- Original Research
- Open Access
Point-of-care capillary HbA1c measurement in the emergency department: a useful tool to detect unrecognized and uncontrolled diabetes
International Journal of Emergency Medicine volume 9, Article number: 7 (2016)
Inpatient hyperglycaemia and diabetes mellitus (DM) are common and are associated with an increased risk of complications and mortality. The severity of hyperglycaemia determines the rate of complications in patients treated in the emergency department (ED). Our aim was to examine whether determination of the capillary haemoglobin A1c (HbA1c) is a reliable method for detecting unknown diabetes and poor glycaemic control in the ED.
A prospective observational study was conducted in adult (>18 years) patients treated in a single-centre ED. We compared the results of HbA1c levels measured by Bio-Rad in2it point-of-care device on a capillary blood sample and by the hospital laboratory.
A total of 187 ED patients with an average age of 57.1 ± 19.2 years were studied. The mean HbA1c value was 5.78 ± 1.26 % by capillary POC testing and 6.10 ± 1.12 % by the hospital laboratory (correlation = 0.712, P < 0.001). A total of 17.1 % of cases had a prior diagnosis of DM. The diagnosis of DM (plasma glucose > 126 mg/dL and/or HbA1c > 6.5 %) was made in ten (5.4 %) additional cases (prior undiagnosed DM) for a total prior DM prevalence of 22.5 % (95 % CI 16.4–28.5 %). Capillary HbA1c detected 11 additional cases of unknown DM (5.9 %). A capillary HbA1c value greater than 6 % has a sensitivity of 85.7 % and specificity of 85.3 % for the screening of DM.
Determination of the capillary HbA1c in the ED is a reliable, fast, and simple system for the screening of unknown or uncontrolled DM.
Hyperglycaemia is a risk marker of morbidity and mortality in the emergency department (ED) and subsequent hospital admission, both in people with and without a history of diabetes mellitus (DM) [1–4]. Detection and management of hyperglycaemia in the ED, however, remains insufficient . Hospital and ED admissions could become windows of opportunity for an early diagnosis of people with diabetes and for improving glycaemic control. Providing information on previous glycaemic control of patients treated in the ED can be useful for stratifying a risk of complications and in tailoring antidiabetic treatment.
Determination of the glycated haemoglobin A1c (HbA1c) value is the best reference measure of glycaemic control in individuals with diabetes. It reflects the glycaemic environment in the past 2 to 3 months. International guidelines recommend measuring the HbA1c value every 3 months in patients that are off target and/or after a therapeutic change, and two annual measurements should be conducted in all patients . In many cases, HbA1c monitoring falls short of the recommended frequency. The HbA1c value is a major indicator to start an intensified medical treatment in patients with DM. In addition, international guidelines have included HbA1c measurement among the methods for diagnosis  (Table 1). Silverman et al. proposed HbA1c cut-offs for screening of diabetes in the ED . Using 6 % as the cut-off, the sensitivity of HbA1c was 76.9 % and the specificity was 87.3 %. Even lower HbA1c levels (5.7 %) have been proposed as a useful tool for the screening of prediabetes.
In recent years, several methods have become available to quickly and easily measure HbA1c from the capillary blood. There are several available easy-to-use point-of-care systems that can be used in the ED. While they are no substitutes for certified HbA1c measured by the laboratory to confirm the diagnosis of DM, their accuracy is highly satisfactory for estimating the degree of control of DM  and detecting undiagnosed diabetes  or at-risk subjects . In this study, we propose to test the usefulness of a point-of-care HbA1c system in the ED for the detection of undiagnosed and poorly controlled DM.
This prospective observational study was conducted at Nuestra Señora de Sonsoles Hospital of Ávila, Spain, a second-level hospital with an ED that covers the needs of a population of nearly 200,000 inhabitants, with approximately 37,000 adult ED visits annually (650 visits a week). We obtained an institutional review board approval prior to initiation of the study.
