulin.

Factors supporting the validity of our study include the selection of a large patient sample representative of the patients seen in a medical ward, a careful review of the clinical histories together with the systematic collection of all capillary blood glucose values during the hospital stay, and the evaluation of the factors that determine both average inpatient glycemic control and its variability. Our study also has limitations, such as the impossibility of collecting the clinical history data of 10% of the patients, the non-inclusion of patients with type 1 diabetes (which was probably due to the exclusion of discharges from the endocrinology department, although an error in the codification of the medical histories at hospital discharge cannot be excluded), and the non-inclusion of surgical patients.

In conclusion, we have identified an opportunity for improving the quality of care of patients with DM admitted to the internal medicine departments of our hospital. Therapeutic inertia needs to be overcome by implementing management protocols that increase the use of scheduled insulin to improve metabolic control in patients, starting at the time of admission from the emergency room. Factors that should be included in the protocol for deciding initial treatment and its changes during the hospital stay include prior outpatient treatment, HbA1c, baseline plasma blood glucose, mean blood glucose in the first 24h after admission, whether or not oral intake is possible, and the use of corticosteroids.

Conflict of interest

The authors state that they have no conflicts of interest.

References
[1]
International Diabetes Federation.
IDF diabetes atlas.
5th ed., International Diabetes Federation, (2011),
http://www.idf.org/diabetesatlas
[2]
F. Soriguer, A. Goday, A. Bosch-Comas, E. Bordiú, A. Calle-Pascual, R. Carmena, et al.
Prevalence of diabetes mellitus and impaired glucose regulation in Spain: the Di@bet.es Study.
Diabetología, 55 (2012), pp. 88-93
[3]
A. Pérez-Pérez, P. Conthe-Gutiérrez, M. Aguilar-Diosdado, V. Bertomeu-Martínez, P. Galdos-Anuncibay, G. García de Casasola, et al.
Hospital management of hyperglycemia.
Med Clin (Barc), 132 (2009), pp. 465-475
[4]
F. Carral, G. Oliveira, M. Aguilar, J. Ortego, I. Gavilán, A. García, et al.
Hospital discharge records under-report the prevelence of diabetes in inpatients.
Diabetes Res Clin Pract, 59 (2003), pp. 145-151
[5]
G.E. Umpierrez, S.D. Isaacs, N. Bazargan, X. You, L.M. Thaler, A.E. Kitabchi.
Hyperglycemia an independent marker of in-hospital mortality in patients with undiagnosed diabetes.
J Clin Endocrinol Metab, 87 (2002), pp. 978-982
[6]
S. Clement, S.S. Braithwaite, M.F. Magee, A. Ahmann, E.P. Smith, R.G. Schafer, et al.
Management of diabetes and hyperglycemia in hospitals.
Diabetes Care, 27 (2004), pp. 553-591
[7]
M.H. Murad, J.A. Coburn, F. Coto-Yglesias, S. Dzyubak, A. Hazem, M.A. Lane, et al.
Glycemic control in non-critically ill hospitalized patients: a systematic review and meta-analysis.
J Clin Endocrinol Metab, 97 (2012), pp. 49-58
[8]
G.E. Umpierrez, R. Hellman, M.T. Korytkowski, M. Kosiborod, G.A. Maynard, V.M. Montori, et al.
Management of hyperglycemia in hospitalized patients in non-critical care setting: an Endocrine Society Clinical Practice Guideline.
J Clin Endocrinol Metab, 97 (2012), pp. 16-38
[9]
S.E. Inzucchi.
Management of hyperglycemia in the hospital setting.
N Engl J Med, 335 (2006), pp. 1903-1911
[10]
B. Draznin, J. Gilden, S.H. Golden, S. Inzucchi.
Pathways to quality inpatient management of hyperglycemia and diabetes: a call to action.
Diabetes Care, (2013), pp. 1807-1814
[11]
G. Maynard, D.H. Wesorick, C. O¿Malley, S. Inzucchi.
Subcutaneus insulin order sets and protocols: effective design and implementation strategies.
J Hosp Med, 3 (2008), pp. s29-s41
[12]
D.J. Wexler, J.B. Meigs, E. Cagliero, D.M. Nathan, R.W. Grant.
Prevalence of hyper and hypoglycemia among inpatients with diabetes.
Diabetes Care, 30 (2007), pp. 367-369
[13]
C.E. Mendez, K.T. Mok, A. Ata, R.J. Tanenberg, J. Calles-Escandon, G.E. Umpierrez.
Increased glycemic variability is independently associated with length of stay and mortality in noncritically ill hospitalized patients.
Diabetes Care, 36 (2013), pp. 4091-4097
[14]
D. Kansagara, F. Rongwei, M. Freeman, F. Wolf, M. Helfand.
Intensive insulin therapy in hospitalized patients: a systematic review.
Ann Intern Med, 154 (2011), pp. 268-282
[15]
M. Kosiborod, S.E. Inzucchi, H.M. Krumholz, L. Xiao, P.G. Jones, S. Fiske, et al.
Glucometrics in patients hospitalized with acute myocardial infarction.
Circulation, 117 (2008), pp. 1018-1027
[16]
C.B. Cook, K.E. Wellik, G.L. Kongable, J. Shu.
Assessing inpatient glycemic control: What are the next steps.
J Diabetes Sci Technol, (2012), pp. 421-427
[17]
G. Su, S.H. Mi, H. Tao, Z. Li, H.X. Yang, H. Zheng, et al.
Impact of admission glycemic variability, glucose and glycosylated hemoglobin on major adverse cardiac events after acute myocardial infarction.
Diabetes Care, 36 (2013), pp. 1026-1032
[18]
E. Saigí, A. Perez.
Management of glucocorticoid induced hyperglycemia.
Rev Clin Esp, 210 (2010), pp. 397-403
[19]
T.W. Costantini, J.A. Acosta, D.B. Hoyt, S. Ramamoorthy.
Surgical resident and attending physician attitudes toward glucose control in the surgical patient.
Am Surg, 74 (2008), pp. 993-996
[20]
D.J. Cobaugh, G. Maynard, L. Cooper, P.C. Kienle, R. Vigersky, D. Childers, et al.
Enhancing insulin-use safety in hospitals: practical recommendations from an ASHP Foundation expert consensus panel.
Am J Health-Syst Pharm, 70 (2013), pp. 1404-1413

Please cite this article as: Sáenz-Abad D, Gimeno-Orna JA, Sierra-Bergua B, Pérez-Calvo JI. Factores predictores del control glucémico promedio y de su variabilidad en pacientes diabéticos ingresados en el hospital. Endocrinol Nutr. 2015;62:257–263.

Copyright © 2014. SEEN
Article options
Tools
es en pt

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?

Você é um profissional de saúde habilitado a prescrever ou dispensar medicamentos

es en pt
Política de cookies Cookies policy Política de cookies
Utilizamos cookies propias y de terceros para mejorar nuestros servicios y mostrarle publicidad relacionada con sus preferencias mediante el análisis de sus hábitos de navegación. Si continua navegando, consideramos que acepta su uso. Puede cambiar la configuración u obtener más información aquí. To improve our services and products, we use "cookies" (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here. Utilizamos cookies próprios e de terceiros para melhorar nossos serviços e mostrar publicidade relacionada às suas preferências, analisando seus hábitos de navegação. Se continuar a navegar, consideramos que aceita o seu uso. Você pode alterar a configuração ou obter mais informações aqui.