For seven consecutive days during 2012, informed consent was requested from all adult individuals (over 18 years) who were treated at the ED of Nuestra Señora de Sonsoles Hospital of Ávila. They were asked via a questionnaire whether they had received a prior diagnosis of diabetes. Those that had used corticosteroids in the past 2 months were excluded. The remaining patient’s clinical data was obtained from medical history, including detailed verification of the existence of a prior diagnosis of diabetes or the result of a diagnostic biochemical tests in accordance with the American Diabetes Association (ADA) criteria  (Table 1).
A capillary blood sample was taken from all patients in the finger pad. An Accu-Chek Aviva™ (Roche Diagnostic, Indianapolis, IN, USA) glucometer was used for in situ measurement of capillary blood glucose. The capillary HbA1c value was determined in situ by an in2it™ system (Bio-Rad, Hercules, CA, USA). Additionally, a venous blood sample was obtained by venepuncture and was sent to the hospital’s central laboratory for measurement of plasma glucose and HbA1c levels.
SPSS 11.0 software was used for the statistical analysis of the data. Correlation coefficients were estimated between and among values obtained in the ED and the laboratory by both the Pearson and Spearman methods (given the asymmetries in the variables). The linear fit was assessed by the R 2 value. P values were estimated using a one-tailed test. Diabetes diagnoses were determined in the laboratory and ED, and the sample was stratified by age (>/<40 years). The diagnostic efficacy of the HbA1c value and glucose level in the blood was assessed for different cut-offs using ROC curves, and the sensitivity, specificity, and positive and negative predictive values were estimated. The degree of agreement was established with the kappa coefficient. The relationship was assessed with a chi-square test.
A total of 187 patients that fulfilled inclusion criteria agreed to participate and, with all data available, were examined. Of them, 101 were men, with a median age of 64, interquartile range (IQR) of 66 years, and mean age of 57.1 ± 19.2 years (95 % CI 54.3–59.9 years) with a range of 18–84 years. Their demographic and clinical data are described in Table 2.
Glucose and HbA1c
The capillary blood glucose level (mean ± SD) was 114.8 ± 44.9 (95 % CI 108.3–121.3) mg/dL. The laboratory blood glucose level (mean ± SD) was 117.3 ± 42.1 (95 % CI 110.9–123.7) mg/dL. The capillary HbA1c value was 5.78 ± 1.26 % (95 % CI 5.60–5.97 %), with a range of 4.00–12.30 %, and the laboratory level was 6.10 ± 1.12 % (95 % CI 5.94–6.26 %), with a range of 4.50–11.50 %. The correlations between the measurements in the ED and laboratory were very high in both cases (glucose: R 2 = 0.739, P < 0.0001; HbA1c: R 2 = 0.789, P < 0.001) and are described in Fig. 1.
Diagnosis of diabetes
Thirty-two subjects (17.1 %) stated a prior known diagnosis of DM. Using the 2015 ADA criteria for the diagnosis of DM (plasma glucose > 126 mg/dL and/or HbA1c > 6.5 %) and previous clinical history data, ten more cases were revealed (prior undiagnosed DM) (5.4 %). The total number of patients who met any criteria for prior DM was 42, for a 22.5 % prevalence rate (95 % CI 16.4–28.5 %). The capillary HbA1c values were useful in detecting 11 additional cases of unknown DM (5.9 %).
Degree of control of prior DM
In patients with prior DM (n = 42), the HbA1c values were >7 % (61.9 %), >8 % (28.6 %), and >9 % (16.7 %) by the hospital laboratory and >7 % (47.6 %), >8 % (19.0 %), and >9 % (11.9 %) by capillary testing.
Diagnostic efficacy of HbA1c and glucose levels
The ROC curve of the capillary HbA1c value measured in the ED had an area under the curve (AUC) of 0.897 (95 % CI 0.835–0.960, P < 0.0001) (Fig. 2). Using 6 % as the cut-off and cross-referencing with the DM diagnosis, we obtained a sensitivity (S) of 85.7 %, positive predictive value (PPV) of 63.2 %, specificity of 85.3 % (122 of 143), and negative predictive value (NPV) of 95.3 % (122 of 128) (χ 2 = 76.83; 1 gl; P < 0.000; kappa = 0.631). According to coordinates of the curve, the 6 % cut-off seemed to best balance the sensitivity and specificity, with both greater than 85 % (Table 3). A previous study in acutely ill patients with random hyperglycaemia at hospital admission showed that a 6.0 % laboratory HbA1c level was 100 % specific (14/14) and 57 % sensitive for diabetes diagnosis .
An early diagnosis of DM remains an unsolved challenge. Scientific data repeatedly confirm the need for an early treatment of DM to avoid the development of complications. However, the Di@bet.es study has shown that the prevalence of DM in the adult Spanish population is 13.8 %, with a distribution of 7.8 % of known DM and 6 % of unknown DM . This rate of undiagnosed DM is clearly higher than that described in other countries  and highlights the importance of screening and early detection . Current guidelines recommend diabetes screening in adults over 45 years of age and/or those who have a body mass index greater than 25 kg/m2 and with any additional known risk factor for DM . The recommended tests for screening are the same as those used for diagnosis and include fasting plasma glucose (FPG), HbA1c level, and/or an oral glucose tolerance test (OGTT). The latter test is inconvenient, expensive, and complex, for which it is seldom used in clinical practice. Determination of blood glucose is limited by the influence of multiple factors (prior intake, drugs, stress, etc.). HbA1c measurement, although more expensive than blood glucose determination, provides information on the glycaemic environment during the previous 2 to 3 months.
Several epidemiologic studies and meta-analyses have reported that diabetes screening in the population over 40 years of age and in high-risk individuals (those with a family history of diabetes and/or hypertension) is cost-effective . Determination of HbA1c is considered the best test, with an intermediate cost between measuring the FPG and the OGTT . However, screening access by HbA1c testing is limited, especially in socioeconomically deprived areas, unless it is performed at primary care consultations . Taking advantage of situations in which people come to the hospital can be a “window of opportunity” for the detection of DM. The population of adults treated in the ED, as described in our sample, would mostly belong to this high-risk group of undiagnosed DM patients . In our sample, 22.5 % of adult patients who were treated at the ED met the DM criteria and another 5.9 % met the criteria for newly diagnosed DM. This finding is consistent with previous studies [7, 18, 19]. HbA1c levels are also an indicator of future DM and may be useful for defining personalized prevention strategies .
Hyperglycaemia is a risk marker for hospital mortality . However, stressors and lack of fasting may influence the interpretation of a single measurement of blood glucose in the laboratory and/or ED. In addition, optimal thresholds for DM screening in ED have not been determined. Capillary HbA1c levels in the ED can be used to detect patients with poor glycaemic control who are at risk of acute and chronic complications [21–23]. Among the patients with DM who were included in our study, 47.6 % had an HbA1c value above 7 % (the accepted target of good control) and 19.0 % had clearly insufficient DM control (over 8 %). In the entire sample, 10.8 % of the patients showed levels above 7 % and 4.3 % had levels higher than 8 %.
Our study indicates that measurement of HbA1c by capillary point-of-care testing could provide considerable advantages in the detection of DM, including immediacy of the result, and the possibility of using the information to guide medical treatment. Clinical practice guidelines recognize that there are no randomized trials demonstrating improved outcomes using HbA1c levels to assist in the diagnosis of diabetes or to guide the glycaemic management of inpatients with known diabetes. However, they unanimously agreed on the practical utility of this strategy based in the consensus opinion . The issue of a possible cost reduction can be addressed by future studies. These advantages are especially applicable to care in an ED. HbA1c information could modify the treatment indicated for the symptoms that brought the patient to the ED. Additionally, in patients with both known DM and newly diagnosed DM, capillary HbA1c information can be useful in determining the most appropriate antidiabetic treatment (i.e., use of oral agents vs. insulin).
Major limitations of our study are the inclusion of a single medical institution and the possibility of a selection bias. To limit this constraint, we recruited all adult patients with no limitations other than their consent or use of hyperglycaemic medications such as corticosteroids. In addition, haemoglobinopathies are a known cause of erroneous results in determining HbA1c levels; however, the prevalence of these diseases in our population was very low.
Determination of capillary HbA1c levels in the ED is a reliable, fast, and simple system for the detection of unknown and poorly controlled DM. Due to the need to improve the early detection of DM and the appropriate selection of antidiabetic therapy, our results suggest that capillary HbA1c represents an important diagnostic tool in the ED. Future studies should investigate if the use of capillary HbA1c measurement in the ED can improve clinical outcome, reduce complications, and guide treatment selection in patients with hyperglycaemia and diabetes.
fasting plasma glucose
oral glucose tolerance test
Zelihic E, Poneleit B, Siegmund T, Haller B, Sayk F, Dodt C. Hyperglycemia in emergency patients—prevalence and consequences: results of the GLUCEMERGE analysis. Eur J Emerg Med. 2015;22:181–7.
Pulsinelli WA, Levy DE, Sigsbee B, Scherer P, Plum F. Increased damage after ischemic stroke in patients with hyperglycemia with or without established diabetes mellitus. Am J Med. 1983;74:540–4.
Weir CJ, Murray GD, Dyker AG, Lees KR. Is hyperglycaemia an independent predictor of poor outcome after acute stroke? Results of a long term follow up study. BMJ. 1997;7090:1303–6.
Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. The Journal of Clinical Endocrinology and Metabolism. 2002;87:978–82.
Vinagre I, Mata-Cases M, Hermosilla E, Morros R, Fina F, Rosell M, et al. Control of glycemia and cardiovascular risk factors in patients with type 2 diabetes in primary care in Catalonia (Spain). Diabetes Care. 2012;35:774–9.
American Diabetes Association. (2) Classification and diagnosis of diabetes. Diabetes Care. 2015;38(Suppl):S8–16.
Silverman RA, Thakker U, Ellman T, Wong I, Smith K, Ito K, et al. Hemoglobin A1c as a screen for previously undiagnosed prediabetes and diabetes in an acute-care setting. Diabetes Care. 2011;34:1908–12.
Petersen JR, Omoruyi FO, Mohammad AA, Shea TJ, Okorodudu AO, Ju H. Hemoglobin A1c: assessment of three POC analyzers relative to a central laboratory method. Clin Chim Acta. 2010;411:2062–6.
Marley JV, Oh MS, Hadgraft N, Singleton S, Isaacs K, Atkinson D. Cross-sectional comparison of point-of-care with laboratory HbA1c in detecting diabetes in real-world remote aboriginal settings. BMJ Open. 2015;5:e006277.
Rowan CP, Miadovnik LA, Riddell MC, Rotondi MA, Gledhill N, Jamnik VK. Identifying persons at risk for developing type 2 diabetes in a concentrated population of high risk ethnicities in Canada using a risk assessment questionnaire and point-of-care capillary blood HbA1c measurement. BMC Public Health. 2014;14:929.
Greci LS, Kailasam M, Malkani S, Katz DL, Hulinsky I, Ahmadi R, et al. Utility of HbA(1c) levels for diabetes case finding in hospitalized patients with hyperglycemia. Diabetes Care. 2003;26(4):1064–8.
Soriguer F, Goday A, Bosch-Comas A, Bordiú E, Calle-Pascual A, Carmena R, et al. Prevalence of diabetes mellitus and impaired glucose regulation in Spain: the Di@bet.es study. Diabetologia. 2012;55:88–93.
Selvin E, Parrinello CM, Sacks DB, Coresh J. Trends in prevalence and control of diabetes in the United States, 1988-1994 and 1999-2010. Ann Intern Med. 2014;160:517–25.
Chatterton H, Younger T, Fischer A, Khunti K, Programme Development Group. Risk identification and interventions to prevent type 2 diabetes in adults at high risk: summary of NICE guidance. BMJ. 2012;345:e4624.
Waugh NR, Shyangdan D, Taylor-Phillips S, Suri G, Hall B. Screening for type 2 diabetes: a short report for the National Screening Committee. Health Technol Assess. 2013;17:1–90.
Goyder E, Wild S, Fischbacher C, Carlisle J, Peters J. Evaluating the impact of a national pilot screening programme for type 2 diabetes in deprived areas of England. Fam Pract. 2008;25:370–5.
Ginde AA, Delaney KE, Lieberman RM, Vanderweil SG, Camargo Jr CA. Estimated risk for undiagnosed diabetes in the emergency department: a multicenter survey. Acad Emerg Med. 2007;14:492–5.
Ginde AA, Cagliero E, Nathan DM, Camargo Jr CA. Point-of-care glucose and hemoglobin A1c in emergency department patients without known diabetes: implications for opportunistic screening. Acad Emerg Med. 2008;15:1241–7.
Charfen MA, Ipp E, Kaji AH, Saleh T, Qazi MF, Lewis RJ. Detection of undiagnosed diabetes and prediabetic states in high-risk emergency department patients. Acad Emerg Med. 2009;16:394–402.
Cheng P, Neugaard B, Foulis P, Conlin PR. Hemoglobin A1c as a predictor of incident diabetes. Diabetes Care. 2011;34:610–5.
Inzucchi SE. Clinical practice. Diagnosis of diabetes. N Engl J Med. 2012;367:542–50.
United Kingdom Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837–53.
The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977–86.
Umpierrez GE, Hellman R, Korytkowski MT, Kosiborod M, Maynard GA, Montori VM, et al. Management of hyperglycemia in hospitalized patients in non-critical care setting: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2012;97(1):16–38.
Ferrer Internacional ceded the point-of-care capillary HbA1c system for this study. José-Manuel García de Cecilia (3datos tratamiento estadístico) performed the statistical analyses and was supported by the Hospital General de Segovia Research Commission. No other external funding was received for this study, which was conducted thanks to the unconditional effort of all participating investigators.
F. Gomez-Peralta is a member of the Advisory Panel of AstraZeneca Pharmaceuticals LP, Sanofi, and Novo Nordisk, Inc., and of the Speaker’s Bureau of Sanofi, Novo Nordisk, Inc., Boehringer Ingelheim Pharmaceuticals, Inc., AstraZeneca Pharmaceuticals LP, Bristol-Myers Squibb Company, and Lilly. C. Abreu is a member of the Speaker’s Bureau of Sanofi, Novo Nordisk, Inc., Boehringer Ingelheim Pharmaceuticals, Inc., AstraZeneca Pharmaceuticals LP, and Bristol-Myers Squibb Company. Leonor Andreu-Urioste, Ana Cristina Antolí, Carmen Rico-Fontsaré, David Martín-Fernández, Rosa Resina-Rufes, Juan Jesús Pérez-García, Ángela Negrete-Muñoz, and Daniel Muñoz-Álvarez declare that they have no competing interests. G.E. Umpierrez is a meber of the Advisory Panel of Merck, Novo Nordisk, Inc., Regeneron, and Boehringer Ingelheim Pharmaceuticals and received research support to Emory University from Merck, Boehringer Ingelheim Pharmaceuticals, Inc., Novo Nordisk, Inc., and AstraZeneca Pharmaceuticals LP.
All authors approved the final article; their individual contribution to the article was as follows: FGP: conception and design of the study; acquisition, analysis, and interpretation of the data; drafting the article and revising it critically, and approval of the final article; CA: analysis and interpretation of the data, revising the article critically, and approval of the final article; LAU: acquisition of the data, revising the article critically, and approval of the final article; CRF: acquisition of the data, revising the article critically, and approval of the final article; DMF: acquisition of the data, revising the article critically, and approval of the final article; RRR: acquisition of the data, revising the article critically, and approval of the final article; JJPG: acquisition of the data, revising the article critically, and approval of the final article; ANM: acquisition of the data, revising the article critically, and approval of the final article; DMA: revising the article critically and approval of the final article; and GU: revising the article critically and approval of the final article.
An institutional review board approval prior to initiation of the study was obtained. Study accepted by the Clinical Research and Ethics Commission of the Hospital Nuestra Señora de Sonsoles, Ávila, Spain.
About this article
Cite this article
Gomez-Peralta, F., Abreu, C., Andreu-Urioste, L. et al. Point-of-care capillary HbA1c measurement in the emergency department: a useful tool to detect unrecognized and uncontrolled diabetes. Int J Emerg Med 9, 7 (2016) doi:10.1186/s12245-016-0107-6
- Point-of-care HbA1c
- Emergency medicine
- Public health